Diagnosis and Treatment2
Diagnosis, Treatment, and Vaccine Candidates
“The physician must be able to tell the antecedents, know the present, and foretell the future―must mediate these things, and have two special objects in view with regard to disease, namely, to do good or to do no harm.”
Diagnostic Devices and Related Techniques
The use of effective diagnostic devices for the detection of SARS-CoV-2 in biofluids is essential for the rapid identification of COVID-19, which serves both to protect the newly diagnosed individual, who can be isolated and begin receiving prompt treatment for as long as the disease is manifested, and the susceptible general public, who can be informed of accurate case counts to take necessary or reasonable precautions to avoid unnecessary spread of disease. Furthermore, thorough contact tracing protocols may be carried out to further isolate potentially contagious carriers unaware of their exposure to the disease. For this reason, the regular and widespread use of reliable methods of SARS-CoV-2 detection through rapid laboratory testing, such as polymerase chain reaction or isothermal amplification, are paramount.
It is well documented that a large portion of COVID-19 carriers are asymptomatic and some may have never become aware of their infection, with some estimates as high as 40-80% of the infected in certain demographics. Others are in pre-symptomatic stages of the disease and may be contagious. Both groups pose a substantial public health risk. Therefore, the prevalence of symptoms should not be the only driver of testing. Even in locations with low documented caseloads, regular population screening for the prevalence of SARS-CoV-2 should be conducted on random subject pools to continually monitor for the prevalence of the virus.
No laboratory testing method is perfect, and the sensitivity and specificity of many diagnostic methods can leave much to be desired. Sensitivity, the percentage of actual positive cases that are correctly identified as positive for a given disease, is particularly low in polymerase chain reaction methods and can be substantially decreased in more rapid forms of such testing. Meanwhile, optimizing the specificity, the percentage of actual disease negative individuals that are correctly identified as negative, is also an important concern. Due to the slow speed and low sensitivity of some of the viral detection methods, other methods of diagnosis have been recommended to be used in conjunction with viral detection, such as the use of chest CT-scans. However, these methods are more likely to be used for symptomatic individuals, who are more likely to seek testing.
As previously noted, a substantial proportion of COVID-19 carriers are asymptomatic and may have recovered, having no prior knowledge of their infection. Due to low availability of testing during various stages of the pandemic, many symptomatic individuals who did contract COVID-19 did not meet necessary criteria for testing and were therefore not properly documented or reported. Serological assays, which detect the presence of SARS-CoV-2 directed antibodies in blood sera present an effective means of tracking COVID-19 cases in such individuals. For this reason, these methods are essential in tracking and documenting the actual prevalence of recent COVID-19 cases.
Polymerase Chain Reaction
Most current nucleic acid tests for SARS-CoV-2 utilize nasal or throat swabbing to collect samples to test for the presence of the SARS-CoV-2-RNA genome. Collected samples are first transported to a lab and must be processed to extract their RNA content. The presence of SARS-CoV-2 RNA is tested using a reverse transcriptase polymerase chain reaction (RT-PCR), a technique that is used to amplify a specific RNA sequence that may be present in the sample. It does so by rapidly copying target regions of the virus’s RNA that are present in known strains of the virus, such as the nucleotides that specifically code for proteins in the virus’s nucleocapsid. If the target region is present in the sample (which means the sample contains the virus), then the target sequence can be amplified and can be visualized using gel electrophoresis, for example, thereby confirming the presence of the virus and infection in the tested individual. Different versions of the RT-PCR tests simply amplify different target regions present in the viral genome, but they generally do so through a repeated process of thermal cycling, which often takes hours to complete.
There are several disadvantages of currently available PCR diagnostic methods. While PCR is effective for testing for the presence of SARS-CoV-2, the sensitivity of these tests can be quite low, as low as 59% found in a study of 1014 patients with COVID-19 (Ai et al., 2020). A more optimistic study measured the sensitivity of the test at 71% in a group of 51 patients (Fang et al., 2020). Regardless, the reported range of sensitivity for such methods (approximately 60-70%) is not optimal, leading to a substantial number of possible false negatives (~30-40% of people who have the virus that test negative on the RT-PCR test). The primary reason for low test sensitivity comes from the method of sample collection; if the swab does not pick up the virus because it comes in contact with too limited a sample, the virus cannot be amplified using PCR. Furthermore, sample collection is currently carried out by medical professionals using PPE that often needs replacement many times throughout the day. In an effort to preserve scarce PPE supplies that are needed to protect medical professionals treating patients, the U.S. government is working with several organizations to develop tests where samples can be taken at home and sealed for safe delivery to labs.
On August 4, 2020, the FDA granted emergency use authorization for the use of a saliva-based test for COVID-19 that had been developed at the Yale School of Public Health with funding from the National Basketball Association and the National Basketball Players Association. A description of the test, its advantages compared with other tests, and initial results were provided in non-peer-reviewed fashion in medRxiv on August 4, 2020 (Vogels et al., 2020). This test, termed SalivaDirect, when compared with earlier tests using nasopharyngeal (NP) swabs, has multiple potential advantages: 1. Use of saliva rather than NP swabs simplifies collection, allows patients to collect their own specimens, and does not require PPE use by medical professionals during collection; 2. Handling of saliva without the need for preservatives for up to 7 days; 3. Avoidance of nucleic acid extraction, which in prior tests was often time consuming, expensive, and dependent on adequate supplies of specific reagents with the use of proteinase K and heat; 4. Use of dualplex RT-qPCT targeting the N1 region of the SARS-CoV-2 nucleocapsid with a human RNase P control. This allowed for reduction in the number of RT-qPCR tests to one per sample.
Tests were conducted with reagents obtained from multiple vendors for both proteinase K and RT-qPCR kits as well as RT-qPCR instruments were investigated, and the majority of combinations gave similar results with some variation in the lower limit of detection. Comparison of results of Ct values from the SalivaDirect test with the modified CDC assay (using nucleic acid extraction and singleplex RT-qPCR) showed slightly weaker detection with a false negative rate of 7.3%. False positives were not observed in 30 tested samples.
The advantages of this test include the simplicity of sample collection, speed, lack of dependence on a single vendor for reagents, and cost of less than $5.00 per test. Large trials are presently being conducted to compare the results of SalivaDirect with conventional nasal swab, nucleic acid extraction, and approved RT-qPCR assays. Tests are also ongoing to investigate the use of SalivaDirect as a tool for pooled assays, in which multiple saliva specimens from asymptomatic individuals are pooled and tested together, with individual testing performed only if the pooled sample tests positive. This technique, if prospectively validated, has the potential to greatly increase testing, with reduced costs and rapid reporting of results, which should be of great benefit in considerations of activities such as opening schools and businesses, not to mention attending basketball games (see Pooled Sampling).
Whereas RT-PCR is a slow process accomplished through a repetitive thermal cycling procedure (cycles of heating and cooling to denature and re-anneal new nucleotides to the template nucleic acid strands), isothermal amplification methods, such as Loop-Mediated Isothermal Amplification (LAMP) can amplify RNA segments specific to the virus at a constant temperature, usually around 60-65°C. Instead of using heat cycles to separate RNA strands, RT-LAMP uses a powerful polymerase to both add nucleotides and separate the strands of nucleic acid. In addition to a polymerase, RT-LAMP relies on specific primers as reagents to amplify different regions of a particular gene and to speed up the amplification process. As amplification progresses, a by-product of the process, pyrophosphate, is produced, which can bind to magnesium ions in solution to form a white precipitate that can be seen by the naked eye. Other methods, such as directly measuring the turbidity of the resulting solution, can also be used to confirm and quantify the amplified RNA. Because the process can all be done at one temperature, RT-LAMP provides a cheaper, faster, and more portable method of viral detection when compared to RT-PCR.
The speed of RT-LAMP methods can be further enhanced when used in conjunction with CRISPR-Cas12. Cas12 is an endonuclease that cleaves specifically targeted nucleic acid sequences. Broughton et al. (2020) discuss a novel detection method using such a combination that the team dubbed DETECTR (DNA Endonuclease-Targeted CRISPR Trans Reporter), which uses reverse transcription and RT-LAMP methods to amplify RNA regions of the virus. Cas12 is then used to identify specific sequences unique to the SARS-CoV-2 genome, which it cleaves at a site with a FAM-Biotin reporter molecule that enables visual detection of the virus. The authors reported ability to detect the virus with a strong visual signal in less than five minutes.
On March 28, 2020, the FDA approved under emergency use authorization a point-of-care rapid testing device that tests for the presence of SARS-CoV-2 in naso-pharyngeal or oro-pharyngeal swabs. The portable test known as IDNOW, which was produced by Abbott Laboratories, is claimed to deliver a positive result within 5-13 minutes and a negative result within 13 minutes. It is based on a platform that is already used for detecting other viruses such as influenza, and it uses isothermal amplification of the nucleotides that encode the RdRP, with a limit of detection of 125 copies/mL. Basu et al. (2020) tested the efficacy of IDNOW in comparison to the Cepheid Xpert Xpress SARS-CoV-2 RT-PCR test, which amplifies nucleotides that encode the N2 and E proteins. The latter test has a runtime of 45 minutes and a limit of detection of 250 copies/mL. A total of 202 samples were taken using nasopharyngeal swabs from a single patient, 101 of which were transported in viral transport medium and 101 on dry swabs. Abbott IDNOW had a significantly higher proportion of false negatives. It missed a third of the samples detected positive by Cepheid Xpert Xpress when using swabs in viral transport medium, and it missed over 48% of samples detected positive by Cepheid Xpert Xpress when using dry swabs. Overall, the Basu et al. indicate that the test has relatively low sensitivity.
Ai et al. (2020) report that noncontrast chest CT scans may be a more sensitive approach for COVID-19 diagnosis. Their study included 1,014 patients in Wuhan, China who took paired RT-PCR tests and underwent chest CTs (always one day apart at most). The findings highlight the increased sensitivity of the approach, reported as 97%, which used observations of ground-glass opacity, consolidation, thickening of the interlobular septa, and other features to determine diagnosis. Compared to RT-PCR, which was reported to have a sensitivity of 59% in this study, CT scanning presented a more reliable and faster approach as a diagnostic procedure that is relatively non-invasive. Radiologists analyzing scans were given data on the patient’s symptoms, which was used to make a final assessment in diagnosis. Unfortunately, the approach had relatively low specificity (25%), and the overall accuracy was approximately 65%.
Raptis et al. (2020) reviewed this and two other studies that had reported that CT scans might be sufficiently sensitive and specific to use for COVID-19 screening and/or diagnosis. They concluded that the existing reports were based on selected retrospective data, provided little data on patient characteristics and criteria for evaluating scans, and in some cases scan specificity showed marked variability among multiple observers. Clear criteria for training observers was not provided. They concluded that more data on well-defined patient populations, scanned and interpreted in uniform fashion, are needed before CT scans can be considered to have a role in COVID-19 screening or diagnosis and that at present their role is properly restricted to evaluation of complications of COVID-19 pneumonia or assessment of possible alternative diagnoses. These are in line with current recommendations from the CDC, the WHO, the American College of Radiology, and the Fleischner Society for Thoracic Imaging (Chou et al., 2020; Rubin et al., 2020).
Lin et al. (2020) reported a comparison of CT imaging findings of a cohort of patients with pneumonia due to COVID-19 (n = 52) and compared these with a group of patients with pneumonia due to influenza (n = 45). Scans were performed using three different scanners with differences in scan thickness and reconstruction thickness. All COVID-19 patients had positive detection of nucleic acid testing. It is important to note, however, that both the COVID-19 and influenza groups excluded patients with “hypertension, diabetes, tumor, chronic obstructive pulmonary disease, bronchiectasis, lung cancer, and other lung diseases,” which may have limited the study by excluding a large number of the sickest patients. Interpretation of images were performed by two chest radiologists who, if disagreeing on interpretation, reached a final decision by consensus. Inflammatory lesions were also analyzed using AI software (FACT version 188.8.131.52, Dexin Medical Imaging Technology), which could evaluate the volume and mean density of the lesions. There is no mention of blinding to diagnosis at the time of review, and presumably this was not done.
While statistically significant differences (p < 0.05) were seen in the proximity of the largest lesion to the pleura, presence of mucoid impaction, presence of pleural effusion, and axial distribution of lesions, there was substantial overlap in these categories. Other characteristics such as the properties of the largest lesion, presence of ground-glass opacities, presence of consolidation, and a number of other features showed no significant differences between the two groups. The authors concluded,
However, CT manifestations of COVID-19 pneumonia and influenza virus pneumonia have a large amount of overlap, and that even with the characteristics evaluated using AI software, no significant differences were detected. Distinguishing between these two types of viral pneumonia with imaging alone is difficult. Therefore CT examination needs to be combined with clinical indicators for comprehensive evaluation, the more important role of CT in the pandemic is in finding lesions and evaluating the results of treatment.
None of these recommendations negate the important role that chest CT may have in management of patients with COVID-19 as determined by molecular testing, particularly with better attention paid to consistency of imaging parameters and the application of more sophisticated methods of image analysis including deep learning and radiomic analysis. They do argue strongly that the primary methods for population screening and clinical diagnostic analysis should remain molecular analysis for viral antigen and/or virus-specific antigen (Chou et al. 2020; Raptis et al.2020; Rubin et al. 2020).
Recent generation of a serological enzyme-linked immunosorbent assay (ELISA), using recombinant antigens from the spike protein of SARS-CoV-2 (Amanat et al., 2020), offers hope for a more sensitive testing approach than the PCR analyses. Unlike PCR methods, which test for the presence of the virus (which eventually clears after infection), serological assays test for the presence of SARS-CoV-2 antibodies that persist long after infection has passed. Even though serological assays have limitations with regard for the detection of acute cases, they can be used for determining seroprevalence in particular populations where previous exposures and identifying highly reactive human donors offer utility for vaccine and other immunotherapy development. As the authors point out, “Sensitive and specific identification of Coronavirus SARS-CoV-2 antibody titers will also support screening of healthcare workers to identify those who are already immune and can be deployed to care for infected patients minimizing the risk of viral spread to colleagues and other patients…” Clearly, breakthroughs in the development of ELISA test kits and other advances in precision omics, including and beyond proteomics, will advance our understanding of the science of COVID-19, as well as support in silico screening as well as new model development for the better design and implementation of both precision and personalized therapies and preventatives.
Gordon et al. (2020) have suggested the mining of the SARS-CoV-2 protein interactome to better predict disease state and to discover druggable proteins or already approved FDA drugs that can be used in precision treatments for COVID-19 and other coronavirus infections. This, in combination with ELISAs for spike proteins from SARS-CoV-2 (Amanat et al., 2020) and the use of machine learning for large scale screening of datasets from e.g. the White House and National Institutes of Health COVID-19 Open Research Dataset (CORD-19) initiative, offer hope for more personalized and precision treatments for a heterogeneous population of patients who possess unique omics and distinctly challenging, coexisting comorbidities.
A June 2020 study (Premkumar et al.) has also helped the cause of more reliable serologic testing by way of focusing on the receptor binding domain (RBD). The recombinant SARS-CoV-2 RBD antigen was found to be highly sensitive for antibodies induced by this and possibly other SARS coronaviruses.
The aforementioned studies have largely assumed that diagnostic testing will be performed on the sample taken from a single individual, in which case the implications of a positive, negative, or indeterminate test are obvious. Individual testing is, however, quite time and resource consuming and while appropriate for diagnosing modest numbers of individuals, it is impractical when testing larger numbers of people (e.g. the workforce of a factory, students in a college dormitory, etc.). This is especially true when testing must be repeated on a frequent basis. To circumvent these problems, testing of pooled samples has been proposed as a resource efficient alternative. In some cases, pooled testing has been implemented albeit without firm data as to its efficacy.
Both the FDA and CDC have issued statements providing interim guidance on pooled testing (Centers for Disease Control. 1 Aug 2020; US Food and Drug Administration. 24 August, 2020). They have considered its use in three settings:
- diagnostic testing of individuals thought likely to have been exposed to SARS-CoV-2 on the basis of symptoms of contact with other individuals known to be infected
- screening testing of individuals with no particular reason to expect exposure in order to detect infection in asymptomatic individuals and/or assist in contact tracing
- surveillance testing of populations to obtain aggregate data on prevalence, population effects of measures such as social distancing, masking, hand-washing, etc
Pooling is defined as “combining respiratory samples from several people and conducting one laboratory test on the combined pool of samples to detect SARS-CoV-2.” If the pooled test is negative, it is assumed that each of the specimens that contributed to the pool is also negative. If the pooled test is positive, the individuals who contributed to the pooled sample must then be tested individually to determine which one or ones are positive. Pooling should be performed only when the estimated probability of test positivity is low, and since most specimens will be negative, the sensitivity of pooled testing will be less than that of individual testing.
A number of universities including the University of Arizona, University of North Carolina, and University of Virginia have taken this one step further and are testing sewage from university dormitories, with planned testing of individual residents should a positive pooled test arise. The CDC has established the National Wastewater Surveillance System (NWSS) to provide a central mechanism for aggregation of such data.
Anticipating a second wave of SARS-CoV-2 infections, Fogarty et al. (2020) have presented a proposal for testing of groups of individuals who work together (e.g. hospital staff) which combines three key elements:
- use of saliva rather than nasopharyngeal swabs, which will facilitate frequent testing and minimize the need for PPE
- Pooled testing
- Obtaining two specimens of saliva at each encounter (Specimen A and Specimen B). The A specimens would be pooled and tested, with the B specimens being available for immediate individual testing should the pooled A test positive, rather than having to track down individuals to obtain another specimen. This latter approach is derived from sports drug testing where two specimens are routinely obtained but the second tested only if the first appears positive.
The researchers do note that this approach runs the risk of detecting late non-viable viral shedding and that the optimum number of specimens to pool and cycle for PCR will depend on the pool size and COVID-19 prevalence.
Current Clinical Management
The National Institutes of Health of the United States (NIH) has characterized five degrees of increasing severity of individuals infected with SARS-CoV-2:
- Asymptomatic or Presymptomatic Infection: Individuals who test positive for SARS-CoV-2 by virologic testing using a molecular diagnostic (e.g., polymerase chain reaction) or antigen test, but have no symptoms.
- Mild Illness: Individuals who have any of the various signs and symptoms of COVID 19 (e.g., fever, cough, sore throat, malaise, headache, muscle pain) without shortness of breath, dyspnea, or abnormal chest imaging.
- Moderate Illness: Individuals who have evidence of lower respiratory disease by clinical assessment or imaging and a saturation of oxygen (SpO2) ≥ 94% on room air at sea level.
- Severe Illness: Individuals who have respiratory frequency >3 0 breaths per minute, SpO2 < 94% on room air at sea level, ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO2/FiO2) < 300 mmHg, or lung infiltrates > 50%.
- Critical Illness: Individuals who have respiratory failure, septic shock, and/or multiple organ dysfunction.
|Self-isolation at home for 10 days from date of initial positive test and 3 days after becoming afebrile||None recommended||None recommended||Contact tracing if short turn-around of testing|
|Generally may be managed in an ambulatory setting using telemedicine or remote visits.||None recommended||Insufficient data to recommend for or against||Close monitoring indicated as some patients may worsen rapidly|
|Hospital infection prevention and control measures. If patient will be undergoing aerosol-generating procedures, placement in airborne infection isolation rooms (AIIR) is desirable. Hospital staff should use N95 respirators during such procedures.||CBC, Metabolic panel including liver and renal function. Pulmonary imaging by CXR, US, or CT.||Refer to sections and tables on Antiviral and Immune-Based Therapies for discussion of investigational agents.||Inflammatory markers including CRP, D-dimer, and ferritin may have prognostic value but are not part of standard care.|
|Hospital infection prevention and control measures. If patient will be undergoing aerosol-generating procedures, placement in airborne infection isolation rooms (AIIR) is desirable. Hospital staff should use N95 respirators during such procedures.||CBC, Metabolic panel including liver and renal function. Pulmonary imaging by CXR, US, or CT.||If secondary bacterial pneumonia or sepsis is suspected, administer empiric antibiotics. Refer to sections and tables on Antiviral and Immune-Based Therapies for discussion of investigational agents.||Inflammatory markers including CRP, D-dimer, and ferritin may have prognostic value but are not part of standard care.|
|These patients will be undergoing aerosol-generating procedures, placement in airborne infection isolation rooms (AIIR) is desirable. Hospital staff should use N95 respirators during such procedures.||As above plus additional tests for evaluation and management of cardiac, hepatic, renal, and CNS disease.||See recommendations from Surviving Sepsis Campaign1. Refer to sections and tables on Antiviral and Immune-Based Therapies for discussion of investigational agents.||Successful management requires attention not only to COVID-19 but also to comorbidities and nosocomial complications. Address goals of therapy with patient(s) (if possible) or family.|
1.Alhazzani, W., Moller, M. H., Arabi, Y. M., et al. Surviving Sepsis Campaign: guidelines on the management of critically ill adults with coronavirus disease (COVID-19). Intensive Care Med. 2020. https://www.ncbi.nlm.nih.gov/pubmed/32222812
The NIH COVID-19 Treatment Guidelines also include reviews of standard and investigational antiviral and immunotherapeutic agents and are updated regularly.
Managing patients during a pandemic can be complicated further by comorbidity, that is, with one or more pre-existing or underlying health conditions including, in particular, cancer. Treatment of cancer patients raises a number of issues with the appropriate management of patients with and without a known COVID-19 infection as well as the appropriate allocation and use of hospital resources and staff who are likely to be in limited supply or absent entirely and distressed at times of peak outbreak and onslaught of ER admissions. These issues may involve surgery, radiation therapy, and systemic therapy (chemotherapy, targeted therapy, immunotherapy), and may also apply to patients being treated with curative and palliative intent.
Management of Patients with Cancer
Liang et al. (2020) reviewed data on a prospective cohort of 2007 patients with COVID-19 from 31 regional hospitals in China. All patients were diagnosed with laboratory confirmation of infection and were admitted to the hospital. Because of inadequate documentation of previous medical history, 417 were excluded leaving 1590 for further analysis. Of these, 18 (1%) had a history of cancer, which the authors felt was higher than the incidence in the overall Chinese population (0.29%). Patients with cancer were older (mean age 63.1 years [SD 12.1] vs. 48.7 [SD 16.2]) and more likely to have a history of smoking (4 of 18 (22%) vs. 107 of 1590 (7%)). Differences in sex, other baseline symptoms or comorbidities, or baseline appearance of CXR were not seen. Patients with a history of cancer were more likely than others to have a severe event, defined as admission to the ICU for ventilatory support or death (7/18 vs 124/1572). Patients who had undergone chemotherapy or surgery within a month of the diagnosis of COVID-19 were at particularly high risk. The authors concluded that patients with a history of cancer might have a greater frequency of COVID-19 than those lacking this factor and, if they did develop COVID-19 and were admitted to the hospital, were more likely to suffer severe consequences. The authors proposed three tentative strategies for cancer patients in the current pandemic and future attacks of infectious disease including 1.) intentional postponing of adjuvant chemotherapy or elective surgery, 2.) stronger personal protection for patients with cancer or cancer survivors, and 3.) more intensive surveillance and treatment for cancer patients/survivors who have COVID-19, particularly if they also have other comorbid factors including advanced age.
Zie et al. (2020) commented on this, noting that the conclusion that cancer patients/survivors were at a higher risk for COVID-19 infection was flawed, as the cancer patients might have been under closer follow up and thereby more likely to be diagnosed, and that a proper determination would require comparing the frequency of COVID-19 in a sample of cancer patients rather than the incidence of cancer in a population of COVID-19 patients. They also noted the marked heterogeneity among the 18 cancer patients/survivors in type of malignancy, length of time since treatment, and disease status. They concluded that the current data were “insufficient to explain a conclusive association between cancer and COVID-19”.
Sidaway (2020) reviewed in Nature Reviews Clinical Oncology the experience of Liang and two other reports of the initial Chinese experience with cancer patients and COVID-19. Yu et al. (2020) described the experience from the Department of Radiation and Medical Oncology at Zhongnan Hospital of Wuhan University from December 30, 2019 through February 17, 2020. Of 1524 patients with COVID-19, 12 had cancer (0.79%), which was higher than the cumulative incidence of COVID-19 cases reported in Wuhan, China in the same time frame (0.37%). Ten of the patients were male. Non-small cell lung cancer was the most common diagnosis (7/12). Two patients were without known disease (one two-years NED and another receiving adjuvant breast irradiation following surgery); the remaining 10 had known active disease. Five of 12 were or had recently received chemoimmunotherapy or radiotherapy. As of time of submission of the manuscript, 3 patients had died, 3 remained hospitalized, and 6 had been discharged alive.
Zhang et al. (2020) reviewed laboratory confirmed COVID-19 patients with cancer in three hospitals associated with Tongji College of Medicine in Wuhan from 13 Jan 2020 to 26 Feb 2020. Of 1276 patients with COVID-19 admitted to hospital, 28, or 2.2%, also had cancer. The most common types of malignancy were lung, esophageal, and breast. Mean age of patients was 65. Survival was significantly worse for patients whose chest CT scans showed patchy consolidation or who had received anti-cancer therapy (e.g., chemotherapy, immunotherapy, radiotherapy) less than 14 days before the diagnosis of a COVID-19 infection.
In summary, Sidaway concluded that despite small patient numbers, retrospective data collection, and limited follow-up, there appeared to be a strong suggestion that patients with cancer, especially those receiving recent treatment, were both at higher risk than the general population and more likely to suffer poor outcomes. He urged caution in the conduct of routine follow-up and treatment visits that might increase the risk of cancer patients to those infected with SARS-CoV-2.
Several Oncology Societies representing surgical, medical, and radiation oncologists have provided guidelines on the management of cancer patients guidelines from national organizations. The focus of these guidelines has been to maintain adequate treatment and follow-up of patients with known cancer, particularly but not exclusively for patients with curable malignancies or acutely life-threatening complications (e.g. vena cava compression, spinal cord compression), while minimizing the risk that cancer patients will become infected during the course of their treatment, limiting risk to hospital personnel, and minimizing use of potentially scarce resources such as PPE. These include:
- American Society of Clinical Oncology (ASCO, www.asco.org) includes separate sections for Patient Care Information, Provider and Practice Information, Meeting and Program Updates, Government, Reimbursement, and Regulatory Updates, and Questions.
- American Society of Hematology (ASH, www.ash.org) has established a website with extensive information on the management of patients with a variety of benign and malignant (e.g. leukemias, lymphomas) and COVID-19. They have also established an international registry to collect data on patients with hematological malignancies and COVID-19. Data will be de-identified and review of the registry by the Western Institutional Review Board has deemed it exempt. Further data are available on the ASH website.
- American Society of Therapeutic Radiology and Oncology (ASTRO, www.astro.org) provides general information on practice management, specific guidelines on hypofractionated regimens, which will limit exposure of patients to the clinic setting, and provides specific guidelines that have been shared by a number of Cancer Centers. They have also provided links to other organizations that provide relevant and appropriate information on management of patients with cancer during the COVID-19 Pandemic (See Hyperlinks).
- Society of Nuclear Medicine and Molecular Imaging (SNMMI, http://www.snmmi.org) has developed a page with information pertinent to this community and with a large listing of relevant resources:
- American Association for Cancer Research (AACR, www.aacr.org) recently (27-28 April, 2020) conducted the first half of its usual annual meeting, which had been scheduled to meet in San Diego, as a virtual annual meeting. Sessions were and remain freely available. The morning plenary session on 28 April was devoted to studies of COVID-19 in cancer patients, addressing epidemiology, prognosis, possible interactions between specific cancer treatments, particularly immunotherapeutic ones, and the interaction of socioeconomic factors on treatment of both COVID-19 and cancer. A second session on COVID-19 and cancer later that day addressed key issues in funding of these trials and their impact particularly on early career investigators.
- The Society for Surgical Oncology (SSO, www.surgonc.org) has general updated resources and recommendations on the management of cancer patients and COVID-19 and specific recommendations for patients with breast, colorectal, endocrine, gastrointestinal and hepatobiliary, melanoma, peritoneal surface, and sarcomatous tumors.
- The American College of Surgeons (ACS, www.facs.org) issued guidelines on triage of patients being considered for thoracic surgery, including patients with known or suspected lung cancer. These were issued in March 2020, and with the ongoing duration on COVID-19 particularly in the United States, may warrant re-assessment to minimize undue delay of appropriate diagnosis and treatment of patients.
In addition to these policy statements, several institutions which have seen a relatively large number of cancer patients with COVID-19 have reported their own guidelines and early experience. Filippi et al. (2020) reported their experience and recommendations from several radiation oncology departments in Northern Italy during their first wave of patients with the COVID-19 pandemic. They defined five areas to prioritize in the management of cancer patients with known or suspected COVID-19: 1.) ensuring radiation therapy delivery to cancer patients, 2.) ensuring safety of staff, 3.) management of cancer patients known or suspected to be COVID-19 positive, 4.) staff reorganization to reduce time in clinic, reducing close contact, working from home, conducting conferences by video or phone, etc., and 5.) reducing patient contact with radiation therapy facilities by postponing or using telemedicine for follow-up visits, using hypo-fractionation when possible, delaying start of non-urgent treatment, and exploring non-radiotherapy methods of palliative treatment.
