MRD-negative duration following latest line of therapy predicts long-term PFS in real-world patients with multiple myeloma

TO THE EDITOR:

Minimal residual disease (MRD) is a highly sensitive measure of depth-of-response and a powerful predictor of progression-free survival (PFS) in multiple myeloma (MM). In the era of modern myeloma therapies, MRD status continues to be an independent predictor of long-term clinical outcome even in patients who have achieved a complete response (CR).^1,2 Therefore, MRD negativity, defined as <1 clonal plasma cell in 100 000 cells in the bone marrow by next-generation flow (NGF) or next-generation sequencing (NGS), has been incorporated into the International Myeloma Working Group (IMWG) response criteria since 2016.³ Initial clinical trials evaluating MRD status typically assessed MRD at end of induction or ∼3 months after autologous stem cell transplant and demonstrated improved PFS with a singular MRD-negative result.^1,2,4-8 More recently, clinical trials have used serial MRD measurements to demonstrate the added prognostic benefit of sustained MRD negativity, including 6 months,⁹ 12 months,¹⁰ and 24 months^11,12 from either the start of treatment or start of CR. However, few clinical trials report MRD data beyond a duration of 2 to 3 years,^11-13 and real-world data to corroborate clinical trial findings are lacking. Furthermore, the optimal duration of MRD negativity required to discontinue maintenance therapy is not yet known. A recent observational study reported a 10-year PFS of 88% for patients MRD negative for >3 years after total therapy,¹⁴ highlighting the potentially excellent long-term prognosis associated with durable MRD negativity. Therefore, using our single-center real-world data, we estimate the long-term PFS of patients with MM patients with durable MRD-negative remission ≥5 years from the start of latest line of therapy and evaluate the prognostic impact of differing durations of MRD negativity.

Routine MRD assessment was introduced for patients with MM in remission at Icahn School of Medicine at Mount Sinai in 2017. We systematically reviewed consecutive MRD results for patients in the Icahn School of Medicine at Mount Sinai (ISMMS) MM registry between July 2017 and May 2023 (Figure 1A). NGF was used to detect MRD from 2017 to 2021, with the addition of NGS from 2021 onward. Most patients (75%) had only NGF results available, in which MRD negativity was defined as <1 cells in 10⁵; the remaining 25% of patients had additional NGS results available in which MRD negativity was defined as <1 cells in 10⁶. Patients were considered “MRD negative” for the calendar year if they had at least 1 negative and no positive MRD results. Missing MRD years in-between first and last MRD-negative results were allowed, providing patients remained in CR and had negative imaging throughout. Patients with ≥3 “MRD-negative years” were included in the study. These patients underwent detailed chart review. Patients with evidence of active myeloma by IMWG criteria during their MRD-negative period were excluded (n = 31). To mitigate potential bias from differing durations of CR before MRD negativity, we excluded patients who had started their latest line of therapy before January 2014, that is, >3 years before routinely available MRD testing. Whole-body positron emission tomography and magnetic resonance imaging were used to evaluate bone or extramedullary disease. PFS was evaluated using Kaplan-Meier curves and survival estimates.

During the study period, we identified 168 patients with ≥3 MRD-negative years at any time point in their disease course (Figure 1A); 120 patients met the additional inclusion criteria of concurrent CR and imaging negativity. Of this group, 76 patients were in CR, and were imaging and MRD negative ≥5 years from the start of their latest line of therapy.

Our initial analysis focused on this cohort of 76 patients in MRD-negative remission ≥5 years from the start of their latest line of therapy. With a median follow-up time of 6.5 years from start of latest line of therapy (range, 5.1-9.3), the 6-year PFS in this cohort (Figure 1B) was 95.9% (95% confidence interval [CI], 0.91-1.0) and the projected 9-year PFS was 88.8% (95% CI, 79.0-99.8). Most patients were International Staging Score stage I (54.0%), Immunoglobulin G isotype (54.0%), with standard-risk cytogenetics at diagnosis (76.3%; Table 1). Over half of patients (57.9%) had received frontline therapy whereas 42.1% had received ≥2 lines of therapy. Types of latest therapy in this cohort are outlined in Table 2. Median duration of latest line of therapy was 5 years (range, 0.6-9.3) and 73.6% of patients were off therapy at latest follow-up, with a median time off therapy of 23.6 months (range, 0.9-105). In total, 12 patients (15.8%) converted to MRD positive in the study period, 4 patients by NGF and 8 patients by NGS. Five patients within the cohort clinically relapsed (6.6%) within the study period; 4 patients relapsed with biochemical disease, and 1 patient with isolated extramedullary disease. Of 5 clinical relapses, 4 were preceded by conversion to MRD positivity, 6 months to 1 year prior. The remaining 8 patients converted to MRD positive within 1 year of the end of our study period, therefore longer follow-up may be required to determine timing of clinical relapse in these patients.

All patients in this cohort had a minimum of 3 consecutive MRD-negative years. Specifically, the number of patients with 3, 4, and 5 consecutive MRD-negative years was 17, 40, and 19 respectively. Of note, all clinical relapses occurred in patients with only 3 consecutive MRD-negative years. We therefore proceeded to interrogate the extent that MRD-negative duration, independent of time from start of therapy, also affects PFS.

