Biomarker-Directed Therapy in Multiple Myeloma Signals the Future of the Field

The biology of multiple myeloma has an evolving role in guiding research efforts involving diagnostic and prognostic capabilities of available biomarkers and related molecular pathways. “Myeloma cells often have genetic abnormalities, and we often think about them as primary abnormalities, which are seen at all stages of the disease,” Shaji Kumar, MD, said during a recent OncLive Insights^® program, a video editorial series featuring in-depth reports from leaders in oncology specialties.

Kumar was joined by Jonathan Kaufman, MD, to discuss recent advances in molecular biomarkers in multiple myeloma. “There is an unmet need in the entire field of myeloma right now,” Kaufman said. “It would be great if we had a biomarker that signaled [which patients] can have a 3-drug induction therapy and not 4-drug, or…that signals [who should receive] doublet maintenance vs single-agent maintenance….We don’t know the answers to any of these questions and using a biomarker would be extremely helpful, once we develop that.”

In the discussion, Kumar and Kaufman reviewed the advances with translocations (t) such as t(11;14), treating practices, and emerging markers in the field.

Kumar is a hematologist at Mayo Clinic Comprehensive Cancer Center in Rochester, Minnesota. Kaufman is medical director and section chief, ambulatory infusion centers, Winship Cancer Institute of Emory University, associate vice chair for quality hematology and medical oncology at the Winship Cancer Institute of Emory University, and a professor, Department of Hematology and Medical Oncology, Emory University School of Medicine, Atlanta, Georgia.

Kumar: We have been using a fair number of biomarkers, and quite a few of them are evolving as well. Particularly, some of those biomarkers or the biological changes that we see in the myeloma cells have lent themselves to be predictive markers for responses and other therapies that are evolving.

Kaufman: Recurrent translocations such as t(11;14) and the less common t(6;14) occur where there’s translocation of the cyclin proteins with chromosome 14. What we know is that certain translocations are associated with high-risk disease. But we say that in a very broad way, without really understanding…what we’re learning more about, particularly in t(11;14).

For example, we’re getting to the point where we can have predictive medicine for patients who present with t(11;14), as they are much more likely to respond to BCL2 inhibition. Clinically, what we’ve used to date is venetoclax [Venclexta], a BCL2 inhibitor.

There are additional abnormalities in terms of copy number. Trisomies are the most common abnormality in myeloma. Uniquely, the trisomies have only happened in the odd chromosomes and are associated with a better prognosis or standard-risk disease. There is some evidence that patients with trisomy are more sensitive to the immunomodulatory agents lenalidomide [Revlimid] and pomalidomide [Pomalyst]. [Additionally], there are enough data that we really can solidly say that 1q abnormalities and 17p deletion are associated with a higher-risk disease.

There is a lot of controversy within the field regarding how much 17p deletion [del(17p)] there must be for a patient to be [stratified as] high risk. The [Intergroupe Francophone du Myélome] [is] advocating that del(17p) is greater than 55% of plasma cells, and where TP53 sits on 17p [and] you have a complete abrogation of the apoptotic pathway. If you have loss in one and p53 mutation in the other, that is associated with significantly worsened outcome and increased risk.

When we think about oncogenic pathways that are involved, MYC is commonly involved as it is proliferative. Another pathway that we’re learning more about…and gaining a better understanding of its biology in myeloma is the RAS pathway. There are a lot of different pathways that are abnormal and to date, we’ve only been able to use 1 as predictive in helping us decide what treatment to do, which brings us back to t(11;14).

Kumar: Among the primary abnormalities, approximately 40% of patients have translocations that involve the immunoglobulin heavy chain region of chromosome 14, with approximately 5 other recurrent partner chromosomes. One is t(11;14), which can be seen in approximately 15% to 20% of patients with multiple myeloma. Outcomes for this subgroup of patients have been studied in detail and it appears that these patients have outcomes that are intermediate between those with trisomies, which are good prognostic markers, and high-risk markers such as t(4;14) or t(14;16).