Ueda et al. (2020) reported experience and policies developed at the Seattle Cancer Care Alliance, Fred Hutchinson Cancer Research Center, and the University of Washington which were the first sites in the U.S. to see a large number of patients with cancer and COVID-19. They addressed issues of general infection control, reduction of hospital-based staffing to the minimum required for quality care with the majority of staff members (e.g. physics, dosimetry, many physicians and nurses, working from home, use of telemedicine for many follow-up and some initial consultations, deferral of consultations for second opinions, discussion of treatment options for patients with low risk prostate and breast cancer whose radiotherapy might be deferred during initial neoadjuvant hormonal management, the use of hypo-fractionated treatment for both definitive and palliative cases, and proactive discussion with patients about appropriate palliative and end-of-life goals). The impacts of these rapid changes on employee and leadership well-being are likely to be significant, and pre-emptive attention to issues such as burnout, dealing with illness and death of colleagues, development of policies for furloughs, mandatory isolation, compensation and provision for child care, and rotation of leadership positions are addressed.
Wu et al. (2020) reported experience from Wuhan, China in January and February 2020. In early January, when neither the extent nor mode of transmission of COVID-19 was clear, treatments were given as usual with no particular attention to mask wearing, hand hygiene, or linear accelerator disinfection. Around January 20, 2020, person to person transmission was reported. Departments were closed between January 23-27, 2020 for Chinese New Year, and after re-opening, many departments closed again for several days because of infections of patients and staff. The Hubei Cancer Hospital, the only hospital specializing in cancer treatment in Wuhan, China did not reopen then but spent three days disinfecting treatment machines and vaults and implementing strict infection control policies for staff and patients before reopening on January 30, 2020. These guidelines included: 1.) patient screening for COVID-19, 2.) health education for patients and re-consenting of patients regarding the risk of infection, 3.) screening of staff for COVID-19, 4.) staff education including proper use of PPE (staff are shown wearing gowns, N-95 masks, gloves, and protective eyewear), 5.) zoning the department into clean/semi-soiled/and soiled zones with specific protocols for allowed activities and disinfection schedules, 6.) special modification of immobilization equipment (e.g., clear wrap, thermoplastic masks, etc.), and 7.) modification of workflow to limit patient-patient and patient-staff contact. The authors report that between January 30, 2020 and the time of the writing of this report (submitted March 17, 2020) there was no documented transmission of COVID-19 between patients and staff.
In departments treating both cancer patients thought to be free of COVID-19 and those with known or suspected infection, it has been suggested that the daily treatment be scheduled to treat first those patients at highest risk for poor outcome with infection (the elderly, those with asthma, COPD, cardiac disease, or diabetes) than treat lower risk but COVID-19 negative patients, and treat COVID-19 positive patients only at the end of the day with reduced staffing.
In some cases these recommendations for alteration in what had been the conventional patterns of treatment in many radiation departments, with relatively lengthy fractionated treatment of many palliative regimens (e.g. bone metastases treated in 10 fractions over two weeks) have been criticized as “brutal” (Johnson 2020). It will be important to indicate to patients in these circumstances that there are a number of well-established, prospective, randomized trials which have shown that shorter treatment regimens, both for palliation of bone metastases, as well as definitive treatment of malignancies such as cancers of the breast and prostate, have shown that shorter regimens are as effective and no more toxic than longer ones (Wright, 2020; Yeramilli, 2020). Some of the reluctance to adopt these, at least in the U.S., has been based on reimbursement patterns which have perversely rewarded radiation oncologists for keeping to the older more protracted regimens. Shorter fractionation patterns reduce the time and economic disruption for the patient, allow more patients to be treated on a limited number of machines, and cost less. Sometimes it takes a pandemic to convince us to do the right thing.
Many of the above recommendations have come from relatively well-resourced cancer centers. As the extent of the COVID-19 pandemic increases to include more patients treated with poorer baseline healthcare, often with low and middle-income, and in rural hospitals, which may have significant limitations in equipment and staffing compared with larger cancer centers, these recommendations may require modification. Some suggestions for these settings have been proposed by Pino et al. (2020).
The initial concern about cancer patients and COVID-19 was to see if these patients had higher incidence and morbidity rates than the general population or a proper age and comorbidity adjusted group and to develop ways to minimize exposure of cancer patients receiving treatment to situations with a high risk of contracting COVID-19. A secondary concern was to develop ways to treat COVID-19 infected cancer patients receiving both definitive or palliative treatment. Initial studies did suggest that individuals with cancer were at somewhat higher risk for infection and for more severe outcomes if infected, although data presented at the most recent AACR meeting suggested that these effects were seen more prominently in data reported from Wuhan, China than in data from France and Italy.
Early strategies tried to minimize contact of known cancer patients with the hospital and clinic environments. Elective surgical procedures, which may have included diagnostic biopsies, were deferred, infusion schedules for chemotherapeutic or immunomodulatory agents were often prolonged, and the use of hypofractionated radiotherapy schedules, particularly those previously shown to be equivalent in tumor control or palliation to more protracted ones, were all strongly recommended. Patient contact was minimized, and most follow-up visits and many initial consultations were performed by telemedicine.
These strategies initially seemed reasonable, particularly when many thought that the COVID-19 pandemic might run its course in a few months. However, now that we understand better, barring rapid introduction of effective antiviral or vaccination regimens, we will remain in this situation for at least a year, and there is increasing concern about the impact of these strategies on survival of cancer patients whether or not they are also infected with COVID-19. The competing needs of patients with COVID-19 and cancer may worsen the prognosis of cancer patients in at least three ways: 1.) Risk of severe or fatal COVID-19 infections in cancer patients, particularly those who are older or have other comorbidities. 2.) Delay of cancer diagnosis and effective treatment due to reallocation of hospital resources, leading to cancer upstaging and likely poorer outcomes, and 3.) Loss of insurance by those whose coverage came from work may lead them to postpone or omit screening procedures or diagnostic evaluation for ambiguous symptoms, causing later diagnosis and poorer long-term results.
Cancer, Immune Checkpoint Inhibitors (ICI), and COVID-19
One issue of recent concern has been the possible interaction of patients with cancer who are being treated with agents which abrogate the PD-1/PD-L1 axis who are also infected with SARS-CoV-2. The use of these agents, as well as those interfering with CTLA-4, has become widespread in a number of cancers including lung, genitourinary, and melanoma, and they have become the mainstay of therapy for patients with metastatic disease in many cases. In theory, the use of these agents might be beneficial by enhancing the immune response to the virus, or deleterious by worsening the immune over-reactivity seen in many cases of COVID-19 associated with a cytokine storm. Early reports of clinical experience gave a mixed picture, with some studies reporting worsened survival for cancer patients who had received ICI (Robilotti et al. 2020) and others failing to show any such association (Mehta et al. 2020). Patients in these studies as well as other case reports had a mixture of underlying cancer diagnoses, co-morbidities, ages, and other factors known to impact prognosis.
Luo et al. (2020) have carefully examined a cohort of 69 patients from Memorial Sloan-Kettering Cancer Center with lung cancer and COVID-19 (documented by RT-PCR) treated between March 12, 2020 and April 13, 2020 and followed for a median of 14 days. Of these, 41 (69%) had received PD-1 blockade prior to the diagnosis of COVID-19 with a median time interval of 45 days (range 4-820 days). Overall, they did observe a significant association between prior PD-1 blockade and disease severity or outcome (hospitalization, intubation, death). Unadjusted differences in these were no longer seen when adjusted for prior smoking status. Peak IL-6 levels did not differ between the two groups. They concluded that these findings were encouraging for the continued use of PD-1 blockade for lung cancer patients during the COVID-19 pandemic, but that further studies in this and other malignancies (e.g. GU, melanoma) were indicated to establish generalizability and durability of their findings.
Potential Impact of Delays in Cancer Diagnosis and Treatment on Prognosis
There are at present few good quantitative data on the extent to which these will occur, or the impact they will have on survival outcomes. Lai et al. (2020) have developed a model for estimating excess mortality in patients with cancer, multimorbidity, and COVID-19. This is based on data on referrals for cancer diagnostic procedures and chemotherapy treatments in England and Northern Ireland during March and April 2020. It requires making assumptions, for which there are as yet few data and the authors admit that their estimates are “plausible’, on both the Relative Impact of the Emergency (RIE) and the Proportion of the population Affected by the Emergency (PAE). Under what they consider to be conservative assumptions that only incident cases will be affected, a PAE of 40% and an RIE of 1.5, they calculate an excess 6,270 deaths in England and 33,890 deaths in the United States due to the impact of COVID-19 on cancer treatment. While the exact figures are open to question and will require more data to refine the estimates of RIE and PAE, it seems clear that the COVID-19 Pandemic will impair survival of cancer patients including those not previously infected by the virus.
Mehta et al. (2020) have reported case fatality rates in cancer patients with COVID-19 treated at Montefiore Health System in New York between March 18 and April 8, 2020. They identified 218 patients with cancer and COVID-19, 164 with solid tumors and 54 with hematologic malignancies. Sixty-one (28%) of these patients died, with the rates highest for those with lung cancer (55%), hematologic malignancies (37%), and gastrointestinal cancers (colorectal (38%), pancreas (67%), upper GI (38%), and gynecologic malignancies (38%). An age and gender matched control group of COVID-19 patients without cancer from the same hospital system has a mortality of 14% during the same period. Active chemotherapy and radiotherapy were not associated with mortality, and few patients were receiving immunotherapy. These patterns of increased mortality with a predominance of patients with lung cancer or hematologic malignancies are similar to that reported by Dai et al. (2020).
It will be necessary in the near future to develop policies which can minimize the risk that cancer patients will be exposed to COVID-19 infection, particularly while under active therapy with chemotherapy, while minimizing delays in their diagnostic and staging procedures, facilitating appropriate surgery, and allowing essential adjuvant radiation, chemotherapy, or immunotherapy with minimal disruption. Waterhouse et al. (2020) have in press in JCO Oncology Practice a survey of ASCO members addressing specifically the changes and challenges impacting clinical trial programs but which also addresses other key elements of oncology practice in the context of COVID-19. They note that the development of robust mechanisms for telehealth, use of electronic signatures, allowing remote lab and imaging facilities to collect data, direct shipment of oral drugs to patients, and standardization of policies and procedures, all of which have been discussed by Cooperative Groups for decades, have now become standard of care in a few months. While borne of tragedy, some of the changes to health care and its documentation may be of great general value. It will be our responsibility to ensure that these improvements are distributed equitably throughout our healthcare system which may be a major challenge in the coming years until we develop a more unified healthcare system.
While entry of new patients on cancer-related clinical trials is likely to be infrequent or nonexistent during the COVID-19 Pandemic, a number of patients currently enrolled on trials will require further treatment, assessment of response, and follow-up. For these patients, the changes in their usual treatment, follow-up, and imaging schedules mandated by the pandemic may understandably result in what would ordinarily be considered protocol variations or violations. Guidelines for appropriate management of these situations, including provision for expedited review of protocol amendments and guidelines to Institutional Review Boards and Protocol Review Committees have been proposed by the FDA, CTEP, and several Cooperative Groups (O’Dwyer, 2020; You, 2020). In some cases there may be potential conflicts for use of scarce agents such as tocilizumab, an IL-6 antagonist, which is used to manage cytokine release occurring following CAR-T therapy and is in investigational use in patients with COVID-19.
The April 17, 2020 issue of The Cancer Letter was devoted largely to interviews with the NCI, ACS, and other major US agencies exploring the interaction between the need for cancer treatment during the COVID-19 Pandemic. The authors provided data indicating that accrual to clinical trials from the National Clinical Trial Network had declined significantly between February 3-9, 2020 and March 23-29, 2020. Comparing the last week of this period to the average of the preceding 7 weeks, accrual for the Intervention step trials decreased by 44% and for the Screening step declined 42%. While the initial impact of the COVID-19 pandemic on accrual to clinical trials has been negative, a number of investigators have viewed this as an opportunity to make some long-needed changes in trial efficiency, approval and accrual processes, availability to individuals not living in major urban centers, and limiting data collection to what is needed to evaluate the endpoints of the trial (Bailey 2020; Borno 2020; Shuman 2020). In the future, these changes may result in more efficient and equitable clinical trials not only in oncology but in medicine in general.
Several registries have been established to collect data from national and international sources on COVID-19 patients with cancer. These include:
- ASCO Survey on COVID-19 in Oncology (ASCO) Registry (email@example.com).
- The COVID-19 & Cancer Consortium (ccc19.org). US Cancer Centers and other hospitals
- The Global COVID-19 Observatory and Resource Center for Childhood Cancer, St. Jude Global (firstname.lastname@example.org) in collaboration with the Societe Internationale Oncologie Pediatrique (SIOP).
- TERAVOLT (Thoracic Cancers International COVID-19 Collaboration) (AACR Abstracts Online 2020; Garassino M.C., TERAVOLT: First results of a global collaboration to address the impact of COVID-19 in patients with thoracic malignancies). Coordinated through the International Association for the Study of Lung Cancer and Vanderbilt University (email@example.com).
Other disease and population specific registries are rapidly being formed and the parent oncologic organizations should be consulted for additional options.
Management of Patients with Diabetes
One of the most severe complications associated with diabetes is diabetic ketoacidosis (DKA), a life-threatening condition induced by low levels of insulin. In the absence of insulin, relatively higher levels of glucagon will trigger the liver to produce glucose from its glycogen stores, thereby elevating circulating blood glucose, which may contribute to increased urination and dehydration. At the same time, the body will also switch over to fatty acid catabolism, producing acidic ketones. The increased production of this byproduct contributes to lowered blood pH (acidosis).
DKA is more common in individuals with Type 1 diabetes, but it may also occur in Type 2 patients, particularly in those who have poor blood sugar management and concurrent infections. Viral infections are especially associated with elevated risk of DKA, and so glucose levels, blood pH, as well as urine or blood ketone levels should be closely monitored in diabetics with COVID-19. Common DKA symptoms include dehydration, increased urination, vomiting, abdominal pain, and in severe cases, a fruity breath odor, confusion, loss of consciousness, and deep gasping breathing (known as Kussmaul breathing), a form of hyperventilation that may help increase blood pH. It is particularly important that diabetes patients are kept well-hydrated with carbohydrate-free fluids to help prevent this possible complication. DKA can make fluid and electrolyte management particularly challenging, which may contribute to elevated risk of sepsis, a serious complication associated with COVID-19.
Management of Pregnant Patients
From the first reports of COVID-19 infection, there has been understandable concern regarding the impact of infection on maternal and fetal health. Initial reports from Wuhan, China suggested that pregnant women did not seem to be at risk for more severe disease, and further experience in Europe and the U.S. appears to have borne this out (Breslin 2020; American College of Obstetricians and Gynecologists 2020). While data are limited, particularly among the medically underserved population who make up a large percentage of the U.S. cases, guidelines regarding management of COVID-19 during pregnancy have been issued by a number of relevant professional societies including the CDC (2020), the American College of Obstetrics and Gynecology (2020), and the Society for Maternal Fetal Medicine (2020).
In general, overall management of the pregnant woman with COVID-19 should be directed primarily by the severity of the COVID-19 infection. Management of the pregnancy should be guided more by obstetric considerations unless COVID-19 is severe. A diagnosis of COVID-19 is not generally considered an indication for early delivery. Women diagnosed with COVID-19 late in the third trimester may attempt to postpone delivery (if feasible) until a negative test is obtained to minimize neonatal transmission. SARS-CoV-2 has not been reported in vaginal fluids nor in breast milk but is present in feces, suggesting an increased risk with vaginal delivery rather than Cesarean section. There is limited data on transplacental vertical transmission of SARS-CoV-2. This appears to be uncommon but has been reported in rare cases, as have neonates who had IgM against SARS-CoV-2 at birth.
Current and Proposed Clinical Trials
The scientific method allows one to formulate hypotheses that are intended to explain phenomena by understanding the cause-and-effect relationship between two measurable variables. These hypotheses can be investigated by designing studies with an appropriate methodological approach. A good study design addresses concerns like biases that represent a threat for the validity and reliability of the results. Studies can be classified into two types: 1.) observational, if there is no intervention being introduced to the subjects by the researcher, or 2.) interventional, if in fact there is. Interventional studies are superior to observational studies in many aspects; in particular, they allow the investigator to have better control over the conditions of study, decreasing the possibility of biases such as confounding.
Among interventional studies in medical research, clinical trials are by far the most reliable source of scientific evidence when investigating cause-and-effect relationships. Essentially, in clinical trials an intervention is introduced into subjects sampled from a population of interest. The effect of the intervention can be evidenced upon completion of the study, when researchers after collecting a sufficient amount of data can perform a thorough analysis according to an appropriate statistical method. This intervention might involve, but is not limited to, for example, a drug, a diagnostic test, or even a survey by questionnaire.
Every clinical trial investigates the effect of an intervention on an outcome (usually multiple outcomes) of interest. Aims of clinical trials may vary depending on the stage of development, which in turn is defined by phases (only for drugs, biological products, or radiation) that differ from each other according to study design characteristics. For example, first-in-human studies are mainly focused on safety to humans. Five different phases are described based on definitions developed by the U.S. Food and Drug Administration (FDA).
Information about clinical trials is publicly available through www.clinicaltrials.gov. Filtering by phase allows to search for studies according to the following categories:
- Preclinical Trial involves testing of a compound in non-human candidates for the purpose of studying efficacy, toxicity, and pharmacokinetic action.
- Early Phase I Trial (formerly listed as Phase 0) involves exploratory trials aiming to confirm what preclinical trials predicted about how the compound will affect the body.
- Phase I Trial (also known as First-in-Human Studies) is mainly focused on safety, as well as side effects (on-target and off-target), pharmacokinetics, posology, etc. and is usually conducted with a small number of subjects, often just including healthy volunteers. One goal is to determine the recommended Phase II dose for use in subsequent studies.
- Phase II Trial (also known as Proof-of-Concept Studies) is focused on effectiveness, but safety is still being evaluated (i.e. short-term adverse events) and may be conducted as a single-arm trial with comparison to historical controls, or as a randomized Phase II design comparing the investigational arm with a standard arm (which may be a placebo if there is no established effective therapy).
- Phase III Trial involves larger studies (often multi-center) with more participants, and is focused on improving the knowledge on safety and effectiveness by studying different populations and different dosages as well as comparing with different compounds. These typically compare an intervention with the current standard of treatment, which in some cases may be a placebo
- Phase IV Trial is the final phase after the FDA has given its approval for marketing. These studies are focused on long-term surveillance for safety, efficacy, and optimal use. Some refer to them as post-approval advertising whose main goal is to get the agent into the hands of prescribing clinicians to encourage its use.
- Not Applicable (Trial) applies to trials without FDA-defined phases, including trials of devices or behavioral interventions.
In planning any clinical trials, careful attention should be paid in its design to the outcome(s) to be assessed, the clinical endpoints which will be measured to assess these outcomes, patient inclusion and exclusion criteria and possible stratification in randomized trials if factors such as age, comorbidities, and severity of disease are felt to be relevant to the outcome(s), and plan for statistical analysis. These should all be defined before patients are entered in the study.
Given the spread and mortality rates of SARS-CoV-2 and COVID-19, significant concerns for humanity arise. Prevention and treatment options are eagerly needed. As for the moment we have no available options, the standard of treatment remains as supportive care. To tackle this threat, several clinical trials are currently being conducted in countries all over the world, and the number of trials is growing each day. As of March 30, 2020, almost 3 weeks after COVID-19 was declared a pandemic by the WHO, 220 clinical studies were listed for public access (see http://www.clinicaltrials.gov), none of which have provided preliminary results. Out of this total, 143 (65%) trials corresponded to interventional studies (actual clinical trials). From the 143 clinical trials, 44 (30.8%) trials were classified as Phase III, and 14 (9.79%) were classified as Phase IV. As of August 23, 2020, out of the ~3,000 clinical trials being conducted concerning COVID-19, ~2,000 were studying treatment options, with ~120 Phase I-IV clinical trials studying potential vaccine candidates (52 in Phase I, 59 in Phase II, 44 in Phase III, and 4 in Phase IV, all four of which were studying the BCG vaccine, an approved vaccine for use in the prevention of tuberculosis).
The development of a novel medication in the U.S. takes on average 12 years, from discovery of target to approval for marketing. For vaccines, from target identification to development of a high quality compound the average time usually lies between 10 to 15 years. These time lapses can be reduced if enough resources and efforts are put together. Some experts expect that good evidence should be available in about 6 months. As of April 1, 2020, the NIH and other U.S. Federal Agencies funded 3 clinical studies; 2 of these for treatment purposes and 1 for the development of a vaccine. The following sub section will summarize the 2 studies being conducted for treatment purposes, as the vaccine is already mentioned later in this section (see Current Vaccine Candidates).
The rapid spread and significant lethality of COVID-19 have understandably given us a sense of great urgency in the development and deployment of effective treatment and preventive strategies. While this is understandable, it may give rise to incautious haste in the design and reporting of clinical trials of these agents. A number of recent agents have had results reported in submitted but not reviewed form, or as press releases, without undergoing peer review. Critical details appear not to be provided. London and Kimmelman (2020) have outlined the core concepts of what they deem “pandemic research exceptionalism” in a recent article in Science. Here, they identify the assumptions driving this practice as 1.) some evidence now is preferable to better evidence later, 2.) the belief that randomization or placebo controls (when there is no standard of care) conflict with the perceived responsibility of the clinician caring for the patient, and 3.) the expectation that researchers and sponsors are free to exercise broad discretion over research design, without the recognition that their work frequently has undergone significant funding and its results a public good. To counter these assumptions, the authors propose five “conditions of informativeness and social value”, namely, 1.) Trials should be important, aiming to detect results that are realistic but also clinically meaningful, 2.) Trials should be rigorously designed, and 3.) Trials should undergo analytical integrity. This includes such elements as specifying trial outcomes and endpoints up-front, rather than choosing them after the fact. 4.) Trials should be reported completely, promptly, and consistently with prespecified analyses. 5.) Trials should be feasible, in terms of timeliness and accrual. While these criteria may seem to run counter to our desire to get rapid answers, they have been well-borne out by decades of clinical research and seem far preferable to the current practice of publication by press release (with details to follow at some unspecified date). There is also the responsibility of the press to report the results of one trial in the context of other trials addressing the same general issue rather than citing its results in isolation.
The global production of literature on coronaviruses, SARS-CoV-2, and COVID-19 has exploded and become a ‘pandemic paper tsunami’ (Brainard 2020). As of April 24, 2020, it is estimated that this literature has grown to ~88,000 published articles and ~24,000 preprints. While a number of journals have made all COVID-19-related publications freely available, not all have and about 20% of these may be behind various paywalls, which limit both human and AI access.
Several institutions have attempted to coordinate these various publications and make them amenable to both human and AI searches. The White House office of Science and Technology Policy launched the COVID-19 Open Research Dataset (CORD-19), which now includes more than 200,000 articles and preprints. However, not all of these are available in full text format, and not all include such terms as ‘coronavirus’ in their titles, abstracts, or keywords.
Another approach, being used by Kate Grabowski and colleagues at Johns Hopkins, is to use careful human curation by 50 or more reviewers of the literature and to select high-value articles in their database, rejecting those which were reviews, commentaries, protocols, or contain poor-quality data. Their results, available as the 2019 Novel Coronavirus Research Compendium, have reviewed and summarized more than 120 high quality papers on a variety of topics related to COVID-19 as of its launch in April and should continue to grow with time. Balsari et al. (2020) have also published guidelines for selecting trustworthy papers from the current unwieldy flood.
Adaptive COVID-19 Treatment Trial
The Adaptive COVID-19 Treatment Trial (ACTT) is an adaptive, multicenter, randomized, double-blind, placebo-controlled Phase III clinical trial. It is funded by the National Institute of Allergy and Infectious Diseases (NIAID). The aim is to evaluate the safety and efficacy of novel therapeutic agents in hospitalized patients with confirmed diagnosis of COVID-19 infection. The population includes people with moderate to severe disease aged between 18 and 99 years, excluding pregnant women, as well as people with hepatic and renal impairment. The primary outcome is defined as the percentage of subjects reporting each severity rating on an 8-point ordinal scale (www.clinicaltrials.gov Identifier: NCT04280705).
Adaptive clinical trials allow investigators to modify parameters of the protocol according to observations related to the outcomes along the course of the study. This is part of interim analyses, which are performed to monitor safety and efficacy of the intervention. The idea is to make decisions during the course of the trial, such as early termination, if needed, either because the data shows enough evidence to justify the next Phase, or to stop the trial due to concerns about risk/benefit ratio.
Leask (2020) has summarized the progression from anecdotal observations to randomized trials with a variety of agents and listed more than 50 current and planned trials as well as provided several links to organizations providing links to trials of both vaccines and therapeutic agents. The NIH provides an up to date listing of COVID-19-related clinical trials (902 trials as of April 23, 20202) which allows filtering by study type, activity, eligibility criteria, phase of trial, funding source, publication status, and other factors (https://clinicaltrials.gov/ct2/results?cond=COVID-19).
The Randomised Evaluation of COVID-19 Therapy (RECOVERY) Trial is a UK randomized clinical trial that aims to test the therapeutic benefit of a variety of possible treatments for hospitalized COVID-19 patients. The primary sponsor for the trial is Oxford University, with funding provided by a grant from the UK Research & Innovation/National Institute for Health Research. As of July 28, 2020, the trial has already published results on the clinical effects of the use of such therapies as lopinavir-ritonavir (see Lopinavir and Ritonavir), dexamethasone (see Dexamethasone), hydroxychloroquine sulfate (see Chloroquines), and clinical trials concerning the use of low-dose dexamethasone, azithromycin, Tocilizumab, and convalescent plasma are ongoing. It has also enrolled over 11,800 patients across 176 NHS hospitals in the UK.
The Solidarity Trial for COVID-19 treatments is a Phase III-IV multinational clinical trial organized by the World Health Organization. It began on March 18, 2020 and has over 100 countries participating. The randomized, clinical trial aimed to evaluate the efficacy and safety of the use of remdesivir, lopinavir/ritonavir, lopinavir/ritonavir with interferon beta, and hydroxychloroquine or chloroquine. On July 4, 2020, the WHO announced that it would continue the hydroxychloroquine and lopinavir/ritonavir treatment arms of the study, citing interim trial evidence that both treatments showed little or no reduction in mortality of hospitalized COVID-19 patients when compared to patients receiving standard care. A separate Solidarity Trial for vaccines was announced in May, 2020.
Preliminary results of this trial have been released as a medRxiv preprint (not peer-reviewed) as of 15 October, 2020 (Pan et al. 2020). A total of 11,266 adults who were hospitalized with COVID-19 were randomized in 405 hospitals in 30 countries Allocation was 27850 remdesivir, 964 hydroxychloroquine, 1411 lopinivir, 651 interferon plus lobinivir, 1412 only interfreron, and 4088 no study drug. Compliance was reported as 94-96% midway through treatment with 2-6% crossover. None of the study drugs definitely reduced in-hospital mortality in unventilated patients or any other subgroup of entry characteristics, nor did they impact initiation of ventilation or duration of hospitalization. The authors concluded that these drugs as used in this patient population had littleor no effect on the predefined study outcomes.
Proposed Repurposed Drug Therapies
As of September 2020, only one FDA-approved drug therapy (Remdesivir) is available for use in the treatment of severe COVID-19 patients (convalescent blood plasma, which was also given emergency use authorization, is discussed in Potential Experimental Treatments), so the identification of existing pharmaceutical drugs that may be repurposed for use in such patients is especially crucial. Moreover, clinical trials for repurposed drugs are much faster than they are for new drug therapies. Repurposed drugs have also gone through previous safety trials, and their potential for adverse reactions has already been well-described. Using AI and large scale compound repurposing (e.g. Riva L et al., Nature, 2020), several new candidate compounds have emerged that could contribute to applications of precision molecular medicine for COVID-19. One finding that came out of this study, looking at isolates of SARS-CoV-2, is the potential for antiviral activity of a highly selective PIKfyve kinase inhibitor. An example of such AI-informed large scale screens of potentially repurposable drugs is the discovery of a compound, LAM-002A, that could have utility for COVID-19 (Bouhaddou M et al., Cell, 2020).
The drugs that follow in this section are compounds that have been approved for use in the treatment of other conditions but may have potential in ameliorating the condition of COVID-19 patients. The mechanisms of their purported actions differ widely, however. Since an understanding into the effects these drugs may have on COVID-19 is only emerging, it is essential to assess the efficacy, specificity, and safety of each individually. For this reason, randomized, large-scale clinical trials on such patients are not only recommended but necessary.
Angiotensin Receptor Blockers
Angiotensin II Receptor Blockers (ARBs) are antagonists of angiotensin II receptors, which are commonly used to treat hypertension. ARBs bind to angiotensin II thereby inhibiting angiotensin II from binding to its receptor. In doing so, ARBs effectively block a physiological pathway responsible for vasoconstriction, which can lead to increased blood pressure and hypertension (see ACE2 Receptor). In addition to being a risk factor for severe COVID-19, hypertension has also been linked to many other risk factors of severe COVID-19, such as endothelial dysfunction, inflammation, and fibrosis. The loss of ACE2, an enzyme which can convert angiotensin II into angiotensin 1-7, during SARS-CoV-2 infection can lead to enhanced levels of angiotensin II, causing enhanced fibrosis and in some cases acute lung injury. Thus, the use of ARBs to lower the action of angiotensin II may be therapeutic in COVID-19 patients, particularly in those with a history of hypertension.
