A roundtable discussion, moderated by Thomas Martin, MD, of the UCSF Helen Diller Family Comprehensive Cancer Center, focused on CAR T-cell therapy considerations in the treatment of multiple myeloma, including data on approved CAR-T options and a look at the pipeline. Dr. Martin was joined by a panel that included Sagar Lonial, MD, FACP; Peter Voorhees, MD; and Shambavi Richard, MD.
In the next segment of the roundtable series, the panel debates whether CAR-T can be moved earlier in the course of myeloma treatment, including a discussion on the KarMMa-2 and KarMMa-3 trials assessing idecabtagene vicleucel in different clinical scenarios and the CARTITUDE-2 cohorts receiving ciltacabtagene autoleucel.
Dr. Martin: Dr. Lonial mentioned that in China they have a dual-targeted chimeric antigen receptor (CAR) with BCMA and CD19. They tried it, it was Juan Du, MD, as frontline therapy. The question then becomes what space should we start moving these CAR T cells? We’ve now seen some recent data on moving these CAR T cells to earlier patient populations. Peter, maybe you can give us a little update on the KarMMa-2 and KarMMa-3 trials. What results we’ve seen so far with using CAR T cells, not in the last-ditch effort, but in the earlier lines of therapy.
Dr. Voorhees: KarMMa-2 is a study looking at idecabtagene vicleucel (ide-cel) in a number of different clinical scenarios. One of the important clinical scenarios that was presented by Saad Usmani, MD, at ASH back in December, is a group of patients with early relapse. These are functionally high-risk patients. This was cohort 2a of the KarMMa-2 trial, and these are patients who had to have progression of their multiple myeloma with 18 months of start of induction therapy, transplant, followed by lenalidomide maintenance therapy. So, 39 patients underwent leukapheresis; 37 patients received the ide-cel product. Not surprisingly, these functionally high-risk patients were enriched for high-risk biology. So, 12 out of the 22 evaluable patients had high-risk cytogenetics and four of them had more than one high-risk cytogenetic abnormality. Importantly, only 24% of the patients had achieved a complete response (CR) as their best response to their induction transplant maintenance course. And 90% of these patients had progressed within one year of transplant. I’ll get back to that in a moment.
In this group of patients, the overall response rate was 84%. The median duration of response was 15.7 months. The median progression-free survival (PFS) was 11.4 months. Now at first glance, I looked at this, and I was somewhat disappointed, but you have to remember the patient population that you’re looking at. There’s really nice data from the MRC 11 trial out of the UK that looks at patients with early relapse within 12 months of autologous stem cell transplant, and their median overall survival is only 26 months. In this particular cohort of patients, the two-year overall survival is 85%. I think you have to think about the disease population, the patient population here, and I actually think that these are really impressive results.
Then there’s the KarMMa-3 study. This was a phase III study that randomized patients 2:1 to either ide-cel or investigator’s choice. Investigator’s choice, and you can quibble with what was chosen, but we’re confined by regulatory and approval constraints, but this consisted of daratumumab/pomalidomide/dexamethasone, elotuzumab/pomalidomide/dexamethasone, daratumumab/bortezomib/dexamethasone, ixazomib/lenalidomide/dexamethasone, and carfilzomib/dexamethasone. Now as far as eligibility criteria, these patients had to have two to four prior lines of therapy. They had to be daratumumab-, immunomodulatory drug (IMiD)-, and proteasome inhibitor (PI)-exposed, and they had to have had progression on or within 60 days of their last regimen. As far as baseline disease characteristics, they had received three median prior lines of therapy, which is a smaller number of lines of therapy relative to the initial KarMMa experience. But 90% of these patients had IMiD-refractory disease, 95% were daratumomab-refractory, yet two of the standard of choice regimens had daratumumab in them, and 75% of patients were proteasome inhibitor-refractory.
When you look at triple-class refractory, two-thirds of the patients in both arms were triple-class refractory. Interestingly, 42% to 46% of patients had high-risk cytogenetic disease. Earlier relapse, but a high-risk patient population. I think we have to take that into consideration, and not surprisingly, ide-cel won with regard to overall response rate, 71% versus 42%. When you look at stringent complete responses, it was 35% versus 5%. When you look at progression-free survival, standard-of-care arms was 4.4 months versus 13.3 months for those that got ide-cel, translating into a hazard ratio of 0.49, reducing the risk of progression or death by 51%.
Again, at first glance I looked at this, and I was a little disappointed that the progression-free survival was not as long as what I might have expected in an earlier relapsed patient population, but this is just barely earlier relapse relative to the KarMMa-1 and KarMMa-2 experiences. I think that this potentially gives us the ability to use ide-cel in an earlier line of therapy, and we can talk about this later, but insisting that patients have four prior lines of therapy before they can access either ide-cel or ciltacabtagene autoleucel (cilta-cel) is incredibly challenging. The ability to get this one or two lines earlier should be a game changer in my view.
Dr. Martin: Yeah, that’s a really big deal. Because once you get past the fourth line of therapy, many patients fall out and say, okay, I’m done. I don’t want any more therapy. You lose a lot of patients that could benefit from such very aggressive and effective therapies.
Similar data, although it’s a little bit behind with cilta-cel. Dr. Richard, do you want to tell us a little bit about the cilta-cel data?
Dr. Richard: CARTITUDE-2 is a multicohort study now looking—similar in parallel with KarMMa’s in a way—in earlier lines of therapy and therefore functionally high-risk population, etc. Cohort A was designed for patients who had progression of disease after one to three lines of therapy, as opposed to two to four with KarMMa. One to three lines of therapy and lenalidomide-refractory.
