ATLAS Trial Signals a Path Toward Customizing Myeloma Maintenance

With the implementation of triplet and quadruplet induction regimens in multiple myeloma, the search is on for new ways to adapt these therapies to best address the needs of patients. In an interview with Targeted Oncology, Andrzej J. Jakubowiak, MD, PhD, director of the University of Chicago’s Multiple Myeloma Program, explained how recent findings from the phase 3 ATLAS trial (NCT02659293) may offer a new strategy to enhance the depth of response yet provide a path to reducing the burden of treatment over time.

The trial investigated carfilzomib (Kyprolis), lenalidomide (Revlimid), and dexamethasone (KRd) as posttransplant maintenance for newly diagnosed multiple myeloma and utilized a risk-stratified, minimal residual disease (MRD)-directed approach in which some patients could be deescalated to lenalidomide alone. Results published in Lancet Haematology confirmed that the KRd arm achieved statistically superior progression-free survival (PFS) and overall survival (OS) compared with lenalidomide alone.¹

In the interview, Jakubowiak emphasized that both high-risk and favorable-risk patients benefited from the intensified regimen. He also highlighted the growing role of MRD in personalizing therapy, suggesting that highly sensitive tools like mass spectrometry will further refine de-escalation strategies. Looking ahead, he anticipates that incorporating bispecific antibodies and chimeric antigen receptor (CAR) T-cell therapies may eventually allow some patients to avoid transplant altogether.

Targeted Oncology: What developments have there been for maintenance therapy in multiple myeloma?

Andrzej J. Jakubowiak, MD, PhD: Over the last few years, a few themes have emerged in treatment of newly diagnosed myeloma. The first theme was importance of intensification of induction by adding antibodies to quadruplet combinations. A few randomized positive studies have been already reported. The second theme is the importance of duration of treatment; it's important that we do not stop therapy too early. A number of studies have shown that, and in line with that, it is also the subject of our report from the ATLAS randomized phase 3 trial, where we showed that intensification and extended duration of multidrug posttransplant treatment—which we still can call a form of maintenance posttransplant—is important.

In the ATLAS trial, we added the proteosome inhibitor carfilzomib to more traditional maintenance with single-agent lenalidomide, but there are other ongoing efforts where the intensification of treatment may be by way of adding more novel treatments like bispecific antibodies.

In all the data which we have seen so far, while we are losing a little bit on slightly higher toxicity when we add additional agents or extended duration of treatment, the overall balance is that this is potentially worth taking as a risk for the gain which you have in some of these studies. We have clearly seen prolongation of PFS, but there are some data, including the ATLAS trial, [suggesting] that even OS can be prolonged by some of these strategies.

What led you to investigate KRd maintenance in the ATLAS study?

What we have learned from the 2012 publication of the original KRd trial and later in 2020 the KRd followed by transplant trial: first of all, KRd is highly effective, appearing in those studies and by their results to be superior to what we have seen at that time with triplets and alternatives like VRd [bortezomib (Velcade), lenalidomide, and dexamethasone].^2,3 Despite some concerns that carfilzomib may have cardiovascular toxicities, it appears that actually counterintuitively, KRd over a longer period of time is for most patients very well tolerated. Because of that, patients can take extended treatment.

What we have observed in those 2 studies was that, first of all, that tolerability for extended time is good. And No. 2, the results were very promising. No. 3, we have clearly documented based on stringent complete response [sCR] as an end point, or MRD, in the newer study, that we continue to see deepening of the response beyond 8, 9 and 12 cycles. The biggest decline of tumor burden was in the first few months, but when patients reached 8 or 9 months, they continue to convert their level of response to deeper level of response, which justified the design of the ATLAS trial, in which we have used an extended treatment with KRd in the experimental arm as an alternative strategy and compared to a control arm of posttransplant standard of care, which currently is single-agent lenalidomide.

In the ATLAS trial, in the KRd arm, we almost identically recapitulated KRd treatment from our KRd and transplant trial, a single-arm trial, except that we diverged treatment in this arm after 8 months into 2 different pathways. All enrolled patients in both arms received, prior to initiation of trial, autologous transplant. In both arms, we conducted evaluation of response after 6 months, including MRD using clonoSEQ, and we adopted an MRD-directed and risk stratified strategy in the KRd arm to limit unnecessary exposure to KRd.

We have given an option of converting therapy for patients with favorable cytogenetic risk factors and for patients who achieved MRD negativity using clonSEQ method at 6 cycles, to lenalidomide alone. This group of patients, after 8 cycles of KRd posttransplant, were converted to lenalidomide alone. Based on these characteristics, close to 50% of these patients were deescalated to lenalidomide alone after 8 months.¹ The remaining patients—either if they had one of the high-risk cytogenetic features, or if they did not achieve MRD negativity, or both—continued KRd for total for up to 36 months, or as long as they tolerate it. Most patients could tolerate 36 months of KRd therapy.