However, the use of ARBs in COVID-19 patients has been the subject of some controversy. In March 2020, Fang et al. suggested that the use of ARBs may upregulate ACE2 expression, which could enhance SARS-CoV-2 entry into host cells, potentially leading to a more severe clinical course from higher successful rates of early viral replication. A response from the European Society of Cardiology on March 13, 2020 claims that there is no evidence to substantiate this claim concerning the use of ARBs, and in fact, prior animal studies have shown that these very medications may be protective against serious lung complications. Below we outline information on current clinical trials that are testing the safety and efficacy of using ARBs in the treatment of COVID-19 patients and provide some study results on the drugs’ impact in COVID-19 patients.
Current Clinical Trials
There are a wide variety of ARBs that have been studied for their effect in the treatment of COVID-19 patients. Some common ARBs include Losartan, Valsartan, Olmesartan, and Telmisartan. A substantial portion of COVID-19 clinical trials concerning ARBs are focused on the use of Losartan, particularly those being carried out in the U.S.
As of July 28, 2020, there are 45 clinical trials studying the effect of ARBs in the treatment of COVID-19 registered internationally, 10 of which are being carried out in the U.S. For the U.S. trials, one is in Phase I, five are in Phase II, two are in Phase III, and one is in Phase IV. Only one registered trial in Ireland has been suspended, and one registered trial in China concerning the impact of hypertension and hypertension treatments in COVID-19 patients was completed on March 30, 2020.
***Mehra et al. (2020) conducted an observational study of 8,910 hospitalized COVID-19 patients from 11 countries located on three continents. The authors report that neither ACE inhibitors or ARBs were associated with increased risk of death or increased severity in COVID-19 symptoms for all patients studied. The same lack of association was found when only patients with hypertension were included in the analysis. The study also found that use of ACE inhibitors or statins (used to treat high cholesterol) was associated with decreased death when compared to non-use of the drugs, but since the study was not randomized, limited conclusions can be drawn. [Concerns over the source of the data presented in this paragraph have been recently expressed by the editors of the New England Journal of Medicine, and the conclusions of this work have been called into question.]*** (Servick et al. (2020))
In a study of COVID-19 patients from the Lombardy region in Italy, Mancia et al. (2020) found similar conclusions. A comparison of 6,272 COVID-19 patients with 30,759 controls matched for age, sex, and municipality of residence revealed that neither ACE inhibitors or ARBs were associated with any differences in prevalence of SARS-CoV-2 infection. The drugs were also not associated with increased severity of COVID-19.
Reynolds et al. conducted an observational study that analyzed the health records of 12,594 patients admitted into the New York University Langone Health System who were tested for COVID-19. Only 5,894 patients tested positive, and of those, only 1,002 showed signs of severe illness. The study found no association between ACE inhibitor or ARB use with increased likelihood of SARS-CoV-2 infection or with increased severity of illness, even when the analysis was isolated to just patients with hypertension.
Chloroquines (with or without Azithromycin)
On March 28, 2020, the FDA issued an Emergency Use Authorization to allow for hydroxychloroquine sulfate and chloroquine phosphate in adult hospitalized COVID-19 patients who were not participating in another investigational clinical trial. However, on June 15, 2020, the FDA revoked the authorization, citing that both chloroquine phosphate and hydroxychloroquine sulfate showed little to no efficacy in treating COVID-19. Furthermore, the use of the drug had been previously linked to an increase in serious cardiac adverse events, as well as other serious adverse effects, showing definitively that the FDA no longer had confidence that the benefits of using the drug outweigh any potential risks.
Initial interest in the use of the two medications in treating COVID-19 stemmed from the widespread availability of the drugs and also their previously reported antiviral properties. Zinc ionophores like chloroquine phosphate and hydroxychloroquine sulfate show increased transport of Zn2+ cation into the intracellular space (Yao et al., 2020), which inhibits RNA-dependent RNA polymerase, thereby inhibiting the replication of the virus (te Velthuis et al., 2010). Both drugs are approved for use to prevent and treat malaria. While the two drugs are structurally similar, hydroxychloroquine sulfate is also used in the treatment of lupus and rheumatoid arthritis. Both drugs have shown efficacy in lowering inflammatory response, and the mechanism for their action is thought to stem from their chemical properties. Both drugs are weak bases that can easily diffuse through lipid membranes. When they enter acidic lysosomes, they become protonated, which in turn inhibits their ability to cross lipid membranes, causing them to accumulate in lysosomes. As a result, the pH of the lysosomes increase, which decreases the lysosomes’ proteolytic activity. It is believed that this process decreases innate immune cell activity, which is why hydroxychloroquine sulfate may be effective in treating some autoimmune conditions.
Most adverse responses to the medications are relatively mild, but retinopathy is a more serious side effect that may develop, particularly with chronic use. Liver damage and liver failure are also possible adverse effects from long-term use, and both medications are contraindicated for patients with glucose-6-phosphate dehydrogenase deficiency, psoriasis, porphyria, anemia, as well as other conditions. The use of these medications may also be associated with elevated risk of heart problems in patients with COVID-19, particularly arrhythmias associated with QT prolongation.
Both medications were tested early on in the pandemic as COVID-19 therapies in China and South Korea, where they showed some efficacy in reducing symptoms and in improving patient outcome (Todaro et al., 2020). The CDC also reported that in cell culture, when administered 24 hours prior to introduction of the virus, chloroquine phosphate significantly reduced SARS-CoV-1 infection (Vincent et al., 2005). Nevertheless, many large-scale meta-analyses have provided evidence that these drugs have limited efficacy in the treatment of patients with severe COVID-19. Some studies have also highlighted the potential for adverse effects in COVID-19 patients, possibly contributing to increased mortality.
Current Clinical Trials
As of September 26, 2020, there were 26 clinical trials studying the effect of chloroquine phosphate in the treatment of COVID-19 registered internationally, two of which are/were being carried out in the U.S. For the U.S. trials, one is in Phase II, and one is currently suspended, but was in Phase III. There are ~200 clinical trials studying the effect of hydroxychloroquine sulfate in the treatment of COVID-19 registered internationally, 53 of which are being carried out in the U.S. For the U.S. trials, 5 are in Phase I, 28 are in Phase II, 21 are in Phase III, 6 are in Phase IV. Nine U.S. trials have been terminated, withdrawn, or suspended, and six have been completed.
Results from Individual Trials
Results from a double-blind, randomized, Phase IIb clinical trial indicate that dosing of chloroquine phosphate is an especially important consideration when treating COVID-19 patients, as higher dosing was associated with higher mortality risk (Borba et al., 2020). The study sought to evaluate the clinical course of 81 hospitalized patients in Manaus, Brazil who were either administered a high dose of chloroquine phosphate (600 mg twice daily for 10 days) or a low dose (450 mg once daily daily for 5 days except on the first day when two doses were administered) in conjunction with ceftriaxone and azithromycin. For all 81 patients tested, the case fatality rate was 13.5%, and the 95% confidence interval (6.9-23.0%) overlapped with that of two other major studies that used patients not receiving chloroquine phosphate. Therefore, the authors were not able to conclude that the drug was associated with reduced COVID-19 mortality. The researchers found elevated mortality in the patients treated with the higher dose, and once observed, all patients receiving the high dose treatment were immediately switched to the low dose treatment for the remainder of the trial.
On March 16, 2020, Raoult discussed results from a successful trial of hydroxychloroquine sulfate and azithromycin tested on 26 patients with COVID-19 infection from southern France (16 other COVID-19 patients were not administered the treatment). A daily 600 mg dose of hydroxychloroquine sulfate not only improved patients’ clinical symptoms and disease outcome, but after 6 days, only 25% of the patients of the patients receiving hydroxychloroquine sulfate remained contagious, whereas the patients receiving standard treatment remained contagious after the same time period (Gautret et al., 2020). The French government plans to scale this testing on COVID-19 patients in other hospitals in the country. On March 27, 2020, this group subsequently reported on a larger trial of 80 patients (6 of whom were included in the earlier report) treated with hydroxychloroquine sulfate and azithromycin. They noted rapid falls in nasopharyngeal viral load and improvement in clinical course with all but two patients being discharged from the ICU (Gautret et al., 2020). This was, however, an uncontrolled study, and we await the results of prospective large-scale randomized controlled trials, both for treatment of active disease with this regimen or its possible use in preventing infection in high-risk individuals (e.g. professionals working with COVID-19 patients).
On March 28, 2020, the FDA issued an Emergency Use Authorization (EUA) for the use of hydroxychloroquine sulfate and chloroquine phosphate for hospitalized COVID-19 patients. Magagnoli et al. (2020) have reported a large but retrospective study of 368 veterans hospitalized in U.S. Veterans Affairs (V.A.) Hospitals with COVID-19. Studied patients (17 women were not analyzed because of small numbers) were all male of median age 69, and 64% were African American. Patients received supportive treatment (no HC, 158), hydroxychloroquine (HC, 97), or hydroxychloroquine plus azithromycin (HC+AZ, 113) at the discretion of their care team. Primary endpoints for the analysis were death and the need for mechanical ventilation. Rates of death in the no HC, HC, and HC+AZ groups were 11.4%, 27.8%, and 22.1%, respectively. Rates of mechanical ventilation were 14.1%, 13.3%, and 6.9%, respectively. The authors concluded that their results showed no evidence of benefit for these drugs and that further use should await results of controlled prospective trials. It is important to note that the study was not randomized, and so the different death rates among the treatment groups do not indicate increased elevated risk of complication from drug treatment.
On March 30, 2020, Molina et al. published results of a small, non-randomized study of 11 COVID-19 patients in France that aimed to test the efficacy of the same hydroxychloroquine sulfate and azithromycin treatment that was administered by the group led by Gautret. By Day 5 of treatment, one patient had died and two had been transferred to the ICU. By Day 6, eight of the remaining ten patients still tested positive for the SARS-CoV-2 virus using RT-PCR, a result which stands in stark contrast to the viral clearance reported by Day 6 in the study conducted by Gautret et al. One of the surviving patients in Molina’s study had to discontinue hydroxychloroquine sulfate use because of a prolongation of the QT interval after treatment, from 405 ms before treatment to 460-470 ms after treatment, indicative of cardiac arrhythmia. The authors conclude that the use of hydroxychloroquine sulfate with azithromycin did not result in significant improvement in clinical outcome or in viral clearance for COVID-19 patients.
Another small study conducted in China on just 30 patients with COVID-19 suggests that these treatments may be less promising: five daily doses of 400 mg of hydroxychloroquine sulfate administered alongside standard treatment was found to be no more effective in treating the disease than standard treatment on its own (Chen et al., 2020). This result was evaluated by comparing the median time of hospitalization, the median time for fever reduction, and the results of CT scans evaluating pulmonary symptoms between the treatment and control groups. The researchers also tested the subjects for the presence of SARS-CoV-2 by pharyngeal swab, and found that by Day 7 of the study, 13 of the 15 patients receiving hydroxychloroquine sulfate treatment had no detectable viral RNA and 14 of the 15 patients receiving standard care tested negative for the virus. While these results may seem discouraging, it is again important to note that the study size was very small, which limits any broader conclusions from being drawn. In fact, rather than concluding that hydroxychloroquine treatment is not effective in treatment, the authors conclude that these preliminary results seem more indicative of an overall promising prognosis for all patients with COVID-19.
A randomized study of 62 COVID-19 patients at Renmin Hospital at Wuhan University aimed to evaluate the efficacy of hydroxychloroquine sulfate in the treatment of hospitalized patients with mild COVID-19. A group of 31 patients received standard treatment along with 400 mg doses of hydroxychloroquine sulfate administered daily for five days, while the other 31 received only standard treatment. For those patients with fever at the onset of the study, it took one day less on average for body temperature to return to normal for those receiving hydroxychloroquine sulfate (Chen, Z. et al., 2020). Furthermore, chest CTs of the subjects revealed that patients receiving hydroxychloroquine sulfate showed higher rates of improvement in pneumonia (25 of 31 vs. 17 of 31 in the control group). The authors also report that cough remission time was significantly improved in the group receiving treatment and that only two patients receiving hydroxychloroquine reported mild adverse reactions (rash or headache), while all four patients that progressed to severe disease were in the control group. These results may show some limited promise in the use of the drug, but the authors note that larger-scale randomized studies need to be conducted to draw any broader conclusions.
On July 15, 2020, Horby et al. posted pre-print preliminary results of a multi-center, randomized, controlled clinical trial concerning the outcome of hospitalized COVID-19 patients treated with hydroxychloroquine sulfate. The trial was part of the Randomized Evaluation of COVID-19 Therapy Program (RECOVERY) in the U.K.. A total of 4,716 hospitalized patients were selected for the study; 1,561 were randomly selected for treatment with hydroxychloroquine, and 3,155 received standard care without hydroxychloroquine treatment. Results indicated that patients receiving hydroxychloroquine were slightly less likely to be discharged from the hospital alive and were slightly more likely to require mechanical ventilation or die within 28 days of randomization. There was no significant increase in cardiac arrhythmia associated with hydroxychloroquine use. Approximately 26.8% of patients receiving hydroxychloroquine treatment died within 28 days of study initiation compared to 25.0% of patients in the control group; this difference was not statistically significant however. Use of hydroxychloroquine was associated with a slightly longer median hospitalization time of 16 days compared to a median time of 13 days for those in the control group. Overall, the authors conclude that the use of hydroxychloroquine appears to have no discernible benefit for hospitalized COVID-19 patients. These results persisted when only specific subgroups were considered, such as those based on sex, age, baseline predicted risk, among other factors.
***Mehra et al. (2020) performed a meta-analysis of 96,032 COVID-19 patients across 671 hospitals across the globe to determine how the use of chloroquine phosphate or hydroxychloroquine sulfate with or without a macrolide affected clinical course and outcome. Patients receiving remdesivir treatment were excluded from the study as were patients receiving mechanical ventilation before treatment with chloroquine phosphate or hydroxychloroquine sulfate. For those patients in the treatment group, defined as the group of patients receiving chloroquine or hydroxychloroquine, only patients who began treatment within 48 hours of COVID-19 diagnosis were included in the study. These groups were compared to a control group, i.e. the group of patients not receiving chloroquine phosphate or hydroxychloroquine sulfate. Patients were evaluated for their degree of severity of the disease by their qSOFA score (a score below 1 is indicative of lower severity) and by oxygen saturation at baseline (a level below 94% is indicative of increased severity). After controlling for multiple cofounding factors, including sex, age, and comorbidities, the authors report that patients in the treatment group were associated with greater case fatality rates and with increased ventricular arhythmia during hospitalization, a condition which has been previously linked to chloroquine and hydroxychloroquine use. Table 3.2 lists the number of patients with specific baseline characteristics, the number of patients that developed ventricular arrhythmias, and the number of survivors and non-survivors for each treatment and control groups. Various clinical outcomes for each of these groups are also included.
On September 18, 2020, Axfors et al. released results of an international collaborative meta-analysis of randomized clinical trials that aimed to assess the survival rates of COVID-19 patients treated with hydroxychloroquine sulfate (HCQ) or chloroquine phosphate (CQ). After identifying potential trials that qualified for study, the researchers reached out to principal investigators and identified 26 trials for inclusion in the meta-analysis. Of these 26 trials, 16 had unpublished results, five had published results, and five had results available in pre-print form. The use of HCQ was evaluated in 24 of the trials (n = 7,659), while the use of CQ was evaluated in four of the trials (n = 307). Two of the studies looked at mortality outcomes in patients treated with HCQ vs. placebo or in patients treated with CQ vs. placebo (n = 63). For the 24 HCQ trials, 499 of the 3,020 patients treated with HCQ died, corresponding to a death rate of 16.5%, while 874 of the 4,639 patients in the control groups died, corresponding to a slightly elevated death rate of 18.8%. However, the 95% confidence interval for the odds ratio (OR) of the death rate of patients treated with HCQ to that of patients in the control group showed that there was no conclusive survival benefit from the use of HCQ (95% confidence interval for the OR was 0.99-1.18). For the published results from randomized clinical trials studying HCQ, there was a statistically significant increased harmful effect found in the use of HCQ in patients (95% CI for this OR was 1.07-1.13). Quite curiously, among the unpublished trials, there was no such statistically significant increased harmful effect found (The 95% OR was 0.71-1.30). For the four trials where the use of CQ was investigated, 18 of 160 patients treated with CQ died, corresponding to a death rate of 11%, while 12 of 146 patients in the respective control groups died, corresponding to a death rate of 8%. The 95% odds ratio for the death rates was 0.15-21.3, again showing that CQ had no added benefit for survival rates in COVID-19 patients.
Axfors et al. astutely highlight that for the HCQ trials, the predominant source of data came from the RECOVERY trial (see Results from Clinical Trials), which concluded no benefit for the use of HCQ in COVID-19 patients and instead showed a longer hospitalization period required for patients administered this treatment, as well as a higher risk of mechanical ventilation and death. Other published studies showed similar results. Meanwhile, studies that found no conclusive evidence for increased mortality with the use of HCQ were more likely to go unpublished. Therefore, the meta-analysis highlights an important bias that is observable from only considering the published studies alone. Null studies, which are well known to have a higher likelihood of going unpublished, must be taken into account to accurately assess the mortality risks for the use of any particular drug in the treatment of any disease.
[Concerns over the source of the data presented in this paragraph and table have been recently expressed by the editors of the Lancet, and the conclusions of this work have been called into question (Servick et al., 2020). NB: The aforementioned papers in both the Lancet and NEJM have been withdrawn by the authors as of 6/4/2020.]***
This retraction of two high profile COVID-19 related papers, including the paper from which the data of Table 3.2 were collected, led to careful consideration of what had gone wrong with the generation and authorship of these two papers, their review, and the urgent decision to publish them, followed within days by widespread concern about their veracity and eventual retraction by the journals. This situation has been reviewed by Catherine Offord in The Scientist (1 October, 2020), with a review of the past rather checkered history of Surgisphere as well as Dr. Desai, the lead author of the studies. It appears that an initial intent to deceive followed by inattention to detail, such as review of primary data, by coauthors led to this clinical disaster which led to significant delays of several trials as well as further public doubt about the accuracy of medical science.
Unfortunately the retraction of the Surgisphere studies is far from unique. Retraction Watch (2020) keeps a running tally of retracted papers in a variety of fields, and as of September 16, 2020, had noted 33 retracted papers between February and September 2020, 3 temporarily retracted, and 2 reviewed with expressions of concern. The urgency of the pandemic is not an excuse for careless or self-serving publications. Continuing episodes such as this will only worsen an already worrisome public suspicion of science, with devastating impact (Horton, 2020).
Dexamethasone and Other Corticosteroids
The use of corticosteroids such as prednisone, methylprednisolone, or dexamethasone has also been considered for patients with COVID-19, particularly in the late stages of infection where a pattern of sustained pulmonary inflammation predominates and is often the cause of death. Such drugs are widely used in management of other inflammatory conditions, are available generically at low cost, and have well-established and modest toxicity profiles when used briefly.
Current Clinical Trials
As of September 26, 2020, there were 25 clinical trials studying the effect of dexamethasone in the treatment of COVID-19 registered internationally, three of which are/were being carried out in the U.S. For the U.S. trials, one is in Phase II, one is in Phase III, and one was a retrospective study completed on June 24, 2020.
On June 16, 2020, investigators at Oxford University reported in a press release the preliminary results of the Randomized Evaluation of COVID-19 Therapy (RECOVERY). On July 17, 2020 these results were published in the New England Journal of Medicine. In this trial more than 11,500 patients with COVID-19 from 175 NHS hospitals in the U.K. were enrolled and randomized to a number of therapies. Accrual to the Dexamethasone arm (6 mg daily x 10 days given orally or by intravenous injection) was halted on June 8, 2020 by the trial steering committee because of evidence of benefit compared with patients receiving usual care. At this time 2,104 patients had been randomized to dexamethasone and 4,321 to usual care. In patients receiving usual care, 28-day mortality was 41% for patients requiring ventilation, 25% for patients requiring supplemental oxygen but not ventilation, and 13% for those requiring no ventilatory assistance. These rates were reduced in the dexamethasone group by one third (HR 0.65, CI 0.48-0.88, p = 0.0003) for the group requiring mechanical ventilation and by about one fifth (HR 0.80, CI 0.67-0.95, p = 0.0021) for patients on supplemental oxygen but not mechanical ventilation. Benefit was not seen for patients not requiring respiratory support (p = 0.14). The authors note that this would result in preventing one death for treatment of 8 patients on ventilators or 25 patients on oxygen.
These reported results with a reduction not merely in time to recovery, as has been reported with other agents such as Remdesivir, but in 28-day mortality, are encouraging. We await full details of the trial including patient population and management of the standard care group, as their 28-day mortality rates seem rather high. The ready availability of Dexamethasone, its well-defined and modest toxicities, and its low cost (30 4 mg tablets currently priced in the $15-25 USD) make it an appealing candidate if these results are substantiated and confirmed. It should also be noted that there are no data that Dexamethasone would be useful as a preventive agent or one for treating infected but asymptomatic patients, and more chronic use can result in toxicities including hyperglycemia, hypertension, aseptic necrosis of bone. Routine use of this agent is not indicated.
On September 2, 2020, Sterne reported for the WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group results of a prospective meta-analysis pooling data from 7 randomized trials of corticosteroids (e.g. dexamethasone, hydrocortisone, or methylprednisolone) versus usual care or placebo in critically ill patients with COVID-19 (Sterne, 2020). The primary endpoint of the meta-analysis was 28-day all-cause mortality. Of 1,703 patients analyzed, 57% had been participants in the RECOVERY trial, the others came from 6 other trials. Dexamethasone was the most commonly used corticosteroid (1,282 patients) followed by hydrocortisone (374 patients) and methylprednisolone (47 patients). The summary OR was 0.66 (95% CI 0.63-0.82; P<0.001) for all-cause mortality favoring the administration of corticosteroids versus usual care or placebo. There was no indication of an increase in adverse events associated with these corticosteroid regimens. The magnitude of the benefits were similar for dexamethasone and hydrocortisone. These data provide further support for the use of corticosteroids in critically ill patients with COVID-19.
Famotidine (the active compound in Pepcid®) is a generic medication that is used to treat peptic ulcer disease and gastroesophageal reflux. It acts as an inhibitor of H2 receptors in parietal cells, blocking histamine from binding to the receptor, which would otherwise activate proton pumps to produce stomach acid. A Phase III clinical trial to study the potential of famotidine to treat COVID-19 was initiated on April 7, 2019, and its enrollment is expected to reach 1,170. The randomized, double-blind comparative trial will test the efficacy of treatment with hydroxychloroquine sulfate and intravenous famotidine (10 mg/mL concentration) as compared to treatment with hydroxychloroquine sulfate and placebo. The total daily dosage for patients in the treatment group will be 360 mg of famotidine per day for a maximum of 14 days.
Callahan, an infectious disease expert and physician, first called attention to the drug’s potential use as a treatment after an extensive review of Chinese medical records of COVID-19 patients, which showed that patients on the drug when compared to patients treating their heartburn with proton pump inhibitors, such as Omeprazole, were dying at a substantially lower rate. Freedberg et al. (2020) later conducted a retrospective study of 1,620 COVID-19 patients, of which 84 (5.1%) had received famotidine within 24 hours of admission. Use of famotidine was associated with reduced risk of death or intubation and with reduced risk of death specifically. Proton pump inhibitors, such as Omeprazole, did not exhibit the same association, however. Modeling of the SARS-CoV-2 papain-like protease, which enables viral replication, shows that famotidine may effectively bind to the protease, thereby inhibiting its action in viral replication (Borrell, 2020).
Favipiravir (also known as Avigan, T-705, or Favilavir) is an antiviral drug produced in Japan by Toyama Chemical. The drug was originally developed for the treatment of influenza. The drug is an analog of the nitrogenous base guanine, and it has shown broad spectrum antiviral activity against a wide range of RNA viruses (Furuta et al., 2009). In vitro studies have revealed that the compound increases both the rate of specific nucleotide transversion mutations action in the influenza A H1N1 genome and the overall mutation frequency in the virus’s genome (Baranovich et al., 2013). Favipiravir has shown efficacy as an inhibitor of viral replication (possibly by inhibiting the action of RNA-Dependent RNA Polymerase, an enzyme required for the replication of an RNA-virus) in Ebola virus infection in a mouse model of the disease, where it reduced viral load, improved disease condition, and increased recovery rates to 100% for infected mice (Oestereich et al., 2014).
Two Chinese clinical trials to test for the efficacy of Favipiravir in treating COVID-19 were registered in early February (De Clerq, 2020). At a press conference on March 17, 2020, Zhang Xinmin, director of the National Center for Biotechnology Development in China, announced the successful results of two Favipiravir drug trials, one on 240 COVID-19 patients in Wuhan, China and the other on an 80 COVID-19 patient cohort in Shenzhen. Xinmin reported that in the Shenzhen trial, patients treated with Favipiravir (treatment group) tested negative after a median of 4 days after testing positive, while those treated with Lopinavir/Ritonavir (control group) tested negative after a median of 11 days. Furthermore, 72% of patients in the treatment group had fevers gone within 2 days compared to 26% of patients with the same fever recovery time in the control group. Chest X-ray revealed that 91% of patients in the treatment group showed improved lung conditions, compared to 62% in the control group. Xinmin recommended the use of the drug in treatment of COVID-19 patients. The Chinese government has granted approval to a Chinese firm to mass produce the drug for use in treatment against COVID-19 disease. These preliminary results must still be followed up with further trials and peer-review to verify the treatment efficacy of the drug.
Interleukin-6 Receptor Antagonists
Interleukin-6 receptor antagonists (tocilizumab, sarilumab, siltuximab) have been proposed as ways of treating the cytokine release syndromes, CRS (or cytokine storm syndrome, CSS) often seen in patients with severe COVID -19 infection (Moore, 2020). These agents were initially introduced and approved for treatment of rheumatologic conditions. Tocilizumab has been used successfully in managing the CRS which may be seen following CAR-T therapy. Xu (2020) reported experience in treating 21 patients with severe or critical COVID-19 infection with standard therapy (lopinavir, methylprednisolone, other symptom relievers and oxygen plus tocilizumab 400 mg in a single dose. They reported marked clinical improvement in temperature elevation and SaO2 within the first day, decline of previously elevated C-reactive protein, and improvement of chest CT scans. Nineteen of the 21 patients were successfully discharged from the hospital. This was a small and uncontrolled trial but the results in this high risk population are intriguing. A number of prospective randomized trials of IL-6 blocking agents (both tocilizumab and sarilumab) have been designed and should begin accrual shortly (Goldberg 2020). In centers also conducting treatment of patients with cancers with CAR-T therapy, careful consideration should be given to the use of tocilizumab both for CAR-T related CRS as well as for COVID-19 to ensure adequate supplies.
Somers et al. (2020) have reported in a non-peer-reviewed preprint in medRix their experience at the University of Michigan using tocilizumab (8 mg/kg) for treatment of mechanically-ventilated patients with COVID-19. This was a large observational study of 154 patients not eligible for other concurrent trials underway at the institution. Of these, 78 patients received tocilizumab, and 76 did not at physician discretion. Baseline clinical characteristics were similar although the tocilizumab patients were younger, less likely to have COPD, and had lower D-dimer levels at the time of intubation. In propensity score weighted models the tocilizumab treated patients had a lower hazard of death (HR 0.55 (95% CI 0.33-0.90)) and improved status on an ordinal outcome scale , with odds ratio of a 1-level increase 0.59 (CI 0.36-0.95). Tocilizumab treatment was associated with an increased proportion of superinfections (54% vs. 26% p < 0.001) but this did not appear to impact 28-day mortality. While observational, these data await the results of currently on-going randomized trials for confirmation. They may also prompt trials of the use of tocilizumab earlier in the course of the disease.
Lopinavir and Ritonavir are antiretrovirals that act as HIV-protease inhibitors. They are often administered in conjunction (higher dose Lopinavir with a lower dose of Ritonavir) as a therapy in HIV infection. The drug combination has been used to treat patients in South Korea and China with COVID-19 infection, where doctors have preliminarily observed that the treatment may be associated with decreased viral load and improved prognosis in patients. Rigorous clinical trials are currently underway to test the efficacy of these medications in treating COVID-19.
A randomized, controlled trial involving 199 severe COVID-19 hospitalized patients in China found that treatment with lopinavir-ritonavir resulted in no significant change in time to improvement of clinical symptoms but some improvement in mortality (Cao et al., 2020). Overall, median time to improvement was one day less for patients treated with the antiretrovirals than for patients assigned to standard treatment. Mortality at 28 days was slightly improved in the lopinavir-ritonavir treatment group (19.8% vs. 25.0% for standard care), and while adverse gastrointestinal events were more common in the antiretroviral treatment group, serious adverse effects were more common in patients receiving standard care (Cao et al., 2020).
On June 29, 2020, pre-print, preliminary results from the COVID-19 lopinavir/ritonavir RECOVERY trial were released by its chief investigators. The lopinavir/ritonavir arm of the trial had enrolled 1,596 COVID-19 patients, who were compared to a group of 3,376 patients receiving standard care. The preliminary results indicate that there was no significant difference in 28-day mortality, which was 22.1% for the group receiving lopinavir/ritonavir and 21.3% for the group receiving standard care. We await published results which detail more of the data collected, but the authors concluded no clinical benefit for COVID-19 patients receiving the treatment.