For this, we analyzed all 120 patients with ≥3 consecutive MRD-negative years, in CR and imaging negative, regardless of timing from the start of their latest therapy (Figure 1A). We conducted a landmark survival analysis from the start of MRD negativity and compared differing durations of MRD negativity. MRD-negative duration of 3 years corresponded to a 5-year PFS of 66.4% (95% CI, 0.938-1.0). In comparison, longer MRD-negative durations of 4 and 5 years were associated with a significantly increased PFS of 97.9% (95% CI, 0.938-1.0) and 100%, respectively (Figure 1C), thus suggesting the additional prognostic value of increasing MRD-negative durations beyond 3 years.

In conclusion, in our single-center retrospective observational study, patients with MM in CR and imaging and MRD negative ≥5 years from start of therapy had an estimated 11% risk of disease relapse at 9 years, with 71% being off therapy. This low rate of relapse in patients who have achieved long-term MRD-negative remission is concordant with other studies¹⁴ and demonstrates that a proportion of patients with MM do not suffer multiple relapses. From our data, these patients were more often standard risk by International Staging Score and cytogenetics and achieved deep responses to frontline therapy. A smaller proportion of patients were able to achieve ≥5 years MRD negativity even after salvage therapies and this may be increasingly possible in the era of T-cell redirection therapies.

Furthermore, our data demonstrate that the number of consecutive MRD-negative years can also affect PFS. Current IMWG criteria define sustained MRD negativity as 2 consecutive MRD-negative results at least 1 year apart³; however, our data suggest that patients with 3 consecutive MRD-negative years may still carry a substantial relapse risk of ∼33% at 5 years from the start of MRD negativity. The risk of long-term relapse reduces significantly with increasing MRD-negativity duration; suggesting that persistent MRD-negative durations >3 years may be required to guide discontinuation of maintenance therapy. This observation is supported by recent studies demonstrating the PFS benefit of continuing lenalidomide maintenance after autologous stem cell transplant beyond 3 years.¹⁵ 

Limitations of our study include heterogeneity in MRD testing frequency and availability amongst our patients, and further analysis of large real-world MM cohorts with longer follow-up is needed to validate these findings. In addition, the majority of MRD results analyzed in this cohort were by NGF and further studies are required to assess the prognostic value of higher-sensitivity methods such as NGS in similar populations. Nevertheless, our data highlight that patients in CR with imaging and MRD negative ≥5 years from start of latest line of therapy have an excellent long-term prognosis and that MRD-negative durations of >3 years may be required to truly predict long-term PFS in real-world patients with MM.

This study was reviewed by the institutional review board at Icahn School of Medicine at Mount Sinai and approved under institutional review board STUDY-24-00439.

Contribution: L.Y.C. wrote the manuscript; L.Y.C. and S.T. performed the data analysis; L.Y.C., S.T., S.P., and S.J. conceptualized the project; and all authors made substantial contributions to data acquisition, critically revised the manuscript, and gave final approval for manuscript submission.

Conflict-of-interest disclosure: A.C. reports research support from Janssen and consulting for AbbVie, Adaptive, Amgen, Antengene, Bristol Myers Squibb, FORUS, Genentech Roche, GlaxoSmithKline, Janssen, Karyopharm, Millennium Takeda, and Sanofi Genzyme. H.J.C. reports employment with the Multiple Myeloma Research Foundation and research support from Genentech Roche, Bristol Myers Squibb, and Takeda, unrelated to this project. L.J.S. reports consulting and advisory board roles for Janssen. C.R. reports research support from Amgen, Janssen, Bristol Myers Squibb, and Teneobio; consulting for Amgen, Janssen, Bristol Myers Squibb, Karyopharm, Sanofi, and Artiva; speakers bureau roles for Bristol Myers Squibb; and scientific advisory board roles for Bristol Myers Squibb, Janssen, Sanofi, and Artiva. A.C.R. reports consulting for Johnson & Johnson, Bristol Myers Squibb, Karyopharm, Adaptive, and Sanofi. S.R. reports honoraria received from Janssen and Bristol Myers Squibb; steering committee roles with Gracell Therapeutics and Bristol Myers Squibb; research support from Janssen, Bristol Myers Squibb, C4 Therapeutics, Gracell Therapeutics, and Heidelberg Pharma; and advisory board roles for Genentech and Janssen. S.P. reports research support from Bristol Myers Squibb, Grail, and Caribou; and advisory board roles for Grail. S.J. reports consulting for Janssen, Bristol Myers Squibb, Caribou, Legend Biotech, Regeneron, Takeda, Sanofi, and Poseida; advisory board and data monitoring committee roles for Genmab, Sanofi, and Janssen; and advisory board roles for GlaxoSmithKline and Bristol Myers Squibb. The remaining authors declare no competing financial interests.

Correspondence: Sundar Jagannath, The Tisch Cancer Institute, 1 Gustave L Levy Place, Box 1185, New York, NY 10029; email: [email protected].