Post-Transplant KRd Maintenance May Allow for Superior Outcomes in Newly Diagnosed Multiple Myeloma

Although continuous lenalidomide (Revlimid) maintenance has been well established as the standard of care (SOC) after induction and autologous stem cell transplant (ASCT), its extended use in combination with carfilzomib (Kyprolis) and dexamethasone (KRd) could provide a viable post-ASCT alternative for patients with newly diagnosed multiple myeloma, according to Andrzej Jakubowiak, MD, PhD. Longer follow-up analysis is necessary to further confirm the regimen’s feasibility, necessary duration, and connection to minimal residual disease (MRD) negativity in this population.

Several previous studies evaluated pre- and post-ASCT KRd for the treatment of patients with newly diagnosed multiple myeloma, including the phase 2 FORTE trial (NCT02203643), which showed that KRd plus ASCT improved responses vs 2 other carfilzomib-based treatment approaches.¹

The phase 3 ATLAS trial (NCT02659293) aimed to build off these findings while addressing the need for more definitive research on alternative maintenance therapies in this setting. Unlike previous investigations of maintenance combinations, however, ATLAS directly compared KRd with lenalidomide maintenance to challenge its superiority.

Interim analysis occurred after 59 of 96 expected events were observed at a median follow-up of 33.8 months. Preliminary results demonstrated a significant improvement in median progression-free survival (PFS) with KRd, at 59.1 months vs 41.4 months with lenalidomide alone (HR 0.51; 95% CI, 0.31-0.86; P = .012). Notably, a 49% risk reduction was observed with KRd despite a shortened duration of 8 cycles of KRd treatment for patients with standard-risk disease and MRD negativity at cycle 6. Moreover, high MRD negativity in the KRd arm was associated with improved PFS.²

“This difference in PFS between KRd and lenalidomide alone is a key finding. Moreover, we were able to show that a planned 36-month duration of post-transplant KRd treatment may not be required for all patients,” said Jakubowiak, a professor of medicine and director of the Myeloma Program at the University of Chicago in Illinois. “Overall, the interim data from the ATLAS trial support our hypothesis that post autologous stem cell transplant treatment with KRd may improve treatment outcome compared to standard of care lenalidomide maintenance.”

In an interview with OncLive®, Jakubowiak highlighted several key studies that influenced the development and design of ATLAS, introduced conclusions from the initial efficacy data, and discussed how ongoing analysis could further contextualize and inform post-transplant treatment approaches in multiple myeloma.

OncLive^®: What previous data inspired you to challenge the post-transplant maintenance SOC of lenalidomide monotherapy with KRd?

Jakubowiak: Several developments influenced our study design and development. First, we learned from the phase 3 FIRST trial [NCT00689936], which investigated first-line lenalidomide for 18 months vs indefinite administration, that extended treatment is better than cutting intensive therapy after a certain period. I always wondered whether pre-transplant induction treatment for 3 to 4 cycles followed by transplant and single-agent lenalidomide maintenance was good enough for younger patients who wanted to extend their lives as much as possible.

What prompted us specifically were the results from our previous two KRd trials. The first phase 1/2 study investigating KRd without transplant or with deferred transplant in patients with newly diagnosed myeloma (NCT01029054) published in 2012, showed that responses to KRd continued to deepen after cycle 8 during an extended treatment with KRd up to 24 months. The second phase 2 trial (NCT01816971) investigating KRd with ASCT was published in 2020 showed that the rate of stringent complete responses continued to increase during the post-transplant extended KRd treatment in a similar way as in the prior KRd study without transplant, although we could not exclude that some of this could have been related to delayed transplant effect.

We correlated results from both trials and found that KRd deepens patient responses during the first 4 months post-transplant in the majority of patients and in an additional 20% to 30% of patients even beyond that time, where we would not expect to see delayed transplant effects.

What additional studies investigating post-ASCT maintenance strategies in multiple myeloma influenced the hypothesis and trial design of ATLAS?

The impetus for these various study design deliberations came from 2 important studies: the phase 3 STaMINA trial [NCT02322320] conducted in the US and the phase 3 EMN02/HO95 trial (NCT01208766) conducted in Europe. Both studies explored differences in patient outcomes with additional post-transplant treatments. STaMINA did not show any difference between lenalidomide alone vs 4 cycles of consolidation [treatment with bortezomib (Velcade), lenalidomide, and dexamethasone (VRd)] or second transplant.

