Processes for Scientific and Ethical Approval of Research Proposals: Better Oversight May Be Necessary.

Sir,

Research proposals pass through two levels of scrutiny—one, for scientific merit and the other, for ethical standards. These processes are not independent; scientific reviewers also examine ethical issues and ethics committees (ECs) also consider scientific issues. Both processes are expected to be rigorous because clinical research involves time and money, patient comfort and safety, and the extension, with integrity, of scientific knowledge.

In this context, we draw the attention of readers to the retrospectively registered protocol of an apparently completed industry-sponsored randomized controlled trial (RCT) that is publicly available at the Clinical Trials Registry—India (CTRI), bearing registration number CTRI/2013/11/004127. This protocol was discussed as an activity in eJCIndia[1] in early April 2020.

The RCT examined the efficacy, tolerability, and safety of vilazodone in Indian adults with major depressive disorder (MDD). According to details available at the CTRI website, the study protocol was approved by the Drug Controller General of India (DCGI) in September 2013, and by the ECs in five centers between October 2013 and January 2015; in four other centers, the protocol was still under EC review at the time of CTRI registration, although the study recruitment status was recorded as having been completed. This means that the trial was conducted at only five of the nine listed sites because the remaining sites could not have recruited patients without EC approval.

The study was an 8-week RCT. There were three arms: vilazodone, dosed at up to 40 mg/day, escitalopram, dosed at up to 40 mg/day, and placebo. Several serious scientific and ethical concerns are immediately apparent from a reading of the protocol. First, with a stated sample of 375 patients, the study was strikingly overpowered for an efficacy RCT (vilazodone vs placebo, with escitalopram as the internal control for assay sensitivity); this means that the study was unethical because more depressed patients were exposed to placebo than was scientifically necessary. However, if the study was powered for a noninferiority design, it would still be unethical because a placebo group is not part of such a design.

Second, the authors dosed escitalopram at a target of 40 mg/day; this is unconscionable because the maximum licensed dose of the drug in MDD is 20 mg/day. Last but not least, the RCT was stated to be open-label. Does this mean that patients in the placebo group were actually told that they were receiving a placebo?

This is an example of a situation where multiple layers of oversight failed. What needs to be done to prevent the recurrence of such failures among investigators, in ECs, and under the supervision of the DCGI is a matter for perhaps more than just introspection.