Multiple myeloma (MM) is the main type of hematological malignancy originating from plasma cells. MM is currently an incurable disease. With conventional chemotherapeutic regimens, such as melphalan and prednisone (MP), vincristine plus adriamycin and dexamethasone (VAD) and high-dose dexamethasone, the overall response rate (RR) of MM patients is 60%, with only <5% achieving a complete response (CR). The median survival is 2–3 years (1). High-dose chemotherapy followed by autologous stem cell transplantation (ASCT) may achieve a higher RR and longer progression-free survival (PFS) and overall survival (OS) (2–4). Over the last decade, the introduction of the proteasome inhibitor bortezomib and the immunomodulatory agents thalidomide and lenalidomide, has revolutionised MM treatment (5,6). The high RR and CR rate achieved by these novel agents have also raised the question whether ASCT should still be considered as first-line therapy in MM. Several phase 3 trials comparing chemotherapy with first-line ASCT have reported an improved PFS, but no difference in OS (7,8), while others support ASCT as part of the treatment strategy (9,10). It is clear that ASCT enhances the response, even after the most active first-line regimens, including bortezomib plus thalidomide and dexamethasone (VTD) and lenalidomide plus bortezomib, pegylated liposomal doxorubicin and dexametasone (RVDD) (11,12). ASCT is currently considered the standard of care in younger MM patients according to the current National Comprehensive Cancer Network guidelines (13).

However, in China, the majority of the patients may be less likely to undergo transplantation due to multiple reasons, and data on which group of patients may benefit more from ASCT are currently limited. In this study, we performed a retrospective analysis of 114 MM patients, aged <65 years, who were treated with or without ASCT following novel agent-containing induction therapy in our hospital.

Multiple myeloma is currently considered incurable. The aim of the treatment of this disease is to prolong PFS, and eventually OS. The majority of the available studies indicate that achievement of CR was associated with prolonged PFS and OS (6,15). With conventional chemotherapy, the CR rate is currently 5–8%. High-dose therapy followed by ASCT increases the response rate and improves response to treatment, achieving a CR rate of 22–44% (16,17). More recently, novel agent-containing induction therapy has achieved a high response rate and CR rate. In the phase 3 VISTA study of bortezomib plus MP (VMP) vs. MP alone, the CR rate was 28% in the VMP group (18). In other bortezomib-containing induction regimens, the CR rate reached 22–47%, with a 1-year PFS of 83–100% (5). The improved outcome of the novel agents may challenge the role of ASCT in the treatment of MM.

In this retrospective study, CR rate reached 28% (32/114)with the novel-agent induction therapy. Further benefits were obtained in the ASCT group, with the CR rate increasing from 31% (pre-ASCT) to 69% (post-ASCT). The survival analysis also revealed that PFS and OS were significantly prolonged in the ASCT group, supporting the beneficial role of ASCT in younger MM patients in the era of novel agents (19,20). However, different results were reported by other studies. Boccadoro et al (7) compared MP/lenalidomide (MPR) vs. ASCT plus lenalidomide maintenance or no maintenance and indicated that ASCT improved PFS, while the effect on OS was insignificant, suggesting that lenalidomide maintenance may balance the OS in the two groups. The improved OS in our study may be due to a higher number of cases achieving CR in the ASCT group (n=29 case, 69%).

Thalidomide, commonly used as MT and sequential therapy, was shown to improve PFS and OS (21,22). In our study, the multivariate analysis revealed that MT was an independent factor of improved PFS and OS in the non-ASCT as well as the ASCT group. However, a study conducted by Barlogie et al (23) reported a beneficial effect on PFS, but not OS and suggested that the similar OS between the two groups is partially due to the shorter survival following relapse in the thalidomide maintenance group. In addition, Attal et al (22,24) reported an OS advantage from thalidomide maintenance with a follow-up of 39 months, while the OS advantage was lost when the follow-up was prolonged to 5.7 years. In our study, the median OS time has not yet been reached, and elucidating the survival advantages of MT requires a longer follow-up and further investigation.

CR achievement was another independent factor in our study, which was consistent with previous studies (4,6). Attaining a CR has been the major objective in the management of younger patients. However, the data on whether patients who achieved CR following induction therapy may further benefit from ASCT are currently limited. We performed a subgroup analysis according to the response status following induction therapy. Considering the important role of MT on survival, we analyzed ASCT as well as MT. Interestingly, in the subgroup of CR, with or without ASCT/MT, no statistically significant difference in PFS or OS was observed. It was previously demonstrated that stringent complete response (sCR) was an attainable goal following ASCT, which significantly improved survival outcome compared with conventional CR (25). As shown in our study, a better OS (5-year OS rate of 100%) may be achieved in the ASCT/MT arm, which may be due to more sCRs obtained, although the difference was not statistically significant. Our findings support the role of ASCT and MT in deepening the CR.

In the VGPR subgroup, ASCT and MT prolonged the PFS (P=0.010), but without an OS benefit (P=0.582). However, through constructing Kaplan-Meier curves, modest differences were observed among arms. Since the median OS time has not been reached, we hypothesized that different findings may emerged after long-term follow-up. In the PR subset, the patients undoubtedly benefited the most from ASCT and MT in terms of survival (P<0.01). The median PFS of the ASCT/MT arm was 42 months, with a 3-year OS rate of 100%, whereas in the non-ASCT/non-MT arm, the median PFS was only 12 months, with a 3-year OS rate of 35%. A poor PFS (17 months) and a 3-year OS rate of 40% was noted in the ASCT/no-MT arm. In this arm, the sample was too small (n=5) and 3 patients remained in PR following ASCT. Similar to other reports, any response less than VGPR following ASCT was always associated with a poor outcome (26,27).

In summary, novel-agent induction therapy resulted in a higher response rate and response quality. ASCT further increased the CR rate and improved PFS and OS, particularly in patients achieving a PR following induction therapy. In the patients who achieved CR following induction therapy, the role of ASCT requires further investigation. As an independent prognostic factor of better PFS and OS, MT has become an important part of the treatment strategy in MM.