Nation-Wide Retrospective Analysis of Allogeneic Stem Cell Transplantation in Patients with Multiple Myeloma: A Study from Korean Multiple Myeloma Working Party (KMM1913)
Article information
Abstract
Purpose
The role of allogeneic stem cell transplantation (alloSCT) in multiple myeloma (MM) treatment remains controversial. We conducted a retrospective, multicenter, nationwide study in Korea to evaluate the outcomes of alloSCT in Asian patients with MM.
Materials and Methods
Overall, 109 patients with MM who underwent alloSCT between 2003 and 2020 were included in this study. Data were collected from the Korean Multiple Myeloma Working Party Registry.
Results
The overall response rate and stringent complete response plus complete response (CR) rates were 67.0 and 46.8%, respectively, after alloSCT. At a median follow-up of 32.5 months, the 3-year probability of progression-free survival (PFS) and overall survival (OS) rates were 69.3% and 71.8%, respectively. The 3-year probabilities of OS rates in the upfront alloSCT, tandem auto-alloSCT, and later alloSCT groups were 75.0%, 88.9%, and 61.1%, respectively. Patients who achieved CR before or after alloSCT had significantly longer OS (89.8 vs. 18 months and 89.8 vs. 15.2 months, respectively). Even though patients who did not achieve CR prior to alloSCT, those who achieve CR after alloSCT had improved PFS and OS compared to those who had no achievement of CR both prior and after alloSCT. Patients who underwent alloSCT with 1-2 prior treatment lines had improved PFS (22.4 vs. 4.5 months) and OS (45.6 vs. 15.3 months) compared to those with three or more prior treatment lines.
Conclusion
AlloSCT may be a promising therapeutic option especially for younger, chemosensitive patients with earlier implementation from relapse.
Introduction
Autologous hematopoietic stem cell transplantation (ASCT) and the development of new agents with potent antimyeloma activity have considerably improved the survival of patients with multiple myeloma (MM) [1,2]. However, MM remains incurable for most patients. Allogeneic stem cell transplantation (alloSCT) through graft versus-myeloma effect has gradually emerged as a potential cure for MM [3]. Although it is a potentially curative approach [4,5], the role of alloSCT in MM treatment is still controversial, primarily due to considerable toxicity, immunosuppression, and subsequent infections, risk of graft-versus-host disease (GVHD), and thus, a potentially high nonrelapse mortality (NRM) [6]. In this regard, alloSCT should be considered an appropriate therapy for any eligible patient with early relapse (< 24 months) after a primary therapy that included an ASCT and/or high-risk features, as it was considered within a clinical trial setting in patients relapsing after primary therapy [7].
Despite a high remission rate of up to 50% in retrospective analyses, early approaches in the 1980s and 1990s with high-dose myeloablative conditioning regimens were limited to younger patients with relapsed/refractory disease due to the high therapy-related toxicity, with NRM rates of 40%-60% [8]. NRM was reduced through improved supportive care and a more rigorous patient selection, but long-term survival was only achieved in 10%-25% of patients [8]. Recently, the occurrence rates of treatment-related mortality (TRM) and GVHD have been reduced through advanced maintenance strategies, better supportive care, more suitable patient selection, and strategies for GVHD prophylaxis. Therefore, alloSCT may be well tolerated and be an effective way to cure MM in the future.
Currently, a majority of the available data on alloSCT in MM originates from Caucasian populations. This retrospective, multicenter, nationwide Korean study evaluated the outcomes of alloSCT in Asian patients with MM.
Materials and Methods
1. Patient description and data source
Patients with MM who underwent alloSCT between 2003 and 2020 were included. The clinical data of patients with MM were collected from the Korean Multiple Myeloma Working Party (KMMWP) registry (part 1), which encompasses data on MM. More detailed stem cell transplantation data were collected in a previous retrospective study (section 2). KMMWP planned the study, which was approved by the scientific committee (study number: KMM1913). Fourteen centers in Korea participated in this study, and each center’s Institutional Review Board approved the study.