Remdesivir (RDV; GS-5734™) is a single diastereomer monophosphoramidate prodrug designed for the intracellular delivery of a modified adenine nucleoside analog. Intracellularly, this prodrug is converted to its triphosphate active form, an adenine nucleotide analog that inhibits the viral RdRp (RNA Dependent RNA Polymerase). That is, the medication acts by inhibiting the viral RNA replication from its template strand, preventing viral replication. Remdesivir was developed by Gilead Sciences, Inc. initially to treat the Ebola virus infection, but over the past few years it has been reported to have antiviral activity against other filoviruses such as Marburg virus, several coronaviruses as well as certain pneumoviruses and paramyxoviruses. The FDA granted Remdesivir emergency use authorization on May 1, 2020. This announcement shortly followed the release of results from a study that showed a decreased time to recovery associated with the use of Remdesivir but it did not show a significant mortality benefit. We refer the reader to the following published reports: [doi:10.1021/acs.jmedchem.6b01594, 10.1038/nature17180, 10.3390/v11040326, 10.1073/pnas.1922083117][doi:10.1038/s41467-019-13940-6, 10.1074/jbc.AC120.013056, 10.1128/mBio.00221-18, 10.1016/j.antiviral.2019.104541], [doi:10.1038/srep43395].
Remdesivir has shown to inhibit SARS-CoV-1 viral replication and MERS-CoV replication in human lung epithelial cell culture (Sheahan et al., 2017). In vivo studies in a mouse model of SARS-CoV-1 infection also revealed that treatment with Remdesivir was associated with a significant reduction in viral load, as well as improved respiratory conditions and symptoms associated with SARS disease progression (Sheahan et al., 2017). The drug is currently undergoing several clinical trials to test its efficacy in treating patients with COVID-19 infection (see Current and Proposed Clinical Trials).
The Expanded Access Remdesivir is a clinical study funded by the U.S. Army Medical Research and Development Command. The population includes people with similar characteristics as the ACTT. Expanded access or “compassionate use” studies are intended to provide access outside of clinical trials to people with life-threatening conditions to a new investigational medical product that has not been approved by the FDA (www.clinicaltrials.gov Identifier: NCT04302766).
Since January 25, 2020, Gilead Sciences, Inc. has been accepting requests for the compassionate use of Remdesivir, which is only administered to patients with SARS-CoV-2 infection (confirmed by RT-PCR) and either an oxygen saturation level below 94% while breathing ambient air or in need of oxygen support. Treatment for approved cases included a first day dose of 200 mg of Remdesivir, followed by 100 mg daily doses of Remdesivir for the following 9 days. Grein et al. (2020) compiled results from 53 patients who underwent at least one dose of Remdesivir between January 25 and March 7, 2020 who were approved for compassionate use and were subsequently evaluated for progression of symptoms. There were originally 61 in the study, but 8 were not considered because they had missing information in clinical follow-up. The total number of patients requesting compassionate use of Remdesivir during this period is not specified. Of the 53 patients, 40 received the full 10 day treatment, 10 received the treatment for 6 of 10 days, and 3 received fewer than 6 days of treatment. At the initiation of treatment, 30 were receiving mechanical ventilation, and 4 were receiving ECMO. Patients’ clinical symptoms were reevaluated anywhere between 13 and 23 days after the initial dose of Remdesivir. These follow-ups resulted in 36 of the 53 patients demonstrating improvement in the category of oxygen support needed, while 8 of the 53 patients showed worsening. All of the patients breathing ambient air or receiving low-flow supplemental oxygen showed improvement. Moreover, 17 of the 30 patients receiving mechanical ventilation support were extubated, and 3 of 4 of the patients receiving ECMO no longer needed it. Seven of the 53 patients died, including 6 who were receiving invasive ventilation (mechanical ventilation or ECMO). Mortality risk was higher in patients over the age of 70 and with higher serum creatinine at the beginning of the study. Four patients had to discontinue Remdesivir early because of adverse reactions, which may have resulted from the use of the drug. The small number of subjects, limited followup, lack of pre-specified endpoints, and lack of a control group all limit any extrapolation of these results to the general population of severe COVID-19 patients. It should also be noted that the study was funded by Gilead Sciences, Inc., who also authored the initial draft of the report. However, the authors contend that when this group is compared to other cohorts studied during the same time period, the results indicate that Remdesivir may improve the clinical course of COVID-19 patients.
Preliminary results from some of the first COVID-19 patients recruited by the University of Chicago in a Phase III clinical trial were released on April 16 in a video recording obtained by STAT Medicine (NB: Not in pre-print form). Of the 125 recruited patients (113 with severe disease), all of whom were receiving daily infusions of Remdesivir, most were discharged from the hospital within six days, and only two had succumbed to the disease. The Phase III trial is a multi institutional effort that is expected to recruit 6,000 patients to assess the safety and antiviral activity of the drug, and additional results on the first 400 patients recruited in the Phase III trial are expected to be released imminently. Mullane, the infectious disease specialist overseeing the trial at the University of Chicago Hospital, exercised caution concerning these optimistic results, stating that the trial on severe patients had no control (placebo) group for comparison. She also noted that fevers fell quickly for the subjects treated and that some patients were taken off ventilators within a single day of treatment.
Preliminary results of this trial were released by Gilead Sciences, Inc. and by the National Institute for Allergy and Infectious Disease (NIAID) on April 29, 2020. On independent review by the Data Safety Monitoring Board (DSMB), Remdesivir was found to be significantly superior to placebo in time to recovery (being well enough for hospital discharge or returning to normal activity) (median 11 vs. 15 days, p < 0.001) and trended better for mortality (8.0% vs. 11.6%, p = 0.059). Full details, including breakdown of these differences by age, comorbid status, and severity of illness, as well as toxicities of the two arms, await publication. This is an encouraging signal, and is likely to lead to accelerated FDA approval of Remdesivir, but a more complete understanding of the role of Remdesivir should be based on consideration of all relevant trials.
On April 29, 2020, Wang, Y. et al. published results of a randomized, double-blind, placebo-controlled multicenter trial conducted in China that tested the efficacy of Remdesivir use in COVID-19 patients. Only 237 patients were recruited into the study, which was conducted between February 6, 2020 and March 12, 2020, during a time when China was experiencing a decline in reported cases. These data showed that of 158 patients randomized to Remdesivir and 79 control, there was no significant difference in time to clinical improvement (21 days for patients receiving Remdesivir and 23 days for the placebo group), mortality at 28 days (14% for patients receiving Remdesivir and 13% for patients in the placebo group), or time to virologic clearance. These results do not indicate a benefit for Remdesivir at these doses in this population of patients.
The results of the Adaptive COVID-19 Treatment Trial (ATTC-1) were published in peer-reviewed form in the New England Journal of Medicine on May 22, 2020 (Beigel et al. 2020). This was a randomized, placebo-controlled, double-blind study comparing intravenous remdesivir with placebo. Planned dose was 200 mg of remdesivir on Day 1 followed by 100 mg on Days 2-10 or until hospital discharge or death. Other treatments for patients were as per hospital policy, except that experimental or off label use of medications to treat COVID-19 were not permitted during the period of remdesivir administration or for two additional weeks, although they were allowed prior to enrollment in the trial. The initial primary endpoint of the trial was to compare the change of the two groups on an eight-category ordinal score of disease severity on Day 15 of the trial. This was changed on April 2, 2020 at a time when only 72 patients had been enrolled on the trial, to a comparison of time to recovery up to Day 29, as it became more evident that patients with COVID-19 infections often ran a longer course. The recommendation to change the primary endpoint was made by the trial statisticians who were not aware of treatment assignment or outcome of patients.
Inclusion criteria for the trial required evidence of SARS-CoV-2 infection by a positive RT-PCR test and symptoms of lower respiratory tract infection is indicated by radiographic infiltrates, SpO2 < 94% on room air, or requiring supplemental oxygen, mechanical ventilation, or extracorporeal membrane oxygenation. Exclusion criteria included elevation of ALT or AST >5 times ULN, impaired renal function or need for hemodialysis or hemofiltration, allergy, pregnancy, or anticipated discharge from the hospital within 72 hours from enrollment.
Enrollment was begun on February 21, 2020. Patients were stratified by location site and disease severity (moderate or severe) and randomized 1:1 between remdesivir and placebo. On April 27, 2020, the data safety and monitoring board reviewed results (a planned interim analysis) and recommended that the preliminary primary analysis report and mortality data be communicated to the trial team members from the National Institute of Allergy and Infectious Disease (NIAID). These initial results on time to recovery and mortality were then made public by the NIAID as well as in a press release by Gilead, the manufacturer of remdesivir.
A total of 1,107 patients were assessed for eligibility, of whom 1,063 were enrolled and randomized between remdesivir (n = 541) or placebo (n = 522). Demographic and disease characteristics appear well-balanced between the two groups. Analysis of the new primary outcome showed that the time to recovery was shorter in the remdesivir than the placebo group, 11 days vs. 15 days (RR 1.32, 95% CI 1.12-1.55; p < 0.001). Analysis of the initial primary endpoint (improvement in the ordinal disease score at Day 15) were also improved in the remdesivir group (ratio for improvement 1.50; 95% CI 1.18-1.91; p = 0.001). Mortality was less in the remdesivir group than in the control group, but this difference was not statistically significant (hazard ratio for death 0.70; 95% CI 0.47 to 1.04). Kaplan-Meir estimates for 14 day mortality were 7.1% and 11.9%, respectively. Serious adverse events were reported in 21.1% of the patients in the remdesivir group and 27.0% of the placebo group. The benefit in terms of reduction of days to recovery was most pronounced in those patients initially on supplemental oxygen (the largest subgroup of patients) as compared with those not requiring O2 or those requiring high-flow O2, non-invasive mechanical ventilation, intubation, or ECMO. This may have been due to the larger sample size of this group, as a test for interaction between baseline status and treatment benefit was not significant.
This study provides reasonable data for the safety of remdesivir and its ability to provide a modest benefit in time to recovery of patients with at least moderately severe COVID-19. It suggests that earlier use of remdesivir, before patients require mechanical ventilation, is better than waiting. It did not address nor should it be used to support the use of remdesivir in patients with earlier stages of disease, as a preventive strategy, or in patients known to be infected but not yet symptomatic who have been shown to be capable of spreading virus to others. Finally, this was an early publication of important data but without full follow-up of all entered patients and the authors acknowledge that full understanding of this trial will await such analysis.
Data is still emerging on Remdesivir from clinical trials that still offer challenges in interpretation because of the small size of patient samples or small numbers of randomized, double blind studies. It is therefore noteworthy that another nucleoside analog inhibitor, the orally bioavailable EIDD-2801, shows promise in animal studies for reducing virus titer, and improving lung function and weight control (Sheahan et al., 2020).
Other Pharmaceutical Drugs
Camostat mesylate is a serine protease inhibitor that acts to inhibit TMPRSS2, the serine protease that must modify the S protein of SARS-CoV-2 so that the virus can bind to ACE2 receptors and enter the cell. In vitro, the protease inhibitor has reduced the amount of SARs-CoV-2 infection of Calu-3 lung cells (Hoffman et al., 2020). It is a medication approved for the treatment of pancreatitis in Japan and is currently being tested on mice infected with SARS-CoV-2.
Work from the Baric Laboratory at the University of North Carolina (Sheahan et al., 2020) describes a very promising prophylactic and therapeutic oral antiviral candidate drug (a bioavailable ribonucleoside analog, β-D-N4-hydroxycytidine, NHC, EIDD-1931) that shows early efficacy in infected mice with reduce viral titers and improved pulmonary function.
The Netherlands, Australia, Greece, and U.K. will initiate clinical trials to test the use of the BCG vaccine as a treatment against COVID-19 infection, specifically in high-risk groups and in healthcare workers (de Vrieze, 2020). The vaccine was originally developed to prevent tuberculosis, a bacterial infection. However, over time, various researchers have observed that the vaccine may be effective against a broader range of diseases, suggesting that the vaccine may enhance innate and adaptive immunity. In 2018, a research team led by Netea was able to show that administration of the BCG vaccine was protective against infection with yellow fever, a viral disease. Furthermore, they found that monocytes, immune cells that produce antibodies that are part of the adaptive immune system, were specifically targeting the virus. They also found that administration of the vaccine enhanced production of interleukin-1β, a cytokine the researchers theorized may have a wider epigenetic effect that may modulate the monocytes. The results suggest that the vaccine may be protective against a broad range of viruses through this proposed mechanism.
Ivermectin is a drug used to treat various parasitic diseases and infestations. It is a relatively safe and inexpensive drug that is approved for human use by the FDA, and it has been demonstrated to inhibit viral replication, including for HIV-1, as well as for several RNA viruses. Caly et al. (2020) show that it also has potential to be repurposed as an antiviral agent in the treatment of SARS-CoV-2 infection. After 24 hours of treating a cell culture of Vero/hSLAM cells infected with SARS-CoV-2 with a single dose of 5µM ivermectin, the researchers tested for the presence of SARS-CoV-2 RNA using RT-PCR. When compared to the infected cells treated with DMSO (control condition), there was a 99.8% decrease in viral RNA detected. After 48 hours of treatment, the amount of SARS-CoV-2 RNA had decreased 5000-fold when compared to the cells in the control condition. This study demonstrates that the drug is a powerful SARS-CoV-2 inhibitor in vitro. Caly et al. (2020) propose that this inhibition may be explained by the drug’s possible inhibition of IMPɑ/β1-mediated translocation of viral proteins into the cell nucleus through the nuclear pore complex.
Acalabrutinib (ACP-196) is a bruton tyrosine kinase (BTK) inhibitor which has been approved by the FDA for treatment of patients with mantle cell lymphoma which has progressed after prior therapy or for patients with CLL/SLL and is marketed as CalquenceTM by AstraZeneca. In vitro studies and early clinical data indicated that the decrease in inflammation produced through BTK inhibition may reduce the severity of inflammation and COVID-19 associated respiratory distress. AstraZeneca will be initiating a global randomized Phase II trial (CALAVI) comparing best supportive care with or without acalabrutinib. This is scheduled to begin accrual in late April 2020 and accrue 428 patients.
Roschewski et al. (2020) released results in Sci. Immunol. of an off-label trial of acalabrutinib in 19 patients with COVID-19 who were hospitalized with evidence of hypoxemia (SpO2 94% or less on room air) and inflammation (CRP >10 mg/dl and/or ferritin > 500 ng/ml and/or lymphopenia (ALC <1000 cells/microliter.)) Of these, 11 patients were receiving supplemental oxygen, and 8 were receiving mechanical ventilation. Acalabrutinib was given orally or by enteric feeding tube twice-a-day for 10 days to patients on supplemental oxygen and for 14 days for patients on mechanical ventilation. Improvement on oxygen requirements and clinical status was seen in both the supplemental oxygen and intubated groups of patients, as were normalization of measures of inflammation such as CRP, elevation of IL-6 levels, and lymphopenia. At the completion of acalabrutinib treatment 8 of 11 (72.7%) of the patients in the supplemental oxygen cohort had been discharged on room air, and 4 of 8 (50%) of the intubated patients had been successfully extubated. No toxicities attributable to acalabrutinib were reported, and cardiac arrhythmias, grade 3 or higher bleeding, diarrhea, or opportunistic infections were not observed. While this is a small uncontrolled trial, the results are encouraging and support enrollment of patients in the aforementioned randomized trial.
Baricitinib (Olumiant®) is a drug approved for use in the U.S. as a treatment against moderate to severe rheumatoid arthritis, specifically for those patients that have not been able to manage their illness with tumor necrosis factor (TNF) antagonists. It acts as a janus kinase (JAK) inhibitor, inhibiting JAK1 or JAK2 in particular. By blocking the action of janus kinases, baricitinib interferes with the JAK-STAT signalling pathway, which in turn reduces cytokine signalling. With the assistance of BenevolentAI, Richardson et al. (2020) searched for drugs that might inhibit SARS-CoV-2 infection, and baricitinib was identified as a possible therapy that could hinder SARS-CoV-2 from infecting AT2 alveolar cells via endocytosis, which is mediated by AP2-Associated Protein Kinase 1 (AAPK1). Baricitinib was identified as one of six AAK1 inhibitors that antagonistically binds to the kinase with high affinity. It also can bind to the cyclin G-associated kinase, which also regulates endocytosis. The group singled out baricitinib as a relatively safe drug that has both antiviral and anti-inflammatory potential for the treatment of COVID-19. On May 8, 2020, NIAID announced that a combination treatment of remdesivir and baricitinib would undergo a randomized, double-blind, controlled clinical trial to evaluate its safety and efficacy in the treatment of severe COVID-19. More than 1,000 participants are expected to be enrolled in the trial. It is considered an arm of the NIAID’s Adaptive COVID-19 Treatment Trial.
Additional drugs for potential treatments in various stages of clinical trials include, but are not limited to, the following:
- Interferon alpha-2B (cytokine treatment for viral infection and cancers)
- Mepolizumab (anti-CD147 humanized antibody undergoing clinical trial in China)
- Methylprednisolone (corticosteroid)
- Tissue Plasminogen Activator (serine protease that catalyzes the conversion of plasminogen to plasmin used to break up blood clots, such as those arising from strokes)
- Umifenovir (Arbidol)
Current Vaccine Candidates
There are several aims for a successful vaccine candidate. Firstly, and perhaps most importantly, the vaccine should allow for an individual exposed to SARS-CoV-2 to generate a robust adaptive immune response upon exposure. In particular, exposure to specific regions of the virion should trigger a potent neutralizing antibody response from the host. It should also trigger a response from immunoglobulin antibodies, which can activate a pathway that culminates in the destruction of the virion itself. Furthermore, the vaccine should generate an adaptive response that is lasting in the individual, so that the vaccinated individual remains immune to the virus for a prolonged duration.
Others have questioned the adequacy of initial antibody response as an adequate surrogate for vaccine efficacy and duration of action (Hellerstein M, 2020). SARS-CoV-2 infection can result in the development of CD4 and CD8 T-cell responses to multiple epitopes in addition to the spike protein, with significant responses to membrane and nucleocapsid proteins. It will be important to assess the intensity and durability of the protective immunity induced by these various epitopes in comparing the results of natural infection and vaccines, those limited to a single protein (typically spike) as the immunogen. A vaccine which induces a strong but transient immune response is unlikely to be a clinical success and may worsen the public acceptance of vaccines in general.
Of course, assessing the safety of vaccine use is a chief concern, which is why adverse reactions should be closely studied and monitored during clinical trials. Of particular concern in the case of potential vaccinations is antibody-dependent enhancement (ADE). ADE occurs when antibodies specific to the virus can enhance the ability of a virion to gain entrance into a host cell. While ADE primarily occurs in cell cultures in vitro, it has been shown to occur in vivo in infections with the Dengue virus, feline coronaviruses such as SARS (suggesting that past coronaviral infections may also be a risk factor in severity of COVID-19), as well as HIV, where patients who are reinfected have worsened severity of symptoms. Because of ADE, effective vaccination, which generally leads to the production of antibodies, may in some cases also lead to increased severity of a viral infection. Thus, human trials are paramount for ensuring the safety of using such vaccines.
In mid-May 2020, the U.S. federal government unveiled a public-private partnership entitled Operation Warp Speed, which aims to accelerate the manufacturing, delivery, and development of COVID-19 vaccines as well as other therapeutics and diagnostics, with a stated goal of delivering 300 million doses of safe, effective COVID-19 vaccines by January 2021. The partnership has a budget of 10 billion USD and is a collaborative effort between the Center for Disease Control, the Food and Drug Administration, the National Institute of Health, the Biomedical Advanced Research and Development Authority, the Department of Defense, among others. As of June 2020, seven biomedical firms had been selected and prioritized for funding the development of their vaccine candidates. These firms include Johnson and Johnson, AstraZeneca-University of Oxford, Pfizer-BioNTech, Moderna, Merck, Vaxart, and Inovio. Of these, there are at least two promising vaccine candidates for COVID-19 being produced in the U.S. that are in clinical trials as of July 4, 2020: INO 4800, a DNA-based vaccine that codes for the SARS-CoV-2 spike protein and mRNA-1273, an mRNA vaccine that codes for the full-length spike protein of SARS-CoV-2. Other U.S.-produced vaccine candidates that have received funding from the project, such as the oral vaccine candidate produced by Vaxart, are in preclinical stages and are expected to enter clinical trials by late-summer 2020.
As of August 2020, there were a variety of different methods tested for vaccine delivery. Genetic vaccines, which include the vaccine candidates mRNA-1273, INO-4800, and BNT162, deliver mRNA or DNA that encode a protein of the SARS-CoV-2 virion that is known to trigger an immune response. Since previous studies have identified regions of the spike protein (particularly its receptor binding domain and the N-terminal portion of the protein) as potent regions of the SARS-CoV-2 epitope that trigger a robust antibody response (see SARS-CoV-2 Epitope), the genetic material delivered through these vaccines often codes for the spike protein itself or a modified version of the protein. Viral vector vaccines, such as ChAdOx1, Ad26-COV2-S, and Ad5-nCoV, deliver a virus that contains specific SARS-CoV-2 genes. These genes encode a protein or protein fragment that will activate an adaptive immune response. Generally, the protein encoded is the spike protein itself or a modified version. Protein-based vaccines, such as PittCoVac and NVX-CoV2373, deliver a SARS-CoV-2 protein or fragment, generally a modified version of the spike protein, directly in an attempt to trigger an adaptive immune response. Finally, some vaccines, such as CoronaVac, aim to deliver a weakened or inactivated version of the SARS-CoV-2 virus itself to trigger an adaptive immune response.
On August 11, 2020, Russia president Vladimir Putin announced that the Russian Health Ministry had officially registered a SARS-CoV-2 vaccine (see Sputnik V), an action tantamount to having received regulatory approval for public use of the vaccine. This approval was met with widespread criticism from the international scientific community as it was granted despite failure to complete any large-scale clinical trials for the vaccine. By that time, only two non-randomized Phase I/II clinical trials had officially been registered, each with 38 healthy participants between the ages of 18 and 60, and no results or preliminary results had yet been released. Despite having limited evidence, Murashko, the Russian Minister of Health, claimed in a press release from the same day that the vaccine was both safe and efficacious, conferring 2 years of immunity to SARS-CoV-2. Cohen of Science Insider reports that the registration certificate allows for the vaccine to be used on “a small number of citizens from vulnerable groups,” including medical staff and the elderly. The certificate also specifies that the vaccine cannot be used widely until January 1, 2021.
RNA Vaccine Candidates
The mRNA-1273 vaccine encodes the SARS-CoV-2 spike glycoprotein along with a transmembrane anchor and an intact S1-S2 cleavage site, together known as the S-2P antigen. The name for the antigen comes from its unique conformation, which is stabilized by two consecutive proline residues substituted in the central helix of the S2 subunit of the spike protein. The vaccine was developed by Moderna Therapeutics, a biotechnology company focused on vaccine delivery through messenger RNA (mRNA). To effectively deliver the nucleic acid contents, the mRNA that encodes the antigen is enveloped in a lipid nanoparticle capsule. The vaccine candidate is stable for 6 months at a temperature of -4℉ (-20C) and can be stored for 30 days with standard refrigeration temperatures. As of December 5, 2020, there are four clinical trials underway to test the efficacy, safety, and immunogenicity of the vaccine. All four trials will take place in the U.S.: one is a Phase I trial, one is in Phase II, one is in Phase III, and one is both a Phase II and III trial. On December 18, 2020, the FDA authorized the emergency use of the vaccine candidate. The following day, the CDC’s Advisory Committee on Immunization Practices voted unanimously to recommend mRNA-1273. On Wednesday, December 23, Health Canada officially approved the use of mRNA-1273 for the prevention of COVID-19, making Canada the first country to completely approve the vaccine candidate for its residents.
Phase I U.S. Trial
The phase I clinical trial of mRNA-1273 (Moderna) began at the Kaiser Permanente Washington Health Research Institute on March 16, 2020. The study originally enrolled a total of 80 healthy human subjects between the ages of 18 and 55, who would receive two intramuscular injections of either the vaccine or placebo on days 1 and 29 of the trial. Five different doses were tested (10 μg, 25 μg, 50 μg, 100 μg, and 250 μg) and the subjects will be evaluated regularly over the course of 12 months after the administration of the second injection. These evaluations are meant to test both the safety of the vaccine and immunogenicity of the vaccine (i.e. whether exposure to the antigen will trigger an immune response), the latter of which will be evaluated using immunoassay methods, namely IgG ELISA, which tests for the presence of IgG antibodies. In April, 2020, the trial was expanded to allow for 40 older adults above the age of 55 to participate.
One July 14, 2020, Jackson et al. reported initial results of the phase I trial that were published in the New England Journal of Medicine. Anti S-2P IgG antibody response was positively correlated with dosage: the geometric mean of immunosorbent assay antibody titers on Day 29 (28 days after the first dose) were 40,227 for the subjects receiving 25 μg dose, 109,209 for those receiving the 100 μg dose, and 213,526 for those receiving the 250 μg dose. On Day 57 (28 days after the second vaccination), the geometric means of the antibody titers had increased further: 299,751 for the group receiving the 25 μg dose, 782,719 for those receiving the 100 μg dose, and 1,192,154 for those receiving the 250 μg dose. Furthermore, all subjects had detectable IgG antibodies by two weeks after the first injection. Neutralizing antibodies were also assessed and found to be robust and positively correlated with dosage. The authors note that both binding and neutralizing antibody responses produced after the first injection were comparable to those found in convalescent patients. After the second injection, however, the median geometric mean antibody titer was in the upper quartile of corresponding antibody titers found in convalescent patients. Given the wide variation in antibody response in convalescent patients, this result supports the use of a two-dose administration of the vaccine. Subjects will continue to be tested for antibody response over the next year to assess the longevity and strength of such responses over time.
The trial also sought to assess adverse effects from the two-dose vaccine administration as well as how these effects may vary between dosage groups. More than half of the subjects experienced fatigue, chills, headache, muscle pain, and pain at the injection site, and systemic adverse effects were more common after the second dose, more pronouncedly so for the group receiving the highest dosage. No subjects experienced fever after the first injection, while 40% receiving the 100 μg dose did after the second injection, and 57% of those receiving the 250 μg dose reported this symptom after the second injection. Twenty-one percent of the group receiving the highest dosage experienced one or more serious adverse events. No participant reported any severe adverse effect after the first injection.
On September 29, 2020, Anderson et al. (2020) published results concerning the safety and immunogenicity of the vaccine candidate when administered to the 40 older participants enrolled. These 40 participants were stratified by age: 20 were between the ages of 56 and 70, and the remaining 20 were all 71 or older. All 40 participants received two doses of either 25 μg or 100 μg; one dose was administered on Day 1 and the other on Day 29 of the study. Local and systemic adverse effects were reported up through Day 57 on the study, which these results summarize, but participants will continue to be monitored for adverse events up to a year after the second dose administered. In this report, no serious local adverse events were reported in either age group for either dose level. The most commonly reported adverse events were headache, fatigues, muscle ache, chills, and pain at the injection site, and these effects were more commonly reported after the administration of the second dose and tended to appear on the day of or the day after the administration of a dose of the vaccine candidate. Most resolved quickly, but three patients experienced mild erythema (skin redness) that lasted for 5-7 days. One other participant reported mild muscle aches that appeared on Day 3, and the aches lasted for 5 days. Overall, there were only two systemic adverse effects that were classified as severe: one instance of fever in a participant in the 56-70 year age group after receiving a second 25 μg dose, and one instance of fatigue in a participant in the 71+ year age group after receiving a second 100 μg dose. A total of 71 unsolicited adverse events were reported, of which only 17 were thought to be related to the administration of the vaccine candidate. All but one of these adverse events was mild, but there was a report of one moderate adverse event, which was decreased appetite reported in a participant in the 56-70 year age group receiving the 25 μg dose. The number of adverse effects reports did increase with increasing dose. Overall however, both doses of the vaccine candidate were well tolerated in both age groups.
Binding antibody responses were more robust for the two older age groups than for the younger cohort previously reported on in July, 2020, and again, responses were dose-dependent. On Day 57, the GMT for binding IgG antibodies to S-2P receiving the 25 μg dose was 323,945 in the 56-70 year age group and 1,128,391 in the 71+ age group. For the 100 μg dose, the GMT for IgG antibodies specific to S-2P was even higher: 1,183,066 for the 56-70 year age group and 3,638,522 for the 71+ age group. All GMTs were considerably higher than the 138,901 GMT found in a panel of convalescent blood sera. Neutralizing antibodies were undetectable before administration of the vaccine candidate and steadily rose in a dose-dependent manner, but this time titer was independent of age. Neutralizing antibody responses to the 614D SARS-CoV-2 strain (the initial strain found in Wuhan but no longer the predominant 614G strain observed around the world) stayed at a high level for at least four weeks after the second dose (through the end of the period studied in this report). The researchers also tested for T cell immune responses to spike protein peptide sequences following administration of the vaccine candidate. They reported that participants aged 56-70 receiving the 25 μg dose and both age groups receiving the 100 μg dose showed a strong CD4+ response. CD8+ responses were markedly lower in all groups tested and were only observable at low levels after the second dose of the 100 μg dose administered for both age groups.
