Success in Clinical Trials: Putting Patients First

Dr Mark Kris discusses clinical trial design and the need to put patients first.
www.medscape.com/viewarticle/...
-- TRANSCRIPT --
I'm Mark Kris from Memorial Sloan Kettering, speaking about a letter to the editor in The New England Journal of Medicine in the paper edition on January 25, 2024. In this paper, three investigators from the Netherlands - Dr Dingemans, Dr Smit, and Dr Hendriks - discuss a paper that appeared in the fall of 2023 about a randomized trial comparing selpercatinib with chemotherapy.
They make the point that the benefits in the phase 2 studies for selpercatinib far exceeded what you could expect from chemotherapy. The time to the response rate with selpercatinib in this lung cancer population with RET fusions was 84% and the progression-free survival was 22 months. These are really outstanding values. There's no chemotherapy regimen with or without neoadjuvant immune checkpoint inhibitor that would get close to that.
Despite that, a randomized trial was done. The authors raised the questions of whether there was equipoise in doing this and whether this was an ethical trial. I don't want to get into that aspect of it, but I do want to get into the need to rethink how our trials are done for these highly efficacious medications that have a broader use and good safety data. As you know, selpercatinib is used in other cancers as well. There were safety data available, but with a very clear benefit in the phase 2 setting.
I have to say that, in the past, our regulators - I'll talk now about the Food and Drug Administration in their approval of crizotinib - basically approved crizotinib with any line of therapy, based on approximately 250 patients. They did that because there was such a clear benefit vs any other treatment that was out there. They thought about that in a very clever way. They put patients first and they put this drug out there.
SUGGESTED FOR YOU
We can argue about why these randomized trials are done, but the truth is they're done because investigators like myself put patients on these trials. We agree to do these trials. I reach out to everyone to consider that when there are very dramatic differences between the two treatment arms, you have to be really circumspect and ask whether it's necessary to do this trial to give us the information we need in prescribing this.
It's kind of interesting. The response rate in phase 2 and phase 3 of these two trials was identical and the progression-free survival was nearly identical. I think it was 22 vs 25 months. You have to wonder what the years of effort and expense of this randomized trial really got us. We need to think differently about drugs.
Why is this drug so different? You had a target. You had a target that we could accurately measure. It was a DNA-based target that we could measure very well. Positive tests were always positive. We had a highly efficacious drug against the mutant kinase brought on by that fusion protein. It's a little different from comparing two different chemotherapy drugs, for example.
In summary, I think we need to look very critically at how we design trials for patients where we're testing drugs that have very clear evidence of benefit. I'm fond of reminding people that I don't think there ever was a randomized trial showing that cisplatin was an effective drug for the treatment of testicular cancer. Why? Because it was so clear that it worked. I do believe there are situations where it's worked. Yes, you have to clearly find it. Yes, the benefit has to be very clear.
I'd put out as a rule of thumb that you need at least a 50% response rate to even think that this kind of strategy should be sufficient for getting drugs approved. When you have that, you really need to think about a different way to streamline it, to get it into the pharmacies around the world, and to not put people at risk.
Yes, a crossover design is helpful, staggered, and 2-to-1 randomization is fine. However, it still means that people are exposed probably unnecessarily, in my mind, to these very specific situations. I'm not saying that everything should be done like this, but when you have a highly active drug target and a very clear story, it can be done.
For the crizotinib story, it got that drug approved right away, made it available, and really helped many patients. It's because a large amount of thinking went into that, and I urge that we do that again.
I thank the Dutch authors for writing that letter. It makes us all think about how we do our job and how we can do a better job.
Transcript in its entirety can be found by clicking here:
www.medscape.com/viewarticle/...