EHA 2025 | Evaluating clinical trial results in hematology: the importance of the overall survival outcome

So I think when you look at a randomized controlled trial in hematology, you’ve got to look in a stepwise fashion: is the treatment population similar to the patient I look after in the clinic, is the control arm treatment the appropriate standard of care for that patient, are the trial design elements favorable to a fair and reasonable result, and then I think very importantly, you’ve got to look at the overall survival outcome. Now, as we now have many diseases where patients live a long time, think chronic lymphocytic leukemia, multiple myeloma, follicular lymphoma, these are all diseases where it’s not really feasible anymore to have the primary endpoint of the clinical trial to be overall survival. But that doesn’t mean that overall survival isn’t really important. Firstly, it’s important as a safety endpoint. Overall survival is the ultimate safety endpoint because if something happens, we’re not sure if it’s related to the treatment or not, but we’ve randomized, it will tell us if the treatment is truly safe or not. So I think overall survival is very important from a safety point of view, but also to look at efficacy to make sure that there’s not something that’s missing in the clinical trial. What do I mean by that? Well, let me give you a couple of examples. So if you look, if you compare the SHINE trial and the TRIANGLE trial, both of these trials are in frontline treatment of mantle cell lymphoma. They both look at adding ibrutinib, a Bruton’s tyrosine kinase inhibitor, to frontline chemoimmunotherapy for mantle cell lymphoma. One in the elderly population with bendamustine rituximab, the other in the younger, fitter population with chemoimmunotherapy and autologous stem cell transplant. And I think it’s really important to look not just at progression-free survival, the primary endpoints of failure-free survival, progression-free survival, the primary endpoint, but to go on and look at what happens in overall survival. And we see that in the SHINE trial, only about a third of patients in the control arm crossed over to receive ibrutinib after progression. Whereas in the TRIANGLE trial, about 80% of patients who went on to receive a subsequent treatment received ibrutinib as post-progression care. And why is this important? Well, because in the SHINE trial, there was no overall survival benefit. In fact, there is potential that there might have been an overall survival decrement. But we can’t know that because there was insufficient crossover of those patients in the control arm to receive ibrutinib after progression, which is what they should have done under standard of care. Compare that with the TRIANGLE trial where there was quite good crossover. It wasn’t built into the trial, but it happened as part of standard of care. And despite that good crossover, there was still an overall survival benefit. What that tells us is that this is really a practice-changing study that we really should think very hard about moving that Bruton tyrosine kinase inhibitor into the frontline setting to ensure that those patients get the overall survival benefits. So the post-progression care is critical to understanding the overall survival outcome, and of course, overall survival is one of the most important things to patients. So another example of the importance of post-progression care in overall survival assessment is highlighted by two of the major CAR T-cell therapy trials, the Zuma 7 trial and the TRANSFORM trial, both randomized controlled trials in the second line setting, aggressive lymphoma, comparing novel CAR T-cell products, axi-cel and liso-cel respectively to chemoimmunotherapy followed by transplant in patients responding to chemoimmunotherapy. But if you look at them simplistically, you might say, well, in Zuma 7, there’s a statistically significant overall survival benefit, but in TRANSFORM, there’s an overall survival trend but not a statistically significant benefit, and you would say that prompts you to ask the question, you know, is one product better than the other, and I think this is a very flawed question. Let me give you one example of why. In the Zuma 7 trial, crossover, that is patients in the control arm crossing over after progression to receive axi-cel, was not incorporated in the trial; in fact, you had to come off the trial and be lucky enough to live in a place that already had access to CAR T-cell therapy. Conversely, in TRANSFORM, the patients in the control arm, not only was crossover allowed in the trial, but they were all collected for those CAR T-cells before they were at the time of randomization. So those CAR T-cells were ready to go if they happened to have disease progression later, and that was shown if you compare the number of days it took the patients in the intervention arm to get CAR-T with those patients who crossed over from the control arm in TRANSFORM to receive CAR-T cell therapy at progression; they got their CAR-Ts much faster. So not only was the proportion of patients about 80% in the control arm of TRANSFORM compared to about 50% in the control arm of Zuma 7, so a much higher proportion of control arm patients crossed over to receive CAR-T after progression, but they also got that CAR-T cell therapy much faster. And that is important because the less disease you have at the time of CAR-T, we know the better the chance that CAR-T has of killing a lymphoma. So I think if you look at the Zuma 7 and TRANSFORM trials side by side, I think one must be very cautious when saying is one product better than the other, because there are multiple trial aspects I haven’t touched on all of them, but one of those aspects is the cross-over that helps us better understand how to interpret that overall survival data.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.