Will the FDA Change the Way They Approve Treatments?

The website BioSpace posted a really interesting article a coupe of weeks ago called "How Will FDA’s Pivot to Overall Survival Affect Cancer Drug Development?" 

In my last post, I mentioned that I was a little surprised that the FDA was considering accelerated approval for the bispecific Epcoritamab, and that I had been doing some reading lately about the FDA potentialy changing the way they approve cancer treatments. The BioSpace article is a nice summary of the kind of things I have been reading.

As the article points out, over the summer, there was a workshop at the American Association for Cancer Research conference, co-sponsored by the FDA, that looked at the issue of Overall Survival (OS) in cancer clinical trials. OS is the gold standard in clinical trials -- if the developer of a new treatment can show that it keeps patients alive longer than currently available treatments, it has an excellent chance of being approved (assuming the side effects are not awful). As patients, this is what we want, too -- a treatment that will keep us alive longer.

The problem is, OS is hard to measure in a reasonable time. As cancer treatments have improved over the years, OS has gotten larger. That means newer treatments have more work to do. In other words, if a new FL treatment was going to be approved, it would need to show a better OS than what we have now -- maybe 15 or 20 years. But in order to do that, a trial would have to last at least that long. How else can you show that patients will survive for 20 years without following them for 20 years? 

The answer is to use a "surrogate endpoint." That means finding a different measurement that would signal that OS will be long. For most trials, that surrogate endpoint is Progression Free Survival (PFS). This is a measure of how long a patient goes before the cancer comes back or gets worse. The idea is, if a new treatment can show a longer PFS than current treatments, that's good enough to approve it. 

And that's where there are problems. PFS is often measured in just a few years, maybe even in months -- certainly fewer than the 8 or 10 years that we might get for an OS in Follicular Lymphoma. For some cancers, PFS can be measured in months. But the makers of new treatments don't want to wait that long. They want their treatments to be available as soon as possible, so they can begin making back the money they put into the development of the treatment, plus a big profit.

To be fair, it's not just the drug makers who want the treatment available quickly -- patients want that, too. And that's why the FDA developed accelerated approval procedures. If a new treatment can show an improvement in PFS, it can potentially get approved based on the results of a phase 2 trial (it usually takes a phase 3 trial for approval),with the understanding that a phase 3 trial will come later. Final approval will depend on the phase 3 trial results.

But, again, the problem is that approving a treatment too soon can create some problems. The biggest one is, there may be side effects that come out years after approval. And this happens, unfortunately. In Multiple Myeloma, a treatment was withdrawn a few years ago by the FDA after accelerated approval because over time, it was found that the treatment had a significant increase in the chance of death. And of course, if you've been reading the blog for a while, you know that a similar (though not so drastic) issue came up with PI3K Inhibitors for FL. Problems come out over time.

Some FDA officials (according to the article) have signaled that they would like OS to be more commonly used, with less of a reliance on PFS.  Ideally, a treatment would be approved after there was an Overall Survival benefit, with no safety issues, based on a randomized clinical trial. That is, there would be a fairly large trial with two groups of patients. One group would get the currently used treatment (called the "Standard of Care") and the other would get the new treatment. They would be followed for potentially many years, until an OS was determined. And then the data would be submitted to the FDA for approval.

Will that happen? It's hard to say. The FDA has a tough job. They have to make decisions based on the best interests of patients. But the question is -- is the best interest of patients to have a treatment become available as soon as possible? Or is it to make absolutely sure that it increases survival?

It will be very interesting to see which direction the FDA goes in. Certainly, based on the Epcoritamab decision last week, they are still considering accelerated approval and not relying on OS completely. And in my opinion (for what it's worth -- I'm not a doctor or a cancer researcher), it will probably be something they consider for each individual application. Some cancers are so rare and have so few treatment options that it will be worth the risk to approve something based on PFS and get it into doctors' hands as soon as possible. And for others, maybe less rare and with other options already available, maybe they will be more cautious and want to see the full trial results. 

The question for us is, where does Follicular Lymphoma fall on that scale?