The patients were divided into three groups according to the timing of alloSCT. The upfront alloSCT group included patients who underwent upfront alloSCT after the induction therapy. The tandem ASCT-alloSCT group included patients who received alloSCT after an upfront single ASCT as autoallo tandem transplantation, with a maximum interval of 8 months between ASCT and alloSCT. The later alloSCT group included patients who received alloSCT at a later time after one, two, or three ASCT with at least 8 months between the first ASCT and alloSCT, mainly as second-line treatment and beyond.
2. Definitions and study parameters
The study parameters were the overall response rate, response duration, progression-free survival (PFS), overall survival (OS), and NRM according to the three alloSCT groups: prior treatment line, response status before or after alloSCT, conditioning regimens, and incidence of acute and chronic GVHD after allSCT. We also evaluated the response rate and duration of new agents after alloSCT relapse. Acute and chronic GVHD were graded according to previously published criteria [9]. The incidence of chronic GVHD was evaluated in patients who survived for at least 100 days. Treatment response and disease progression were evaluated according to the International Myeloma Working Group uniform response criteria [10]. High-risk MM was defined as having high-risk chromosomal abnormalities: del(17p), t(4;14), t(14;16), or chromosome 1q abnormalities on fluorescence in situ hybridization [11].
3. Statistical analysis
PFS was defined as the time from alloSCT to disease progression or death from any cause, whereas NRM was defined as death without disease progression or relapse. Disease-free survival (DFS) was determined from the date of complete response (CR) to relapse, death in CR, or last follow-up. OS was defined as the time from alloSCT until the date of death or last follow-up. The probabilities of PFS and OS were estimated according to the Kaplan-Meier method and compared among the groups using the log-rank test. p < 0.05 was considered statistically significant.
Results
1. Patient characteristics
The patient characteristics are summarized in Table 1. Overall, 70 (64.2%) males were included, with a median age at diagnosis of 49 years (range, 22 to 61 years). Complete cytogenetic analysis was performed in 38 of 109 patients. Among these, 13 patients (34.2%) had high-risk cytogenetic abnormalities.
Patient characteristics on diagnosis
The median age of the patients at alloSCT was 51 years (range, 23 to 63 years). Patients received a median of three (range, 1 to 8) lines of prior therapy, excluding ASCT. Ninety-five patients (87.2%) underwent ASCT. Fourteen patients (12.8%) underwent upfront alloSCT, nine patients (8.2%) underwent tandem ASCT-alloSCT, and 86 patients (78.9%) underwent later alloSCT. Of the 109 patients, 51, 29, 6, and five underwent human leukocyte antigen (HLA)–matched sibling, HLA-matched unrelated, HLA-mismatched unrelated, and haploidentical donor transplantation, respectively. Eighteen patients had no donor information in their registry. Table 2 shows the MM-treatment parameters before alloSCT.
Patient characteristics prior to and at allogeneic stem cell transplantation
2. Transplantation outcomes
Seventy-two and 67 patients were evaluated for grades IIIV acute and chronic GVHD, respectively. Of those patients, 32 (44.4%) developed grade II to IV acute GVHD, while 26 (38.8%) developed chronic GVHD. Nineteen patients (17.4%) developed NRM (Table 2). Of them, eight patients died due to septic shock, four patients due to pneumonia, three patients due to veno-occlusive disease, two patients due to acute GVHD, one patient due to infection, and one patient due to myocardial infarction, respectively.
The disease status before alloSCT included 26 stringent CR (sCR) plus CRs (23.9%), 16 very good partial responses (VGPR) (14.7%), 34 partial response (PR) (31.2%), and 12 cases of stable disease (SD) (11.0%). After alloSCT, we observed the following best responses: 51 sCR plus CR (46.8%), eight VGPR (7.3%), 14 PR (12.8%), nine SD (8.3%), and two minimal responses (1.8%). Ten patients (9.2%) developed progression (Table 2). The overall response rate was 67.0% after alloSCT. The proportions of CR or better and VGPRs or better were 23.9% and 38.5% before alloSCT and increased to 46.8% and 67% after alloSCT, respectively.
