CoMMpass Data Continue to Shape Understanding of Genomic Diversity in Multiple Myeloma

Sagar Lonial, MD, FACP

Insights from the prospective, longitudinal, observational CoMMpass study (NCT01454297) have continued to advance the understanding of genomic diversity in multiple myeloma, according to Sagar Lonial, MD, FACP.

“Even though the [CoMMpass] dataset is now over 10 years old, it still has relevance in helping us to understand some of these key drivers and potential new therapeutics in myeloma,” Lonial said in an interview with OncLive®.

CoMMpass was a longitudinal, observational study that included patients with newly diagnosed multiple myeloma (n = 1143) who had tumor samples collected at baseline and subsequent relapses. These samples underwent whole-genome sequencing, whole-exome sequencing, and RNA sequencing, and investigators aimed to identify genes that were the target of recurrent gain-of-function and loss-of-function events.

Findings showed that 8 copy number and 12 expression subtypes of multiple myeloma were identified. Notably, 25.5% of patients transitioned to a high-risk expression subtype at disease progression.

In the interview, Lonial highlighted the critical gaps the CoMMpass aimed to fill, including the lack of genomic data for myeloma compared with other cancers. He expanded on the study’s key findings and the implications that are still affecting the multiple myeloma landscape.

Lonial currently serves as Professor and Chair of the Department of Hematology and Medical Oncology at Emory University School of Medicine.

OncLive: What was the rationale behind the CoMMpass study? What knowledge gaps or unmet needs were you aiming to address with this research?

Lonial: When we first came up with the idea behind the CoMMpass study, what we were trying to do was give the world and the myeloma community a global roadmap for the genomic diversity at the time of diagnosis for patients with multiple myeloma. The National Cancer Institute [NCI] had supported many Cancer Genome Atlas projects looking at, for instance, brain tumors, lung cancer, colorectal cancer, or breast cancer. That data was already available. What was the genomic diversity of those diseases at the time of diagnosis? However, multiple myeloma didn't make the top 10 list for the NCI, and the Multiple Myeloma Research Foundation [MMRF] saw that this was a unique opportunity for them as a patient-based organization focused on research to [design] a trial that would create a database to provide a resource for physicians, patients, and researchers all over the world.

What was the design and methodology of the trial? What methods were used for tumor characterization?

The eligibility criteria were pretty loose. Any patient with newly diagnosed multiple myeloma who is going to receive either a proteasome inhibitor or an immunomodulatory drug [IMiD] as part of their initial induction therapy. We had a lot of discussion around whether we should limit what the induction therapy choices would be, and several of the members of the MMRF and the Multiple Myeloma Research Consortium argued against that. They wanted to [enroll] all comers and understand what the impact of these genetic abnormalities may be based on these different regimens that patients receive.

All [patients needed] to have was [treatment with] a proteasome inhibitor and an IMiD as part of their induction. They also needed to agree to serial follow up, meaning that at the time of first relapse, we would get another potential bone marrow [sample] that would be sent [for analysis], and then again at subsequent relapses. There were many patients who had 5 to 6 serial samples in the registry that we have now used to look at changes in genetic composition over time.

The test that we used was whole genome sequencing and whole exome sequencing. In some patients, RNA sequencing was done if we had sufficient cells. It is a robust data set, and although the focus was on genetics and genomics, we have been able to use some of those samples to further evaluate the immune profile of patients at initial diagnosis and then at subsequent relapses. There is a paper currently under review looking at that series of patients and a subset of patients with immune composition being the primary question.

What were the baseline characteristics of the patient population? Given the trial's diverse enrollment, how might this impact the findings?

We had over 1000 patients [with newly diagnosed multiple myeloma]. Patients did not come only from the United States. There were some Canadian patients, Italian patients, and Spanish patients that were enrolled. That diversity allows us to ask questions not just about genetics and genomics; we can break the data down potentially by differences in region or ethnicity. All of that expands the richness of the dataset across the board.

What specific genetic alterations or mutations were found to be the target of the recurrent gain of function and loss of function events?

When we first embarked on CoMMpass, what we had hoped, what we would find was something like chronic myeloid leukemia or Waldenström macroglobulinemia, where there was one common dominant mutation that could be targeted to make the disease go away.

What we began to see in the first few hundred patients who were sequenced is that the genetic mutation pattern of patients with myeloma was not that different from many other solid tumors. We saw abnormalities in RAS and P53. We saw the same players that we see in general with many other cancers. Interestingly, we did see things like IDH mutations, which are often seen in leukemia. We also saw BRAF mutations, which are commonly seen in melanoma, lung cancer, and other cancers. There are small percentages of patients that have potential targetable mutations that are seen in other solid tumors; therefore, we're co-opting those treatments where appropriate, such as MEK inhibitors, RAS inhibitors, and BRAF inhibitors.

