36 is the new 18: Functional high‐risk multiple myeloma in the modern era

Short abstract

In this issue of Cancer, a threshold of 36 months (rather than 18 months) is established as the optimal threshold for the identification of functional high‐risk multiple myeloma. Early use of T‐cell–redirection therapies may improve outcomes for these patients.

In this issue of Cancer, Ravi and colleagues describe their analysis of over 300 patients with newly diagnosed multiple myeloma (MM) who were eligible for autologous stem cell transplantation (ASCT) and received quadruplet induction therapy, including a CD38‐targeted monoclonal antibody, a proteasome inhibitor, and an immunomodulatory drug. ¹ They focus specifically on the kinetics and postprogression outcomes of patients with early relapses, delving deep into the incompletely understood concept of functional high‐risk MM (FHRMM). They identify 36 months as the optimal cutoff to identify patients with FHRMM who are at high risk of poor outcomes thereafter: namely, overall survival (OS) ≤2 years after progression. Their research suggests that moving immediately to T‐cell–redirected therapies (TCRTs), for example, chimeric antigen receptor T‐cell (CAR‐T) therapies or bispecific antibodies (bsAbs), may mitigate this risk of worsened outcomes once FHRMM is identified.

With any treatment modality that is not reliably curative, there will always be patients with unusually short or unusually long durations of responses (DORs). Therefore, it is unfortunate but not surprising that some patients with MM will have relapses within a few years of initiating induction therapy. What has been sobering, however, is the disproportionately poor progression‐free survival (PFS) and OS that patients who have early relapses face with subsequent therapies versus patients who have slower relapse kinetics. ² , ³ , ⁴ Even late‐line CAR‐T therapy cannot close this gap once FHRMM is identified, ⁵ illustrating the need for better therapeutic modalities and treatment sequencing options for this patient population. To make matters more complex, FHRMM cannot reliably be predicted in advance using existing tools such as high‐risk cytogenetic abnormalities. ⁶ , ⁷ , ⁸ Transcriptional and genomic profiling can potentially identify patients with standard‐risk cytogenetics at diagnosis who are likely to develop FHRMM, ⁹ but these testing modalities are not routinely available in clinical practice. Thus patients with FHRMM often remain in our blind spots until their relapses begin to accumulate with each postprogression line of therapy.

Identifying FHRMM patient populations for dedicated analyses and trials is an essential first step toward improving their outcomes, but dichotomizing any continuous variable such as the frontline DOR into two discrete populations is inherently an imperfect science. Various studies and expert groups have recommended a DOR ≤18 months (with either the first cycle of frontline induction or the date of ASCT as the starting point) as the optimal cutoff to define FHRMM. ² , ³ , ⁴ , ⁹ , ¹⁰ , ¹¹ , ¹² Among patients who receive triplet induction, the specific type of induction regimen—for example, whether or not lenalidomide was incorporated—does not affect postprogression OS once FHRMM is identified. ¹³ However, a limitation of almost all existing studies in this space is their derivation from the era of triplet induction. Because multiple quadruplet regimens have outperformed triplet regimens in modern trials, our FHRMM definitions should be re‐examined in tandem. As shown in Figure 1, phase 3 trials have consistently demonstrated smaller proportions of patients who have relapses ≤18 months when treated with quadruplet versus triplet induction regimens. Relying on an 18‐month threshold would mean that only about 10% of quadruplet‐treated trial participants would have FHRMM; although undoubtedly high‐risk, this narrowly defined subset likely would be too small for dedicated future studies.

FIGURE 1.

Estimated FHRMM proportions in selected phase 3 trials. FHRMM estimates, using thresholds of 18 months or 36 months after frontline induction, are presented using visual approximations from published Kaplan–Meier PFS curves from all of the listed phase 3 trials (which were selected given their patient population and sufficient follow‐up). PFS includes other events of interest, such as nonresponse, censoring, or death; therefore, these proportions are not exactly equivalent to FHRMM proportions. CASSIOPEIA indicates ClinicalTrials.gov identifier NCT02541383; CEPHESUS, ClinicalTrials.gov identifier NCT03652064; FHRMM, functional high‐risk multiple myeloma; GMMG‐HD7, ClinicalTrials.gov identifier NCT03617731; IMROZ, ClinicalTrials.gov identifier NCT03319667; PERSEUS, ClinicalTrials.gov identifier NCT03710603; PFS, progression‐free survival; quad, quadruplet induction therapy.

