Multiple myeloma (MM) is a plasma cell dyscrasia of the bone marrow (BM) that accounts for around 10% of all hematologic malignancies1 and remains incurable, despite recent advances. The current standard of care for newly diagnosed patients below age 70 with no significant comorbidities is triplet induction therapy, followed by autologous stem cell transplant (ASCT) using high-dose melphalan conditioning, and then risk-adapted maintenance with lenalidomide alone or in combination with bortezomib or other agents2. The use of ASCT for selected patients along with the introduction of novel agents has, over the past two decades, markedly improved the median overall survival (OS) of patients to around 7 to 10 years in younger patients3. However, given the remarkable progress in the treatment of newly diagnosed myeloma, it is increasingly important to determine the optimal induction and maintenance regimen for a given patient, either with or without ASCT.
Numerous trials have investigated the efficacy of ASCT vs. non-ASCT treatment approaches, and consistently demonstrated that ASCT improves progression-free survival (PFS)4-10. However, in most recent studies, upfront ASCT has not significantly improved OS when using modern novel triplets as part of induction remission therapy, and especially when combined with prolonged lenalidomide maintenance7. In this context, it is important to note that between 70-80% of patients randomized to delayed ASCT did eventually undergo ASCT as part of salvage therapy. Given the acute and long-term toxicities of ASCT, the equivalent OS outcome observed with upfront versus delayed ASCT, and the continued progress in developing highly efficacious salvage treatments that are non-ASCT based, a key area of research is to accurately identify those patients who benefit most from ASCT, and conversely, those who may not.
We have recently performed next-generation sequencing on the peripheral blood from a cohort of 629 MM patients who underwent ASCT at Dana-Farber/Brigham and Women's Cancer Center (DF/BWCC) and detected myeloid malignancy-associated somatic mutations in 22% of patients at time of transplant. This entity is currently termed clonal hematopoiesis of indeterminate potential (CHIP) and has been associated with increased risk of developing hematologic malignancies, cardiovascular disease and a higher all-cause mortality11-13. Importantly, we found CHIP to be associated with a worse OS and PFS following ASCT in MM patients. Furthermore, of these 629 patients, 21 (4% overall) eventually developed a secondary leukemia at a median of 4 years post-ASCT and with subsequent lenalidomide maintenance. Additional analysis revealed that in 80% of these cases, the mutation associated with secondary leukemia was already detectable in the peripheral blood at time of transplant. A critical question is whether these patients would have developed secondary leukemia if they had received only conventional novel agent-based therapy without transplant. Current data would suggest this is less likely, given the well-established genotoxic effects of melphalan, and particularly when high-dose melphalan is followed by lenalidomide maintenance, but this remains to be confirmed. Moreover, when evaluating an additional cohort of 986 newly diagnosed treatment-naïve MM patients from the Multiple Myeloma Research Foundation (MMRF) database, we found CHIP prevalence to be similar to that of the post-induction therapy DF/BWCC cohort. This suggests that induction therapy did not have any effect on the emergence of these mutations and thus makes these findings highly provocative.
ASCT and high-dose melphalan remains a standard of care for younger, fit newly-diagnosed MM patients. Given the concern about generating mutated stem cells and the selection for mutant clones to grow into secondary leukemia in a small but significant number of patients, our goal is to further refine treatment choices and tailor therapy to better define those who are at higher risk for developing secondary malignancies14. This may allow us in turn to predict who is best served by keeping ASCT in reserve and only employed if needed15. This is particularly relevant in the current era, where very active and well tolerated novel agent combinations are capable of producing responses and clinical benefit with PFS and a negative minimal residual disease (MRD) status that matches or might exceed that of ASCT16. Yet, since these are preliminary data and testing for CHIP is not yet readily available, we need to fully assess the consequences of CHIP and its true clinical value in the context of ASCT, and outside of it, before using it to determine treatment outcomes for our patients. On a more exciting note, the recently completed DETERMINATION study (led by DF/BWCC) reflects one of the largest efforts to date in this area, and preliminary results are anticipated later this year or early in 2020, with no difference in outcome seen to date between early or delayed ASCT, despite the median follow-up now exceeding 4 years.
Therefore, we suggest all newly-diagnosed patients be referred to an academic medical center for evaluation and a personalized decision be made about the type of induction remission therapy and whether to use up front ASCT in eligible patients. While ASCT remains an important and generally valuable treatment option for younger patients with MM15, we look forward to results of randomized trials which aim to answer questions regarding the optimal treatment regimen for newly-diagnosed MM, with an emphasis on examining the option of keeping ASCT in reserve, and so avoid the potential long-term drawbacks of this approach, as well as identifying those who may benefit most.