In STaMINA, several patients in both the transplant and consolidation arms did not receive intended treatment, potentially contributing to no difference in outcome in these 3 arms in an intention-to-treat patient population. A later, less stringent, based on as treated evaluation of additional interventions like second transplant or consolidation did show benefit in this study.

That analysis correlated with data reported from EMN02/HO95, which showed that 2 cycles of VRd and transplant benefited patients, particularly those with high-risk disease. We concluded that longer treatment duration is better.

Since we have already tested KRd post-transplant, we knew that we could deliver it and that other than the inconvenience of coming to a center to receive treatment, patients do not appear to experience more toxicities than with other therapies. That was why we evaluated this strategy in this randomized phase 3 ATLAS trial.

In ATLAS, we selected patients after any type of induction regimen, in contrast to those in other ongoing or reported studies. The strategy was similar to that of the phase 3 CALGB-100104 trial [NCT00114101] of lenalidomide vs placebo, which was one of 2 key studies leading to the approval of lenalidomide maintenance therapy. Patients in CALGB-100104 were eligible for maintenance therapy after receiving any induction therapy followed by transplant, provided they did not have disease progression. We applied similar eligibility criteria in ATLAS to allow for historical comparisons to CALGB-100104.

The FORTE trial, which was initiated after ATLAS, was primarily designed to investigate 2 induction regimens and then compare KRd with or without transplant, but also had a second randomization for maintenance with carfilzomib and lenalidomide (KR) vs lenalidomide (R). Updated data from the FORTE were just published in 2022. One-third of patients randomized to KR were after an extended 12 cycles of induction. The remaining two-thirds were patients who received induction for 4 cycles, transplant, and 4 cycles of either KRd or KCd (carfilzomib, cyclophosphamide, and dexamethasone). While FORTE trial had a different design than ATLAS, both similarly showed that KR (i.e carfilzomib added to lenalidomide) or KRd improve PFS compared with lenalidomide alone.

Cross-study comparison is tricky, but the results from ATLAS and FORTE both document that extended treatment with carfilzomib and lenalidomide combinations improves PFS, with a HR in ATLAS of 0.51. FORTE had a slightly higher HR. However, when you look at the Kaplan-Meier curves, the comparison is striking. There is a similar effect of KR in FORTE and KRd vs lenalidomide in ATLAS.

Could you expand on the methodology and patient population of ATLAS?

ATLAS was supported by Amgen, Celgene, and Bristol Myers Squibb, but was investigator led and coordinated by the University of Chicago, with key contributions from the Polish Myeloma Consortium. As an investigator-initiated study, we aimed to enroll the necessary number of patients to evaluate primary end-point – the difference in PFS between KRd and lenalidomide alone. We used historical data from CALGB-100104 for our statistical modeling for the lenalidomide arm, and data from the KRd with transplant trial for our KRd arm. Based on this model, we concluded that at 4 years, we should be able to see a statistically significant difference in PFS between the 2 arms.

In addition, based on prior result, we hypothesized that we could de-escalate the duration of KRd treatment for a group of patients with standard cytogenetics who achieved MRD negativity at 6 cycles of KRd post-transplant. In total, 44% of patients in the KRd arm were de-escalated to lenalidomide alone maintenance after 8 cycles of KRd according to these criteria.

What initial PFS data did you find in ATLAS? 

In total, 61% of projected events required for final analysis had been seen, and we have already observed a difference in PFS between the KRd and lenalidomide arms. The median PFS was 59.1 months for the KRd arm and 41.4 months for the lenalidomide arm, which represents 49% reduction of progression or death. Bearing in mind that ATLAS trial data come from an unplanned interim analysis, the results are indicating the benefit of post-transplant therapy with combination therapy (KRd) compared to lenalidomide alone.

In this interim analysis, patients who received only 8 cycles of KRd and then were de-escalated to lenalidomide alone treatment were doing as well as the patients who received KRd maintenance for 36 months. It’s a bit of an unfair comparison because these patients are selected based on low risk and MRD negativity, and the other group has a mix of patients. Still, this group does not appear to drive PFS down, although we cannot prove this without a control arm.

We need to wait for confirmation of these data from a final analysis, which will come in about a year and a half. However, we felt it was important to report these interim data earlier than planned.