At the time of analysis, 42 of 109 patients were alive with a median follow-up of 32.5 months (range, 0.1 to 201 months) after the date of transplantation. The median PFS was 10 months (95% confidence interval [CI], 5.3 to 14.7), and the probabilities of PFS at 1 and 3 years were 45.4%±0.049% and 23.6%±0.042%, respectively. The median OS was 32.5 months, and the probabilities of OS at 1 and 3 years were 66.9%±0.046% and 47.6%±0.05%.
For the upfront alloSCT, tandem auto-alloSCT, and later alloSCT groups, the median PFS was 44 (95% CI, 0 to 89), 30.1 (95% CI, 18.7 to 41.5), and 6.9 months (95% CI, 3.5 to 10.3), respectively. The probabilities of PFS at 1 and 3 years were 75.0%±0.153% and 60.0%±0.182% in upfront alloSCT group, 88.9%±0.105% and 44.4%±0.166% in tandem auto-alloSCT group, and 86.3%±0.038% and 17.2%±0.043% in later alloSCT group, respectively. The median OS was 58.5 (95% CI, 4.1 to 113), 107.1 (95% CI, 16 to 198.3), and 23.5 months (95% CI, 7.1 to 39.9), respectively. The probabilities of OS at 1 and 3 years were both 75.0%±0.153% in the upfront alloSCT group, both 88.9%±0.105% in the tandem auto-alloSCT group, and 86.3%±0.038% and 61.1%±0.056% in the later alloSCT group, respectively. The cumulative incidences of NRM at 1 year were 25.0%, 11.1%, and 20.6% in the upfront, tandem auto-, and later alloSCT groups, respectively (Fig. 1).
Progression-free survival (A), overall survival (B), and nonrelapse mortality (NRM) (C) rates in the three allogeneic stem cell transplantation (alloSCT) groups.
Patients who underwent alloSCT with 1-2 prior treatment lines had improved PFS (22.4 vs. 4.5 months, p < 0.001) and OS (45.6 vs. 15.3 months, p=0.006) compared to those with three or more prior treatment line (Fig. 2).
Progression-free survival (A) and overall survival (B) according to prior treatment line.
When distinguishing between patients who did and did not achieve CR, including sCR prior to alloSCT, the PFS for patients achieving CR prior to alloSCT had statistically significantly longer compared with those who did not achieve CR with median PFS of 41.3 vs. 6.3 months, respectively (p < 0.001) (Fig. 3A). Similarly, in patients showing CR prior to alloSCT, median OS was significantly prolonged with 89.8 months compared to 18 months in those without achievement of CR prior to alloSCT (p=0.005) (Fig. 3B).
Progression-free survival (PFS) and overall survival (OS) according to disease status prior to allogeneic (A, B) and after allogeneic stem cell transplantation (alloSCT) (C, D). (E, F) PFS and OS according to changes in disease status during alloSCT. CR, complete response; NA, not available.
When comparing patients achieving CR with those without achievement of CR after alloSCT, in patients achieving CR, the median PFS was statistically significantly prolonged with 30.1 months compared with only 4.7 months in those without achievement of CR after alloSCT, respectively (p < 0.001) (Fig. 3C). Patients achieving CR after alloSCT also had superior OS compared with those who did not achieve CR after alloSCT (median OS, 89.8 months vs. 15.2 months, p < 0.001) (Fig. 3D). Even in patients who did not achieve CR prior to alloSCT, if those patients achieved CR after alloSCT, PFS and OS were improved compared with those who had no achievement of CR both prior and after alloSCT (median PFS, 17.2 vs. 4.7 months; median OS, 58.5 vs. 15.2 months) (Fig. 3E and F).
We also compared the survival outcomes according to the conditioning regimen. The median PFS was 12.1 and 3.7 months in the FluMel and FluBu groups, respectively (p < 0.001) (Fig. 4A). The median OS was 37.7 and 10.8 months in the FluMel and FluBu groups, respectively (p < 0.001) (Fig. 4B).
