ABSTRACT
Introduction: The urgency of finding effective treatments for severe forms of COVID-19 has necessitated adaptations to clinical trial designs. Almost all of the initial randomized trials had to be open-label and placebo-free. We were able to seize the opportunity to conduct a double-blind placebo-controlled trial by integrating ourselves into an ongoing trial and offer here some lessons from this experience. Material and methods: The CAPE-COD trial (NCT 02517489) sought to show a decrease in day 28 mortality in patients admitted to critical care with community-acquired pneumonia and receiving hydrocortisone compared with placebo. The inclusion of patients with COVID-19 was possible but required urgent responses to several problems: i) choose between subgroup analysis or segregation of COVID patients from others;ii) simultaneously include in both trials or discontinue the parent trial;iii) adapt methods to achieve a rapid response, including changing the primary endpoint;iv) define the maximum size and stopping boundaries of a sequential trial in the absence of solid data on disease progression;v) quickly obtain ethical and regulatory approvals and funding;vi) conduct the trial in overcrowded centers;vii) not jeopardize the rigor of the parent trial despite an interruption in its recruitment [1]. These elements will be discussed orally. Results: 150 patients were included before the DSMB suggested that inclusions be stopped due to the pre-publication of RECOVERY. Hydrocortisone did not significantly reduce the primary endpoint, treatment failure (defined as death or persistent respiratory support) at day 21, but a post-hoc analysis showed that relative mortality at day 21 was 46% lower (14.7 vs. 27.4%) with hydrocortisone compared to placebo, a non-significant difference (p = 0.057) but with an effect size comparable to that observed in RECOVERY. Tolerance was good [2]. Discussion/Conclusion: The rapid response to many methodological problems allowed a trial with low risk of bias to be conducted despite the pandemic pressure.