In this interview in honor of Multiple Myeloma Awareness Month, Dr. Rahul Banerjee, an Assistant Professor at the University of Washington and Fred Hutchinson Cancer Center, and member of the Oncology Data Advisor Editorial Board, speaks with Dr. Manni Mohyuddin about the importance of designing clinically relevant end points for clinical trials in multiple myeloma.

Rahul Banerjee, MD: Welcome to Oncology Data Advisor, a digital resource for the multidisciplinary cancer team. My name is Dr. Rahul Banerjee. I'm one of the editorial board members. Because March is Multiple Myeloma Awareness Month, I'll be interviewing several myeloma experts across the country to discuss important topics in myeloma, for both patients and providers, that we often don't talk about as much. Today, it's my honor to speak with Dr. Manni Mohyuddin, an Assistant Professor at the University of Utah Huntsman Cancer Center, who is an expert in clinical trial design and patient end points as they pertain to clinical trials in myeloma. Dr. Mohyuddin, it's very nice to have you here. Thanks for taking the time to do this.

Manni Mohyuddin, MBBS: The pleasure is mine. I would not consider myself an expert, but thank you for those kind words.

Dr. Banerjee: Of course. Well, I think you're taking on a topic that is very important to all of us in terms of clinical trials and making sure that the clinical trial results are as relevant to our patients as they can be. I don't know of anyone who really taken on this issue with as much passion as you have.

My first question I can ask you is, when you see a clinical trial in multiple myeloma that's published—or let's say if you're advising a patient or patient advocate who's reviewing a trial that just got published in some journal somewhere—what are some things that go through your mind when you're reviewing a clinical trial in terms of things that make a trial good, things that make it not so good, things that make it relevant to patients or not so much, or so forth? What's your rubric for going through a clinical trial?

Dr. Mohyuddin: Right, that's such a great question. I have a different rubric and a different set of standards for a randomized phase 3 or even a large randomized phase 2 trial compared with a smaller trial, but let's say it is a randomized trial. One of the first things that I look at is the question being asked and what the study is powered to show. Obviously, what the end point is is very relevant. We can talk more about this, but the end point generally in myeloma is progression-free survival or response rate; very rarely is it overall survival. Obviously, that is one important thing.

The second thing that I look at is what is it being compared with? I think there's a lot of context that goes into that, because one of the unfortunate realities of our clinical trials in myeloma is that we don't know how to sequence therapies. We don't know, amongst all the options that we have, what is the best to use? I'll give you an example. When patients relapse—let's say they got VRd (bortezomib/lenalidomide/dexamethasone), and they relapse—we have so many different trials of three versus two drugs that have shown that three drugs are better for the end point being studied. But we have zero trials that compare different three-drug regimens at first relapse. It sort of does put us in a fix where we have all of these different options. I don't personally get excited if I see long-term follow-up of three versus two or if I see a new three versus two design for first relapse.

Obviously, what the control arm is, is very pertinent and very, very relevant. There are a lot of caveats that go into that, which we can spend a lot of time talking about. Then it is also very important to consider what the inclusion criteria are. Myeloma is a very heterogeneous disease. Sometimes the inclusion criteria will be in such a way that a certain disease biology will be excluded. Sometimes it's not even written in the inclusion criteria, but just how the trial is designed disincentivizes people from putting certain patients on it. Looking at the patient characteristics of what sort of patients were enrolled and looking at the inclusion criteria on what patients they were intending to enroll are very important in helping us contextualize what the benefit is.

Then there are a few things that I feel like our field doesn't emphasize enough, but post-protocol therapy is also particularly important and a topic that's close to my heart. There is attrition in myeloma, and I do recognize that giving good treatments upfront is important and makes a difference. But there's also a problem that in our clinical trials, when people do progress, they don't get standard treatments—standard meaning what could have been given in the US, where the company's trying to seek approval for it. This becomes particularly important when it comes to overall survival, because that's very dependent on what sort of therapy is given. This is by no means an exhaustive list of the things that I look at, but they're definitely some starting points, and we can sort of go on about this for a long time.

Dr. Banerjee: You're exactly right about the post-protocol therapy. The way that I've learned to understand it—for example, with daratumumab or a CD38 antibody—really anything that you shouldn't be testing, giving it versus not giving it at all, you really should be giving it now versus giving it later. Is that a good way of approaching things in a randomized trial?

