TouchHAEMATOLOGY coverage of data presented at EBMT 2024:
Previous analyses of the DETERMINATION study (NCT01208662) revealed that the significant difference observed in progression free survival (PFS) between patients with multiple myeloma treated with lenalidomide-bortezomib-dexamethasone (RVd) alone versus those treated with RVd + autologous stem cell transplantation (ASCT) varied between African American (AA) and White patients, and that progression free survival PFS prognostic factors, including race and body mass index differed by treatment regimen. In this interview, Dr Paul Richardson (Dana-Farber Cancer Institute, MA, USA) reports the results from his group’s additional exploratory analyses of these findings.
We discuss how the data suggests a potential meaningful PFS benefit in AA patients from RVd-alone. Additionally, we explore safety findings, which reveal a similar safety profile between AA and White patients despite a substantial Duffy-null prevalence in the AA patients, but that neuropathic toxicity may be higher in AA patients. Considering all these findings, we finish by discussing how these clinical insights could influence treatment decisions for newly diagnosed multiple myeloma patients.
Questions:
- Could you provide a brief summary of the DETERMINATION study and explain the key findings from the multivariable analysis? (0:39)
- What did the study reveal about the efficacy outcomes when comparing African-American and White patients treated with RVD-alone versus RVD+ASCT? (8:20)
- What were the observed differences in safety between the two study populations? (17:30)
- How might these findings influence treatment decisions for newly diagnosed multiple myeloma patients? (20:24)
Disclosures: Paul Richardson discloses acting on advisory committees for BMS/Celgene, GSK, Karyopharm, Oncopeptides, Adaptive and Sanofi; and received research funding from BMS/Celgene, Karyopharm and Oncopeptides.
Associated abstract:
Richardson PG, et al. Outcomes in african-american (AA) and white patients with newly diagnosed multiple myeloma receiving RVd-alone vs RVd+ASCT in determination: Efficacy, safety, and second primary malignancies (SPMS). EMBT 2024. Abstract OS10-05
This content has been developed independently by Touch Medical Media for touchHAEMATOLOGY and is not affiliated with the EBMT 50th Annual Meeting, 2024. Unapproved products or unapproved uses of approved products may be discussed by the faculty; these situations may reflect the approval status in one or more jurisdictions. No endorsement of unapproved products or unapproved uses is either made or implied by mention of these products or uses by Touch Medical Media or any sponsor. Views expressed are the speaker’s own and do not necessarily reflect the views of Touch Medical Media.
Transcript:
Hello, everyone. My name is Dr Paul Richardson, and I serve as the, Director of Clinical Research and a Clinical Program Leader of the Jerome Lipper Multiple Myeloma Center, at Dana Farber Cancer Institute in Boston, Massachusetts. And I’m also the RJ Corman Professor of Medicine from Harvard Medical School. And it’s truly my pleasure to be with you, to discuss some of the findings from our analysis of African American patients with newly diagnosed multiple myeloma receiving either RVd alone or RVd + early transplant as part of the DETERMINATION study.
DETERMNATION study & initial findings:
The premise of the DETERMINATION was as follows, that we wanted to evaluate delayed vs early transplant with lenalidomide maintenance, as part of anti-myeloma triplet therapy, given that when this study was designed in 2009, RVd had arrived as a triplet of choice in the upfront setting based upon phase 1/2 studies that had shown 100 percent response rate with excellent tolerability, without transplant.
And in the same context, obviously, transplant remained or remains a standard option in younger patients who are transplant eligible. So this was a partnership with some of my outstanding colleagues in the IFM in particular led by Dr. Michel Attal, and Michel and I designed the trial as RVd, with three cycles of induction emission therapy then stem cell collection.
And for our arm A of the study, transplant was delayed. So there were a further four cycles of RVd given thereafter, so cycles four through to eight, and then lenalidomide maintenance followed. I should emphasize that when we mobilized stem cells, we used hydrocyclophosphamide and GCSF support in both arms. For arm B, transplant was given immediately.
So there were three cycles of RVd given, stem cell harvest transport, and then two cycles of RVd as consolidation followed by lenalidomide maintenance. Now, a key difference between us and our French partners driven primarily by, the directives of the French regulators. There was that lenalidomide maintenance could only be given for one year in the French study because of concerns regarding second cancers. And in our study, based upon the strength of the CLGB 100104 study, which had shown both progression free and survival benefit, overall survival benefit to the use of lenalidomide maintenance until progression, we deployed lenalidomide maintenance until progression.