What were the results of the trial?

Despite the fact that we reduced the treatment intensity for the good-risk group with no high-risk features, this overall KRd group had superior outcome compared to lenalidomide. In fact, the group which was deescalated had the best outcomes, probably because they had disease which was sensitive to KRd and a combination of 8 cycles of KRd and maintenance was good enough for this group of patients to contribute to superiority of KRd in this trial.

The study was reported earlier as positive based on preliminary results which we published in Lancet Oncology in 2023.⁴ Now we published final results in Lancet Haematology which confirmed highly statistically superior PFS of KRd vs lenalidomide in posttransplant therapy.¹ In addition, even if it was not powered to seek this end point, we found superior OS for the KRd arm. We always argue that PFS may not be enough if a patient can be rescued, but if we show there is survival benefit, one has to pause when we choose what to do with patients with transplant by showing there's not only longer remission, but longer survival.

We have shown that high-risk patients benefit from extended KRd. But we also showed that that benefit is also highly significant for favorable-risk patients (ie, patients without high-risk cytogenetics). Based on this result, we propose not limiting KRd to only high-risk patients as maintenance, because the good-risk patients were benefiting as well.

How did MRD contribute to this trial?

Deescalation was possible because we have used MRD evaluation in our studies very systemically at consistent landmark time-points for a long period of time. We generated a lot of experience with the clonoSEQ assay for many years and for multiple of our studies supported by our publications. We have learned that patients with good risk and MRD negativity are doing very well, even in studies when we had no transplant. So, based on these observations, we felt fairly comfortable that deescalation would not be harmful. We feel good with the ATLAS results, as they were generated in a well-conducted international phase 3 study, showing as a principle that this strategy can work and it adds to our work like MRD2STOP [NCT04108624], where we actually deescalate completely in patients who are MRD negative after maintenance.⁵ We can follow these patients for many years without treatment for the majority of them in remission. Deescalation is an important part of the study design and strategies we have right now.

What are the next steps for this research?

No. 1, we are on the verge of defining a group of patients who may need extended treatment, but not necessarily with transplant as part of the treatment strategy. The standard of care remains that eligible patients after induction receive consolidation with high-dose melphalan and autologous transplantation, followed by maintenance. But transplant with high-dose melphalan has 2 main risks and important considerations. First of all, the idea of having such an intensive treatment with admission and toxicities so early in the treatment course is not liked by many patients, Patients go with the recommendation for transplant, but this is with a lot of trepidation for many of our patients. Even more important is that melphalan and transplant add to the risk of developing second malignancies like leukemia or other cancers, and if it is not absolutely needed for all patients, why take this risk?

To add to that, if we can define a group of patients who can avoid transplant, that may be to their benefit. The randomized phase 3 MIDAS trial [NCT04934475] already reported that after quadruplet induction for 6 cycles, patients who achieve MRD negativity do not seem to statistically benefit from transplant compared with extended quadruplet treatment with isatuximab [Sarclisa] and KRd.⁶ In MIDAS, KRd is also used as induction and extended treatment. We may need to reach a more generalizable set of data that achieving MRD negativity after induction not only with Isa-KRd, but with different regimens, can be used to forgo transplant. Maybe 6 cycles is too early. Maybe we can look for a different time point where MRD is predictive of good outcome at the time when transplant is not adding any meaningful prolongation of remission, notwithstanding survival.

What other areas of research have potential to modify the structure of frontline treatment?

Considering more open-ended areas of research, we are very much still attached to triplets like VRd and KRd, which we use as a backbone in quadruplets with daratumumab [Darzalex] or isatuximab, the anti-CD38 antibodies. We already have data that we can instead of anti-CD38 antibodies incorporate T cell engagers like targeting BCMA [B-cell maturation antigen] at induction or later in maintenance, and that it appears to be generating wonderful results. In MajesTEC-4 [NCT05243797], we have seen 100% progression free at some point, which probably requires a bit longer follow up to see that it holds over time.⁷ This is in my opinion the next wave of development, going to some of these strategies, and part of this will be also using CAR Ts as a consolidation of past induction like in CARTITUDE-5 [NCT04923893] or CARTITUDE-6 [NCT05257083] which have completed enrollment, and some other studies earlier in development, which may potentially eliminate transplant and even potentiate what we are achieving by extended induction or instead of transplant.