Dr. Mohyuddin: That is an excellent question. When a drug is already proven to be as effective in the relapsed/refractory setting—and daratumumab undoubtedly is effective in that setting—the relevant question to ask becomes, is it better to give this drug earlier versus is it better to give it later? The question should not be, is it better to give this drug early versus not at all. What unfortunately has happened in several myeloma trials looking at daratumumab upfront is that these trials ran after we knew daratumumab was effective. It's within the sponsor's ability to give daratumumab upon progression. The trials should have been data in the newly diagnosed setting versus daratumumab at relapse, right?

Dr. Banerjee: In the protocol, not per investigator discretion.

Dr. Mohyuddin: Exactly, with guaranteed access to those patients who progressed and were still alive and still illegible for receiving treatment. That's what should have been done, but unfortunately that's not what was done. If you look at, for example, the ALCYONE trial—which is daratumumab, bortezomib, melphalan, and prednisone versus bortezomib, melphalan and prednisone—only about 10% of patients who progressed on bortezomib, melphalan and prednisone received daratumumab, which is a very dismally low number.

Even if you look at the MAIA trial—which is, perhaps I would argue, more relevant to the United States, which compared lenalidomide and dexamethasone to daratumumab, lenalidomide, and dexamethasone—in the lenalidomide/dexamethasone arm, when patients progressed, the top three most commonly used regimens of progression were all non–daratumumab-containing regimens. The other interesting thing is we got an overall survival advantage, and we saw that early. We got an overall survival advantage so quickly for these newly diagnosed trials, but we still haven't had published results showing an overall survival advantage in the relapsed/refractory setting. That's crazy. You'd think that you'd see an advantage much earlier, and all of this just boils down to post-protocol therapy.

Because in POLLUX and CASTOR, which are the trials of daratumumab in the relapsed/refractory setting, there was very high crossover. They found out that the drug worked, so they gave it to the control arm upon progression. I have not been able to come up with a good answer for why they didn't really do that in the newly diagnosed setting. It's a very complicated topic. I think the answers to these are complicated as well. I think the solutions to this are complicated, and many people can come up with many different excuses. How I think about this is that the problem is very simple; even if the answer is complicated, it's important to recognize the problem. The problem is that our patients don't get the therapy that they could have gotten upon progression in these technical trials. As a field, we historically have been agnostic to post-protocol therapy.

As a perfect example, nobody doubts how good VRd is, right? It's been our de facto standard. I'm just giving you an example of how agnostic we are towards post-protocol therapies. When the long-term follow-up of this trial was published, there was a separate paper looking at long-term follow-up. It didn't have a very simple, basic fact about how many people in the lenalidomide/dexamethasone arm got bortezomib upon progression. That was nowhere in the supplement. I looked really hard before I went public with this fact in my research paper, but nobody knows. As a field, as far as the reviewers or the editors—or maybe it wasn't even collected—this just goes to show that it's not a priority for us.

I am passionate about raising awareness for this. Hopefully there's a piece that will come out, but there are a lot of vulnerable populations in oncology, like racial minorities. There's so much work that needs to be done to improve outcomes for those patients and to level the playing field. There are centuries and decades of injustices that we have to correct. But I also think that the patients on the control arm of clinical trials are a vulnerable population. Despite all of the bureaucracy, the institutional review boards (IRBs), all of those things, those patients remain vulnerable. I feel like not enough people talk about their rights and advocating for them. I would like to consider myself an advocate for all patients, but especially for those that are on the control arm of randomized trials that get overlooked when it comes to what the control arm is getting and what post-protocol therapy they get.

Dr. Banerjee: I totally agree. Again, you are the expert in this topic, and you have written and published extensively on it, so thank you. Let me pivot to something you briefly touched on. Another one of your papers was just about end points. I think myeloma is unique among malignancies, at least in my mind, in that our end points are not just radiographically based. They are very complicated measures of plasma protein surrogates, what the myeloma cells are doing, and so forth. Overall survival is not just what we have typically used. Can you talk through what your thoughts are for not just myeloma experts, but for other health care providers or for patients who are looking at a trial like this, or an ongoing trial? What should they make of that progression-free survival (PFS), time to next treatment, overall response rate, and so forth? Do your thoughts about these surrogates, or putative surrogates, depend on the type of trial you're looking at?

Dr. Mohyuddin: That's such a great question with so many different aspects to that answer. Historically, the gold standard—and that's what we've been taught, that I very strongly believe—is that our treatments should make people live better or live longer. Live longer means the treatment should prolong overall survival; live better means that it should improve quality of life. Now at first thought, you'd think that if a treatment improves progression-free survival—which is defined as the time from trial initiation to when the disease progresses—is based arbitrarily on a certain M-protein value going up or light chains going up or a time from treatment initiation to progression or death, whatever the cause of death is. You would think that a treatment that delays progression improves quality of life, but I think that it's a lot more complicated than that.

