COVID-19, caused by SARS-CoV-2, represents a serious and urgent unmet medical need. Clinical trials are essential to provide evidence for the efficacy of any new therapy against COVID-19. However, clinical trials in epidemic situations present several challenges.
In case of COVID-19, the selection of drugs for the trials would be from agents used in SARS-CoV or MERS-CoV because of their apparent in vitro activity, e.g. hydroxychloroquine, lopinavir-ritonavir as well as other anti-retrovirals, remdesivir, favipiravir and the new anti-cytokine IL-6 receptor antagonist, tocilizumab. However, supporting lab or animal data against COVID-19 are unlikely to be available.
Some of these drugs could be repurposed for COVID-19, but may not work against the new virus in the approved dosage regimen. For example, the failure of lopinavir against SARS-CoV-2 in the LOTUS China trial could be due to the high concentration required to inhibit viral replication, compared to the relatively low concentration required in HIV patients. Higher doses and longer duration of treatment could be justified if there is support from lab and animal studies. But such doses also increase the risk of adverse drug reactions.
In an infection with high mortality, clinical trials are conducted in hospitalised, severely affected adult patients. However, such patients are less likely to respond to any therapy, as they have suffered considerable multi-organ damage. Delayed initiation of treatment could also be a reason for the lack of demonstrated efficacy in the LOTUS trial. Hence, a clinical researcher should include COVID-19 patients who have had the disease for a shorter duration. Of course, this will require a rapid turnaround of lab data – viral load, organ function tests, imaging, ECG, etc. — and a well-trained site team available at odd times to complete the study procedures for the enrollment of patients.
If a study is planned in a paediatric population, deciding the dose would be difficult, as dose-finding and safety data in children may not be available. In such a case, a weight-based dose could be extrapolated from the adult dose range.
In epidemics, the acceptable design would be a randomized comparison between investigational therapy + standard therapy and standard therapy alone. As quality placebos are not available, such studies employ an open design. Hence, there would be a potential for information bias in the assessment of clinical efficacy endpoints. Viral load reduction may also be difficult to assess because of the limits in sensitivity and specificity in the new diagnostic tests and any discordance between detection of virus from nasal and oropharyngeal swabs.
As randomization poses challenges for the patients to consent to be in the non-drug group, the trial could be a single arm, open-label and non-randomized one in which all patients receive the investigational therapy along with the standard care. However, the results of such non-randomized study do not provide evidence for the efficacy of the investigational therapy.
The detection and assessment of adverse drug reactions in the trial of COVID-19 is also difficult because of the severity of illness, the use of a variety of medical procedures, underlying comorbidities, and the use of concomitant treatments.
Therefore, the success of a clinical trial in COVID-19 depends on addressing basic questions of science related to the choice of therapy, the selection of patients and the validity of the study design.
Writer is a consultant on clinical research & development from Mumbai.