Phase II U.S. Trial
A phase II randomized, observer-blind, placebo-controlled trial was initiated on May 28, 2020, which aims to assess the safety and immunogenicity of two different doses of the vaccine in 600 participants over the age of 18. There are four experimental groups: one for adults aged 18-54 receiving a 50 μg dose of mRNA-1273, one for adults over the age of 55 receiving a 50 μg dose of mRNA-1273, one for adults aged 18-54 receiving a 100 μg dose of mRNA-1273, and one for adults over the age of 55 receiving a 100 μg dose of mRNA-1273. Each treatment group will be compared to a controlled group receiving saline placebo in lieu of a vaccine dose. Subjects will be monitored over the course of a year for any adverse effects; they will also be tested for SARS-CoV-2 neutralizing antibody titer during that time period to assess immunogenicity.
Phase II/III U.S. Adolescent Trial
On December 2, 2020, a Phase II/III randomized, observer-blind, placebo-controlled clinical trial to assess the efficacy, reactogenicity, and safety of mRNA-1273 in healthy adolescents between the ages of 12 and 18 was officially registered. It has a projected enrollment of 3,000 participants, who will be divided into two arms of the trial. Volunteers in the experimental group will receive two 100 μg intramuscular injections of the vaccine: one on Day 1 and one on Day 29 of the trial. Those in the placebo group will receive intramuscular injections of 0.9% saline solution on the same days of the trial. The study will track the number of participants with local and systemic adverse reactions up to 7 days after the first dose and up to 7 days after the second dose. Unsolicited and serious adverse events will also be noted, and the number of participants who reach a pre-defined minimum serum antibody level by Day 57 (four weeks after the second dose) will also be tracked. The geometric mean titer (GMT) values of the serum neutralizing antibody at Day 57 will also be compared to the Day 57 GMT of serum neutralizing antibody in the Phase III trial underway on adults (see Phase III U.S. Trial). Secondary outcome measures include recording the GMT values of SARS-CoV-2 Spike protein-specific neutralizing antibody and the SARS-CoV-2 specific neutralizing antibody on Days 1, 57, 209, and 394. The number of participants that develop a SARS-CoV-2 infection from Day 57 - 394 will also be tracked as will the number of participants that develop COVID-19 between Day 29 and Day 394.
Phase III U.S. Trial
The phase III randomized, observer-blind, placebo-controlled clinical trial (known as COVE) assessing the efficacy, immunogenicity, and safety of the vaccine candidate in adults over the age of 18 was initiated on July 27, 2020. It has an estimated enrollment of 30,000 subjects and will be conducted over the course of two years. Subjects are assigned to the experimental or placebo group. All participants in the experimental group will receive two intramuscular injections of 100 μg, one on Day 1 and one on Day 29 of the study. Subjects in the placebo group will receive two intramuscular injections of 0.9% saline solution, one on Day 1 and one on Day 29. The study aims to measure the number of participants that develop COVID-19 starting 14 days after the second injection, the number of patients with adverse effects or reactions, the geometric mean titers and geometric mean fold rise of SARS-CoV-2 specific neutralizing antibodies in participants receiving the mRNA-1273 vaccine, and the geometric mean titers and geometric mean fold rise of S-protein specific binding antibodies.
On November 15, 2020, an NIH-appointed independent data and safety monitoring board overseeing the Phase III trial of mRNA1273 shared preliminary trial data and analysis. Among volunteers enrolled in the study, 95 had developed COVID-19, of which 90 were in the placebo group and 5 were participants that received the vaccine candidate, suggesting a 94.5% efficacy rate, a result which was statistically significant. Out of the 95 that developed COVID-19, only 11 developed a severe form, and all 11 were in the placebo group. On November 30, 2020, Moderna announced in a press release that the vaccine candidate demonstrated 94.1% efficacy. By this point, 196 cases of COVID-19 were confirmed in the participants: 185 were in the placebo group and 11 received two doses of mRNA-1273. No severe cases were reported in the treatment group whereas 30 were reported in the placebo group.
BNT162 is a set of four candidate vaccines developed by BioNTech in collaboration with Pfizer; the four potential vaccines are BNT162a1, BNT162b1, BNT162b2, and BNT162c1, all of which are mRNA-based and encode some portion or all of the SARS-CoV-2 spike protein. In particular, BNT162b1 encodes a trimerized version of the SARS-CoV-2 RBD, while BNT162b2 encodes a prefusion membrane-anchored full-length SARS-CoV-2 spike protein. A limiting factor to its distribution is that the vaccine candidate must be stored at a temperature of -70℃, or -94℉, in order to maintain viability for up to 6 months.
As of November 15, 2020, there are five registered clinical trials testing the vaccine candidate. A Phase I/II clinical study conducted in Germany was approved on April 22, 2020 by the Paul Ehrlich Institute. The trial aims to test the safety and efficacy of four candidates, and a range of doses from 1-100 μg were to be administered to determine optimal dosing regimens. On April April 29, 2020, a Phase I portion of a U.S. trial was initiated, and on July 27, 2020, Pfizer and BioNTech announced an expansion of the study to a Phase II/III study to be conducted in the U.S. and various South American and European countries, aiming to recruit 30,000 volunteers. This number was later expanded to 43,000 volunteers on September 12, 2020, with an enrollment of 43,538 participants reported in a November 9, 2020 press release. In August 2020, China registered a Phase I study of the BNT162b1 vaccine candidate.
On Monday, November 30, 2020, just a few weeks after announcement of promising preliminary Phase III trial results, Pfizer formally applied to the FDA for emergency use authorization of the BNT162b vaccine candidate. On Wednesday, December 2, 2020, the United Kingdom approved the vaccine candidate for emergency use. On December 11, 2020, The U.S. FDA authorized the vaccine candidate for emergency use, the first ever granted to a COVID-19 vaccine candidate in the United States. Pfizer currently has an agreement with the U.S. to supply 100 million doses by March, 2021; all shots are to be free to the public. As of December 21, 2020, three countries have granted the vaccine full approval: Bahrain, Saudi Arabia, and Canada.
Phase I/II/III International Trial
On April 29, 2020, the Phase I portion of a randomized, placebo-controlled, observer-blind Phase I/II/III global trial was launched. The purpose of the first phase was to determine if there was any preferred candidate vaccine to use in further trials and, if so, to determine its optimal dosing regimen. Specifically, the safety and immunogenicity results from the use of the two candidates BNT162b1 and BNT162b2 were compared. Ultimately, BNT162b2 was selected for use in the latter phases of the clinical trial because of a more favorable set of adverse reactions reported.
On July 1, 2020, Mulligan et al. published initial results from the Phase I study on vaccine candidate BNT162b1, a lipid, nucleoside-modified mRNA vaccine that encodes the SARS-CoV-2 spike protein receptor binding domain (RBD). The nucleoside modification involves the usage of 1-methylpseudouridine, which has been shown previously to increase mRNA translation rates in vivo (Kariko et al., 2008). Forty-five participants were randomized and vaccinated: 12 subjects received a dose of 10 μg on Days 0 and 21, 12 received a dose of 30 μg on Days 0 and 21, 12 received a single dose of 100 μg on Day 0, and nine received placebo. Adverse effects, which included chills, fever, muscle pain, among others, increased with dose level and were reported more frequently after second doses. Severe adverse effects were reported in two participants and included elevated fever and sleep disturbances following vaccination. IgG antibody concentrations specific to the RBD, as well as neutralizing SARS-CoV-2 antibody concentrations increased with dose and also increased after a second dose. Furthermore, the geometric mean for neutralizing antibody titers was 1.8-2.8 times greater than that of those found in a group of tested COVID-19 convalescent human sera.
On August 28, 2020, Walsh et al. published further preliminary results of this initial phase, where data supporting the use of BNT162b2 for future phases of the study were highlighted. The study used 13 groups of 15 participants (n = 195) where 12 received the vaccine and 3 received placebo; each group had a unique combination of vaccine used (only BNT162b1 and BNT162b2 were tested), vaccine dose, and age range of participants tested. Doses of either vaccine were 10 μg, 20 μg, or 30 μg, all delivered through two doses, administered 3 weeks apart. Younger participants between the ages of 18 and 55 were randomized and grouped together; they were kept separate from older participants between the age of 65 and 85, who participated in their own groups. Younger participants reported mild to moderate local reactions, mostly pain at the injection site, and reactions were more common after the second dose. Both local reactions were milder in older patients. Local reactions were fairly similar between the groups receiving BNT162b1 and BNT162b2. However, the frequency and severity of systemic adverse reactions were much lower in participants receiving BNT162b2. For younger participants receiving BNT162b1, 75% reported a fever over 38℃ after the second dose of 30 μg. For older participants receiving the same vaccine, reports of fever were fewer, with only 33% reporting fever over 38℃ after the second dose. However, the frequency of patients reporting fever was significantly lower in groups receiving BNT172b2: only 17% of younger participants and 8% of older participants reported a fever over 38℃ after the second dose. Moreover, no severe systemic adverse events were reported in older participants receiving the vaccine. The frequency of local and systemic adverse events increased with increasing dose, were higher after the second dose, and were all temporary. The immunogenic response elicited from vaccination by both vaccine candidates was robust and similar. Generally, responses were lower in older adults. Geometric mean titer of neutralizing antibodies for both vaccine candidates measured in older participants a week after the second dose ranged from 1.1-1.6 the geometric mean titer found in convalescent blood plasma. For younger participants, the geometric mean titer of neutralizing antibodies was 2.8-3.8 times the level found in convalescent blood plasma. Based on these results, latter phases of the trial will proceed in testing the BNT162b2 vaccine at a 30 μg dose level, a recommendation spurred by the dramatic reduction in adverse events reported with the use of BNT162b2 over BNT162b1.
On November 9, 2020, Pfizer and BioNTech announced preliminary results testing the safety and efficacy of BNT162b2 from the Phase III portion of the study in a press release. They reported that for a total of 38,955 healthy participants with no prior SARS-CoV-2 infection who had received two doses of BNT162b2, the vaccine candidate was over 90% effective in preventing COVID-19 disease at Day 28 of the study, 7 days after the administration of a second dose. By November 8, 2020, a total of 94 trial participants had been diagnosed with COVID-19 and no serious adverse effects from the administration of the two doses had been reported in participants. By November 18, 2020, a total of 170 participants had been diagnosed with COVID-19: 162 were in the placebo group vs. 8 participants that received the vaccine, suggesting that the vaccine candidate was 95% effective. Furthermore, BNT162b was 94% effective in preventing COVID-19 in adults over the age of 65. As of November 18, 2020, a total of 43,538 participants have been enrolled in the study. No serious grade 3 events were reported in frequencies over 2% except for 3.8% of participants reporting fatigue and 2.0% percent of participants reporting headaches.
Phase I/II German Trial
On April 23, 2020, an interventional, non-randomized, dose-escalation Phase I/II trial to test the safety and immunogenicity of the BNT162a1, BNT162b1, BNT162b2, and BNT162c1 vaccines was initiated. The trial recruited 456 healthy participants and included two parts: Part A was an initial phase with the purpose of testing dose escalation and de-escalation in participants aged 18-55 and in older participants between the ages of 56 and 85, and Part B was intended to study larger cohorts of participants using dose levels that were determined from the analysis of data collected from Part A.
On Jul 20, 2020, Sahin et al. reported preliminary results from the phase I/II trial clinical trial testing antibody and T-cell responses in healthy adults aged 18-55. A total of 48 subjects were administered two doses of BNT162b1: 12 received doses of 1 µg, 12 received doses of 10 µg, 12 received doses of 30 µg, and 12 received doses of 50 µg. These doses were administered on Day 1 and Day 22 of the study. An additional 12 subjects received one single dose of 60 µg on Day 1 of the study. No serious adverse effects were reported among the subjects. Most adverse events were mild, which included fatigue, headache, and pain at the injection site, while some moderate events included fever, chills, and muscle pain. Concentrations of RBD-specific IgG antibodies and SARS-Cov-2 neutralizing antibodies were measured at the beginning of the study, and on Days 7 and 21, to measure the immunogenic effect of the first dose. For those subjects that received a second dose, antibody concentrations were also measured on Days 29 and 43 of the study. The geometric mean concentrations of RBD-specific IgG antibodies increased with dosage, and on Day 21 of the study were in the range of 265-1672 U/mL. For those patients receiving a second dose, the geometric mean concentrations had increased to 2,015-25,006 U/mL on Day 29, which changed to 3,920-18,289 U/mL by Day 43. For the group that received one single 60 µg dose, the geometric mean concentration of RBD-specific IgG was 1,058 U/mL, indicating a booster dose was more effective at eliciting a robust immunogenic response. These concentrations were all significantly higher than the concentration of the same antibody found in samples taken from 38 convalescent COVID-19 patients, which had a geometric mean concentration of 602 U/mL Geometric mean titers of SARS-CoV-2 neutralizing antibodies were observed to follow similar trends, generating robust immunogenic responses that exceeded those found in convalescent patients. Furthermore, RBD-specific CD4+ T cell responses increased with increasing RBD-specific IgG and SARS-CoV-2 antibody concentrations.
DNA Vaccine Candidates
INO-4800, a DNA-based vaccine candidate that codes for the SARS-CoV-2 spike protein, entered a non-randomized, open-label, phase I clinical trial on April 7, 2020. The trial aims to evaluate the safety and immunogenicity in a small sample of 120 healthy human participants. Participants were divided into three treatment groups with 40 participants each; each group was delivered a different dose of the vaccine: Each participant in Group 1 received one intradermal injection of 1.0 mg of the vaccine on Day 0 and Week 4 of the study, participants in Group 2 received two 1.0 mg intradermal injections on Day 0 and Week 4 (a total dose of 2.0 mg per dosing visit), and participants in Group 3 received one 0.5 mg intradermal injections of Day 0 and Week 4 of the study. Formal results of the trial are expected to be published in July 2020.
Smith et al. (2020) published results demonstrating the INO-4800’s potential efficacy in both mice and guinea pigs receiving the vaccine. In the sera of mice vaccinated, IgG antibodies showed binding to the SARS-CoV-2 spike protein, a response which was considerably diminished in the presence of SARS-CoV-1 spike protein. Mice vaccinated also showed markedly elevated concentrations of SARS-CoV-2 neutralizing antibodies when compared to controls. Similar results were demonstrated in guinea pigs injected with the vaccine candidate. Furthermore, sera from both mice and guinea pigs inoculated with INO-4800 inhibited binding of the SARS-CoV-2 spike protein to the ACE2 receptor.
Viral Vector-Based Vaccine Candidates
On March 16, 2020, CanSino Biologic’s Recombinant Novel Coronavirus Vaccine, Ad5-nCoV (a recombinant adenovirus Type-5 vectored vaccine), which expresses the full-length SARS-CoV-2 spike glycoprotein, became the first Chinese vaccine candidate to receive approval to begin a Phase I clinical trial, becoming one of the first vaccine candidates in the world to be tested on humans. The vaccine uses a replication-defective adenovirus that encodes the SARS-CoV-2 spike protein, which, when transcribed and translated, is intended to generate an immunogenic host response. The Academy of Military Medical Sciences was the sponsor of the Phase I trial, which took place in Wuhan, China. On April 10, 2020, a Phase II trial was registered in China, and after the release of promising preliminary results from these early trials, China’s Central Military Commission approved the use of the vaccine on June 25, 2020 for a period of up to one year. As of September 7, 2020, the vaccine candidate is restricted for use by the Chinese military only.
International trials outside of China have also been pursued during this time. On May 16, 2020, Justin Trudeau, the prime minister of Canada, announced that Health Canada had approved the vaccine for human trials and had initiated a Phase I/II clinical study to be carried out in Halifax, Nova Scotia, sponsored by the Canadian Center for Vaccinology at Dalhousie University. However, news reports dating back to mid-July, 2020 began to emerge stating that the vaccine candidate was not approved by Chinese customs to export to Canada, and on August 27, 2020, Canada officially terminated the trial. Nevertheless, a larger scale international trial was registered on August 26, 2020, this time a global, multicenter Phase III clinical trial. The study endeavors to recruit 40,000 volunteers from Saudi Arabia, Russia, Mexico, and Pakistan, among other countries.
Phase I Chinese Trial
On March 18, 2020, Ad5-nCoV was officially registered to be tested in a non-randomized Phase I trial that aimed to assess the safety and immunogenicity of various dosages of the vaccine. Study participants were sequentially enrolled into one of three treatment groups. A total of 108 healthy adults between the ages of 18 and 60 were to receive intramuscular injections of the vaccine into the deltoid muscle, 36 of them receiving a low dose (5 x 1010 viral particles), 36 receiving a medium dose (1011 viral particles), and 36 receiving a high dose (1.5 x 1011 viral particles). The study subjects were evaluated for adverse events and immunogenicity on specific days following vaccination during a 6-month period. Persistence of antibodies and cellular immunity would also be tested over the same 6-month period.
On May 22, 2020, initial results of the Ad5-NCoV Phase I trial were published in the Lancet (Zhu F., et al.). At least one adverse reaction was reported in the first seven days after vaccination for 30 subjects that received the low dose, 30 subjects that received the medium dose, and for 27 subjects that received the high dose of the vaccine. Pain at the injection site was the most common localized adverse effect, and systemic adverse reactions were also frequently reported, including 46% of participants reporting fever, 44% reporting fatigues, and 39% reporting headache. No serious adverse effects were reported within 28 days after the vaccination. Overall, the authors concluded that the vaccine was tolerated in all dosage groups but more severe reactions were observed in the highest dose group, so the authors recommend the low and middle dosages be further assessed in the Phase II clinical trial.
Fourteen days after vaccination, all participants’ blood sera showed rapid antibody response to the receptor binding domain of the spike protein. During the 28-day duration of the study, titers of both SARS-CoV-2 neutralizing antibodies and antibodies specific to the spike protein RBD peaked on Day 28, and antibody titers were positively correlated with dosage. More specifically, on Day 28, the geometric mean titer (GMT) of RBD specific IgG antibodies was 1,445.8 in the high-dose group, 806.0 in the middle-dose group, and 615.8 in the low-dose group. Neutralizing antibody titers were all negative by Day 0 for all groups, rose somewhat by Day 14, and peaked by Day 28. By Day 28, the GMT of neutralizing antibody titer for 34.0 for the high-dose group, 16.2 for the middle-dose group, and 14.5 for the low-dose group, again demonstrating that neutralizing antibody titer was positively correlated with dose. T cell responses were also significantly higher in the high-dose group than in the low-dose group. Furthermore, interferon-γ was detected from CD4+ and CD8+ cells on Days 14 and 28 after vaccination in all dose groups.
Phase II Chinese Trial
On April 10, 2020, a Phase II, randomized, double-blind, and placebo-controlled clinical trial of the use of the vaccine in 508 healthy adult subjects (who had never had SARS-CoV-2 infection) was officially registered in China. The phase II trial aimed to evaluate the immunogenicity and safety of the vaccine when used in participants receiving two different doses. The placebo-control group was comprised of 126 individuals who received a 1.0 mL intramuscular injection in the deltoid muscle on Day 0. The high-dose group was made up of 253 individuals who received 1.0 x 1011 viral particles administered in a 1.0 mL intramuscular injection on Day 0. The low-dose group was made up of 129 individuals who received 5 x 1010 viral particles in a 1.0 mL intramuscular injection on Day 0. Levels of Anti-SARS-CoV-2 spike IgG and neutralizing antibody response were assessed throughout the course of the 6-month study. The occurrence of adverse reactions were also monitored throughout the study.
On July 20, 2020, Zhu F. et al. released preliminary results of the Phase II clinical trial. Blood samples were collected from participants on Days 0, 14, and 28 to assess the vaccine’s immunogenicity. Starting on Day 14, RBD-specific IgG antibodies were detected in the participants receiving the vaccine. The high-dose group had a GMT of 94.5 and the low-dose group had a GMT of 85.1 by this point. By Day 28, the RBD-specific IgG GMTs had risen even further, at 656.5 in the high-dose group and at 571.0 in the low-dose group. Subjects in the placebo group showed no increase in antibodies from baseline. Significant neutralizing antibody response to live virus was induced in blood samples collected from subjects on Day 28. The GMT of neutralizing antibodies by this point was 19.5 for the high-dose group and 18.3 in the low-dose group. Baseline T cell responses were negative in 99% of participants, but by Day 28, 90% of the high-dose group and 88% of the low-dose group showed spike-specific IFN-γ responses, a response mediated by T cells.
Participants were also observed for adverse reactions within the first 14 days after vaccination and in the first 28 days after vaccination. During the first 14 days after vaccination, 72% of the high-dose participants, 74% of the low-dose participants, and 37% of the participants in the placebo group reported at least one adverse reaction. The most common local reaction was pain at the injection site (57% for the high-dose group and 56% for the lower-dose group), and the most common systemic reactions reported were fatigue (42% in the high-dose group vs. 34% in the low-dose group), fever (32% in the high-dose group and 16% in the low-dose group), and headache (29% in the high-dose group and 28% in the low-dose group). Most reactions were mild or moderate, but a total of 24 (9%) of the participants receiving the high-dose reported severe adverse reactions, a percentage that was significantly higher than for the low-dose group. The severe adverse reactions were most commonly fever, and all resolved within 72-96 hours without medication. The authors note that pre-existing Ad5 immunity was correlated with significantly lower occurrence of fever after vaccination.
Phase III Global Trial
On August 26, 2020, a double-blind, placebo-controlled, randomized, global Phase III trial of Ad5-NCoV was registered officially. The estimated enrollment of the study is currently 40,000, with half of participants assigned to the placebo group and half of participants receiving a single, intramuscular dose of the Ad5-NCoV vaccine. Study participants will all be adults over the age of 18. The study aims to assess the safety, immunogenicity, and efficacy of the vaccine.
AZD1222 (ChAdOx1 nCoV-19)
AZD1222, previously known as ChAdOx1 nCoV-19, is a SARS-CoV-2 vaccine candidate that uses a chimpanzee adenovirus, ChadOx1 (Chimpanzee adenovirus Oxford 1), which expresses the SARS-CoV-2 spike protein. ChadOx1 has previously been used as an adenovirus vectored vaccine candidate for other human coronaviruses, such as MERS-CoV and SARS-CoV-2. In particular, the MERS-CoV vaccine candidate ChadOx1-MERS, which encodes the MERS-CoV spike protein in the same simian adenovirus vector, was shown to both protect non-human primates from MERS-CoV disease and be safe and well tolerated in a previous Phase I clinical trial at three different doses. At the highest dose tested, the vaccine candidate produced a strong immunogenic effect against MERS-CoV a month after vaccination.
ChadOx1 nCoV-19 was originally developed at the University of Oxford Jenner Institute, and on April 30, 2020, AstraZeneca entered into partnership with Oxford to develop and distribute the vaccine. Three weeks later, the pharmaceutical company received over a billion dollars in funds from the US Biomedical Advanced Research and Development Authority (BARDA) for the development, production, and distribution of the vaccine. On August 14, AstraZeneca entered into an agreement with the European Commission to supply up to 400 million doses of the AZD1222 vaccine at no profit during the pandemic, allowing for EU member states to access the vaccine, provided it receives regulatory approval. As of November 27, 2020, there are seven clinical trials underway for the study of the vaccine candidate (of which one was suspended), with one Phase III trial initiated in the U.S. on August 31, 2020, supported in part by the public-private partnership Operation Warp Speed and another Phase III trial initiated in Brazil on June 20, 2020.
According to a STAT News report, on Tuesday, September 8, 2020, Soriot, CEO of AstraZeneca, shared with investors on a private conference call that the Phase III clinical trial in the U.S. had been halted because of a serious adverse reaction reported in a participant receiving the vaccine. The participant may have experienced this condition due to an unrelated condition however, but since the individual received the vaccine, the trial has been halted for safety reasons. On September 9, the University of Oxford announced that enrollment in the international Phase II/III trial had also been temporarily halted. On September 12, 2020, the Phase II/III trials were resumed after Medicines Health Regulatory Authority officially confirmed that it was safe to do so. The FDA granted authorization for the restart of the Phase III I.S. trial on October 23, 2020, claiming it was safe to do so.
Pre-Clinical Trial Results
Munster et al. (2020) demonstrate a strong immunogenic response to the ChAdOx1 nCoV-2019 potential vaccine in mice and in rhesus macaques. IgG serum titers for the S1 and S2 subunits of the SARS-CoV-2 spike protein were compared between controls and mice that received the potential vaccine. Control mice had below detectable levels of the antibodies whereas all mice that received the potential vaccine had detectable titers. The same result was found when comparing the titers of viral neutralizing antibodies between the two groups. Furthermore, the authors report a strong Th1 (T-helper or CD4+ cell) response post vaccination. The authors also report a strong immunogenic response in six rhesus macaques that received ChAdOx1 nCoV-2019. Spike specific antibodies were detected in these animals within 14 days of vaccination, and virus neutralizing antibodies were detected in all 6 animals receiving the vaccine versus the detection of no virus neutralizing antibodies in 3 control animals. All nine animals were challenged with SARS-CoV-2 20 days post-vaccination, and control animals fared worse clinically, while no animals that received ChAdOx1 nCoV-2019 showed signs of pneumonia. The authors also report that the animals receiving the vaccine demonstrated a significantly lower viral load detected in their bronchoalveolar lavage fluid and respiratory tract tissue compared to control animals. Munster at al. report no evidence of antibody-dependent enhancement post vaccination for the six animals studied.
Graham et al. (2020) tested the immunogenic effect of a second dose of ChadOx1 nCoV-2019 in mice and pigs and compared it to the antigen-specific antibody response derived from a single dose in these animals. A single dose of the vaccine was enough to produce an antigen-specific and T-cell response in both mice and pigs. However, a secondary booster immunisation increased the overall SARS-CoV-2 neutralizing titers found in both animals, but did so more pronouncedly in pigs.
Phase I/II U.K. Trial
In March 2020, The Jenner Institute of the University of Oxford, the King Abdullah International Medical Research Centre (KAIMRC), and Vaccitech announced a Phase I/II, single-blinded, randomized, placebo-controlled, multi-center human trial (NCT04324606) to study the efficacy, toxicity, and immunogenicity of ChAdOx1 nCoV-19 against COVID-19 in approximately 1,112 healthy U.K. adults aged 18-55. Volunteers are divided into four different groups, each of which will either receive the ChAdOx1 nCoV-19 or the MenACWY vaccine intramuscularly. MenACWY is a vaccine that protects against 4 strains of the meningococcal bacteria―A, C, W and Y and will act as an active comparator in this clinical trial study. Volunteers will be blinded and will not know if they have received the ChAdOx1 nCoV-19 or the MenACWY vaccines. The experimental groups 1, 2 and 4, will receive a single dose of 5 × 1010 viral particles (vp) of ChAdOx1 nCoV-19. Some subjects in Group 2 however, will receive two different doses, a single dose of 5 × 1010 vp of ChAdOx1 nCoV-19 followed by a boost dose of 2.5 × 1010 vp of ChAdOx1 nCoV-1. The efficacy of the candidate ChAdOx1 nCoV-19 against COVID-19 will be assessed within a 6 month time frame by counting the virologically confirmed PCR positive symptomatic cases, whereas the safety of the vaccine will be tested by measuring the occurrence of serious adverse events (SAEs).
On July 20, 2020, Folegatti et al. published preliminary results of the trial, which enrolled a total of 1,077 participants during the 28-day period from April 23, 2020 to May 21, 2020. Half (n = 533) of the subjects received ChadOx1 nCoV-19 vaccine candidate at a dosage of 5 × 1010 vp, and half (n = 534) were in the placebo group, receiving the MenACWY vaccine candidate instead. Additionally, ten of the subjects that received ChadOx1 nCoV-19 received a follow-up dose of 5 × 1010 vp ChadOx1 nCoV-2019 28 days after the initial vaccination. Of these, 67% of subjects receiving ChadOx1 nCoV-2019 reported pain after vaccination compared to 38% of subjects receiving MenACWY. For the group receiving ChadOx1 nCoV-2019, 70% reported fatigue (48% in the MenACWY group), 68% reported headaches (41% in the MenACWY group), 60% reported muscle ache, 56% reported chills, and 51% felt feverish; 18% of the subjects in the treatment group recorded a fever over 38℃, compared to less than 1% in the placebo group. The severity of the adverse reactions tended to peak on Day 1 after the vaccination. Only one serious adverse reaction was noted, and it occurred in a subject receiving MenACWY who subsequently developed hemolytic anemia. Overall, adverse reactions were more common in the experimental group, but these preliminary safety results indicate that larger scale trials could be recommended. For the group receiving ChadOx1 nCoV-2019, anti-spike IgG antibodies peaked at 28 days (median of 157 ELISA units) after the administration of the vaccine candidate and stayed elevated until Day 56 with a median of 119 ELISA units. For those subjects that received two doses, a median of 639 ELISA units was recorded at Day 56 of the study. Spike-specific T-cell responses peaked on Day 14. Furthermore, 91% of subjects receiving a single dose of the vaccine candidate were able to achieve 80% virus neutralization compared to 100% of subjects who received two doses. Using a Marburg VN assay, the researchers also demonstrated that 62% of subjects receiving a single dose of the vaccine were able to induce complete inhibition of the cytopathic effect of SARS-CoV-2 by Day 56, compared to 100% that received two doses. Overall, the vaccine showed a strong immunogenic effect that was boosted with a secondary dose.
