Double Vs Single Autologous Stem Cell Transplantation After Bortezomib-Based Induction Regimens For Multiple Myeloma: An Integrated Analysis Of Patient-Level Data From Phase European III Studies

In the novel agent era, the role of double autologous transplantation (ASCT) as up-front therapy for MM still remains undefined. Recently, several European cooperative groups prospectively compared bortezomib-based vs non-bortezomib-based induction regimens before ASCT for newly diagnosed myeloma (MM). By study design, patients enrolled into these trials were prospectively assigned to receive either a single or double ASCT. A multivariate regression analysis revealed that the leading factors independently associated with shorter PFS and OS were ISS 3, presence of high-risk cytogenetic abnormalities (hr-cyto), including t(4;14) and/or del(17p) by FISH, failure to achieve CR after induction therapy and assignment to receive a single ASCT. In comparison with patients for whom a single ASCT was planned by study design, those who were assigned to receive a double ASCT had significantly longer PFS (median: 38 vs 50 months, p<0.001) and OS (5-years estimates: 63% vs 75%, p=0.002). To evaluate the impact of double vs single ASCT on outcomes, an integrated analysis of patient-level data from these studies was performed. The intent-to-treat (ITT) population included 606 patients who were randomized to bortezomib-based induction regimens and for whom a single (n=254) or double (n=352) ASCT was planned at time of study entry. Based on the presence or absence of one, two or three adverse prognostic variables (ISS 3, hr-cyto and failure to achieve CR after induction therapy), four groups of patients with different risk of progression and/or death were identified. In group 0 were included patients with ISS 1-2 MM, lack of hr-cyto and who achieved CR after induction therapy; these patients represented 13% of the overall population. Patients in group 1 (61%) were identified by the presence of a single adverse variable. Group 2 included patients (23%) with two adverse variables. Patients in group 3 (3%) were identified by the presence of all the three adverse variables. These groups of patients had different clinical outcomes in terms of PFS and OS, and differently benefited from double ASCT. Median PFS was 61 months for patients in group 0, 56 months for group 1, 36 months for group 2 and 26 months for group 3 (p<0.001). A Cox regression analysis adjusted for the number of preplanned ASCT(s) showed the following hazard ratio (HR) values for PFS when group 1 (HR=1.7, p=0.02), group 2 (HR=3.2, p<0.001) and group 3 (HR=6.6, p<0.001) were compared with group 0. In comparison with a single ASCT, prospective assignment to receive a double ASCT was associated with longer PFS for patients with a single (group 1) or two (group 2) adverse prognostic variables (median, 54 vs 43 months; HR=0.70, p=0.006). The greatest PFS benefit with double vs single ASCT was seen for patients in group 2 who had two adverse prognostic variables (median, 41 vs 20 months; HR=0.52; p=0.003), in particular for those with a hr-cyto profile at baseline and who failed CR after bortezomib-based induction regimens (median: 42 vs 21 months with a single ASCT; HR=0.41, p=0.006) (Fig.1). The Cox proportional hazards model for the ITT population confirmed that the presence of two (HR=4.8, p<0.001) or three (HR=9.0, p<0.001) adverse prognostic variables conferred a progressively shorter OS compared to the lack of all the three adverse variables (group 0). Consistently with results of PFS analysis, patients in group 2 who had two adverse variables and by study design were assigned to receive a double ASCT had significantly longer OS in comparison with the same group of patients for whom a single ASCT was planned (median, 67 vs 31.5 months; HR=0.32, p<0.001). OS benefit with double ASCT was particularly relevant for patients who failed CR after bortezomib-based induction therapies and who presented with hr-cyto (5-year estimates: 70% vs 17% with a single ASCT; HR=0.22, p<0.001) (Fig.2) or ISS 3 MM (HR=0.42, p=0.033). To the best of our knowledge, this is the first analysis so far reported comparing double vs single ASCT applied after the gold standard, bortezomib-based, induction regimens. Results suggested a possible beneficial role of double ASCT in improving outcomes for newly diagnosed MM patients with poor prognosis, in particular for those who failed CR after exposure to bortezomib as part of induction therapy and who had a hr-cyto profile. These data need to be confirmed by prospective phase III clinical studies which are currently ongoing.