This has been shown across other cancers, as well, that just delaying progression doesn't necessarily mean that it also improves quality of life. That, in part, reflects the arbitrary cutoffs that we have. I'm really glad that we have uniform definitions of response and progression, but we have to recognize that they are arbitrary. When you move from a stable disease to a partial response, there's nothing magical about that which would make a patient feel better. Having understood that the historical gold standard is overall survival and quality of life, we also have to recognize that overall survival takes a long, long time. As much as I advocate for overall survival, I do recognize that for newly diagnosed myeloma, if you want to design a certain intervention, it's going to take a really, really long time to get to your median. If you're looking at a median overall survival of, 7, 8, 10 years, that's a really, really long time to wait.

I recognize that there is a need for surrogate end points that will give us quicker answers. I also recognize that industry is heavily incentivized to have those end points and get their products earlier. I also recognize that end points like progression-free survival, which I just defined earlier, or response rate, or measurable residual disease (MRD)—which is a highly sensitive indicator of how much cancer is left behind—none of these basically account for post-protocol therapy, which is important in myeloma, where people can live through multiple lines of therapy. It's a complicated situation.

In newly diagnosed myeloma, I recognize that it is impractical at times to use overall survival, and surrogates are needed. But I think it's also important to recognize that these surrogates are imperfect, and they are a high-profile example where improvement in a surrogate has not led to improvement in overall survival. It's a tough question. I recognize that in newly diagnosed settings, overall survival may not be practical. Now if you look at patients with heavily relapsed/refractory disease, it's interesting, and it's almost paradoxical that these papers that are testing heavily relapsed/refractory patients are going to cite a study that will say that survival for this patient population is less than a year; but the same time, they're not going to use an overall survival end point. I think that is a little problematic.

I do think that for heavily relapsed/refractory disease, where people have had multiple lines of treatment, overall survival is practical. The other thing is that when you're talking about sequencing treatments, I think that you should either have overall survival or something called PFS2, which is basically PFS1, which I've already described, plus the time to the second progression or death. It accounts for what treatment is given afterwards. I think that when we're talking about sequencing or bringing a treatment up earlier, PFS-2 is practical, as well.

Dr. Banerjee: Just to segue to your earlier point about post-protocol therapy, theoretically, if patients had access to the experimental drug after progression, PFS2 would account for that.

Dr. Mohyuddin: Right, exactly. Then I see there's a lot of excitement about measurable residual disease (MRD), and I am extremely excited about using MRD to adapt treatment. We've historically treated myeloma pretty much the same. Whether you are an exceptional responder or whether you responded just a little bit, you were treated the same with the same regimen for the same duration of therapy. Using MRD to adapt treatment is really, really exciting. I love the trials that we have ongoing looking at that, and maintenance strategies, et cetera. But I think that there are a lot of problems when MRD becomes the sole goal or the sole objective of treatment. There are some high-profile examples, and you could argue that the MRD from a decade ago is not the MRD of today.

Dr. Banerjee: That's true.

Dr. Mohyuddin: There are some high-profile examples; we always use the IFM 2009 trial, which I know you love and you talk a lot about, as well. That's a trial of bortezomib/lenalidomide/dexamethasone plus autologous transplant upfront versus bortezomib/lenalidomide/dexamethasone alone. Everybody got lenalidomide maintenance. Basically, it was a comparison of using transplant upfront or not. The important thing to know is that 78% of patients that didn't get transplant upfront got transplant later. So what the trial ended up becoming was early transplant versus late transplant. I would argue that's what the daratumumab trial should have been, as well: early dara versus later dara.

Getting back to the point, in this trial, there was an improvement in progression-free survival. For the MRD of that time, which is not the MRD of today, there was actually much more MRD negativity in the patients who received transplant. But the overall survival at the eight-year mark was similar between both arms. It sort of goes to show that in a disease like myeloma, especially for fitter young patients who may go through other lines of therapy, just having more MRD negativity doesn't translate to survival. You could argue that if you're looking at curative strategies for myeloma, your best chance at curing is upfront. Maybe quads plus transplant will make the difference. Maybe we'll have a better overall survival, and maybe the increased MRD negativity in this setting with more sensitive MRD negativities will translate to overall survival.

I think that there are a lot of assumptions here, and I'm not sure I'm jumping fully on this bandwagon. I just want us to recognize that there are a lot of assumptions. I think a historical perspective of myeloma would prove that surrogates have not always translated to improved outcomes. We all know this, but it's always important to highlight that there are some high-profile examples in the relapsed/refractory setting where drugs have improved progression-free survival but worsened overall survival—venetoclax in the BELLINI trial being one example, and then melflufen more recently in the OCEAN trial being another example.