Ravi and colleagues thus sought to compare the traditional 18‐month FHRMM threshold against other thresholds in the era of quadruplet therapy. ¹ Their analysis of over 300 quadruplet‐treated patients included both clinical trial participants and real‐world patients from a prospective registry. As expected, traditional categorizations did not identify early relapses well: 32% and 44% of patients with FHRMM (using the 18‐month and 36‐month cutoffs, respectively) had standard‐risk cytogenetics by 2025 consensus criteria. The authors decided a priori to select the FHRMM threshold that best delineated patients with postprogression OS <2 years, and the 36‐month threshold was ultimately the best fit: the median postprogression OS was 23.8 months (95% confidence interval, 16.2–31.3 months) with the 36‐month threshold versus 8.1 months (95% confidence interval, not reached to 17.1 months) for the 18‐month threshold. The authors thus defined 36 months as the best FHRMM threshold for quadruplet‐treated patients. When summing a frontline DOR of ≤3 years with a median postprogression OS of ≤2 years, this equates to patients with an estimated survival of approximately 5 years or less despite quadruplet induction and ASCT. Because 5 years was once the total expected OS for all patients a decade ago before rapid improvements in MM treatment, ¹⁴ , ¹⁵ this threshold is appropriate to identify the subset of patients who are still being left behind despite these advances.

Although limited by small sample sizes, the authors do find that second‐line TCRTs for FHRMM (as opposed to late‐line TCRTs, which are still associated with poorer outcomes in this population) ⁵ may constitute the best path forward. Only five of their 45 patients with FHRMM using a 36‐month threshold immediately proceeded to CAR‐T therapies or bsAbs, but these five patients performed dramatically better. The postprogression PFS rate at 12 months was 80% for TCRT‐treated patients versus only 23% for non‐TCRT–treated patients, with the receipt of a second‐line TCRT a strong predictor of superior postprogression PFS in multivariable analyses. ¹ Although validation of these findings is needed in larger data sets, this observation aligns with post hoc analyses of the phase 3 CARTITUDE‐4 and MajesTEC‐3 trials (of CAR‐T therapies and bsAbs, respectively; ClinicalTrials.gov identifiers NCT04181827 and NCT05083169, respectively) using an 18‐month cutoff for largely triplet‐treated patients. ¹⁶ , ¹⁷ Smaller single‐arm trial cohorts have similarly demonstrated impressive efficacy with second‐line CAR‐T therapy despite enrolling only patients with FHRMM. ¹⁸ , ¹⁹ , ²⁰ Whereas TCRTs have impressive efficacy across the board regardless of relapse kinetics, they evidently offer disproportionate benefits at first relapse for patients with FHRMM regardless of how many drugs are included in induction.

In conclusion, the work by Ravi and colleagues sets an important new benchmark for the field. As quadruplet induction increasingly becomes the norm for newly diagnosed MM in many countries, patients with relapses within 36 months of treatment initiation should be prioritized for CAR‐T or bsAb therapies. Similarly, as clinical trials of novel combinations in relapsed/refractory MM complete accrual, proportions of patients with FHRMM (using the 36‐month definition) should be reported alongside more traditional baseline characteristics: for example, proportions of patients with high‐risk cytogenetics or markers of higher tumor burden. The authors’ findings, of course, have their limitations, most notably the lack of ASCT‐ineligible patients in their analysis and the limited number of patients who proceeded to second‐line TCRTs after initial relapse. However, their conclusion is clear and convincing (Figure 1): when it comes to FHRMM in the modern era, 36 is the new 18 (Figure 2).

FIGURE 2.

FHRMM definitions based on the type of frontline induction. FHRMM indicates functional high‐risk multiple myeloma.

CONFLICT OF INTEREST STATEMENT

Rahul Banerjee reports grants/contracts from AbbVie, Bristol Myers Squibb Company, Gilead Sciences Inc., Johnson & Johnson, Karyopharm Therapeutics, Novartis, Pack Health, and Prothena; and personal/consulting or advisory fees from AbbVie, Adaptive Biotechnologies, Arcellx, Bristol Myers Squibb Company, Caribou Biosciences Inc., Genentech, Gilead Sciences Inc., GlaxoSmithKline, Janssen Biotech, Karyopharm Therapeutics, Legend Biotech, Pfizer, Poseida Therapeutics, Regeneron Pharmaceuticals Inc., Sanofi, and SparkCures outside the submitted work. Swetha Reddi disclosed no conflicts of interest.