Cohort B was for early relapse, so relapse less than 12 months after initial induction, whether that included just plain induction or transplant in the induction treatment, whether they were transplant-ineligible or -eligible, respectively. Cohort C was for relapsed/refractory myeloma who had had in addition to failures post-PI, IMiD and anti-CD38, was also post-prior-BCMA therapy, so very relevant to today’s age with all these BCMA therapies going on.
Cohort D was suboptimal response, so less than CR after induction and transplant. This cohort actually includes lenalidomide maintenance after CAR-T. Cohort E was for high-risk, newly diagnosed patients, no transplant intended. These E and F were for high-risk and standard-risk, newly diagnosed, essentially replacing transplants, and E because they were high-risk patients, includes lenalidomide maintenance again after the CAR-T. Those were the six arms in CARTITUDE-2. We have some data from the first three arms thus far.
Cohort A, this was initially reported by Adam Cohen, MD, at ASH in 2021 and Herman Einsele, MD, last year at ASCO. In fact, they had one patient on this, I think it was treated as an outpatient, so 20 patients on this study. The results were very similar to CARTITUDE-1, and in fact interestingly this is now being taken to study further in CARTITUDE-4, which I’ll talk about in a second. The median follow-up on this was 17.1 months, and interestingly more than half had one or two prior lines of therapy, 40% for three prior lines of therapy, so really moving to earlier lines of relapse; 95% were refractory to the last line of treatment. Predictably, 40% were triple-class refractory, 35% had high-risk cytogenetics, and overall response rate was very impressive at 95%, deep responses. Very good partial response (VGPR) or better was all of them, 95%, CR or better at 90%, and stringent CR for 85% of patients; 80% of these patients were minimal residual disease (MRD)-evaluable, and they were all MRD negative. Essentially 80% of the overall group was MRD negative. The median duration of response, again, not surprisingly is not reached, but the 15-month progression-free survival was 70% on this arm. Again, the toxicities were as expected. Six of the 20 patients did have neurotoxic events, but 15% were immune effector cell-associated neurotoxicity syndrome and the others were other neurotoxicities. But no Parkinson’s-like syndromes were seen in these neurotoxicities. There were four deaths on this study. Two from progression of disease and two from infections.
Cohort B was presented by Niels van de Donk, MD, at ASCO last year and this was now intended for early relapse as I mentioned, 12 months or less after initial treatment, whether that was transplant or nontransplant induction with functionally high risk. The normal expectation for patients like this would be a median overall survival of less than two years; 19 patients were enrolled on this study with a median follow-up of 13.4 months. Now with CARTITUDE-2, I should mention that management strategies were implemented to minimize risk of what we had seen with the CARTITUDE-1, in terms of Parkinsonian symptoms and things like that. So, 79% of these patients had a prior transplant and now the overall response rate on this arm was 100%, 95% was VGPR or better, 90% was CR or better, and 93% of these patients were MRD negative.
Again, very impressive. Median duration of response, not reached, but 12-month progression-free survival was 90% on this. Again, in this study there was one grade 3 Parkinsonian symptoms, and as I mentioned, from the CARTITUDE-1, we had learned that there were certain risk factors to delay neurotoxicity, including progressive disease at post-bridging, or high tumor burden, or high grades of cytokine release syndrome, and this patient did have all of these risk factors, not surprisingly. There was one death due to progression of disease on this arm.
Cohort C was for the prior BCMA. This is by definition just a much higher, more resistant group of patients; 20 patients enrolled, 13 had a prior antibody-drug conjugate and seven had a prior bispecific antibody. So, 80% of them were refractory to prior BCMA therapy; 90% were triple-class refractory, 55% were pentarefractory. There was a median follow-up of 11.3 months. Again, not surprisingly for this group, overall response rate was less than 60% and the median duration of response was 11.5 months, and the median progression-free survival was nine months. All of this is not unexpected with this very refractory group of patients; 70% of patients were MRD negative, also slightly lower. There were seven deaths on this arm. Three from progression of disease, one from treatment-emergent adverse events, and three from other adverse events that were deemed to be not treatment related as per investigator. Interestingly, there was some information, because this then becomes relevant for future sequencing, is who are the people who responded less well and [it] was found that patients who had shorter duration of prior BCMA therapy and had other non-BCMA therapy intervening between the past BCMA and this study, or a longer interval from the prior BCMA. These were the patients who tended to do a little better.
Responders also had a longer time from BCMA to the apheresis as well. I think all of this was very interesting and important information that we gleaned from this arm thus far. In terms of CARTITUDE-4, again this is a very similar group of patients as in the cohort A of CARTITUDE-2, but this is a randomized, phase III study of cilta-cel CAR-T versus standard of care. There were just two possibilities for this standard of care versus dealer’s choice in terms of KarMMa-3. This was standard of care with either daratumumab/pomalidomide/dexamethasone or pomalidomide/bortezomib/dexamethasone. These patients were lenalidomide-refractory; 419 patients were enrolled on this, all prior PI and IMiD exposed and relapsed after their last treatment line within six months. We don’t have formal published results on this, but earlier this year Janssen did announce that the study had met its primary endpoint of significant PFS improvement at the first pre-specified interim analysis, so eagerly waiting for a formal data presentation.
Dr. Martin: Yeah, I think everybody’s really awaiting those data and that is going to be certainly an earlier population than what Peter presented with the KarMMa-3 study.