When you look at patients with t(11;14), they tend to have [fewer] secondary abnormalities such as del(17p) in their evolution over time. This is a subgroup of patients who appear to have differential outcomes. There seem to be some fundamental biological differences as well. What we have shown is that when you look at these patients with t(11;14), the myeloma cells appear to be more B cell–like. They tend to express some of the proteins on the surface, such as CD20, that you often associate with B cells. Their plasma cells generally tend to be smaller [and] can be more lymphoid compared with the other plasma cells we see in other groups. Additionally, these patients tend to have more bone disease compared with the other subtypes or subgroups in terms of genetics.

Even though the patients may appear similar, it is good to represent different diseases based on the fundamental genetic abnormality that they have. This has implications in terms of how we treat these patients. Increasingly, we’re starting to look at the differential effects of the many drugs we have available for treating patients with myeloma. There have been studies looking at BCL2 inhibition strategies, starting with venetoclax, a drug specifically created to target BCL2, given the significant role that BCL2 plays across many types of malignancies.

As with other cancers, preclinical work has demonstrated that use of venetoclax is associated with activity in myeloma cells, particularly when the myeloma cells carry t(11;14). Phase 1 or 2 early trials have been able to show that in approximately one-third of patients with relapsed multiple myeloma, venetoclax was able to induce a tumor response to therapy. When we looked at it in more detail, it was clear that [most] of these responses appear to happen among patients who have t(11;14). Furthermore, it appeared that among the patients with t(11;14), those with high levels of BCL2 expression appear to experience even more efficacy with venetoclax. Based on preclinical data showing synergy, venetoclax was combined with dexamethasone for treatment of relapsed myeloma. It appears that the dexamethasone makes these cells more sensitive to an inhibitor to the BCL2 pathway.

When we treat myeloma cells with BCL2 inhibitors, we sometimes see other protective mechanisms, such as MCL1 expression, go up. Proteasome inhibitors such as bortezomib [Velcade] appear to have the capability of downregulating MCL1 through different mechanisms, which allow that synergy to happen. Based on all the preclinical work and the results from phase 1 studies, the phase 3 BELLINI trial [NCT02755597]¹ was designed to explore the activity of venetoclax plus bortezomib and dexamethasone. This was conducted in patients with relapsed multiple myeloma who had 1 to 3 prior lines of therapy. Approximately 300 patients were randomly assigned to bortezomib/dexamethasone or bortezomib/dexamethasone with venetoclax.

The addition of venetoclax improved the depth of response and the progression-free survival among the overall group compared with bortezomib/dexamethasone alone. Interestingly, the initial readout of the trial demonstrated that patients who received venetoclax had increased mortality. This was an intriguing finding, which led to several additional analyses of the data. What became very clear was that the activity of venetoclax in combination with bortezomib/dexamethasone was most pronounced in patients with t(11;14), as one could have surmised from that preclinical work and the early clinical trials.

In addition, even among patients with the high expression of BCL2, measured using gene expression studies, those patients appear to benefit from the use of venetoclax as well, even if they did not have t(11;14). The results of BELLINI allowed us to explore 2 different biomarkers—t(11;14) as detected by FISH [fluorescence in situ hybridization] used in the clinic all this time, but also the BCL2 expression, using gene expression studies, which are not yet approved and in the clinic at this time. Both biomarkers allowed us to identify a population of patients likely to benefit from venetoclax and allowed us to exclude a patient population where the addition of venetoclax could potentially lead to a negative effect on outcomes.

I think much has been learned from BELLINI, particularly the type of patients [in whom] the use of venetoclax will have a positive effect. As these assays become available, particularly the BCL2 expression assay, and venetoclax gets approved for multiple myeloma, we will be able to use it in this patient population.

Kaufman: [In development] are flow panels that identify a B-cell phenotype that can help differentiate myeloma cells that are sensitive to venetoclax, because as you mentioned, not all t(11;14) cells are sensitive, and although rare, some non-t(11;14) cells are sensitive. And so, a BCL2 assay or a flow panel can help us make that decision so we can have appropriate predictive medicine.