So the study was designed in 2009, launched in 2010, and actually, we were able to read out our first data on the study in 2022, pointing to the number one extremely important efficacy of the regimen regardless of whichever arm a patient was assigned to, and at the same time, maturity of follow-up, which, was approaching seven years. So with that data in mind, our primary endpoint was progression free survival, and our secondary endpoints were response rates, duration of response, overall survival, quality of life, safety and then a series of subgroup analyses that would be hypothesis-generating, but nonetheless hopefully able to discern who benefit them and who may not.
I should mention that lenalidomide maintenance in our study was started at 10 milligrams per day continuously.
And then after month four was increased to 15 milligrams continuously, based upon again the the strength of the data derived from our pivotal CLGB 100104 study, which showed the clinical benefit of lenalidomide maintenance until progression.
What was particularly important was that our primary endpoint was met. And in this regard, we were able to show that, the RVd followed by early transplant resulted in the median progression free survival of 67.5 months.
In that context, the RVd alone arm performed well but significantly less in terms of its progression free survival estimate at 46.2 months. So there was a difference of almost 21 months between the two arms, which exceeded our expectation of at least a year. So with that in mind, the, hazard ratio generated was striking at 1.53 in favour of early transplant for the the progression free survival endpoint.
What was then very interesting is we looked at the event free survival between early transplant vs RVd alone or delayed.
We are particularly struck by this because in fact, in our delayed transplant arm, only 28 percent of the patients to date at the time of data analysis, had received their delayed transplants. 72 percent had received other therapies.
What we were struck by with the event free survival analysis was that this was much closer. The hazard ratio was 1.23 in favour of early transplant, but the median event free survival for the transplant arm was 47.3 months.
And for the RVd alone, it was 32 months. So event free survival included, obviously, not just progression, but receipt of non-protocol therapy progression, toxicity, or obviously mortality.
Then what was so interesting was when we moved to overall survival, I think this was a key endpoint that we determined. When we looked at overall survival, we showed no overall survival improvement, with RVd early vs RVd alone or delayed. And this was a key secondary endpoint. And with the median follow-up approaching, seven years, we felt really quite struck by this.
And the maturity of data obviously requires further follow-up. But nonetheless, we were able to show that for RVd alone, the five year overall survival estimate was 79.2 percent vs 80.7 for early transplant, and the hazard ratio was non-significant at 1.1. So no difference in survival between the two arms, despite relatively mature follow-up.
Then when we drove deeper and started to look at, MRD and DOR and response rates overall as part of our secondary analyses, we are able to show that MRD rates were higher for early transplant at 56 percent.
But in fact, for RVd alone, we achieved an MRD rate of 39 percent. Now obviously, that was in favor of early transplant, but the key point was that if you achieved, MRD negativity with RVd alone, actually, the progression free survival estimate, was actually quite interesting. It was a five year progression free survival estimate of 59 percent compared to 53.5 percent for transplant early if you’re MRD negative. So this suggested that if you achieved MRD negativity regardless of approach, you could do very well. Obviously, that requires further follow-up. We then dove deeper and started to look at progression free survival estimates by subgroup recognizing obviously that with overall survival, being no difference that that any analysis and overall survival would be premature.
One of the most interesting things about our progression free survival estimates was the emergence of the PFS benefit for African American patients actually being closer to a hazard ratio 1.07, so not significant. And we were intrigued by this observation because in our black patients, we were very pleased that we enrolled almost 20 percent of our study was African American. we sought to take a deeper dive. I should mention that in the same we sought to take a deeper dive. I should mention that in the same context, when we look at, hazard ratios across the trial, it also became clear that for other features such as high-risk vs standard risk, the relative benefit of early transplant vs delayed transplant was different.
Specifically in high-risk disease, there was clearly, particularly for translocation 414, a tremendous advantage to early transplant. But interestingly enough for standard-risk, that same degree of benefit, was less apparent compared to the two arms. So this then prompted us to take a deeper dive on specific subgroups and in particular, our black patients.
Efficacy: AA vs white patients RVd alone vs RVD+ASCT:
In terms of the efficacy outcomes for black versus white patients, the first thing that struck us was that when we looked in an exploratory multivariable model of prognostic factors for PFS and looked at early transplant, we saw the degree of benefit, for black vs white patients, to be significantly less than we would have expected for the group overall in in the context of, the black patients. So the sort of cut to the chase, because we did this in a very comprehensive and detailed manner, and I want to especially acknowledge my statistician, lead biostatistician, Dr Susana Jacobus, who is a really eminent figure, in my opinion, in the statistical world of myeloma.