However, there was no one dominant clone that we saw that represented a majority of patients with multiple myeloma, and that probably represents the fact that multiple myeloma is multifocal in terms of its origin. There are multiple things that may initiate the clonal plasma cell in the first place. It's not going to be as simple as one target that we can go after.

Regarding the subgroups, how did consensus clustering help in identifying the distinct subgroups of multiple myeloma and their underlying molecular characteristics?

When we looked at subgroups by mutation patterns, we did see that those patterns continue to hold up when you look at expression and mutation patterns. In many ways, we validated some of those gene expression profiling using mutations as a way to identify things such as MMSET for instance, which is associated with the t(4;14)translocation.

Interestingly, the t(11:14) translocation group came up and broke down into two different groups, similar to what we saw in the gene expression profiling. There was a hyperdiploid group that came up again. Most of these were critical.

When we look at subsequent relapses, the one common road that many patients go down is the PR pathway. PR represents proliferation, meaning you lose that indolent plasma cell phenotype over time. Patients become much more proliferative and lose many of those plasma cell markers that we often use to target the disease. Drugs like proteasome inhibitors, IMiDs, anti-CD38 monoclonal antibodies, and BCMA-directed therapies all [target] plasma cell markers. As the disease becomes more proliferative, patients may lose expression of some of those key drivers of therapeutic success we've had over the last decade.

For patients who transition to high-risk progression, how might this impact the management of their disease?

To reiterate the idea that patients are moving to the PR phenotype, which represents a more proliferative, high-risk group, that suggests that multiple myeloma is a complicated disease. Every time it relapses, it becomes genomically more complicated.

The strategies around how to optimize treatment should try to maximize the benefit at diagnosis. Even in high-risk multiple myeloma, the disease is never more sensitive than it is at the time of diagnosis. With each subsequent relapse, it becomes genomically more challenging to treat a patient over time, and that was born out in the dataset we have with sequential biopsies.

What are the broader implications of these findings, and how do they contribute to advancing the field of multiple myeloma?

There have been more 300 abstracts [published] looking at CoMMpass data to interrogate specific biomarkers, genes, or potential targets in myeloma. The papers that have been published that defined double hit myeloma, for instance, came out of the CoMMpass data set.

When somebody proposes a new target for myeloma, the first question I and many of my colleagues ask is: what did it look like in CoMMpass? What was the expression of that target in CoMMpass? Are there genomic subtypes that either preferentially over or under express that target in myeloma?

[Using the CoMMpass data] has almost become the second step in the development of a new drug or a new target in myeloma. The first step is thinking about something new; then you go validate it or understand its expression profile using CoMMpass data. At our center, not a week goes by that we don't go back to CoMMpass to ask: what did this mean? What was the implication of this pattern?

Were there any limitations about this study?

As we define risk in myeloma, it is often dependent on the treatment landscape available at the time of initial treatment. Because the treatment landscape has changed so much, understanding, for instance, how does the [profiling data from CoMMpass] predict for sensitivity or resistance to a BCMA-directed CAR T-cell therapy? I don't think we can extrapolate that information [from CoMMpass].

What we can extrapolate is whether there subsets of patients with multiple myeloma who over or under express BCMA, which could tells us who [to select for certain therapies]. With the immunologic revolution in therapy that we've seen in the last 5 years with antibody-drug conjugates, T-cell engagers, and now CAR T-cell therapies, that has impacted the genetic definitions that [have emerged] from CoMMpass. we'll have to reevaluate those again because there may be different drivers of drug resistance than we had expected in CoMMpass.

Are any next steps planned for this research?

What we're trying to do now is look at additional subprojects. For instance, we are interested in the impact of cyclophosphamide at diagnosis. Patients often receive bortezomib [Velcade], cyclophosphamide and dexamethasone [VCd] as opposed to daratumumab [Darzalex] plus bortezomib, lenalidomide [Revlimid], and dexamethasone [D-VRd].

When we first looked at CoMMpass, patients who received VCd had shorter remissions and worse survival compared with patients who received D-VRd. We're trying to validate that and get that data out, because VCd is a cheap, commonly used regimen, and it may be impacting our long-term outcomes. That's just one example of trying to go back and understand how to use some of this data.

There are other projects that are in development. [What is the] impact of being at a major myeloma center as opposed to a community site? What impact does that have on progression-free survival and overall survival? Those are questions that we're going to try to ask in the coming years.

Reference

1. Skerget S, Penaherrera D, Chari A, et al. Comprehensive molecular profiling of multiple myeloma identifies refined copy number and expression subtypes. Nat Genet. 2024;56(9):1878-1889. doi:10.1038/s41588-024-01853-0