Going back to the laboratory, the concept that t(11;14) cells are a little bit more lymphoid-like [and have] a little like a B-cell phenotype, the hypothesis of combining venetoclax with daratumumab (Darzalex) was that the venetoclax was going to address the more B-cell phenotype of the plasma cell, and the daratumumab—the CD38 monoclonal antibody—would target the plasma cell component of it. We moved forward with the t(11;14) patient population in the phase 2 trial [NCT03314181] and the 3-drug combination of venetoclax/daratumumab/ dexamethasone, which showed very high response rates.²

There [are] also very good preclinical data of the combination of venetoclax and proteasome inhibitors, specifically carfilzomib [Kyprolis], venetoclax, and dexamethasone for patients who received 1 to 3 prior lines of therapy [NCT02899052]. Investigators were able to demonstrate safety of that regimen.³ The dose expansion in the trial was 800 mg venetoclax daily with 70 mg/m² of carfilzomib on days 1, 8, and 15 of a 28-day cycle, plus weekly dexamethasone, and high response rates were observed in the 80% to 90% range.³

It is important to remember that all these patients had t(11;14). We have been involved in developing the story, really from the very beginning, of the phase 1 study and it’s an exciting story. There are other biomarkers in myeloma not looking at t(11;14) and the BCL2 story. I want to touch on some of those emerging biomarkers of other biology in myeloma.

Kumar: Certainly, t(11;14) has kicked off the interest in trying to understand the fundamental differences in biology between these different types of myeloma, so to speak. Hopefully, as more therapies are introduced, we can use the best drug for the most appropriate patient.

Results that we get from the whole-genome sequencing or the template sequencing studies on the tumor cells show that one of the most common findings is the mutations involving the KRAS or the NRAS gene, and maybe, though a bit smaller, a proportion of patients with BRAF mutations as well. MEK inhibitors are an intervention that could have significant effect for these patients, as we have [had] several studies demonstrate that you can eradicate or at least get rid of a significant proportion of the myeloma cells that harbor this mutation by using the MEK inhibitors. That includes drugs such as trametinib [Mekinist] and cobimetinib [Cotellic].

We have taken these concepts 1 step forward and are currently enrolling the MyDRUG trial [NCT03732703], which is looking at various genetic abnormalities and trying to tailor therapy.⁴ These are patients with prolapsed myeloma, who have functional high-risk disease, who get sequencing done and based on the predominant, underlying genetic mutation, they get assigned to a specific therapy. For example, if they have a RAS mutation or a BRAF mutation, they get assigned to cobimetinib, which is given for a couple of cycles alone, and then combined with a 3-drug combination: dexamethasone, ixazomib, and pomalidomide. The initial 2 cycles give us a good assessment whether this targeted approach is…worthwhile and at the same time recognizing the fact that just targeting 1 abnormality is probably not sufficient for long-term disease control. Thus, we are adding a combination on top of the targeted agent. For those with t(11;14)…they get assigned to the venetoclax, dexamethasone, ixazomib, and pomalidomide arm.

Some of these emerging biomarkers are helping us decide how best to approach the treatment of myeloma in each patient. There are several other biomarkers that are being developed, too. There’s a lot of interest in circulating tumor DNA, but it is early days for that in the myeloma space.

Dr Kaufman, what are your thoughts in terms of some of the best of these biomarkers that are already in the clinic that you’re using, and some of the new ones that are evolving?