I will not limit it to only T-cell engagers and CAR Ts. There are also antibody-drug conjugates, which are also in this space. So there is a lot of movement in this direction, and a lot of preliminary data which look amazing, in my opinion, which makes me think that triplets with CD38 antibody may not be history, but may not necessarily be a first choice if we get very solid and well documented data from some alternatives with either bi-specifics or CAR Ts or antibody-drug conjugates.

In addition, we are starting to see that we not only can deescalate treatment based on biomarkers, and still have good outcomes, but that we can identify patients who have disease eliminated by some of these strategies. We claimed cure potential for about 33% of patients after CAR T in very advanced relapsed disease. We anticipate and already have encouraging preliminary results to confirm this in earlier lines, it will be a higher proportion of patients who may be considered potentially cured. Some patients may be also achieving cure with combination of induction, transplant, and postinduction consolidation and maintenance. At this time, we are likely curing some patients and identifying patients who are if not cured, can be considered functionally cured, and what are the conditions to potentially offer these patients completely stoppage of therapy, decreasing toxicities, risk of second malignancy, and costs, which is not unimportant.

How can this approach benefit patients in the long term?

I was among those who argued that we are obliged to explore better treatments, even if some felt that they are too costly and risking toxicities. By gaining this knowledge, we can apply more expensive therapies but give an opportunity to actually shorten overall duration of treatment, which otherwise for a majority of patients is for the rest of their lives. I would like to refer back to MRD2STOP trial, in which we showed as proof of principle, supported also by the Greek [Myeloma Study Group] that one can stop maintenance treatment based on MRD negativity and expect with high probability of staying in remission 3, 5, and more years later.^5,8

We have now better tools for measuring residual disease. Bone marrow may not be ideal. We have blood-based mass spectrometry, highly sensitive methods with liquid chromatography combined with mass spectrometry or clonotypic-based mass spectrometry, which I anticipate can add to identifying patients who can stop treatment. We already have generated preliminary results in MRD2STOP that sensitivity at 10^-7 predicts more precisely who will not be progressing after stopping treatment that 10^-6. So we have tools. We just need to get more …data, and we need to look forward with this type of development and with hope for many of us and for our patients.

REFERENCES

1. Jakubowiak AJ, Wrobel T, Jamroziak K, et al. Carfilzomib, lenalidomide, and dexamethasone (KRd) as maintenance therapy after autologous stem-cell transplantation (ASCT) in patients with newly diagnosed multiple myeloma (NDMM): Final analysis of the ATLAS phase 3 trial. Lancet Haematol. 2026. doi:10.1016/S2352-3026(26)00011-6

2. Jakubowiak AJ, Dytfeld D, Griffith KA, et al. A phase 1/2 study of carfilzomib in combination with lenalidomide and low-dose dexamethasone as a frontline treatment for multiple myeloma. Blood. 2012;120(9):1801-1809. doi:10.1182/blood-2012-04-422683

3. Jasielec JK, Kubicki T, Raje N, et al. Carfilzomib, lenalidomide, and dexamethasone plus transplant in newly diagnosed multiple myeloma. Blood. 2020;136(22):2513-2523. doi:10.1182/blood.2020007522

4. Dytfeld D, Wrobel 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;24(2):139-150. doi:10.1016/S1470-2045(22)00738-0

5. Derman BA, Major A, Cooperrider J, et al. Discontinuation of maintenance therapy in multiple myeloma guided by multimodal measurable residual disease negativity (MRD2STOP). Blood Cancer J. 2024;14(1):170. Published 2024 Oct 7. doi:10.1038/s41408-024-01156-x

6. Perrot A, Hulin C, Lambert J, et al. MRD-driven strategy following IsaKRD induction in transplant-eligible NDMM: Primary endpoints of the phase 3 MIDAS trial. J Clin Oncol. 2025;43(suppl 16):7500. doi:10.1200/JCO.2025.43.16_suppl.7500

7. Zamagni E, Silzle T, Špička I, et al. Phase 3 study of teclistamab (tec) in combination with lenalidomide (len) and tec alone versus len alone in newly diagnosed multiple myeloma (NDMM) as maintenance therapy following autologous stem cell transplantation (ASCT): Safety run-in (SRI) results from the Majestec-4/EMN30 trial. Blood. 2024;144(suppl 1):494. doi: 10.1182/blood-2024-205608

8. Malandrakis P, Ntanasis-Stathopoulos I, Kostopoulos IV, et al. Sustained MRD negativity for three years can guide discontinuation of lenalidomide maintenance after ASCT in multiple myeloma: results from a prospective cohort study. Blood. 2024;144(suppl 1):361. doi:10.1182/blood-2024-209826