Phase II/III U.K. Trial
On May 26, 2020, a randomized Phase II/III interventional clinical trial of ChadOx1 nCoV-19 using participant masking was initiated, which aims to assess the safety and efficacy of the vaccine candidate in eleven study groups that consist of approximately 12,230 healthy participants. Groups 1, 7, and 9 consist of adults between the ages of 56 and 60; groups 2, 8, and 10 consist of adults over the age of 70; group 3 consists of children between the ages of 5 and 12; groups 4, 5, 6, and 11 consist of adults between the ages of 18 and 55. Depending on which group a subject is assigned to, the subjects in the treatment group will receive a single dose, two doses spaced several weeks apart, or a single dose followed by a smaller secondary dose. Dosages are determined by the group to which a subject is assigned. These subjects will be compared to subjects receiving one or two doses of MenACWY, which comprise the control group. The subjects will undergo follow-up for 1 year after vaccination to assess for safety, adverse reactions, and immunogenicity of the vaccine.
Phase III U.S. Trial
On August 31, 2020, AstraZeneca announced that a Phase III clinical trial had been registered in the U.S. The randomized, double-blind, placebo-controlled, multi-center interventional trial aims to enroll approximately 30,000 participants. Two-thirds (n = 20,000) of participants will receive two intramuscular doses of 5 × 1010 vp of ChAdOx1 nCoV-19 spaced 4 weeks apart; these subjects will make up the treatment group, while the remaining third (n = 10,000) will receive two doses of saline placebo spaced 4 weeks apart. The primary outcome measures are to test the efficacy of the AZD1222 candidate vaccine in the prevention of COVID-19 in adults, to test the immunogenicity of the candidate vaccine, and to assess the safety and potential adverse effects associated with the candidate vaccine.
Phase III Brazil Trial
On June 20, 2020, Latin America’s first Phase III trial of a COVID-19 vaccine candidate began enrolling study subjects, originally aiming to test 5,000 healthy volunteers. As of November 27, 2020, the randomized, controlled Phase III trial has an estimated enrollment of 10,300 adult subjects divided across four study arms. Participants in group 1a received a single dose of AZD1222 (5 x 1010 virus particles) with paracetamol, and they were compared to participants in group 1b who received a single dose of MenACWY vaccine (a vaccine protecting against four strains of meningococcal bacteria) with paracetamol. Meanwhile participants in group 1c received two doses of AZD1222 spaced 4-12 weeks apart, starting with an initial dose of 5 x 1010 virus particles and a 0.5 mL booster dose of 3.5 - 6.5 1010 viral particles. This group was compared to a control group that received an initial dose of MenACWY vaccine with a 0.5 saline placebo boost. All groups received paracetamol in conjunction with the vaccine candidate, vaccine, or control. The study aims to measure the safety, tolerability, and reactogenicity of AZD1222, as well as test for overall efficacy in preventing COVID-19.
Ad26.COV2.S is a candidate vaccine produced by Johnson & Johnson that delivers a recombinant adenovirus containing the SARS-CoV-2 spike gene. The gene expresses the stabilized pre-fusion SARS-CoV-2 spike protein. It uses a recombinant form of adenovirus serotype 26, which was originally developed by researchers at Beth Israel Deaconess Hospital, as a vector to deliver DNA encoding a particular antigen. A recombinant form of the virus has been produced by Johnson & Johnson previously as a vaccine for Ebola, as well as other candidate vaccines for HIV, RSV, and Zika. The vaccine candidate can be stored for up to two years at -4℉ (-20℃) and up to three months when refrigerated at 36-46℉ (2-8℃).
As of December 20, 2020, there are currently four registered clinical trials underway for the study of the vaccine candidate. On June 18, 2020, a Phase I/II randomized, double-blind, placebo-controlled trial to test the efficacy and safety of the vaccine candidate was first registered. On August 10, 2020 a randomized, double-blind, placebo-based Phase III vaccine trial with an expected set of 60,000 participants over the age of 18 was officially registered. The trial officially began in September, 2020. On August 12, 2020, a Phase I trial of the vaccine candidate was first registered in Japan. On November 16, 2020, another Phase III trial was officially initiated, this time with the primary aim of evaluating the safety and efficacy of using two doses of the vaccine candidate instead of just one.
On August 5, 2020, Johnson & Johnson announced that the U.S. government had reached an agreement with the U.S. government to manufacture and supply 100 million doses of the vaccine candidate following approval for emergency use by the FDA, which at the time had not yet been granted. BARDA (the U.S. Biomedical Advanced Research and Development Authority) agreed to pay one billion dollars to pay for these doses. The European Union reached a similar agreement with the company on October 8, 2020 for the delivery of 2 million doses. COVAX, a global initiative aimed at working with vaccine manufacturers to deliver vaccine candidates equitably around the world, entered an agreement to purchase 500 million doses.
Pre-Clinical Trial Results
Mercado et al. (2020) immunized 52 rhesus macaque monkeys with one dose of adenovirus serotype 26, which either encoded one of seven spike protein variants or a sham control. Twenty of the monkeys were in the control group, while 32 received an adenovirus encoding a spike variant. All 52 subjects were then challenged with a 1.0 x 105 50% tissue culture infectious dose of SARS-CoV-2 six weeks later, either intranasally and intratracheally. RBD-specific binding antibodies were observable in 31/32 monkeys in the treatment group at two weeks post-immunization and in all monkeys in this group by four weeks. Furthermore, neutralizing antibodies were observable in most treatment subjects two weeks after immunization, and the titers increased steadily up through week four. The Ad26-S.PP variant, which encoded a full wildtype leader sequence and a full-length spike protein with a mutation at the furin cleavage site and two stabilizing proline mutations, elicited the strongest pseudovirus neutralizing antibody response, with a median titer of 408, and the strongest live virus neutralizing antibody response, with a median titer of 113. The median titer of neutralizing antibodies was compared to a cohort of nine convalescent macaques and 27 humans after recovery from SARS-CoV-2 and was found to be four times greater, showing evidence of a robust immune response. The vaccine candidate also elicited detectable Spike-specific IgG and IgA responses in bronchoalveolar and nasal samples taken from the monkeys.
Mercado et al. (2020) also used RT-PCR methods to assess SARS-CoV-2 viral loads in bronchoalveolar and nasal swab samples taken from all 52 monkeys after virus challenge. All 20 control monkeys were successfully infected with the virus, with a median peak of 4.89 log10sgRNA copies/mL in bronchoalveolar samples. However, none of the 6 monkeys inoculated with the Ad26-S.PP adenovirus variant had detectable virus in their corresponding samples. The authors note that partial protection was elicited with the other 6 vaccines, but not as strongly as was observed with the Ad26-S.PP treated group. The median peak of viral load in the nasal samples of control monkeys was 5.59 log10sgRNA copies/mL, and only one of the six macaques that received Ad26-S.PP showed low levels of virus in its nasal samples. The monkeys that received the other vaccines that encoded spike variants showed reduced levels of SARS-CoV-2 in nasal samples, but those that received Ad26-S.PP showed the most reduction in levels, which in most cases was completely undetectable.
Phase I/II U.S. and Belgium Trial
A Phase I/II clinical trial to test the safety, reactogenicity, and immunogenicity of Ad26.COV2.S was first registered on June 18, 2020. The trial is a randomized, double-blinded, placebo-controlled study with an estimated enrollment of 1,045 adult participants. Participants were divided into 5 arms of the study. Only adults aged 18-55 were recruited into the first arm of the study, where participants received the vaccine candidate at two dose levels, on a single or two dose schedule within an 8-week interval, or received a matching placebo on Day 1 and Day 57. Participants aged 18-55 were recruited into the second arm of the study, where they either received a single dose of the vaccine candidate or they received placebo on Days 1 and 57. Participants aged 18-55 were included in the third arm of the study, where they received either Ad26.COV2.S or placebo on Day 1, along with booster shots of the vaccine candidate at the same dose or matching placebo at 6, 12, or, 24 months after the first dose. Participants aged 18-55 were included in the fourth arm of the study, where they received either Ad26.COV2.S or placebo on Day 1, along with booster shots of the vaccine candidate at the same dose or matching placebo at 8, 14, or, 26 months after the first dose. A final fifth cohort only included participants over the age of 65 who received the vaccine candidate at 2 dose levels, on a single or 2 dose schedule within an 8-week interval, or received a matching placebo on Day 1 and Day 57.
On September 25, 2020, preliminary results of the trial were first circulated in pre-print. Sadoff et al. (2020) specifically described results pertaining to participants in the first two arms and final arm of the study described above. These results were gathered from 402 participants aged 18-55 and 394 participants over the age of 65. Participants who received Ad26.COV2.S intramuscularly were given a dose level of either 5 x 1010 viral particles or 1 x 1011 viral particles per vaccination, either as a single dose or as two doses given on Day 1 and 57 of the study. Participants receiving placebo injections were matched to participants receiving the vaccine candidate in a 1:1 ratio and received 1 mL 0.9% saline solution on matching days. For participants aged 18-55 receiving the vaccine candidate, local adverse effects were reported in 58% of cases, whereas only 27% of participants over 65 receiving AD26.COV2.S reported local adverse events. The most commonly reported local adverse event was pain at the injection site. Systemic adverse effects were also more common in younger participants, occurring at a rate of 64% versus 36% in participants aged 65 and over. The most commonly reported systemic adverse effects were fatigue, headache, and muscle pain. Fevers were reported in 19% of participants aged 18-55 and in 4% of participants over the age of 65, and grade 3 level fevers were far less common in both groups.
Sadoff et al. (2020) also report seroconversion rates and immunogenicity profiles of the participants. However, it is important to note that seroconversion results were only available for the first 15 participants over the age of 65, so the sample size for this particular group is much smaller. For participants aged 18-55 receiving a single dose of the vaccine candidate, 92% of participants were seropositive for SARS-CoV-2 neutralizing antibodies by Day 29 of the study, with a geometric mean titer of 214 for those receiving the lower 5 x 1010 viral particles dose and a GMT of 243 for those receiving the higher 1 x 1011 viral particle dose. For adults over 65 receiving a single dose of the vaccine candidate at the lower dose level, 100% showed seroconversion for neutralizing antibodies by Day 29 with a GMT of 507, while only 83% demonstrated the same seroconversion by Day 29 at the higher dose level, with a GMT of 127 overall. Spike-specific antibody seroconversion was observed in 99% of participants aged 18-55 receiving the vaccine candidate, with a GMT of 528 and 695 for the groups that received the 5 x 1010 vp dose and 1 x 1011 dose, respectively. For the older participants administered the lower dose levels, 100% had observable Spike-specific antibodies with A GMT of 507. Only 83% of older participants showed the same seroconversion with a GMT of 248. By fourteen days after the vaccination, Spike-specific CD4+ and CD8+ T-cells responses were also robust in the groups receiving the vaccine candidate. Based on these overall findings, the authors recommend the clinical use of the lower 5 x 1010 vp dose for the single dose administration of the vaccine candidate, which had both fewer adverse events and higher seroconversion rates.
Phase III International Single Dose Trial
The first Phase III clinical trial to test the efficacy of Ad26.COV.S in the prevention of moderate to severe COVID-19 was officially registered on August 10, 2020. The trial is a multi-center, international study with an expected number of 265 study locations, many of which are based in the U.S., but other study locations include sites in Central and South America, as well as South Africa. The original estimated enrollment was 60,000 participants, but Johnson & Johnson lowered this number to 45,000 in December. The study divides participants into two arms: those in the treatment group will receive a single intramuscular injection of a 5 x 1010 viral particle dose of Ad26.COV.S on Day 1, and those in the placebo group will instead receive an intramuscular injection of placebo on Day 1. The primary outcome of the study is to determine the number of participants that develop severe/critical COVID-19 after Day 14 of the study. Other secondary aims are to determine the number of participants that develop mild or moderate to severe COVID-19 during the study and also to determine the SARS-CoV-2 viral load in these participants. Seroconversion rates of spike and nucleocapsid protein binding antibodies will also be evaluated, as will SARS-CoV-2 neutralizing antibody titers in the study participants. Local and systemic adverse effects will be monitored in the study groups throughout.
On October 12, 2020, Johnson & Johnson announced that the trial would be temporarily halted due to an unexplained illness in one of the participants. The company did not disclose whether this participant was in the treatment or placebo group. On October 23, 2020, the trial was resumed, and the company stated that the serious medical event in the study volunteer had no clear cause. The first set of preliminary results are expected in January, 2021.
Phase III International Two Dose Trial
A second Phase III trial to test the efficacy of the vaccine candidate was officially registered on November 4, 2020. This randomized, double-blind, placebo-controlled study will aim to test the safety and efficacy of a two dose regimen of Ad26.COV2.S with the goal of enrolling up to 30,000 adult participants. Participants in the treatment group will receive two intramuscular injections of the vaccine candidate, one on Day 1 and one on Day 57 of the study. Participants in the placebo group will instead receive an intramuscular injection of placebo on the same schedule. The study, known as the ENSEMBLE 2 study, will run in parallel with the single dose Phase III trial, with the same stated goals.
Sputnik V, formerly known as Gam-Covid-Vac, is a two-dose SARS-CoV-2 vaccine candidate that gained immediate notoriety when, on August 11, 2020, it became the first vaccine candidate in the world to be registered for widespread use despite not having cleared a large-scale Phase III clinical trial. The first dose of the vaccine candidate delivers the SARS-CoV-2 S protein gene via a recombinant adenovirus vector, Ad26, using a delivery method similar to Johnson & Johnson’s AD26.COV2.S vaccine candidate. The second booster dose delivers the SARS-CoV-2 spike protein via a recombinant adenovirus vector known as Ad5, thereby using a mode of delivery similar to CanSino’s Ad5-nCoV vaccine candidate. Sputnik V was developed by the Gamaleya Research Institute of Epidemiology and Microbiology in Russia, and as of September 7, 2020, it has been registered for study in two non-randomized Phase I/II clinical trials (now both completed) and in one Phase III trial. On December 22, Belarusian health ministry officially registered the Sputnik V vaccine, effectively granting early use or emergency approval. On December 23, the Argentinian government approved the use of Sputnik V, and the first shipment of 300,000 doses arrived just one day later.
Phase I/II Trials
Two non-randomized, small-scale trials were conducted, each testing 38 healthy adults aged 18-60 at one of two Russian hospitals. These trials aimed to assess the safety and immunogenicity of two different formulations of the candidate vaccine (frozen and lyophilized) in participants. In each of the two studies, 9 individuals received a single dose of the 26 recombinant adenovirus delivering the spike gene (rAd26-S), 9 participants received single dose of the Ad5 recombinant adenovirus delivering the spike gene (rAd5-S), and 20 participants received an initial dose of rAd26-S followed by a dose of rAd5-S 21 days later. Participants were monitored over the course of both studies to note any potential adverse effects. For those participants receiving only one component of the vaccine, adverse effects were measured from Day 0 to Day 28 of the study. The participants that received both doses were monitored for adverse effects from Day 0 to Day 42 of the study. In all participants, changes in antibody levels specific to the SARS-CoV-2 spike protein from baseline were evaluated at 42 days after treatment to test for the vaccine’s immunogenicity. Neutralizing antibody tiers were measured on Days 0, 14, and 28 for subjects receiving one component of the vaccine, and on Days 0, 14, 28, and 42 for subjects receiving both doses. Antigen-specific cellular immunity, as measured by T cell immunity and interferon-γ production, were measured on Days 0, 14, and 28 after vaccination. Note that no subjects were assigned to control groups in either of the registered studies.
On September 4, 2020, results of the two trials were published in the Lancet. Logunov et al. (2020) report that both components of the vaccine candidate were well-tolerated by the study participants. The most common adverse effects reported were pain at the injection site (reported in 58% of participants), fever (reported in 50% of participants), headache (reported in 42% of participants), weakness or lethargy (20% of subjects), and muscle and joint pain (reported in 24% of participants). The majority of these symptoms were reported as mild rather than moderate. In participants that received both doses of the vaccine candidate, the authors report that most adverse effects occurred after the second component (rAd5-S) was administered. Furthermore, no serious adverse effects were reported, and no participant needed to withdraw from the study due to any adverse effects.
Moreover, the authors report that the vaccine candidate demonstrated immunogenicity in all participants. By Day 14 of the study, 89% of the participants that only received the rAd26-S component of the vaccine candidate showed detectable levels of SARS-CoV-2 RBD-specific IgG antibodies, compared to 84% of the participants that had only received the rAd5-S component. By Day 21, all participants receiving a single dose had developed detectable levels of these antibodies. For participants receiving both doses of the vaccine, 85% had detectable levels of the antibodies by Day 14 and all had them by Day 21 (with a geometric mean titer of 1629 with the frozen formulation of the vaccine and a geometric mean titer of 951 with the lyophilized formulation). The second dose with rAD5-S on Day 21 led to an increase in titers recorded later in the study: by Day 28, the geometric mean titer (GMT) had gone up to 3,442 with the frozen formulation and up to 5,332 with the lyophilized formulation. On Day 42, the GMT had risen to 14,703 for the frozen formulation and up to 11,143 for the lyophilized formulation of the vaccine candidate. Formation of SARS-CoV-2 neutralizing antibodies was achieved in all participants only in the group receiving both components. The authors note a strong positive correlation between antigen-specific antibody titer and neutralizing antibody titer. Finally, the authors report that 100% of subjects showed the formation of antigen specific CD4+ and CD8+ T cells and an increase in production of interferon-γ secretion in response to challenge with spike protein. The authors do concede some limitations of the study, which include the small sample size and lack of a placebo-control group, and they also mention that the titers of neutralizing antibodies are lower than what was reported for the mRNA vaccine candidates, as well as for the adenovirus vectored vaccine ChadOx1 nCoV-19 (also known as AZD1222). Nevertheless, they emphasize the strong immunogenicity and tolerability of the vaccine in the study participants, and the strong results substantiated a larger-scale, Phase III study to be approved on August 26, 2020.
Phase III Trial
On August 28, 2020, a randomized, double-blind, placebo-controlled, multi-center clinical Phase III trial for Sputnik V was initiated. The trial aims to assess the safety, immunogenicity, and efficacy of the candidate in 40,000 healthy, adult participants over the age of 18. A total of 10,000 participants will be assigned to the control group, where they will receive a 0.5 mL placebo dose on Days 0 and 21 of the study. The remaining 30,000 participants will receive the Gam-COVID-Vac combined vector vaccine candidate, receiving a 0.5 mL dose of rAd26-S on Day 1 and a 0.5 mL dose of rAd5-S on Day 21. Subjects will participate in the study for approximately 180 days after the first dose of placebo/vaccine and will be expected to complete 5 visits (on Days 0, 21, 28, 42, and 180), as well as one additional follow-up visit with a study physician through a remote telemedicine consultation. Blood samples may be collected from certain participants during certain visits; these samples will be measured for antibody levels against the SARS-CoV-2 spike protein, for cellular counts of CD4+ and CD8+ T cells in response to challenge with the spike protein, and for SARS-CoV-2 neutralizing antibody titer. These secondary measures together aim to assess the immunogenicity of the vaccine. To test for efficacy of the vaccine candidate, the study will also measure the percentage of participants who develop COVID-19 within 6 months of the first dose as well as the severity of the disease in those cases where COVID-19 does occur. Finally, adverse effects and their frequency will be tracked to assess safety.
On November 11, 2020, the Russian Direct Investment Fund announced early results from the trial, stating that the vaccine candidate was 92% effective and that a total of 20 individuals in the clinical trial. The group but did not specify how many had received placebo or had received the vaccine candidate. A follow-up report on the vaccine candidate’s efficacy was given in a joint statement from the Russian Health Ministry, Gamaleya, and the Russian Direct Investment Fund on November 24, 2020. Based on an analysis of 18,794 volunteers that had received both doses of the vaccine, the efficacy of the vaccine was estimated as 91.4% by Day 28 (7 days after the second dose). Within this group, 14,095 had received two doses of the vaccine candidate, and only 8 cases of COVID-19 had been reported, while the remaining 4,699 participants who received only placebo had a total of 31 cases of COVID-19. The estimated efficacy was claimed as above 95% by Day 42, 21 days following the second dose, although the number of participants from whom this estimate was inferred was not specified. The statement also cited that no unexpected adverse events had occurred in the participants, and that interim data would be published in a peer-reviewed journal at some point.
CoronaVac is a SARS-CoV-2 vaccine candidate developed by the Chinese company Sinovac Biotech. It is a chemically inactivated form of the whole SARS-CoV-2 virus, and needs refrigeration at a temperature between 2 and 8 degrees Celsius, where it can remain stable for up to three years. On April 13, 2020, the company received approval to have its vaccine candidate enter Phase I/II clinical trials in China. The company has also entered various international trials, including two Phase III trials initiated in July 2020, one in Brazil and one in Indonesia. In July 2020, the vaccine candidate was approved by regulatory health agencies in China for emergency use in individuals at high-risk for SARS-CoV-2 infection, such as medical staff. In August 2020, a Phase III clinical trial to test the vaccine candidate was initiated in Chile, and on September 14, 2020, a Phase III clinical trial was registered in Turkey.
On August 25, 2020, Sinovac committed to producing at least 50 million doses of CoronaVac for Indonesia by March 2021. Following a Phase III trial conducted at the King Abdullah Center for the Saudi Arabian National Guard, The King Abdullah Medical Research Center agreed to purchase 7,000 doses of the vaccine candidate to be used on healthcare workers. In November 2020, Turkey agreed to purchase 50 million doses of the vaccine candidate for delivery through February 2021. In December 2020, Hong Kong ordered 7.5 million doses of CoronaVac, which will begin arriving in January, 2021. On December 16, shortly following promising preliminary Phase III results, the Brazilian federal government agreed to purchase 46 million doses of the vaccine candidate.
Phase II Chinese Trial
On August 10, 2020, Zhang, Y. et al. published preliminary results from the Phase II, randomized, double-blind, placebo-controlled trial. A total of 600 individuals between the ages of 18 and 59 were enrolled in the study and assigned to one of four groups: those receiving two doses of 3 µg/0.5 mL of CoronaVac, those receiving two doses of 6 µg/0.5 mL of CoronaVac, and those receiving a placebo. Each of these three groups were subjected to one of two possible dosing schedules: some received a dose of the vaccine candidate or placebo on Days 0 and 14, while the remainder received a dose of the vaccine candidate or placebo on Days 0 and 28. For the groups receiving doses on Days 0 and 14, adverse effects were reported in 35.0% of the individuals receiving the 6 µg/0.5 mL dose, in 33.3% receiving the 3 µg/0.5 mL dose, and in 21.7% of the placebo group. For those receiving doses on Days 0 and 28, adverse effects were reported in 19.2% of the individuals receiving the 6 µg/0.5 mL dose, in 19.2% receiving the 3 µg/0.5 mL dose, and in 18.3% receiving the placebo treatment. For both dosing schedules, there was no reported significant difference between the adverse reactions reported between treatment groups and the placebo group. The most common adverse reaction was pain at the injection site, and no individual reported a severe adverse reaction. The researchers also assessed the immunogenic response from the administration of the vaccine candidate and found no significant difference in immunogenic response between the two dosing schedules. All individuals were seronegative at the beginning of the study, and by 28 days after the second dose was administered, the geometric mean titer of anti-RBD specific spike protein antibodies had increased to 34.5 and 27.6 for the 3 µg/0.5 mL and 6 µg/0.5 mL doses, respectively. By this time, 92.4% on the Day 0 and 14 schedule had achieved seroconversion, while 97.4% on the Day 0 and 28 schedule had detectable levels of anti-RBD specific IgG. The geometric mean titer for neutralizing antibodies 28 days after the second dose was significantly greater in those who received doses on Day 0 and 28 compared to those who received doses on Days 0 and 14, with no significant difference observed between the two dosages. Furthermore, the authors identified that neutralizing antibody titer significantly decreased with increasing age.
Phase III Brazil Trial
On July 2, 2020, a double-blinded, randomized, placebo-controlled trial testing the safety and efficacy of Coronavac in an estimated 13,060 adult participants was officially registered. The trial, which is currently underway in Brazil, will divide participants into four distinct arms. One group is made up of adults between the ages of 18 and 59 who will receive two doses of the vaccine candidate spaced 14 days apart. The placebo group to which this group will be compared will instead receive two doses of placebo on the same time schedule. A third group will consist of adults aged 60 and over who will receive two doses of the vaccine candidate spaced 14 days apart, and they will be compared to a group of adults over the age of 60 receiving two doses of placebo on the same schedule.
On October 19, 2020, the Butantan Institute, a Brazilian biological research center that is the primary sponsor of the trial, announced preliminary results collected from 9,000 volunteers. Dimas Covas, the director of the institute, states that the vaccine was safe in volunteers but that further results would not be released until a full 13,000 volunteers had gone through the trial. He reported that no severe adverse reactions to the vaccine had been reported and that 20% reported mild pain at the injection site and 15% reported headaches after the initial dose and 10% reported the same symptom after the second dose. Fewer than 5% of participants reported nausea or fatigue. The Brazilian health secretary also reported that the vaccine produced an immunogenic response. Furthermore, the governor of São Paulo, João Doria, announced in a press briefing that the vaccine was the most effective and safest of the five vaccine candidates that at the time were being tested in Brazil.
On November 10, 2020, the Phase III clinical trial was temporarily halted due to a serious adverse event that had occurred on October 29, 2020. Investigative news outlets cited a police report that found a participant had committed suicide on that day, suggesting that this event may have been responsible for the temporary pause. ANVISA (Agência Nacional de Vigilância Sanitária), Brazil’s primary health regulatory agency, announced that the trial had been paused in order to evaluate the data and assess risk, but the trial resumed the following day on November 11, 2020. Dimas Covas stated that the adverse event was an independent event that had no link or relation to the vaccine trial, and Sinovac stated that it was confident in the vaccine candidate.
On November 24, 2020, Dimas Covas reported that a total of 74 trial participants had confirmed cases of COVID-19. The number of these participants that had received the vaccine candidate as opposed to the placebo was not disclosed. Covas later followed up on December 13, 2020 by announcing a total of 170 study participants had contracted COVID-19. Because of these promising results, Covas announced that on December 23, documentation would be submitted to request for full approval of the vaccine candidate from Brazil’s health regulatory agency.
BBIBP-CorV is one of two vaccine candidates currently being tested by China’s state-run Sinopharm. The vaccine candidate was developed by the Beijing Institute of Biological Products, and it is an inactivated form of SARS-CoV-2. BBIBP-CorV was originally developed from the isolation of three strains of the virus, which were obtained from bronchoalveolar and throat swabs of three Chinese COVID-19 patients (Wang, H., Zhang, Y., et al., 2020). The three strains were the HB02, CQ01, and QD01 strains, which the authors suggest spanned the main SARS-CoV-2 phylogenetic clades at the time. The strains were isolated by infection of Vero cells by the throat swab samples. The researchers ultimately chose the inactivated version of the HB02 strain for the BBIBP-CorV vaccine candidate, as this strain generated the highest viral yields when replicating in the Vero cells.
A Phase I/II trial was first registered in April, 2020, and a Phase III trial was initiated in the United Arab Emirates in July, 2020. Phase III trials in Peru and Morocco soon followed in August, 2020. The United Arab Emirates granted the vaccine candidate emergency approval for use on healthcare and government workers on September 14, 2020. On December 9, 2020, after the announcement of promising interim Phase III efficacy results, the United Arab Emirates granted BBIBP-CorV full approval. Bahrain soon followed, granting full approval for the use of the vaccine candidate on December 13, 2020. China granted emergency approval in the summer of 2020, and by November, 2020, Sinopharm announced that over 1 million people in China had been vaccinated with BBIBP-CorV.
Wang, H., Zhang, Y., et al. (2020) report that when mice, rats, guinea pigs, rabbits, monkeys, and macaques were injected with BBIBP-CorV, high levels of SARS-CoV-2 neutralizing antibodies were induced in the weeks following injection. Mice were inoculated with one of three doses of the vaccine candidate: 2 μg, 4 μg, or 8 μg, and neutralizing antibody titers were assessed at 7, 14, 21, and 28 days after administration of BBIBP-CorV. Seroconversion was achieved in all of the animals, regardless of dose, at 7 days after receiving the vaccine candidate. However, neutralizing antibody titers were positively correlated with dose and with time. The researchers also tested mice administered two-dose and three-dose schedules of the vaccine candidate with different time intervals between doses in order to determine which schedule achieved the optimal neutralizing antibody response. The highest levels of immunogenicity were achieved with three doses, followed by two doses, which was still higher than neutralizing antibody titers found for the mice receiving a single dose. Immunogenicity was next tested in mice, rats, rabbits, and guinea pigs at the three dose levels, and neutralizing antibody titers were determined 21 days after injection. Again, all animals had achieved seroconversion by this point.
