Gottlieb and McClellan Critique FDA’s Requirements for COVID Clinical Trials
[Thursday, June 25, 2020]

​In a rare criticism of FDA’s requirements for clinical trials, two former commissioners of FDA, Scott Gottlieb and Mark McClellan, argue that FDA should allow practical trial designs of the kind used to evaluate dexamethasone treatment in COVID patients in the UK where data collection was “limited to most important measures of disease severity and outcomes” rather than the elaborate protocols requiring extensive data collection typically required by the FDA for COVID-19 trials. The dexamethasone trial, called the Recovery trial, compared a range of possible treatments with usual care in patients hospitalized with COVID-19. The open-label, randomized study evaluated dexamethasone 6 mg given once daily for up to ten days vs. usual care alone. The Recovery trial has only one primary endpoint; estimate of the effect of study treatments on all-cause mortality. The rather simplistic design, with limited additional data collection requirements, allowed a team of more than a 1000 physicians, located at 176 clinical centers, enroll 11,320 participants of which more than 2000 were randomized to dexamethasone treatment to find that in the most sick COVID-19 patients, dexamethasone treatment improved outcome by more than 35%, reducing the total mortality rate by 17%. That is very comparable to the effects seen in the same population with remdesivir. Dexamethasone was approved, by the regulators in UK, the same day the results were announced. Gottlieb and McClellan argue that the study design for the Recovery made participation easier and hence yielded quick results (less than 4 months after the start of the trial). They contended that “the protocols of trials in the U.S. are more elaborate, which means less enrollment and results take longer.” It is an easy argument to be made. FDA’s IND requirements would have likely asked for non-clinical safety studies and smaller, more restricted, patient population for the trial along with extensive additional tests and data requirements, since dexamethasone has never been formally tested in COVID patients. FDA did not support the Recovery trial design citing lack of data collection on all the variables that could influence outcomes, and this was a mistake. In an emergency setting where physicians are overwhelmed with medical care tasks, asking to track and record additional data is unreasonable. FDA needs to consider clinician’s perspectives over theoretical risk assessments. Just like the UK, several other clinical trials with various potential therapeutic and vaccines for COVID are ongoing in other countries. In a race to find a cure for the virus or prevent disease, FDA needs to allow clinical trials with lower threshold so we can get to the goal sooner. Otherwise, the US may be playing catch up. Even former FDA chiefs, who know much about the internal workings of the Agency, agree.