Again, I am not saying that we need to make overall survival an end point for everything, but I think we need to recognize the imperfections of some of our end points. We need to call things for what they are and not overhype MRD or overhype PFS. I just want to keep things real and want us to be aware of these limitations.

Dr. Banerjee: Totally agreed. I agree on all fronts. One last point: the final question I'll ask relates to how you described MRD in two ways. One is as an end point, as you talked about seeing it as a surrogate for overall survival. Then you also touched upon what I would call MRD as a middle point: using MRD to decide to intensify or de-intensify treatment. You mentioned, for example, autologous transplant. I think the MASTER trial is a good example of that, albeit single arm. What studies are you excited about that are either currently ongoing or planning to launch soon in terms of using MRD to intensify or de-intensify treatment, and why?

Dr. Mohyuddin: For sure. That's such a great question. The trials that I'm most excited about use MRD to de-escalate treatment for patients who are responding really well. One of the examples for that in the maintenance setting is the DRAMMATIC trial of two years of maintenance therapy with either lenalidomide or daratumumab/lenalidomide depending on what arm you are randomized to. For patients that are MRD-negative, they can be randomized to be put off treatment or continue on therapy. That is really exciting. Within a randomized trial where you are following these patients closely, you're getting answers that will help all of humanity. You are able to take people who are MRD-negative off treatment, whereas the current standard of care in the United States is that patients get lenalidomide long-term after a transplant. I'm incredibly excited about that.

The other thing that I'm incredibly excited about is for patients who achieve a really deep response with induction therapy. You give them quad, and they become MRD-negative. Can we forgo an autologous transplant for that patient population? There are two trials that are looking at this. One is the MASTER-2 trial, which will enroll in the United States. Then the other is the IFM 2020 or the MIDAS trial in France. Both trials are asking multiple questions, and I may not agree with all of the arms being used for all of these trials. But the one thing I love about both these trials is that after induction for people who are MRD-negative, they're randomized to either a transplant-based strategy or just additional cycles of consolidation therapy without a transplant.

That will answer really important questions about whether we can skip an autologous transplant and not subject our patients to the toxicity of high-dose chemotherapy if they're responding deeply. I'm very excited about that. I am a lot more wary about using MRD to escalate treatment, to give more treatment for people who are still MRD-positive, but I am very excited about deescalating treatment for people who are MRD-negative.

Dr. Banerjee: That's a very good nuance, and that's an excellent point. Instead of going above a standard of care, you're trying to find people where you can keep them in a good spot but avoid those toxicities. The one last plug I'll add is for the DRAMMATIC study. If anyone wants to look online, it's SWOG 1803, so it's sponsored by the Southwest Oncology Group. I think it's a good example of scientific investigative clinicians coming together to answer a question like this with multiple randomizations that I think will, as you said, help all of humanity, and most especially help us figure out what to do with our patients.

Dr. Mohyuddin: Exactly. For sure.

Dr. Banerjee: Any closing thoughts you have or anything else you wanted to add?

Dr. Mohyuddin: Well, thank you for inviting me. I think it's a very exciting time for young investigators like us to be in this field. I think there are a lot of things to look forward to. There are also a lot of things that are hyped up, so it's important to be real and mindful about our interventions. I keep saying that I have to be real and not get hyped up, but I'll admit that it is an exciting time to be a myeloma doctor, and it's a privilege to take care of our patients and to be able to offer them these trials and all these exciting interventions.

Dr. Banerjee: Absolutely. Thanks again for everything, Dr. Mohyuddin. Pleasure to speak with you. Again, thank you all for listening to this show for Oncology Data Advisor. Thanks again.

Dr. Mohyuddin: The pleasure is mine.

About Dr. Mohyuddin and Dr. Banerjee

Manni Mohyuddin, MBBS, is an Assistant Professor in the Division of Hematology/Bone Marrow Transplant at the University of Utah Huntsman Cancer Institute, where he specializes in the treatment of multiple myeloma. His research interests focus on quality of scientific methods, drug repurposing, cost effectiveness, and evidence-based myeloma care.

Rahul Banerjee, MD, FACP, is an Assistant Professor in the Division of Medical Oncology at the University of Washington (Seattle, WA); he also holds a faculty appointment at the Fred Hutchinson Cancer Center. He previously completed his Hematology/Oncology Fellowship and Advanced BMT/CAR-T Fellowship at the University of California, San Francisco. His clinical interests are in multiple myeloma, AL amyloidosis, and CAR-T therapy. His research interests are in toxicity management, digital health, and the patient experience.