Progression-free survival (A) and overall survival (B) according to the conditioning regimen. FluBu, fludarabine plus busulfan; FluMel, fludarabine plus melphalan; NE, not evaluable.
In the current study, nine of 109 patients achieved a long-term PFS of > 5 years. We focused on parameters that could contribute to the long-term survival of these patients. We compared the parameters between the nine patients who had a long-term DFS of more than 5 years and the remaining patients as controls. Patients with long-term DFS had a low number of Durie-Salmon stage III disease, chromosome abnormalities, prior chemotherapy lines, and more upfront alloSCT than later alloSCT. Patients who achieved sCR or CR before and after alloSCT were included as long-term DFS survivors. Almost all patients with long-term DFS received FluMel- or melphalan-based conditioning regimens and none received FluBu- or BuCy-based conditioning regimens. The incidence of acute GVHD was lower in long-term DFS survivors than in the control group, whereas the incidence of chronic GVHD was similar in both groups (Table 3). Of nine long-term DFS patients, five achieved sCR or CR prior to allSCT. Of the four patients who achieved less than CR, including progressive disease, all achieved sCR or CR after alloSCT.
Characteristics of long-term survivors (disease-free survival > 5 years) and other patients
Discussion
Despite the advent of highly effective new therapeutic modalities for MM and promising data in immunotherapies (chimeric antigen receptor T-cell therapy [CAR-T], bispecific antibodies), alloSCT remains a method with curative potential.
In this nationwide, multicenter, retrospective study in Korea, almost all alloSCTs were performed for second-line and beyond treatment, not in upfront settings. This might be related to physicians’ concerns regarding the relatively high mortality rate from alloSCT, as well as reimbursement issues by the national health insurance service in Korea.
However, the role of upfront alloSCT in high-risk patients with MM remains unclear. From this perspective, upfront tandem auto-alloSCT demonstrated promising results where debulking by a previous ASCT appears to maximize the graft-versus-myeloma effect by the alloSCT [12]. The current study showed that the tandem auto-alloSCT group had a median OS of 107 months, which was superior compared with that of the upfront alloSCT group, while the PFS rates of both groups were comparable. Upfront alloSCT without prior ASCT is less beneficial, mainly because of a higher NRM of 25% at 1 year, which is similar to the NRM in patients transplanted in later stages of the disease. Recently, new drugs have been combined in triplets or quadruplets, including second-generation IMiDs lenalidomide, second-generation proteasome inhibitor carfilzomib, monoclonal antibodies against CD38, SLAM7, and BCMA induction therapy, and subsequently consolidated with ASCT as well as maintenance therapy, which has improved PFS and OS [13]. Therefore, alloSCT as a frontline therapy should be carefully considered.
Recently, alloSCT has been conducted in relapsed/refractory situations in heavily pretreated patients, mostly showing high-risk disease based on cytogenetic analysis and/or International Staging System (ISS) or Revised ISS. We observed a median PFS of 6.9 months and a median OS of 23.5 months in patients who received later alloSCT. For the cohort with relapsed/refractory disease, a median OS of 13-24 months was recorded in different trials [14,15], with our result of 23.5 months ranking in the upper range of these values, whereas the TRM rate with alloSCT remained relatively high. Regarding the comparison of alloSCT and ASCT after relapse from upfront ASCT, a large registry analysis revealed no advantage of a salvage alloSCT over repeat ASCT [16]. In that study, OS after salvage ASCT was higher than that after alloSCT in the intermediate-risk group. However, several conflicting retrospective studies have been reported. Patriarca et al. [17] reported that the donor group had higher PFS and OS than the no donor group. In another small donor versus no donor comparison, de Lavallade et al. [18] demonstrated that patients with relapsed MM and an HLA-identical sibling who underwent reduced-intensity allograft had a significantly better event-free survival than those without an HLA-identical sibling (46% vs. 8% after 3 years). These studies suggest the existence of a specific population suitable for alloSCT.
