Max Parmar, professor of medical statistics and epidemiology at University College London, unpacks what clinicians and patients can gain from the application of multiarm multistage (MAMS) trials.
Max Parmar, professor of medical statistics and epidemiology, director of the MRC Clinical Trials and the Institute of Clinical Trials and Methodology at University College London, elaborated on the benefits multiarm multistage (MAMS) trials can provide to clinicians and the well-being of patients. Parmar spoke about these models in his session regarding more intelligent trial designs at the European Society for Medical Oncology Congress 2023.
This transcript has been lightly edited for clarity.
Can you provide an overview of the clinical significance and implications of multiarm multistage trials?
The basis for this sort of idea is that our rate of making progress in many cancers, in fact many diseases, is just too slow. And often, that's because we don't show at the phase 3 stage the new intervention is better than our current standard or control group. And the trouble is, that takes us a long time to get there to show that, and after many, many years, and often hundreds of millions of pounds of investment, we find we've made no progress. We need to speed up the way that we evaluate interventions, and more effectively find the ones that work and reject the ones that don't. So the idea of the multiarm multistage trial is to rather than just have 1 new intervention compared against the control, or a standard group, have a number of them. And not to just add a number of them, but at various stages during the course of that project to add new interventions to compare, and drop interventions which don't look promising enough that are likely to show an improvement in outcomes.
One example of that is the one that I'm going to give in prostate cancer, which we’ve been doing over now 18 years as a single protocol and a single project in which we've compared 10 interventions against control. We've shown 4 of them to be better than control, and, as a consequence, improve the outcomes for men with metastatic prostate cancer for a median survival of 3 years to a median survival now of 7 years, which is more than doubling the survival. If you think of any other way that we might have evaluated all of those 10 interventions over that period of time, it would have taken us a generation to do that. We've done that in 18 years.
What medical or scientific problems would you say benefit the most from these designs?
I think a priority for these sorts of desires are those diseases where we've made little or no progress for many decades. So an example was prostate cancer, when we started, we had made no progress in the treatment of prostate cancer for 40 years, we'd had the same treatment for 40 years, and men were dying, on average, 3 years after their diagnosis of metastatic disease. That's just a terrible indictment of where we'd got to; of course, that position has changed now.
Similarly, those areas where we have made little or no progress for many, many decades, are ripe for this simple design. Examples include neurodegenerative diseases, where we know actually there are very, very few disease-modifying agents in disease like motor neuron disease, Parkinson disease, Alzheimer disease, and progressive multiple sclerosis. We’ve made little or no progress in those areas for decades. And so we need to change how we evaluate things. Similarly, in terms of cancer, and specifically which this Congress is aimed at, there are a number of cancers which we've made little or no progress for many decades: pancreatic cancer, not a great deal of progress; esophageal cancer, not a great deal of progress. So we really need to change the game in those areas.