How can real world evidence lead to FDA label changes?

How can real world data lead to FDA label changes?

So this is a holy grail of what I want to see with my research career is that using some of these projects that are very near and dear to my heart, to actually change how the FDA approves things, that's a much higher bar to set.

Right, because the FDA obviously, you know, they are very interested in patient safety. They are very interested in well run clinical trials that answer that question. So do I expect real world data to change FDA practices tomorrow? I don't think so. But as I alluded to earlier, one good example is the question of bortezomib, the drug Velcade, where again, twice weekly bortezomib, velcade was used in clinical trials.

And it continues to be used in clinical trials just by inertia just because the old study did it. Therefore this study does it. Therefore this study does it, even though that means more neuropathy for patients. You know, that means more time in clinic. It's like a lose lose lose situation for patients. And so, you know, if you know, the statistical pure way to fix this would to be do a randomized trial of once weekly versus twice weekly dosing and prove in a randomized study that twice weekly bortezomib is worse.

I wouldn't feel comfortable with that study. My patient would not feel comfortable with that study, because I already know that twice weekly bortezomib is worse. And so we are actually going and we're going to be, you know, in the coming year, working with the FDA, hopefully to show that maybe this is a good decision for we have many, many and the biggest real world data I'm co-first author on that was published two months ago in Blood Cancer Journal, a big journal, open access.

Doctor Vicky Hoff is the first author on the Doctor Carr from UT Southwestern is the senior author on it. You're welcome to look at it. Real world data showing that once weekly bortezomib is a standard of care used by most physicians, and it works just as well as twice a week with fewer side effects.

So we're going to take that to the FDA to say, hopefully for future trials, the FDA should allow future trials to use once weekly bortezomib.

And then if the FDA allows it at a trial, that's what they will allow in the label in the actual package insert once it's approved. I will say for rarer disorders. So very rare disorders like poems or amyloidosis, we still tend to prefer randomized studies and in oncology we often are able to do randomized studies, we are. But in some other domains, like very rare, you know, I can think of example of, you know, very rare neurological disorders where sometimes it's not ethical to give someone, you know, do a randomized trial where someone gets nothing or it's not even practical to do that.

Right. Every patient coming to you, they're so rare, we have to put them onto something. And so the FDA has said that real world control arms so looking historically at patients who were treated before the drug became available, and how they did is a reasonable strategy to pursue. Is that the best kind of data? No. But is, you know, are some data better than no data? Yes.

So I think that not so much in oncology. There are scenarios where the FDA may consider real world data or what's called a synthetic control arm, where it's actually designed from real world data, kind of designed to kind of match the patients who were on the trial. for multiple myeloma and for cancers. I don't see that being something that the FDA or any of us will adopt tomorrow. However, like I said, in terms of changing how an existing drug is used, that's where real world data may be very helpful.