She was able to show, I think, very impressively, that we were able to differentiate that in the, black patients, the multivariable analysis, that there was a PFS for African American patients in terms of the hazard ratio compared to their white counterparts, was striking in that, if you looked at RVd early transplant, the hazard ratio for black patients vs white was just 0.95, so just smack on the median with no obvious difference apparent.
Conversely, if you looked at RVd alone, the hazard ratio for black versus white patients was 0.58 in favor of black patients receiving RVd alone. So this was quite striking and this was part of a multivariable analysis.
So we then took a deeper dive and tried to better understand why this difference could be.
We looked at a series of hypothesis generating analyses, across black vs white patients. We looked at age, sex, BMI, 30 or above, ISS, LDH, high-risk cytogenetics, RISS, haemoglobin, and neutrophil counts because our working hypothesis was that perhaps with low neutrophil counts being more common in black patients, that was a factor.
What we then looked at was this question of Duffy null because Duffy null is such an important pathobiological characterization in black patients that has relevance in oncology broadly but may be particularly important in in patients, African American patients with myeloma.
So when we looked at our white patients receiving RVd alone and versus early transplant in DETERMINATION. African American patients generally were younger.
They were more commonly female.
They more frequently had a BMI greater than 30. And interestingly, they more frequently had an elevated LDH and a haemoglobin less than 10. And, of course, they had a lower baseline median neutrophil count. So we took a dive on Duffy null and looked at this, in a number of patients and in particular in a preliminary subset to get an idea of whether or not we had a Duffy null group, compared to Duffy-wild type that was commensurate to what we’d expect.
So, in the African American population overall, about 60 to 70 percent are Duffy null. Just to explain why that may be, it’s basically a phenotype that protects against the consequences of falciparum malaria and by so doing, protects the patient from the mortality of falciparum malaria. But the red cell itself is actually, as a result of being sort of reinforced to not allow the parasite to enter the red cell, actually is less able to act as a cytokine and chemokine sink, which significantly reduces its ability to act as an inflammatory mediator. That may be very important both in the context of chemokine scavenging, chemokine presentation and as a chemokine reservoir which has implications for inflammation and malignancy overall.
In any event, our Duffy null population was about 60 percent in our preliminary sample that we looked at. So, it looked to us as if the working hypothesis around Duffy null may be very relevant.
Just to expand a little bit more, if I may, on this hypothesis of why the Duffy antigen chemokine receptor and the hypothesis impact of Duffy null status is complicated is because if you’re Duffy null, you may have a reduction of your cytokine sink, you have may have reduced binding to inflammatory cytokines, you may have reduced attraction neutrophils to the periphery, which obviously results in a lower end PNC. But I think this is the key point, there may be an exaggerated inflammatory reaction either with chemotherapy and disruption of cytokine homeostasis and the augmentation of the inflammasome, which may be particularly relevant, in the context of myeloma pathobiology and responses to treatment.
So what we then did was to say, well, hang on a second. Let’s take a look at overall treatment exposure by race within each treatment arm, and just basically take a look at what that looks like. Well, what we showed was that there was similar overall treatment exposure by race, which was good to see. It meant that there weren’t differences differences in treatment exposure that could explain these in African American patients. So that was actually I think a very important finding.
And then when we looked at, best responses in the RVd alone vs RVd transplant arm by race, we found something very interesting.
We actually found that the odds of achieving a complete response were lower for African Americans vs white patients on the transplant early on. That was completely unexpected.
And this odds ratio actually, was quite impressive, at a 1.91 in favour of RVd alone, compared to transplant. So how how did that look in actuarial terms? The, African American patients achieving CR in the RVd alone arm was 46 percent compared to 42 percent in the white population.
But remarkably, in those patients who underwent early transplant, the CR rate was 30 percent compared to 52 percent in the white population. So this was really interesting. Now we had to be careful because these are obviously hypothesis generating findings, but that difference did generate an odds ratio of 0.41, which suggested significance. So certainly worthy of further study. Now what did this mean in terms of progression free survival? Well, I think this is probably the most important part of the presentation.
What we showed was that progression free survival with RVd alone vs RVd + early transplant, basically for RVd alone, the median PFS for black patients had not been reached, whereas that for RVd plus transplant was 61.4 months. For white patients, RVd + transplant generated a median PFS of 67.2 months compared to just 44.3 in the RVd alone arm with a hazard ratio of 1.67. So that difference appeared to be real. And then when we took a deeper dive and actually looked at respective, additional subgroup analysis and their interactions, specifically BMI or sex, we were even more struck.
