As the Food and Drug Administration (FDA) begins its transition to a new administration, attention has focused on how the organization could shorten the process for approving medical treatments. Some suggestions have targeted the FDA’s standards for approving new treatments, saying that the current requirements are too stringent.
Perhaps as a response to this attention, the FDA recently published a paper called 22 Case Studies Where Phase 2 and Phase 3 Trials Had Divergent Results. In these case studies, promising research results for new treatments fell short as the scale of the studies increased and the objectives narrowed.
In these 22 examples, Phase 3 study results—the larger randomized control trials (RCTs) that aim to prove whether something works—differed from the results of smaller, safety and proof-of-concept oriented Phase 2 studies.
Here are some of the products the FDA described.
- A stent and drug combination to relieve emphysema symptoms. The extensive treatment, which included a tube down the throat and inserting a stent in a lung, did no better than a sham surgery.
- A drug aimed at increasing levels of “good” (HDL-C) cholesterol. The drug improved scores, but in larger studies, serious cardiac incidents increased.
- A reformulated treatment for schizophrenia. Safety risks developed when the treatment was redesigned as a long-acting pill instead of an injection.
How Necessary Are Large-Scale RCTs?
Practically speaking, a key element in considering an appropriate standard for approval is determining the need for extensive RCTs. The FDA consistently describes RCTs as the “gold standard” for showing whether something works. But here, “gold” can mean more than quality; it also represents cost.
RCT results depend on statistical evidence, and meaningful statistical evidence depends on getting the numbers right. A protocol might need thousands or tens of thousands of participants to accrue sufficient statistical strength. This scale comes with high cost, not only financially but also in terms of timeline. Standards that allow fewer or smaller RCTs could lead to faster approvals, but the FDA’s paper suggests some concerns there.
The recently passed 21st Century Cures Act (Cures) includes a section (“Utilizing Real World Evidence”, section 3022) that aims to reduce the need for RCTs by defining circumstances under which they may not be necessary. For example, Cures directs the FDA to explore how real-world data—defined as “data regarding the usage, or the potential benefits or risks, of a drug derived from sources other than randomized clinical trials”—could justify a new use for an existing drug. The FDA has three years to assess how those real-world observations and data sets might diminish the current reliance on RCTs.
Is There Still a Place for RCTs?
If we were to reduce the need for RCTs (or simplify the primary objectives for approval), then getting a drug to market would become faster and less expensive. The economics of this sort of shift are undeniable, but the value of large-scale RCTs is critical to consider. The 22 cases that the FDA reported (six of which involved treatments that were already approved) show circumstances where large RCTs provided crucial revelations about both safety and effectiveness.
At best, approvals of those treatments could have placed ineffective treatments into the health care system. At worst, the treatments could have made a condition worse.
While the FDA’s current system is not flawless—areas for improvement doubtless exist—the report’s real-world examples suggest large, well-controlled studies remain an important component of the drug development process.