The present study showed that patients with fewer prior treatment lines had superior PFS and OS compared with those with more prior treatment lines. The OS of patients who underwent alloSCT after 3 prior treatment lines was only 15 months and 45 months after 1-2 prior treatment lines. Our results were comparable with previous studies where 5-year OS was approximately 60% in alloSCT used as a firstline therapy [19,20] and 30% in subsequent lines [21]. Another study revealed that heavily pretreated patients showed significantly worse survival after alloSCT while early intervention had relatively less toxicity, leading to lower NRM [22]. Therefore, earlier implementation of alloSCT may lead to better survival outcomes.
Better clinical outcomes are likely to be associated with myeloma chemosensitivity during transplantation. In the current study, patients who achieved CR before and after alloSCT showed favorable survival outcomes. A Japanese study also revealed that patients who achieved a VGPR or better before alloSCT showed favorable outcomes after alloSCT [23]. Montefusco et al. [24] also described that chemorefractory disease at alloSCT was one of the factors related to a shortened PFS, and a chemosensitive disease group that achieved at least a VGPR before alloSCT was associated with a prolonged PFS. We also confirmed that even in patients who did not achieve CR before alloSCT, long-term survival is a potential advantage.
Few studies have compared the conditioning regimens for alloSCT until now. Generally, FluBu is allocated as a myeloablative regimen, whereas FluMel is allocated as a reduced-intensity conditioning regimen. In the current study, the FluMel conditioning regimen had superior survival outcomes compared to FluBu for alloSCT, which might be related to the higher TRM rate in the FluBu group compared to the FluMel group (22.6% vs. 14.7%). Previous reports also showed that use of myeloablative conditioning resulted in unacceptably high early TRM of up to 40%-50% [25,26]. One of the major concerns of RIC regimens is the increased risk of relapse [27]. However, in our study, relapse rate was not significantly different between the two groups. Furthermore, Bashir et al. [28] showed a lower relapse rate in the FluMel100 arm compared with that in the FluMel140 arm, 43% vs. 70%, respectively). Controversial remains, however, the FluMel conditioning regimen may be more feasible for alloSCT in patients with MM.
We also evaluated the characteristics of patients who achieved long-term survival after alloSCT and found that upfront alloSCT, fewer prior chemotherapy lines, achieving sCR or CR both pre and post-alloSCT, and the FluMel conditioning regimen were associated with long-term survival after alloSCT.
In conclusion, alloSCT can be a beneficial treatment option for a select group of patients with MM, even in heavily pretreated settings, and even in the era of novel agents. Improvements in transplantation techniques and supportive care have reduced the rate of NRM after alloSCT. Moreover, new drugs have enabled better disease control before alloSCT, thereby improving long-term survival outcomes. AlloSCT may be a promising therapeutic option, especially for younger, chemosensitive, and high-risk patients at the first relapse. Recently, CAR T-cell therapy and bispecific antibodies have been widely used after relapse following upfront ASCT. However, these agents are difficult to use in many countries, including Korea. Therefore, in this setting, alloSCT may be a useful approach for patients with selective relapses after ASCT.
Notes
Ethical Statement
This study was reviewed and approved by the institutional review board (IRB) of Pusan National University Hospital (IRB No. 2005-004-090). Informed consent was waived as we used anonymized data.
Author Contributions
Conceived and designed the analysis: Shin HJ.
Collected the data: Kim DY, Kim K, Min CK, Lee JJ, Mun YC, Lee WS, Lim SN, Kim JS, Moon JH, Kim DJ, Bang SM, Won JH, Jo JC, Ko YI.
Contributed data or analysis tools: Kim DY.
Performed the analysis: Kim DY.
Wrote the paper: Shin HJ.
Conflicts of Interest
Conflict of interest relevant to this article was not reported.
Acknowledgements
This work was supported by a 2-Year Research Grant of Pusan National University.
We are grateful to the multicenter physicians for collecting and analyzing the clinical data.