If you had high BMI greater than 30 and you were female, your, median PFS for RVd + transplant early was 58.6 months, but it had not been reached for, those patients receiving RVd alone. So this was really interesting because it suggested actually that in African American patients of female sex with high BMI, RVd alone performed particularly well. And RVd with early transplant, whilst it performed well, nonetheless, overall, it was less than the median PFS for white patients in the same category.
So specifically, if you looked at white patients who received RVd + transplant and they had female sex and high BMI, they did actually quite well. Conversely, if they were black, the median PFS was just 58.6 months. So this was really very interesting. Again, I have to emphasize it’s all preliminary, but and a hypothesis generating, but very important because it might suggest there are real pathobiologic differences that mean, patients undergoing early transplant, may actually perhaps benefit from keeping it in reserve given the strong performance of the RVd alone arm. Again, a hypothesis and one that requires further study and validation, but nonetheless intriguing.
Safety: AA vs white patients RVd alone vs RVD+ASCT
In terms of safety, what we were pleased to see is there were no meaningful differences in neutropenias, thrombocytopenia or anaemia.
Interestingly, in terms of neuropathy, there were significantly higher rates of neuropathy for black patients compared to their white counterparts in both arms, which I think is really interesting because this is an observation that’s been seen in other studies as well.
We also took a look at quality of life and were able to show there, importantly, there was clearly a clinically meaningful decrease in quality of life with early transplant with subsequent recovery. And when we looked at black vswhite patients, we were pleased to report similar patterns of transient decreases in QoL and subsequent recovery, but there was no striking difference that was statistically significant that would lead us to believe that the QoL hit to a black patient was substantial compared to a white patient. That pattern of change was very similar. So I don’t I don’t think one could say that. What was also very important though is we did an updated dive on, second cancers.
And here we’d seen a statistically significant difference, in favour of keeping transplant in reserve because at the time of our first analysis, we only had, well, sorry. I apologize. We had 10 cases of AML MDS in the urine transport on versus none for RVd alone. With our updated look, we had increased the number of MDS AML cases in the early transplant arm to 12, but we had seen three cases emerge of MDS, in the RVd alone, lenalidomide alone arm, which suggested that the lenalidomide obviously was still generating a degree of a potential risk in terms of MDS AML as expected, but not to the same extent as was seen for early transplant patients, in which, again, the rate had continued to increase. And this was statistically significant with a p-value of 0.034.
Interestingly, when we looked at African American patients, so the numbers are are very small. So I think one has to take it with a great deal of caution. But nonetheless, interestingly, for the black patients with RVd alone, we saw no AML MDS to date.
Conversely, however, there were at least three cases for early transplant. So recognizing the numbers are small, and this is hypothesis generating, we could argue that these exploratory analyses suggest that outcomes were similar in African American patients, but potentially driven by specific subgroups of patients.
Duffy null status may be relevant, and we need to explore that further. Haematologic toxicity, may be important to bear in mind, although I think right now our data suggests that’s not the core problem. But I do think the inflammasome and cytokine homeostasis and myeloma pathobiology may be much more relevant.
So, this is something we need to look out further.
Impact of finding on newly diagnosed multiple myeloma patients
Well, I think this analysis, as I say, suggest that the PFS event free survival and overall survival was similar, so I wouldn’t recommend that this radically changes practice because obviously there’s not a marked difference, in these key analyses that would prompt me to say, No, you must not offer transplant to African American patients because it could hurt them. That’s not what we’re saying. What we are saying, though, is that outcomes overall with RVd + transport appear to be similar.
However, exploratory analyses indicate a possible meaningful PFS benefit from RVd alone in certain African American patient subgroups, in particular, women of high BMI. And so this, I think, could further support the strategy of keeping transplant in reserve without loss of clinical benefit in selected patients. I think what’s really interesting is that maintenance duration proved longer in African American patients, which we were not expecting given the working hypothesis of a low white count. But I think overall grade three hematologic toxicities were similar, and the Duffy null presence of about 60 to 65 percent in African American patients was as expected.
What we did find was, and this I think is very interesting, is that grade 3 or greater neuropathy rates were higher in African American patients, and this was consistent with the GRIFFIN study, which we published recently at the subgroup analysis led by my colleague, Dr AK Nooka. And we published this in Blood Cancer Journal about two years ago where we showed that, neuropathy rates were higher in all African American patients in the GRIFFIN trial. We saw the same thing in DETERMINATION.
And the question is, is this a differential effect on the inflammasome? Does this have any bearing on Duffy null and myeloma pathobiology more broadly? And I think ongoing analyses will help us better understand this.
Interviewer/Editor: Gina Furnival
Cite: Richardson P. Promising data in advancing CAR T-cell therapy outcomes. touchHAEMATOLOGY, July 25 2024.