The researchers next tested the immunogenicity and efficacy of the vaccine candidate when used in rhesus macaques. The macaques were immunized on Days 0 and 14; monkeys in the placebo group received intramuscular saline injections while monkeys in the two treatment groups received either a 2 μg dose or an 8 μg dose. On Day 24, the geometric mean titers of SARS-CoV-2 neutralizing antibodies for the two groups were 215 and 256, respectively. On Day 24, the macaques were challenged with SARS-CoV-2, and viral loads of various tissue samples were determined to measure the efficacy of the vaccine candidate. Throat and anal swabs taken from macaques in the placebo group maintained a high viral load throughout the study period, while the macaques in the low dose group had a significantly lower viral load. Monkeys in the high dose group all tested negative for viral load in throat samples throughout the study, and three out of four of these monkeys tested negative for viral load in anal swab samples taken throughout the study. The researchers concluded from further PCR testing on lung tissue from sacrificed monkeys that BBIBP-CorV conferred efficient protection against SARS-CoV-2 in both the high and low doses administered. Adverse effects in all animals studied were also minimal, and the authors concluded that the result supported the evaluation of BBIBP-CorV in clinical human trials.
Phase I/II Chinese Trial
On April 29, 2020, a Phase I/II randomized, double-blinded, placebo-controlled trial to evaluate the safety and immunogenicity of BBIBP-CorV in healthy participants aged 3 and over was formally registered in the Chinese Clinical Trial Registry (Registration Number: ChiCTR2000032459). On October 15, 2020, the first set of interim results of the trial were formally published in the Lancet. There, Xia et al. (2020) report on both Phase I and Phase II results concerning the vaccine candidate’s safety and immunogenicity in the participants enrolled.
For the first phase of the study, a total of 192 participants were enrolled; half of the participants were between the ages of 18 and 59, and half were aged 60 or over. For each age group, participants were further divided into four arms of the study: 24 received a two-dose schedule of 2 μg of the vaccine on Days 0 and 28, another 24 received two doses of 4 μg on the same schedule, another 24 received two doses of 8 μg on the same schedule, and 24 received placebo on the same two days. The most common systemic adverse reactions reported were fever, which were more common in the younger group and more common with higher doses. No serious adverse reactions were reported within 28 days after the second dose was administered. Neutralizing antibody titers were also reported for the groups and were typically higher in the younger group and also dose-dependent. By Day 42 of the study, for participants aged 18-59, geometric mean titers were reported as 87.7, 211.2, 228.7, for the 2 μg, 4 μg, and 8 μg dosed groups, respectively. For the participants over the age of 60, geometric mean titers for the corresponding three groups were 80.7, 131.5, and 170.87, respectively.
In the second phase of the study, an additional 448 participants were enrolled, with a mean age of 41.7. These participants were randomly assigned to one of five arms of the study. A total of 84 participants received a one-time 8 μg dose on Day 0. Another 84 participants received a two dose schedule of a 4 μg dose on Days 0 and 14. Another 84 participants received the same two 4 μg doses but on Days 0 and 21, and another 84 participants received the same two doses but on Days 0 and 28 instead. A final 112 participants received placebo, randomly assigned to one of the four schedules described for the first four treatment arms, so as to run in parallel with the treatment groups. At least one adverse reaction was reported in 23% of the participants receiving BBIBP-CorV (38% for those receiving the single 8 μg dose, 21% for those receiving the two doses on Day 0 and 14, 18% for those receiving the two dose schedule on Days 0 and 21, and 12% for those receiving the two dose schedule on Days 0 and 28). No serious adverse reactions were reported except for a grade 3 fever reported in the placebo group, and the most common systemic adverse reaction was fever, ranging from 1-4% for each treatment group. The two-dose schedules elicited significantly higher SARS-CoV-2 neutralizing antibody titers than the single 8 μg dose alone. On Day 28, geometric mean neutralizing antibody titers were reported as 14.7, 169.5, 282.7, and 218.0 for the single 8 μg dose, the two-dose schedule on Days 0 and 14, the two-dose schedule on Days 0 and 21, and the two-dose schedule on Days 0 and 28, respectively.
Sinopharm-Wuhan Vaccine Candidate
The second vaccine candidate currently being tested by China’s state-run Sinopharm was developed by the Wuhan Institute of Biological Products, and, just like the other Sinopharm vaccine candidate (see BBIBP-CorV), is an inactivated form of SARS-CoV-2. Xia, S., et al. (2020) report that the vaccine candidate came from an inactivated form of the WIV04 strain, originally isolated from a patient in Wuhan, China. The virus was propagated in a Vero cell line, and the infected cells were inactivated with ꞵ-propiolactone. A Phase I/II clinical trial carried out in Hunan Province, China began enrolling healthy adult volunteers on April 12, 2020, and interim analysis of both the Phase I and Phase II trials began on June 16, 2020. Preliminary results were published a month later. Phase III trials are currently being carried out in several countries, including Peru, which temporarily halted its trial in December due to an adverse neurological problem experienced by one of the study’s volunteers.
Phase I/II Chinese Trial
On August 13, 2020, preliminary results from Phase I/II interim analyses were released in JAMA. Xia, S., et al. (2020) report that the Phase I trial, which was carried out on 96 healthy adults, showed relatively low 7-day adverse-reaction rates at different dose levels compared to placebo. Those that participated in the trial received either placebo or one of three different doses (2.5, 5, or 10 μg) of the vaccine candidate on Days 0, 28, and 56 of the study. Participants receiving the three 5 μg doses reported the least amount of adverse reactions, with only 16.7% reporting such events compared to 12.5% of the placebo group. Geometric mean titers of neutralizing antibodies were 316, 206, and 297, for individuals receiving the low, medium, and high doses, respectively.
The Phase II interim results were based on data collected from 224 participants. Participants in the treatment groups only received the 5 μg dose on different schedules, either on Days 0 and 14 or on Days 0 and 21. Participants in the placebo group ran in parallel to the treatment groups and received injections of aluminum hydroxide on the same respective schedules. Adverse effects were reported in 6% of participants receiving the vaccine candidate on Days 0 and 14, compared to 14.3% of the placebo group receiving doses of placebo on the same schedule, while adverse effects were reported in 19.0% of participants receiving the vaccine candidate on Days 0 and 21, compared to 17.9% receiving placebo on the same schedule. Geometric mean titers of neutralizing antibodies at 14 days after the second injection were 121 and 247, for the Day 0 and 14 schedule and Day 0 and Day 21 schedule, respectively.
Protein-Based Vaccine Candidates
Several vaccine candidates that directly deliver SARS-CoV-2 proteins or protein fragments are also under development, including NVX-CoV2373 made by the pharmaceutical company Novavax. The vaccine candidate aims to develop an immunogenic response to the SARS-CoV-2 spike protein by delivering a recombinant form of the protein in conjunction with a patented adjuvant, known as Matrix-M adjuvant, that enhances the production of neutralizing antibodies. As of December 5, 2020, NVX-CoV2373 has four clinical trials underway: a Phase I/II trial in Australia, a Phase II trial in South Africa, a Phase III trial in the United Kingdom, and a Phase III trial in the United States.
Phase II South Africa Trial
On August 31, 2020, a Phase II A/B randomized, observer-blinded, placebo-controlled study to assess the efficacy, immunogenicity, and safety of NVX-CoV2373 in healthy adults in South Africa was registered. As of November 30, 2020, the study is in its Phase IIb stage, and there are a total 4,422 volunteers taking part in the trial, which now includes 245 medically stable, HIV-positive subjects. The subjects are split into four arms. The first arm consists of only HIV-negative subjects that receive two doses of 5μg of the vaccine candidate with 50 μg of Matrix-M adjuvant. Both doses are delivered through intramuscular injection: one dose is administered on Day 1 of the study and the other on Day 21. These participants are compared to a placebo group, which is composed of HIV-negative subjects administered saline injections on Day 1 and Day 21. The other two arms of the study are comprised of HIV-positive participants, and volunteers in each group will receive the same dosage of placebo or vaccine candidate as their counterparts in the HIV-negative cohort.
Phase III U.K. Trial
On September 28 2020, a Phase III, randomized, observer-blinded, placebo-controlled trial to test the efficacy, immune response, and safety of NVX-CoV2373 officially began. The trial has an estimated enrollment of 15,000 participants aged 18-84 in the United Kingdom. At least 25% of the study subjects will be above the age of 64. Subjects are divided into four separate study groups. Participants in one arm of the trial will receive two doses of an intramuscular injection of 5 μg SARS-CoV-2 recombinant spike protein vaccine candidate with 50 μg of Matrix-M1 Adjuvant one Days 0 and 21. Participants in the comparator placebo arm will instead receive two doses of an intramuscular injection of 0.5 mL of 0.9% saline solution on Days 0 and 21. A smaller set of 400 subjects will be used in a sub-study. This secondary cohort will contain an experimental arm that will be comprised of participants that will receive two doses of 5 μg spike protein vaccine candidate with 50 μg of Matrix-M1 Adjuvant on Days 0 and 21. They will also receive a licensed seasonal flu vaccine on Day 0. The comparator placebo arm will consist of volunteers that will receive two doses of saline placebo on Days 0 and 21 in addition to a licensed flu vaccine on Day 0.
Phase III U.S. and Mexico Trial
On November 2, 2020, a phase III, randomized, observer-blind study to test the safety, efficacy, and immunogenicity of NVX-CoV-2373 was first officially registered. The study aims to enroll at least 30,000 healthy participants over the age of 18 in the U.S. and Mexico. Participants will be divided into two arms: those receiving two doses of 5 μg spike protein vaccine candidate with 50 μg of Matrix-M1 Adjuvant on Days 0 and 21. The comparator arm will instead receive two doses of 0.5% saline placebo on Days 0 and 21.
EpiVac Corona is a protein-based vaccine candidate developed by the Vector State Research Center of Virology and Biotechnology in Russia, located in Siberia. More specifically, EpiVac Corona is composed of SARS-CoV-2 protein fragments known as peptides. These synthetic peptide antigens are the chemical agents that are used to trigger an immune response. On August 26, 2020, a Phase I/II clinical trial was officially registered, although some news reports had stated that the trial was already well underway by this stage. On October 14, 2020, Russian President Vladimir Putin announced that Russia had granted regulatory approval to EpiVac Corona, making it the second vaccine to have been granted such approval after the Gamaleya Gam-COVID-Vac (also known as Sputnik V). This occurred before any preliminary results of the Phase I/II trial had been released and before a Phase III trial had been initiated. At this point, the Phase I/II trial had a total of 100 volunteers between the ages of 18 and 60. Russian news agencies reported that a Phase III trial began in November, 2020. On November 24, 2020, it was reported that two trials had been registered, one that aimed to recruit 150 volunteers over the age of 60 and another that aimed to recruit 3,000 volunteers over the age of 18. On December 9, 2020, documents related to the EpiVac Corona vaccine candidate had been submitted to the world Health Organization, and as of December 15, it has been reported that 1,438 volunteers have received EpiVac Corona.
Other Protein-Based Vaccines
Kim et al., 2020 describes a skin patch vaccine, utilizing a microneedle array, composed of 400 needles, as a scratch method to introduce a SARS-CoV-2 vaccine in mice. In this way, the device directly delivers spike protein into the skin. The “PittCoVacc” generated an antibody surge against SARS-CoV-2 in the animals within two weeks. It is still in preclinical stages of development.
DIY Vaccine Efforts
In addition to the many vaccines under development by national programs, pharmaceutical companies, and academic institutions, there has also been a stream of individuals developing vaccines on a do-it-yourself basis and on occasion administering these to themselves, making information for their creation and administration available on the Internet, and in some cases making these products available to others for their administration. (Estep et al. 2020, CoroNope 2020, Complaint, State vs. Stine 2020). The project described by Estep, which involved a number of individuals from the Harvard and MIT communities, was reviewed in some detail by Regaldo (2020). Their vaccine consists of peptides which match part of the spike protein, coated in chitosan and administered nasally. It has reportedly been self-administered to about 70 people, but there are no data on its safety or effectiveness. The scientific, ethical, and legal aspects of such DIY vaccine development have been recently reviewed (Guerrini et al. 2020, Neal 2020). It seems clear that such an approach is highly unlikely to generate enough data to demonstrate that they have a safe and effective product, while negative results may further inflame those who already distrust vaccines.
Recommendations for Future Research
Yuan et al. (2020) provides crucial structural biology data on a highly-conserved, cryptic epitope in the receptor-binding domains of SARS-CoV-1 and SARS-CoV-2 (see Characteristics of SARS-CoV-2 Proteins). Characterizing a neutralizing antibody isolated from a convalescent SARS patient may provide insights into the spike protein receptor binding epitope, which may inform vaccine design for COVID-19.
In light of advances being made daily in vaccine development, as well as ongoing and new clinical trials proceeding internationally with already strong vaccine candidates offered by both public institutions and companies, a truly needy and demanding public deserves an effort that also includes science and medicine without walls. For complete openness and transparency, scientists, clinical investigators and the leadership of all countries that are currently under siege by SARS-CoV-2, the public deserve to know that openness and communication are driving an international effort to treat and cure COVID-19, because this pandemic warrants it. Toward such a cause, a noteworthy article in Science (S Berkley, 2020) has promoted the idea of an international Manhattan Project, a true and reliable stimulus to get all involved better capable for discovery and application in expedience-requiring situations - that is, an ever evolving pandemic as we are currently seeing. A new perspective also published in Science (Corey, Mascola, Fauci and Collins, 2020) also advances a truly genuine need for public-private partnerships during the development and testing of emerging vaccine, immuno- and molecular therapies for this coronavirus, and all other pathogenic infections that could be in the pipeline. It is worth considering advocating for such national and collaborative international efforts, including public-private cooperation that the scientific world has not seen since World War II; where again, human pain and suffering required the thinking and actions of all who bring intersecting fields and expertises to the table.
The efficacy of a vaccine against SARS-CoV-2 will hinge on its immunologic effectiveness in producing an appropriate antibody and T-cell response, safety, widespread availability to the population at risk (see above paragraph), and the willingness of the population to be vaccinated. The last of these may be one of the most difficult obstacles to surmount, particularly in the United States, where relatively poor compliance with vaccination for other diseases (influenza vaccination in 2018 was only 45.3% for adults according to CDC figures) and a worrisome anti-vaccine bias in some is being compounded by the entry of partisan political considerations into announcements of vaccine readiness and availability, in addition to similar assertions followed by retractions on therapies such as hydroxychloroquine and convalescent plasma which have come from the CDC and FDA. This has led to an increasing level of public skepticism about the truth of statements made about vaccines, and a high likelihood that many people who are vaccine candidates will refuse to take it initially because of lack of trust as to statements of its safety and efficacy. Prior examples of roll-outs of flawed vaccines such as that for the Swine Flu in the US in 1976 or dengue in the Philippines in 2017 have given ammunition to this argument, to the extent that the WHO noted in 2019 that vaccine hesitancy was one of the top ten global health threats (WHO 2019). The time to embark on a rational public explanation of what vaccines can and cannot accomplish is now, so that we will not be faced in several months with an effective vaccine that the public is unwilling to take. As noted in a recent editorial in the Lancet “The development of robust scientific evidence takes time, whereas anecdote, sensationalism, and weak science travel quickly.” (Salmon et al. 2020). While we await the results of the multiple Phase III trials of vaccines currently underway, the obligation of helping to improve public understanding of the potential importance of vaccination as part of a multi-pronged effort (which will still include distancing and mask use for some time) to limit the impact of COVID-19 is paramount to an effective strategy for its mitigation.
Toward the cause of the best ways to test vaccine efficacy, in addition to non-human primate and other animal models relying on hamsters or ferrets because of similarities to human in lung and other COVID-19 pathologies, the laboratory mouse has now been adapted via a single nucleotide alteration in the RBD of the spike protein to again be a well-characterized animal model for assaying vaccines as well as emerging antivirals (Gu et al., 2020).
Potential Experimental Treatments and Preventatives
As mentioned above, ferrets have been used as one animal model for testing therapeutics because of, among other things, similar lung pathophysiology from SARS-CoV-2 infection as that seen in humans. Testing a new class of preventatives or prophylactic treatments for COVID-1 9, exploiting insights into the protein composition of the receptor binding domain (RBD) of the virus spike, it has been shown that a nasal spray containing a non-toxic lipopeptide viral-host membrane fusion inhibitor shows promise for inhibiting SARS-CoV-2 infection in ferrets (de Vries et al., bioRxiv, 2020). It is anticipated that more work on the cell and molecular biology of this virus will provide new preventatives that like this first generation nasal spray will help deter infection and transmission. In contrast to current pharmaceutical drugs being tested for use in current or future COVID-19 infections, potential treatments discussed in this section have not yet been approved for any known human use. However, they have been developed through various experimental (and often novel) methods, showing promise in the treatment or prevention of COVID-19. We discuss monoclonal antibodies and other chemical compounds, cell therapies, convalescent plasma transfusions, gene-editing therapies, and radiation therapy.
On August 3, 2020, Baum et al. published a pre-print article characterizing the efficacy of the use of two potent neutralizing antibodies, Casirivimab and Imdevimab (together known as REGN-COV2), that target two different, non-overlapping epitopes in the spike protein both rhesus macaques, which were used as an animal model for mild COVID-19, and golden hamsters, which were used as an animal model of severe disease. In this report, the antibody cocktail was shown to be effective at lowering viral loads in both the upper and lower airways and at reducing pathology and severity of pneumonia when used prophylactically or therapeutically. Baum et al. also highlight that the treatment was the first of its kind to limit weight loss in an animal model of severe COVID-19.
On September 29, 2020, Regereron announced by press release preliminary results of their seamless phase 1/2/3 trial of a combination of two monoclonal antibodies, REGN-COV2, with publication pending (Regeneron 2020). The two component antibodies bind non-competitively to the spike protein, which is thought to reduce the likelihood that mutations in the spike protein, as have been shown to occur, would blunt the effectiveness of the antibody combination. Regeneron’s report described the first 275 patients entered on the trial. All had laboratory confirmed COVID-19 and were being treated in the outpatient setting. Baseline determination of serology to determine antibody status against SARS-CoV-2 was performed (45% seropositive, 41% seronegative, 14% unclear or unknown serology status), but it was not stated whether this was used as a stratification factor in the randomization between placebo, low dose (2.4 g), or high dose (8 g) of REGN-COV2 given as a single infusion. REGN-COV2 significantly reduced the viral load by approximately 75% in seronegative patients through Day 7 in the high dose group (p = 0.03) and nearly did so in the low dose group, with a mean weighted-time average reduction of approximately 70% (p = 0.06) when compared to placebo. In the overall population, the mean weighted-time average reduction was approximately 70% for the high dose (p = 0.0049), a statistically significant result. However, for the low-dose group, the mean weighted-time average reduction was approximately 41% (p = 0.20), which was not found to be statistically significant. Among the seronegative patients, median time to ‘alleviation of symptoms’ was reported as 13 days for the placebo group, 8 days for the high dose group (p = 0.22), and 6 days in the low dose group (p = 0.09). Both doses were well-tolerated, with infusion reactions reported in 4 patients (2 placebo, 2 REGN-COV2) and serious adverse events reported in 3 patients (2 placebo, 1 low dose REGN-COV2). The investigators were encouraged by these data, are continuing accrual of at least 1300 patients to the phase 2-3 component of this trial, and are also conducting trials of REGN-COV2 in hospitalized patients and household contacts of infected individuals.
Before Casirivimab and Imdevimab received authorization for emergency use by the FDA, Regeneron could grant patients access to the antibody cocktail through compassionate use. On October 2, 2020, U.S. President Donald J. Trump received such approval when his medical staff reached out to Leonard Schleifer, the chief executive of Regeneron, for permission to use REGN-COV2, which was approved by the FDA. That National Institute of Allergy and Infectious Disease director, Dr. Anthony Fauci, has stated that there is a good chance that the treatment was at least partially responsible for the president’s swift recovery from COVID-19. On Novememer 21, 2020, the FDA authorized the use of Casirivimab and Imdevimab for emergency use for the treatment of mold to moderate COVID-19 in adults and pediatric patients at risk for progressing to severe COVID-19.
In June, 2020, Bamlanivimab also known as LY-CoV555, a monoclonal neutralizing antibody developed by Eli Lilly, entered a Phase II randomized study aimed at finding the optimal dosage for the drug in treating 452 non-hospitalized patients with mild to moderate COVID-19. Of these, 302 subjects received the medication and were divided into three treatment groups based on the size of the dose received, namely, 700 mg, 2800 mg, and 7000 mg. The remaining 150 subjects were assigned to the placebo group. Eli Lilly released interim results of the Phase II randomized, double-blind, placebo-controlled trial on September 16, 2020. Out of the 302 subjects who received the drug, only 5 (1.7%) were hospitalized, whereas 9 out of the 150 (6%) individuals receiving the placebo were hospitalized. No subject either in the treatment or placebo groups died or required the assistance of mechanical ventilation. Moreover, no serious adverse effects were reported for the patients receiving the drug. Based on these small-scale results, the drug appeared to reduce the risk of hospitalization by about 72%, an encouraging figure that supports a move to later phase clinical trials. The researchers also indicated evidence of the drug’s antiviral properties, which may explain lower viral loads observed in subjects taking the medication.
On November 9, 2020, Bamlanivimab received Federal Emergency Use Authorization from the FDA for the treatment of patients over the age of 12 weighing at least 40 kg with mild to moderate COVID-19 and who are at high risk for hospitalization or for developing severe COVID-19. The treatment is recommended for use within 10 days of symptom onset and shortly after a confirmed positive COVID-19 test. The antibody is not approved for use in hospitalized patients or for those receiving supplemental oxygen therapy, as some clinical evidence suggests the use of monoclonal antibodies may be associated with worsened outcomes in patients using high-flow oxygen supplementation or mechanical ventilation. The decision to grant this authorization was substantiated by evidence from clinical studies showing that use of bamlanivimab was associated with reduced hospitalization rates in patients at risk of developing severe COVID-19. As of November 15, 2020, there are five clinical trials studying the use of the monoclonal antibody in COVID-19 patients.
Other Antibody Treatments
Leronlimab, also known as PRO 140, is an experimental drug that is in clinical trials for its potential medical use in HIV patients and in patients with breast cancer. It is a monoclonal antibody that targets the CCR5 receptor (C-C chemokine receptor type 5) in T-cells where it acts as a receptor antagonist, thereby blocking other substrates from binding to the receptor. Chemokines, a type of proteinaceous cytokine responsible for cell signalling and chemotactic induction (referring to movement induced by chemical signal) in nearby cells, are the typical substrate for this receptor. Since chemokines signal other immune cells to the site of infection, the drug may mitigate the cytokine cascade effect that contributes to end-stage COVID-19 conditions such as ARDS. In mid-March 2020, the FDA approved leronlimab to enter a Phase II clinical trial to investigate its potential therapeutic benefit in the treatment of patients with COVID-19, where it is proposed to lessen inflammation in lung tissue. The first two patients of this clinical trial were enrolled on April 6, 2020.
On March 27, 2020, CytoDyn, the company that produces the drug, announced by press conference the first preliminary results after testing the drug in four patients with severe COVID-19. In addition to improving overall condition of two intubated patients within three days of use, resulting in their extubation and release from the ICU, the results indicate that the drug may be associated with a significant reduction in IL-6 (interleukin-6) and TNF-α (tumor necrosis factor alpha), two potent cytokines that mediate early inflammatory response, in patient lung tissue. As of March 27, 2020, the drug had been administered to seven patients in the New York City area with severe COVID-19 (for use as an emergency IND), and more results are forthcoming.
Wang, C. et al. (2020) identified four antibodies derived from hybrid mice and myeloma cells that exhibited cross-reactivity with the SARS-CoV-2 S1 domain. Of these four, the 47D11 H2L2 antibody demonstrated the ability to neutralize both SARS-CoV-1 and SARS-CoV-2 in vitro. This antibody was selected to be reformatted into a human immunoglobulin, referred to as human monoclonal antibody 47D11 (human mAb 47D11). The human 47D11 antibody was shown to inhibit SARS-CoV-1, with 50% inhibition occurring at an antibody concentration of 0.19 µg/L, and SARS-CoV-2 infection, with 50% inhibition occurring at an antibody concentration of 0.57 µg/L, in VeroE6 cells. The authors report that the antibody targets the receptor binding domain of the S1B domain of both viruses in vitro, and with similar affinities. The human monoclonal antibody 47D11 is the first report of a human monoclonal antibody that effectively neutralizes SARS-CoV-2, albeit in vitro in VeroE6 cells. Furthermore, it binds to a relatively conserved part of the SARS-CoV-2 epitope, as it demonstrates strong neutralizing cross-reactivity with the same region in SARS-CoV-1, which may enhance its attractiveness as a potential therapeutic.
Wrapp, McLellan, et al. (2020) report on two single-domain antibodies, both isolated from a llama immunized with SARS-CoV-1 spike protein, that show potent binding to the SARS-CoV-2 RBD domain. Single-domain antibodies isolated from camelids produce heavy-chain only antibodies, known as VHH fragments. VHHs are much smaller than common antibodies, which commonly contain heavy-chains (VH) and light-chains (VL). After immunizing the llama with SARS-CoV-1 and MERS CoV spike proteins, the research team successfully isolated seven distinct VHH antibodies specific to the MERS-CoV spike protein and five unique VHH antibodies specific to SARS-CoV-1 spike protein. Of these, four of the MERS-CoV antibodies could bind with high affinity to the pre-fusion form (the uncleaved form of the protein before it binds to the ACE2 receptor) of the MERS-CoV spike protein and three of the SARS-CoV-1 antibodies could bind with high affinity to the pre-fusion form of the SARS-CoV-1 spike protein. These VHHs were also able to successfully neutralize MERS-CoV or SARS-CoV-1 virus, respectively. There was no cross-reactivity for the SARS-CoV-1 antibodies to the MERS-CoV spike protein or vice versa. However, one particular VHH isolated, SARS VHH-72, could bind to a region of the SARS-CoV-1 that exhibits a high degree of conservation in its amino acid residues. The research team was able to demonstrate that SARS VHH-72 could bind with high affinity to the RBD of WIV1-CoV, a betacoronavirus found in bats that is closely related to SARS-CoV-1 and that also binds to the ACE2 receptor. SARS VHH-72 also showed cross-reactivity with SARS-CoV-2 RBD, albeit with lower affinity, as demonstrated by a higher dissociation constant (~39 nM). The researchers then engineered two bivalent versions of SARS VHH-72 in an attempt to increase binding affinity and found that both bound successfully to the SARS-CoV-1 and SARS-CoV-2 pre-fusion spike protein RBDs. The authors report that the antibodies successfully reduced the binding of the spike protein to ACE2 in vitro in mammalian cell lines. Since these bivalent variants had such successful binding to a variety of pre-fusion betacoronavirus RBD, they may serve as a promising future treatment for a variety of emerging zoonotic betacoronaviruses.
Potential protection using SARS-CoV-2 neutralizing antibodies is a promising area of research, with most studies focusing on the receptor binding domain of the spike protein. New animal models in addition to feline, ferret and non-human primates are emerging (e.g. Syrian hamsters, Rogers et al., 2020), including the use of humanized mice where the Regeneron group has shown extremely promising results with an antibody cocktail that has advantages that can mitigate selective pressure of single antibody treatments (Hansen et al., 2020).
Hansen et al. (2020) used both humanized mice and convalescent patients to create antibodies against the SARS-CoV-2 Spike protein. From the large panel of antibodies that they generated, they found pairs of antibodies that could simultaneously bind to the receptor binding domain. Because single antibody treatments may give rise to selective pressures that may favor new genetic variants that are resistant to the single antibody therapies, a paired antibody treatment may likely reduce such selection pressures. Baum et al. (2020) studied four of the antibodies identified by Hansen et al. more closely to determine whether this hypothesis may be true. Each of the four antibodies was a potent neutralizing antibody against all current strains of SARS-CoV-2. However, when the virus was exposed to the single antibodies in vitro, new spike mutants that were resistant to the neutralizing effect of the single antibodies rapidly appeared. This effect also occurred when pairs of antibodies that bind to different but overlapping regions were used. However, when the pair of antibodies could bind to regions that did not overlap, no such variants appeared. These results support that the use of cocktails of non-competitive antibodies may substantially reduce the number of escape variants that could arise from exposure to single antibodies.
A bivalent monoclonal antibody with impressive utility, including in a hamster model of SARS-CoV-2, has also been reported (Li W et al., 2020). It is anticipated that monoclonal antibody therapies will be an important part of the treatment arsenal for COVID-19.
ACE2 Peptides and Recombinant ACE2
On March 19, 2020, Zhang G. et al. published results stating that a peptide fragment they had synthesized was able to effectively bind to the receptor binding domain (RBD) of the SARS-CoV-2 spike protein. The peptide is composed of a chain of 23 amino acid residues that is identical in primary structure to a portion of the ɑ1 helix of the ACE2 peptidase domain (PD). The peptide fragment may be a promising novel therapy as it binds to and remains on the surface of the virus’s RBD, the region of the virus’s spike protein that binds to the ACE2 receptor in order to gain intracellular entry. At certain concentrations, the peptide may be capable of covering the RBDs of the spike protein, thereby limiting RBD-ACE2 interaction. Furthermore, it can remain attached to available RBD in solution with high affinity; the team found that the dissociation constant of the peptide to the RBD was 47 nM. The researchers also tested a truncated version of this peptide but found that it had a much higher dissociation constant (and thus lower binding affinity). While the peptide has not been tested in humans, the authors note that it is an endogenous peptide that will likely be tolerated by the human immune system.
