COVID-19 trials and tribulations for treatments

COVID-19 has impelled the scientific community to research and develop new therapies to counter a serious viral pandemic. Since 2020, many old drugs repurposed as treatment for COVID-19 have been studied in clinical trials to assess efficacy and safety.  As per PubMed, 6013 clinical trials on COVID-19 treatments have been published, of which only 22.9 per cent (1377) were randomised controlled trials. However, from these only a few high-quality clinical trials have provided evidence of efficacy of treatments – oxygen, dexamethasone, tocilizumab, remdesivir – for COVID-19. Many other clinical trials of drugs have been publicised by the media. But the quality and validity of most such trials are in question. Conducting clinical trials which meet globally expected scientific and ethical standard is always challenging but more so during a pandemic. The need to rapidly announce results in media seem to have affected the design and conduct of clinical trials of COVID-19 treatments.

For clinical trials of COVID-19 treatments to be valid and useful, the sponsor company or the clinical research investigator should consider the following in planning (USFDA May 20):

• Population – moderate to severely ill patients with RT-PCR positive test

• Standard of care – current accepted of treatments – oxygen, steroids, remdesivir, tocilizumab, intensive care management, heparin, antibiotics etc as background therapy

• Design – Randomised comparison of standard of care + repurposed drug vs standard of care, blinding of the investigator team assessing the efficacy endpoint, placebo control if feasible

• Efficacy endpoints – clinically meaningful based on clear definitions and specific clinical criteria e.g., mortality, need for oxygen or ventilator or hospitalisation or intensive care management, sustained clinical recovery

• Duration – Efficacy assessment after adequate duration of at least 28 days

• Sample size – large enough to provide a reliable answer to the safety and efficacy questions

When we review clinical trials of drugs for COVID-19 conducted in India, we see many deviations from these scientific requirements. Recently, 2-Deoxy-D-Glucose was launched for the treatment of COVID-19. The press release mentioned that the drug helped in faster recovery of patients hospitalised with COVID-19, reduced their dependence on supplemental oxygen, and showed a favourable trend in median time to achieving normalisation of specific vital signs parameters compared to standard of care. However, in absence of a peer-reviewed publication in a reputed journal, it is difficult to judge the validity of these trials and the risk-benefit of this drug in seriously ill patients. Some obvious questions remain regarding the design, and conduct of 2-Deoxy-D-Glucose trial:

1) Definitions of endpoints – clinical criteria

2) Process of Randomisation and allocation concealment in an open trial subject to selection bias from the investigators

3) Avoidance of clinician-observer bias in the evaluation of clinical endpoint

4) Statistical method of estimation of sample size

5) Statistical methods to compare the efficacy and safety endpoints in two arms.

For 2-Deoxy-D-Glucose, one major concern would be safety in COVID-19 patients who can suffer from myocarditis and develop diabetes mellitus. A Phase I trial in patients with advanced solid tumours (Cancer Chemotherapy and Pharmacology volume Dec 2012) had reported serious adverse events – hyperglycemia, gastrointestinal bleeding, and grade 3 QTc prolongation.

In 2020, a meta-analysis of 23 clinical trials of COVID-19 (BMJ Jul 20) reported that the majority – 21 trials had a high risk of bias:

1) during the randomisation process

2) because of departures from the intended intervention

3) due to missing outcome data

4) in the measurement of the outcome and

5) due to selective reporting of results.

The majority of clinical trials of repurposed drugs in India suffer from a similar high risk of bias. Most have been conducted in a small population of mild and moderate COVID-19 patients, using viral clearance, or clinical scale as efficacy endpoints. Viral load reduction, as an endpoint is difficult to assess because of sensitivity and specificity limits of RT-PCR, and discordance between detection of virus from nasal and oropharyngeal swabs. Also, there is no established predictive relationship between the magnitude and timing of viral reductions and the extent of clinical benefit for a patient.

Several therapies approved internationally – dexamethasone, remdesivir, tocilizumab – have become the mainstay of COVID-19 treatment as these high-quality clinical trials conducted in statistically relevant number- over 1000 patients, employed objective clinical important endpoints, and took care to reduce bias.

During the pandemic, the urgency to disseminate clinical trial results in the lay press without waiting for publication in a peer-reviewed scientific journal can impact the health care system adversely. Speedy regulatory approval of a new drug and wide media publicity using superlatives to describe it as a wonder drug can create high expectations of therapeutic benefit in the minds of patients. This puts tremendous medical-social-legal pressure on the physician to prescribe a new anti-COVID-19 drug, which can lead to misuse or overuse of the new drug, add to the cost of therapy, and increase the risk of adverse events. Acceptance of therapies based on low-quality clinical trials also lowers the scientific image of the industry, academia, and regulatory authorities. It would be desirable for us to investigate Indian innovations in high-quality clinical trials planned based on globally acceptable scientific and ethical principles.