Kaufman: From a testing perspective, every patient [with a new diagnosis] should have FISH evaluating for trisomies, evaluating for the 14-related translocations, including t(11;14) and 1q abnormalities. This should be the standard approach. An interesting question is should we do karyotyping? In data presented at the [American Society of Clinical Oncology], we showed that karyotyping in patients we thought were standard risk— because they didn’t have high-risk abnormalities—those who had conventional cytogenetic abnormalities have worse outcomes.⁵

The reality is that those abnormalities are in every patient, but our hypothesis is that if we can identify conventional cytogenetic abnormalities, we can capture some patients we wouldn’t have thought were high risk but really are high risk. In myeloma today, the standard remains measuring the amount of myeloma in the body, using the original biomarker, which is the monoclonal protein as measured by serum protein electrophoresis and urine protein electrophoresis. There’s an enormous amount of work involved in switching from relatively old technology of the electrophoresis and immunofixation to mass spectrometry.

Using flow cytometry to identify MRD [minimal residual disease], using next-generation sequencing to identify MRD, these things are ongoing, and as a community we’re just learning how to use these. I think the question becomes: When do we order the next-generation sequencing or whole-genome sequencing, so we can look for mutations? From my perspective, the time to look is not necessarily at the newly diagnosed stage but after initial, optimal therapy. In this situation, ordering [personalized approaches] like in MyDRUG [may help us] see [whether] we can identify that subgroup of patients [who] might have a targetable or an actual mutation. Outside of that study, we’ve started incorporating the whole-genome sequencing to look for recurrent mutations in that setting.

Kumar: In terms of timing of testing, t(11;14) is important from the therapy standpoint. We know that many FISH abnormalities have contributed to the prognostication even if you may not necessarily differentiate on the initial therapy. Clinical trials and maybe clinical practice may be altered based on some of these prognostic markers, particularly the high-risk FISH abnormalities. It tells you about a choice between 2-drug maintenance vs 1-drug maintenance therapy and so forth. FISH testing at diagnosis is something that every patient with myeloma should absolutely have. We probably don’t need the more extensive and expensive molecular testing at the time of diagnosis, because we are anticipating a good median of 4 to 5 years without progression with modern therapies for newly diagnosed myeloma. Those tests are more likely to be useful at the time of relapse, when new abnormalities may emerge and the results may open opportunities for new therapeutic options, whether part of a clinical trial or outside. If patients have clinical behavior that seems to be quite different than one would anticipate from their baseline FISH findings and their relapse is early, those are times where we would want to do some of those next-generation sequencing types of testing. This is something that needs to be continually reassessed. What are your thoughts in terms of barriers to routine testing in biomarkers?

Kaufman: When I talk to colleagues and oncologists around the country, one of the questions is: If you do a test and you get the information, and you don’t act on that test, then what’s the use of it? The more we correlate these biomarkers with something to do, the more these barriers to testing are going to be broken down.

References

1. Kumar SK, Harrison SJ, Cavo M, et al. Venetoclax or placebo in combination with bortezomib and dexamethasone in patients with relapsed or refractory multiple myeloma (BELLINI): a randomised, double-blind, multicentre, phase 3 trial. Lancet Oncol. 2020;21(12):1630-1642. doi:10.1016/S1470-2045(20)30525-8
2. Bahlis NJ, Baz R, Harrison SJ, et al. Phase I study of venetoclax plus daratumumab and dexamethasone, with or without bortezomib, in patients with relapsed or refractory multiple myeloma with and without t(11;14). J Clin Oncol. 2021;39(32):3602-3612. doi:10.1200/JCO.21.00443
3. Costa LJ, Davies FE, Monohan GP, et al. Phase 2 study of venetoclax plus carfilzomib and dexamethasone in patients with relapsed/refractory multiple myeloma. Blood Adv. 2021;5(19):3748-3759. doi:10.1182/bloodadvances.2020004146
4. Obiekwe D, Joseph N, Hofmeister CC, et al. The impact of complex karyotype identified by conventional cytogenetics on survival outcomes of 1,000 patients with newly diagnosed myeloma (NDMM). J Clin Oncol. 2022;40(suppl 16):8063. doi:10.1200/JCO.2022.40.16_suppl.8063
5. Myeloma-developing regimens using genomics (MyDRUG). ClinicalTrials.gov. Updated September 10, 2021. Accessed August 25, 2022. https://clinicaltrials.gov/ct2/show/NCT03732703