For faster FDA approval of a new drug, do bigger clinical trials help?

July 18, 2017 1:33 PM

Assuming all trials of a new drug show efficacy and side effects well within FDA requirements. All else being equal, do larger trials shorten time-to-approval?
posted by LonnieK to Science & Nature (6 answers total)

Possibly, but possibly not. It is often assumed that a larger N is better and provides more certainty of both effect and safety, and this tends to be true, but it is also possible to over-power a study, and to see an effect when there is not one, which may hamper FDA approval. So the answer is that it is a somewhat complicated statistical issue that depends on the other variables at play.
posted by Lutoslawski at 1:43 PM on July 18, 2017 [1 favorite]

Not necessarily. There's a bunch of other stuff that has to be proven in the process of FDA certification, and I mean like very non-study stuff like manufacturing scaling proof of concept and ingredient sourcing and equipment and an audit of the financials.

Generally you only do as large a trial you absolutely must, to satisfy FDA requirements for your methodology. It's not the size of the trial, it's the numbers you get back, and the bigger the trial the higher the cost, and the bigger the risk. On the other hand, a larger trial might buy you a skin-of-the-teeth pass rate where a smaller trial wouldn't, and the FDA is (at the moment) full of people who know how to math and care how the math gets done.

It is a really ponderous process all around, and the trials are one aspect of it but not the entirety.
posted by Lyn Never at 1:45 PM on July 18, 2017

Trials take a long time. One of the parts that takes a long time is enrolling patients. It takes months to get a trial enrolled, and then you get to wait for the results. For something like cancer trials, it takes months into years to see whether the treatment actually makes people live longer.

The only part of that process that can be shortened is enrollment. In theory, you can throw money at that problem and enroll patients more quickly by working with more centers, but that wouldn't get you a lot. Having more patients on the trial makes enrollment take longer, and it doesn't get you to your endpoints any faster. So if anything, more patients would slow things down.
posted by Winnie the Proust at 2:21 PM on July 18, 2017

If "larger trials" means "a more diverse representation of the patient population," then, yes, there will be fewer questions from reviewers about underrepresentation. If "larger trials" simply means "numerically more volunteers," probably not.

There's quite a lot of inter-reviewer variability at the agency and attempts to identify systematic approaches to shortening time to approval are viewed a bit like the corporate snake oil of the day. If a company is wiling to pay someone to tell them the Secrets of Rapid Approval, an entrepreneur will eventually fill the niche and start selling that company whatever "secrets" they can muster.
posted by late afternoon dreaming hotel at 2:53 PM on July 18, 2017

Response by poster: Whoa .. 2 hours in, 4 great answers. Thanks all.
posted by LonnieK at 4:24 PM on July 18, 2017

Other than the power of the study, a thing under consideration is what the clinical trial is for. It is my (only slightly educated) understanding, that the FDA has a fast track for a rare disease clinical trials (and there might not be enough people in the whole world suffering from the disease to make for a large sample anyway). Orphan Drug Act (sl)
posted by Dotty at 7:49 AM on July 19, 2017