How did MRD status factor into patient outcomes?

In addition to PFS, this trial, along with other reported or ongoing studies, seems to show that MRD may serve as a marker for successful treatment de-escalation that does not adversely affect outcomes. The FDA and other agencies are also looking into the role of MRD as a potential surrogate marker for patient outcomes. At this point, in our analysis, we see a statistically significant superiority of MRD negativity in the KRd arm vs the lenalidomide arm at 6 cycles. MRD negativity was measured by International Myeloma Working Group criteria, meaning 10-5 sensitivity and a complete response or better. We’ll report later MRD results at other landmark time points, but these results indirectly show that MRD may be a predictor of PFS.

How tolerable is this combination compared with lenalidomide monotherapy?

As expected, there was a slightly higher rate of toxicities and adverse events (AEs) of all grades in the KRd arm compared with the lenalidomide arm. Importantly, the incidence of grade 3 and 4 AEs was comparable between the 2 arms for most categories. Most major differences in higher-level toxicities were higher rates of infections in the KRd arm, mostly lower respiratory infections and thrombocytopenia. Slightly higher rates of other toxicities were also observed in the KRd arm. There was also a slight, but not dramatic, difference in serious AEs between the arms. Together considering the clinically meaningful PFS difference, the analysis of AE data do not indicate that putting patients on KRd is an unnecessary risk, or that the patients are inclined not to stay on KRd treatment.

I would not use carfilzomib-based therapy in frail patients or those with with a significant history of cardiac events. However, patients who met the eligibility criteria for any standard lenalidomide-based maintenance therapy were enrolled, and our analysis indicated that this enrollment approach is reasonable also for carfilzomib-based maintenance.

Notably, this paper did not include quality of life [QOL] data, which we have collected and which we will report later. These data may help us find greater granularity between the 2 arms. We expect that those QOL data may be in line with the toxicity trends we have seen so far, as these therapies are well tolerated, with some trends toward higher toxicities and life inconvenience from frequent treatment visits in the KRd arm.

What next steps are planned for this research, and how might they better inform post-transplant treatment selection for these patients?

This study is ongoing, although the enrollment is completed. We will conduct our final analysis when the last enrolled patient has been in follow-up for 4 years, which will come around the fall of 2024. By that time, we may have more data to support what we have reported from this interim analysis.

Although this study was not designed with regulatory objectives in mind, the National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology has already included considerations for using KR post-transplant based on FORTE data. Our initial ATLAS data align with those guidelines and provide additional evidence for carfilzomib-based post-transplant maintenance.

In the meantime, we anticipate increased interest in more intensive post-transplant treatments like KRd based on risk-benefit analyses. The final analysis will include additional data on MRD results from other landmark time points, as well as QOL data. It will help us see whether any patient subsets may benefit more from KRd vs lenalidomide.

As discussed after our oral presentation at the 2022 ASCO Annual Meeting, this regimen may be considered for patients with high-risk disease, although we don’t have that clear signal yet. However, we have strong evidence to support KRd use in patients with standard-risk disease with or without MRD negativity, as we see a striking difference outcomes for stadard risk patients who receive KRd vs those who receive lenalidomide alone. KRd may also benefit patients with standard-risk disease. Conducting a subgroup analysis at later time points will be critical for clarifying those results and helping us navigate post-transplant treatment.

Editor’s Note: Dr Jakubowiak has consulting and advisory board roles with honoraria with ABBVIE, AMGEN, BMS, Gracell, GSK, Janssen, and Sanofi-Aventis.

References

1. Mina R, Musto P, Rota-Scalabrini D, et al. Carfilzomib induction, consolidation, and maintenance with or without autologous stem-cell transplantation in patients with newly diagnosed multiple myeloma: pre-planned cytogenetic subgroup analysis of the randomised, phase 2 FORTE trial. Lancet Oncol. 2023;24(1):64-76. doi:10.1016/S1470-2045(22)00693-3
2. Dytfeld D, Wróbel T, Jamroziak K, et al. Carfilzomib, lenalidomide, and dexamethasone or lenalidomide alone as maintenance therapy after autologous stem-cell transplantation in patients with multiple myeloma (ATLAS): interim analysis of a randomised, open-label, phase 3 trial. Lancet Oncol.2023;12(1):S1470-2045(22)00738-0. doi:10.1016/S1470-2045(22)00738-0