APN01, also known as hrsACE2 (human recombinant soluble Angiotensin Converting Enzyme 2), is a soluble form of recombinant human ACE2 that was originally developed by APEIRON Biologics, an Austrian biotech company, after the SARS-02 outbreak. The drug candidate, which was previously tested in healthy volunteers in a Phase I clinical trial (Haschke et al., 2013) and in ARDS patients in a Phase II clinical trial (Khan et al, 2017), is currently being studied for its potential role as a COVID-19 therapeutic to reduce viral replication by competing with the SARS-CoV-2 spike binding sites of human cells. By binding to the spike receptor binding domain, APN01 prevents the virus from infecting these cells, thereby neutralizing the virus. APN01, as a form of ACE2, may also inhibit the dysregulation of the renin-angiotensin system associated with COVID-19, a key driver of acute lung injury, ARDS, and organ damage (see ACE2 Receptor).
On April 24, 2020, Monteil et al. published results from an in vitro study that found hrsACE2 was capable of reducing SARS-CoV-2 viral growth in Vero E6 cells by a factor of 1000-5000. Furthermore, the researchers report that hrsACE2 was able to effectively inhibit SARS-CoV-2 infection of genetically engineered human blood vessel organoids and kidney organoids. Such promising results spurred clinical trials testing the efficacy of the drug in treating COVID-19 patients.
A pilot clinical trial to treat 24 COVID-19 patients with APN01 was first set to take place in China. The primary purpose of the randomized, controlled study was to determine whether a Phase IIb study was warranted. The clinical trial, which was first posted on February 27, 2020, was later withdrawn on March 17, 2020. On April 30, 2020, a multi-center international randomized Phase II clinical trial testing the drug on 200 COVID-19 patients was initiated. Patients in the treatment group received the drug intravenously twice daily while those in the control group received a placebo intravenously twice daily. The trial aims to compare the 28-mortality rate, the number of ventilator free days up to 28 days or hospital discharge, the time to death, and the levels of lactate dehydrogenase (an indicator of organ damage) at Day 5 between the placebo and treatment groups.
On September 24, 2020, Zoufaly et al. published a case study in the Lancet that documented the use of hrsACE2 in a severely ill COVID-19 patient, who exhibited high fever and hypoxemia. The patient, who was eventually intubated, showed high levels of serum lactate dehydrogenase (636 U/L), ferritin (880 µg/L), C-reactive protein (103 mg/L), and D-dimer (1.3 mg/L), all markers of severe disease. The patient was treated with hrsACE2 intravenously 9 days after symptom onset, and after the first injection of the treatment, the authors report that her angiotensin II, IL-6, and IL-8 levels dropped. The use of soluble recombinant ACE2 not only reversed the dysregulation of the angiotensin-renin system, but it also reduced inflammatory markers that are known to contribute to acute lung injury and cytokine storm. Ferritin, C-reactive protein, and tumor necrosis factor alpha levels also markedly dropped following the administration of hrsACE2. SARS-CoV-2 levels, which were detectable 2 days before and on the day of the first injection, showed drastic drops after the one day of treatment and became undetectable the following day for the remainder of the patient’s hospital stay. The results of the case study, while exciting, cannot be extrapolated to a larger population, as only one subject was studied. We await results from the Phase II clinical trial currently underway to draw broader conclusions concerning the efficacy of the drug candidate in the treatment of COVID-19.
Cao et al. published in Science (9 September, 2020) a report describing the design and in vitro testing of a number of SARS-CoV-2 miniprotein inhibitors. They used two strategies to design these ‘minibinders’: first, incorporating the alpha-helical motif from ACE2, which makes most of the interactions with the spike protein receptor binding domain (RBD), making additional modifications to enhance affinity; and second, de novo design of minibinders from small helical scaffolds followed by rotamer interaction field docking to identify shape and chemically complementary binding modes (Chevalier et al., 2017, Dou et al., 2018). These compounds were screened for binding activity against the RBD displayed on yeast cells, and they identified three candidates from the first approach and 105 from the second. Further optimization of these structures for thermostability and binding to the RBD yielded three candidate structures, AHB2 from the ACE2 helix scaffolded design approach and LCB1 and LCB3 from the de-novo design approach. These were stable at room temperature for more than 14 days. Binding affinity to the RBD was <1 nM. They studied the ability of these compounds to inhibit infection of Vero E6 monolayers with SARS-CoV-2 and observed IC50 values of 15.5 nM for AHB2 and 23.54 and 48.1 pM respectively for LCB1 and LCB3. They noted that, on a molar basis, these were three-fold lower than those of the most potent anti-SARS.CoV-2 monoclonal antibodies described to date (Alsoussi et al, 2020).
These compounds have not yet been studied in humans. They have several potential advantages over the use of monoclonal antibodies to block interaction of SARS-CoV-2 and ACE2. They span a range of binding modes making mutational escape from inhibition less likely. They are thermostable which will facilitate distribution. Their molecular size (56-64 amino acid residues) is dramatically less than that of monoclonal antibodies which should facilitate their production, as they do not require production in mammalian cells to ensure proper post-translational folding. Their small size should facilitate their use in topical intranasal therapies. Finally, the close correspondence of the structures of LCB1 and LCB3 as ascertained by cryoEM and computational design models speaks in favor of this approach to targeted design against both SARS-CoV-2 and other potential future pandemic agents.
Other Experimental Compounds
Qing and Zhang, developers of the QTY code, have suggested a practical application of their method for the development of water-soluble chemokine and cytokine receptors to take up excess blood-circulating cytokines. The QTY code is a method of protein design that replaces the hydrophobic amino acids (e.g., leucine, valine, isoleucine, and phenylalanine) with glutamine (Q), threonine (T), or tyrosine (Y), rendering the protein water soluble. Previously, for the study of transmembrane proteins, such as G-coupled protein receptors, in aqueous solution, the hydrophobic domains needed to be treated with detergents, but the novel protein design results in fully functional 3D protein structure that a detergent would typically denature. In fact, Qing and Zhang show that chemokine receptors designed through this method retain their α-helical structure, thermostability, and ligand-binding-affinity in 50% human serum (Zhang, S. et al., 2018). The research team has even been able to successfully design chimeric detergent-free chemokine receptors that were previously unattainable through detergent use (Qing et al., 2019). They further demonstrate that when these receptors are expressed in E. coli, the receptors can bind to their chemokine ligands with high affinity (dissociation constants were in the nM range), and they have high thermostability in a wide range of temperatures. The QTY code provides a means for developing cytokine receptors which may have widespread potential use in the treatment of autoimmune disorders and the cytokine storms that are typical of viral infections, such as COVID-19. As a water-soluble version of a cytokine receptor, the proteins can effectively travel through the bloodstream, instead of remaining static inside cell membranes, where their natural counterparts are generally found. As a result, the circulating water-soluble proteins have the potential to bind to the cytokines and chemokines that would otherwise drive the inflammation that may lead to ARDS or other end-stage complications of COVID-19.
Following infection of cells by SARC-CoV-2 and translation of the first open reading frame to two polypeptides, proteolytic processing of these by the main protease (Mpro) and two papain-like proteases is required for production of 16 nonstructural proteins which are involved in the production of subgenomic RNAs encoding multiple structural and other proteins. (See Replication Cycle for further discussion.) Mpro is conserved among coronaviruses, with common features at cleavage sites. As there is no human homologue for this enzymatic activity, it is an appealing target for rational drug development. Dai and colleagues (Dai et al., 2020) have reported the development of two covalent inhibitors of SARS-CoV-19 Mpro which were shown by examination of the structure of the crystal structure of compound 11a and Mpro at 1.5 angstrom resolution to bind covalently to cysteine 145 in the substrate binding pocket of the enzyme. This compound was selected for pharmacokinetic studies in mice, rats, and dogs (beagles), where relatively little toxicity was seen at dose levels up to 18 mg/kg in rats and 40 mg/kg in dogs, when given by IV drip over 7 days. The IC50 of this compound was 0.053+/-0.005 𝜇M. Further clinical study of this compound, which potentially inhibits the life cycle of SARS-CoV-2 at a point not attacked by other compounds, would appear warranted.
Several clinical trials registered in China will investigate the use of mesenchymal stem cells (MSCs) in the treatment of COVID-19 in hospitalized patients. This cellular therapy is thought to help in the reduction of inflammation in patients with ARDS and boost tissue regeneration. MSC transplantation has been associated with improved outcome in 7 hospitalized patients with COVID-19 pneumonia in Beijing (Leng et al., 2020). Within two days of the MSC transplant, patients showed significantly improved pulmonary function and overall outcome. Researchers also observed that the numbers of circulating peripheral lymphocytes grew and C-reactive protein, a protein whose expression is increased during inflammation after IL-6 secretion, decreased. Furthermore, any overactive cytokine secreting T and Natural Killer (NK) cells were not detectable within 3-6 days of treatment. The circulating concentration of tumor necrosis factor alpha was also significantly reduced, suggesting a pronounced anti-inflammatory effect.
The FDA has fast-tracked a mesenchymal stem cell therapy using multipotent adult progenitor cells (MAPC) under the name Multistem from Athersys Inc. These cells derived from human bone marrow with both stromal and progenitor cell characteristics have been shown in previous animal model and human clinical studies of graft versus host disease following myeloablative allogeneic hematopoietic cell transplantation (Maziarz et al., 2015) to be a kind of hard reboot for the immune system where a dangerous inflammatory state, akin to a cytokine storm or sepsis, shows promise for mitigation and shorter ICU time and favorable at-risk tissue/organ response.
On July 4, 2020, an article published in the Lancet’s E-Clinical Medicine Journal showed promising results for the use of adipose-derived MSCs in the treatment of 13 COVID-19 patients requiring mechanical ventilation. Sánchez-Guijo et al. (2020) report that the treatment, which was first administered at a median of 7 days after mechanical ventilation, showed no adverse effects in treated patients. At a median of 16 days following the treatment dose, 9 of the 13 patients showed clinical improvement, including seven patients who were extubated and discharged from the ICU. Patients received between one and three doses of the treatment, and the immunological profiles of the patients were monitored throughout the course of the study. For the improved patients, inflammatory markers began to steadily decline after 5 days following the administration of the dosage. A decrease in C-reactive protein was observed in 8 of the 9 patients. A decrease in LDH was observed in all 9 patients, and levels of D-dimer and ferritin were also reduced in 5 of 8 of the patients tracked for these markers. The authors also report that improved patients showed increasing counts of lymphocytes, and all patients showed an increase in CD4+ and CD8+ T cells. Randomized, placebo-controlled Phase II interventional clinical trials are already underway in Spain to further test the safety and efficacy of the use of MSCs in severely ill COVID-19 patients.
On April 2, 2020, the FDA approved Cellularity’s CYNK-001 therapy, a natural killer (NK) cell therapy, for investigational new drug (IND) use in a new Phase I/II trial on 86 patients with COVID-19. CYNK-001 doses contain natural killer cells, a type of white blood cell with cytotoxic properties, that were developed from human placental CD34+ hematopoietic stem cells. NK cells are activated during infection from a broad range of viruses, and researchers propose that the treatment may have efficacy in slowing the replication of SARS-CoV-2. Furthermore, these NK cells may show potential in the targeted destruction of infected cells. CYNK-001 is also currently undergoing clinical trials investigating its use in treating acute myeloid leukemia, multiple myeloma, and glioblastoma multiforme. Some researchers caution that the use of NK cells may have a pro-inflammatory effect however, which may contribute to deteriorating COVID-19 symptoms, and so studies concerning the safety of the therapy are especially important.
Convalescent Plasma Transfusion
On August 23, 2020, the FDA issued Emergency Use Authorization for the use of convalescent blood plasma in the treatment of hospitalized COVID-19 patients. This decision was made after the review of months of data supporting that the benefits of the treatment outweighed any potential risks. Convalescent plasma transfusion is a medical treatment that uses blood plasma extracted from patients that have recovered from a disease as a means of treating patients with the disease. The convalescent blood plasma, which contains antibodies specific to the pathogen, is transfused into the ill patient in an attempt to boost the patient’s immunity. The method has been successful in the treatment of various viral diseases over the past century.
Shen et al. (2020) published results on five critically ill COVID-19 patients with ARDS (all using mechanical ventilation) with continually high viral load despite treatment with antivirals and methylprednisolone (a corticosteroid) who received blood plasma containing SARS-CoV-2 specific IgG antibody. The study was very small with no controls, and so limited conclusions can be drawn from it, but it showed a potential link between the plasma transfusion and reduced viral load, as well as improvement in clinical symptoms. All five patients remained stable after 37 days of the transfusion, and three no longer required mechanical ventilation two weeks following the transfusion. Four out of the five patients had their body temperatures return to normal within three days of the transfusion, and viral loads dissipated to undetectable levels within 12 days, while serum antibody increased. Duan et al. report loss of viremia by 7 days for 10 severely ill COVID-19 patients (median time from initiation of symptoms to blood transfusion was 16.5 days) that received one dose of 200 mL of convalescent plasma. The study was non-randomized and all patients were already receiving antiviral treatment, as well as a standard course of hospital treatment. The authors report that all of the patients showed improvement of clinical symptoms within 1-3 days of the transfusion, including some patients who were extubated from mechanical ventilation or no longer required high-flow nasal cannula. Chest CTs revealed visible signs of reductions in pulmonary lesions, resulting in improved pulmonary function. Five of the ten patients showed an increase in neutralizing antibody titer, and SARS-CoV-2 viral load was undetectable in all but one patient 3 days after the transfusion (the final patient cleared all viral load by 6 days).
Salazar et al. (2020) describe promising findings for the use of convalescent blood plasma in treating 25 patients with severe or life-threatening COVID-19 in Houston, Texas. All patients were receiving oxygen supplementation and many were receiving anti-inflammatory and antiviral treatment. Throughout the study, clinical improvement was assessed on a modified WHO 6-point scale. Seven days after transfusion with convalescent blood plasma, nine patients had improved by at least 1 point on the scale, and a total of seven patients had been discharged from the hospital. Fourteen days after blood transfusion, 19 patients showed at least a 1 point improvement, and a total of 11 had been discharged. The authors report no adverse effects from the blood plasma transfusion for any of the patients. While the results seem promising, the study size was relatively small and the study did not include a control arm, so limited conclusions can be extrapolated.
U.S. Expanded Access Trial
Several clinical trials have investigated the therapeutic benefit of convalescent blood plasma transfusion. On April 8, 2020, the U.S. FDA announced an Expanded Access Program to provide convalescent blood plasma transfusions for COVID-19 patients. The multi-institutional effort is led by the Mayo Clinic, and as of May 15, 2020, 16,498 COVID-19 patients had been administered the treatment. An initial safety report in 5,000 patients from this program was published on May 14, 2020 (Joyner et al., 2020). The incidence of severe adverse effects, including death, reported in the first few hours of blood transfusion was less than 1%, while death was reported in 0.3% of cases. The seven-day case fatality rate for the 5,000 critically ill patients in the study was 14.9%, but the authors conclude that this fatality rate is not excessive given the population pool from which the patients were studied, who were at higher risk of death from COVID-19. The authors conclude that the results indicate that the treatment is safe.
On August 12, 2020, Joyner et al. went on to release preliminary results of the Expanded Access Trial. The report of the Multicenter Trial, which included 2,807 acute care facilities, included results collected from 35,322 adult COVID-19 patients with severe or life-threatening illness who were treated with at least one unit of convalescent blood plasma between April 4 and July 4, 2020. At the time of plasma transfusion, 52.3% of the patients were in the intensive care unit and 27.5% were receiving mechanical ventilation. For patients receiving convalescent blood plasma within three days of COVID-19 diagnosis, the seven-day mortality rate was 8.7%, and for those receiving the transfusion 4 or more days after diagnosis, the seven-day mortality rate was significantly higher, at 11.9%. Thirty-day mortality rates followed a similar trend: 21.6% of patients had died if receiving treatment within three days of diagnosis, while 26.7% of patients died after receiving transfusion on or after 4 days post-diagnosis. The authors also identified a correlation between increasing IgG titer in the blood plasma samples and 7-day and 30-day survival. In particular, patients receiving a high IgG titer, which was defined as having a signal to cutoff ratio greater than 18.5, had an 8.9% 7-day mortality rate. Patients receiving a mid-level IgG titer (with a signal to cut off ratio between 4.62 and 18.45) experienced an 11.6% 7-day mortality rate, and patients receiving a relatively low IgG titer (signal to cutoff ratio < 4.65) experienced a 13.7% 7-day mortality rate.
The CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) Cas system is a widely used gene-editing tool used in molecular biology that has broad consequences in disease treatment. CRISPR itself is comprised of a group of short nucleotide sequences (often 20-30 bp long) that are complementary to a target piece of DNA or RNA. These sequences were identified in the genomes of bacteria and archaea but originated from bacteriophage genomes, the genetic codes of viruses that once infected the prokaryote. CRISPRs are used as a protective form of immunity in prokaryotes against future viral infections. The more widely known CRISPR-Cas9 system uses Cas9, which is an enzyme that has the ability to unwind and cleave DNA. As Cas9 unwinds DNA, it is guided by a CRISPR sequence that is complementary to a specific strand of nucleotides, at which point the enzyme can cut the phosphodiester bond holding the string of nucleotides together. As the enzyme does this, it inactivates various genes in the target sequence, which results in the downregulation of proteins for which the sequences code. For this reason, the system can be used to inactivate DNA viruses. In 2017, F. Zhang of the Broad Institute, developed the CRISPR-Cas13 system, which uses Cas13, an RNA endonuclease that specifically cuts single-stranded RNA, instead of Cas9. This breakthrough broadened the use of CRISPR biotechnology to target RNA sequences and thus cleave and inactivate RNA viruses. The method also shows potential as a diagnostic method for detecting RNA viruses, such as those in the coronavirus family.
On March 14, 2020, Abbott et al. published a result showing that a CRISPR-Cas13 method that the research team dubbed PAC-MAN (Prophylactic Antiviral CRISPR in Human Cells) was able to significantly downregulate the expression of genes in the SARS-CoV-2 and H1N1 genome in a cell culture of infected human A591 epithelial lung cells (note that the cells were infected with either a synthetic form of SARS-CoV-2 or a live H1N1 strain using lentiviral transduction methods). Furthermore, the team identified six crRNA sequences that were highly conserved across all coronaviridae and could be found in 91% of all coronaviruses; they also identified a set of 22 crRNA sequences that have the potential to target all known coronaviruses. For the strains of SARS-CoV-2 that had been sequenced by publication, the team also found one highly conserved crRNA sequence that could be used in all but one of 202 strains of SARS-CoV-2. The method uses Cas13d, a specific form of the Cas13 RNA endonuclease that the researchers identified as most effective for catalyzing the cleavage of SARS-CoV-2 RNA. Since the method uses crRNA specific sequences that are highly conserved across all coronaviridae, the method shows promise as a method of inactivating viral activity and replication across a broad spectrum of coronaviruses, including those that may arise from future mutations.
A February 18, 2020 editorial (Nguyen et al., 2020) has suggested a virus-against-virus approach for treating COVID-19 as well as diseases from new coronaviruses or other viral pathogens. Using an AAV vector and a CRISPR/Cas13d technology, the authors propose to target patient infected lungs early and cleave the SARS-CoV-2 RNA genome and therefore destroy its ability to reproduce and translate.
Many patients with severe pulmonary manifestations of COVID-19 develop a syndrome of cytokine release (CRS) similar to what has been reported in patients with SARS-CoV-1 and MERS-CoV pneumonia as well as patients who have undergone Chimeric Antigen Receptor T-Cell Immunotherapy (CAR-T cell therapy). These involve activation of proinflammatory macrophages which then secrete cytokines such as IL-1/IL-6/TNF-a which may be involved in this severe lung damage. One approach to combat this is to use agents such as tocilizumab which inhibit IL-6.
An alternative strategy is to use low dose ionizing radiation (LDRT) to the lung in these patients. LDRT has been shown to have significant anti-inflammatory activity, and there have been anecdotal reports of its use in treatment of pneumonia in the past (pre-COVID-19). These studies have been recently reviewed and suggestions made to conduct a trial of LDRT (0.5 Gy in a single fraction) to both lungs (Lara et al., 2020; Rödel et al., 2020). This dose is well below that typically associated with lung damage, and its major potential toxicity might be a low rate of induction of cancer (estimated at 1%) with a latency of 10 years or more. An initial trial assessing the acute toxicity and feasibility of such an approach has been initiated at the Emory Cancer Institute in Atlanta, GA (Williams, 2020; ClinicalTrials.gov Identifier NCT04366791). They have recently reported their experience with the first cohort of patients treated on this protocol (Hess et al., 2020). Five patients who were residents of nursing homes, had COVID-19, with pneumonia and requiring supplemental oxygen but not ventilator-dependent received 1.5 Gy to both lungs. Two additional patients were consented but deteriorated rapidly and were not treated. Four of the five showed rapid improvement and had been discharged (3) or were preparing for discharge by Day 14. No toxicities were noted, and the trial continues.
Ameri et al. (2020) have reported their initial experience with LDRT for five patients with COVID-19 pneumonia in the International Journal of Radiation Oncology.Biology.Physics. Their patients were over the age of 60, hospitalized and receiving supplemental oxygen, and all had one or more comorbidities. Four of the 5 had PCR confirmation of infection. They received ‘standard therapy’ (though none received remdesivir, dexamethasone, or hydroxychloroquine) and a single dose of 0.5 Gy radiation to both lungs delivered 1-3 days after hospitalization. Of the five patients treated, one worsened and died on the third day, four showed improvement in O2 saturation and temperature within one day of treatment. One patient opted out of the trial on Day 3. The other three patients were discharged on Days 5, 6, and 7, although one remained on home O2. No acute toxicities were reported. The authors concluded despite the small patient numbers the results of this and the trial at Emory indicate that LDRT ‘may be a successful treatment with severe COVID-19 pneumonia’ and are continuing their trial at a higher radiation dose of 1.0 Gy.
While the results of these two trials are encouraging in not reporting acute disasters, they are hardly adequate to exclude more subtle deleterious effects on the immune response to COVID-19 or to establish significant benefits. As of mid-August 2020, clinicaltrials.gov lists 9 currently accruing trials of LDRT in COVID-19 patients in the U.S., India, Italy, and Spain with another 5 trials active but not currently accruing or pending but not currently accruing. These are predominantly Phase I-II assessing safety and feasibility in this population
The radiobiologic and clinical rationale for these studies has not been universally adopted by other radiation oncologists, who find the reliance on old, largely anecdotal, studies worrisome and are concerned that the impact on LDRT might worsen the ability to respond appropriately to viral infection rather than simply mitigate cytokine storm (Kirsch 2020, Salomaa 2020). They also raise concerns that the use of LDRT is logistically more complex than pharmacologic therapies, may lead patients enrolled in these studies to be ineligible for other potentially more promising agents, and has some risk of inducing long-term harm to the heart and lungs, including a small risk of radiation-induced malignancy. They have suggested additional studies in animal models before extending this therapy to humans and argue that current data do not warrant even limited Phase I-II trials.
Potential Nutritional Supplements
While the following nutritional supplements exhibit certain features that may merit further investigation, no rigorous studies have yet shown any speculated outcomes in COVID-19 patients.
Botanical Polyphenols. There is an extensive literature on alkaloid compounds, including antioxidant polyphenols present in botanicals that can target particular pro-inflammatory gene/protein pathways involved in onset and progression of a variety of degenerative and neoplastic diseases, as well as pathogenic infections (for a review, see Steindler and Reynolds, Adv. Nutr., 2017) including COVID-19. Severe tissue inflammation, especially involving the lungs, has been suggested to be involved in COVID-19, contributing to respiratory distress as well as the attenuation of important immune responses that together lead to cell loss and tissue damage. Likewise, activation of the innate and adaptive immune systems by SARS-CoV-2 can lead to a cytokine storm where particular anti-inflammatory compounds and immune system enhancers can mitigate the dangerous consequences of such an overactive immune response.
Glucosamine. High dosage of glucosamine may upregulate mitochondrial antiviral-signaling protein (MAVS), which is a protein involved in signalling an early immune response to RNA viral infections. At least one study has reported that a diet rich in glucosamine increased survival from influenza infection, caused by another RNA virus, in mice (McCarty et al., 2020).
Lectins (e.g., Elderberry, Stinging Nettle Root, Lentils, etc.). One area of active research involves the potential use of carbohydrate binding proteins present in particular plants that have the ability to bind to receptors present on the virus surface (Keyaerts E. et al., 2007), and that can affect the viral replication cycle and/or cell penetration. Generally recognized as safe compounds from such a list can be strategically designed to act in some instances as frontline, or perhaps more routinely as adjunctive therapies for standard of care or emerging molecular therapeutics while they are being repurposed or approved in human clinical trials.
Melatonin is a hormone released by the pineal gland, but also by other organs, and is known for its role in the regulation of circadian rhythm. As a supplement, it is commonly used as a sleep aid, although its efficacy in this regard is somewhat controversial. Melatonin has been widely reported as a powerful antioxidant that is particularly effective as a free radical scavenger of various oxygen and nitrogen radicals (Galano et al., 2012). Because of its antioxidative properties, it has shown some efficacy in fighting viral infections, as well as in treating sepsis (Srinivasan et al., 2012). These effects are attributed to its inhibition of pro-inflammatory mediators, which it does through a speculated immunomodulating pathway. Melatonin also reportedly inhibits calmodulin, and calmodulin is associated with the inhibition of the removal of ACE2 extracellular proteins, which may participate in the intracellular entry of SARS-CoV-2 (Zhou et al., March 2020). Melatonin’s efficacy in treating COVID-19 has not been studied, but the indirect action it may have on ACE2 could make it a possible candidate for future clinical research.
Vitamins C may boost the production of Interferon and the function of phagocytic immune cells, thereby enhancing antiviral action.
Vitamins D comprises a set of lipid-soluble compounds that are perhaps most well-known for enhancing calcium absorption, as well as the absorption of other essential minerals that support bone health. However, a 2017 review and meta-analysis of 25 studies with a total of 11,321 participants found that vitamin D supplementation is protective against acute respiratory disease infection (Martineau et al., 2017). In particular, the analysis found that supplementation with vitamin D3 (cholecalciferol) or D2 (ergocalciferol) was most protective in individuals with lower initial 25-hydroxyvitamin D—a product of vitamin D3 that is converted by the liver—levels, and supplementation reduced acute respiratory disease risk in all participants studied. Note that because vitamin D can be produced in the skin upon exposure to UVB radiation, individuals with low exposure to sunlight may be at higher risk for vitamin D deficiencies. A review from 2006 highlights that in studies where participants were given the influenza virus, 25-hydroxyvitamin D also prevented an overactive inflammatory response. Furthermore, supplementation with this vitamin was associated with increased expression of antimicrobial peptides in various immune cells and in the epithelial cells of the respiratory tract, bolstering a protective immune response against a variety of respiratory illnesses (Cannell et al., 2006). Taken together, the data supports that Vitamin D supplementation can boost immunity and lower overactive inflammatory response in various respiratory infections. While vitamin D itself has not been shown to have any direct action against viruses, its global effect as an immune booster and anti-inflammatory may prove to be efficacious in the prevention of COVID-19 disease. A large scale data analysis by Daneshkhah et al. (2020) suggests that this may be the case. For patients with severe Vitamin D deficiency, the risk of severe COVID-19 symptoms was 17.3%, whereas for patients with normal Vitamin D levels, risk of severe disease was 14.6%. The data, which was compiled from COVID-19 patients in many different countries, indicates that patients from countries with lower mean 25-hydroxyvitamin D had higher case fatality rates from the disease. This is true even when only countries with similar levels of testing were compared (as countries with high testing rates may show lower case fatality rates since mild cases are more likely to be captured in screening) and when patients within matching age groups were compared.
Zinc Chelate. Zinc inhibits RNA dependent RNA Polymerase, necessary for replication of RNA viruses, such as SARS-CoV-2 (te Velthuis et al., 2010). Some commonly available chelates include zinc gluconate, zinc citrate and zinc acetate. Carlucci et al. (2020) conducted a retrospective observational study of 932 COVID-19 patients admitted to NYU Langone Hospital that were treated either with hydroxychloroquine, azithromycin, and zinc sulfate (411 patients) or with hydroxychloroquine sulfate and azithromycin alone (521 patients). The addition of zinc sulfate was not found to be associated with a number of factors, including length of hospital stay, duration of mechanical ventilation, and oxygen flow rate. However, use of zinc sulfate was found to be associated with a decrease in mortality and a decreased need for the ICU or mechanical ventilation. The authors also report that patients treated with zinc had higher baseline lymphocyte counts. The study demonstrates the first in vivo evidence of the potential efficacy of zinc in treating COVID-19 patients. Future randomized, double-blind studies testing this hypothesis should be conducted to support or refute these initial findings.
- Note that because vitamin D is a lipid-soluble vitamin (and hence can be stored in the body for a longer period of time), a form of overdose known as hypervitaminosis D, while rare, is still possible. Therefore, adhering to proper dosing recommendations from a medical specialist is essential. Foods fortified with vitamin D and sun exposure will not affect vitamin D toxicity (as levels of vitamin D in fortified foods are too low and the body can regulate vitamin D production in the skin), but extremely large doses of vitamin D supplements can lead to hypercalcemia (excess calcium) in the blood, which may have deleterious effects on other systems in the body.