Autologous Hematopoietic Cell Transplantation in Patients with High Risk Multiple Myeloma: Post- Transplant Responses Do Not Translate to Longer Survival

Background: Multiple myeloma remains an incurable disease with a heterogeneous clinical course, somewhat explained by the occurrence of high-risk prognostic markers. The International Myeloma Working Group defined high risk myeloma (HRM) as the presence of del17p13 or t(4;14) with ISS II/III. Conflicting data exist regarding t(14;16), hypodiploidy and chromosome 1 abnormalities (1q21 amplification, 1p deletion and others).

Methods: We analyzed the outcomes of 142 HRM patients with high risk FISH or cytogenetic findings reported to the CIBMTR from 2008-2012 treated with an upfront (within 12 months of diagnosis), melphalan-conditioned autologous hematopoietic cell transplant and compared them to 573 patients with no high-risk markers (NHRM). Patients that received more than 2 induction regimens were excluded in this analysis. The HRM cohort comprised del17p13 (n=27), t(4;14) (n=27), t(14;16) (n=5), chromosome 1 abnormalities (n=42), hypodiploidy (n=13) and ≥2 high-risk markers (n=31). Planned post-transplant therapy was collected. Outcomes of interest included progression-free survival (PFS) and overall survival (OS).

Results: The HRM and NHRM groups were similar to each other except for the following differences: HRM was associated with lower Karnofsky (KPS) (49% vs 36% with KPS<90, p 0.02) and higher stage at diagnosis (41% vs 28% with ISS/DSS III, p 0.008). More HRM patients received induction with bortezomib and immunomodulatory drug (imid) combinations (55% vs 43%, p <0.001) and had a lower complete response rate prior to transplant (12% vs 16%, p 0.04). More HRM patients had planned post-transplant combined bortezomib and imid therapy (27% vs 12%, p<0.0001). Median follow up in the 2 groups was 36 months for HRM and 44 months for NHRM. At 100 days post-transplant, similar numbers of patients had achieved complete and very good partial responses in the 2 groups (Table 1). At 3 years post-transplant, HRM patients had lower PFS (36% vs 50%, p <0.001) and OS (73% vs 85%, p <0.001) compared to NHRM. Univariate outcomes are shown in Table 2 divided by type of HRM. Table 3 shows the results of the multivariate analysis. The figure shows the Kaplan-Meier curves of probability of survival. Among the relapsed patients (HRM = 91, NHRM = 296), the 2 year survival was 48 (35-60)% for HRM and 70 (64-76)% for the NHRM groups, p-value 0.004.

Conclusions: Patients with HRM achieved similar day 100 response compared to NHRM but were unable to maintain this response over time despite being more likely to receive post-transplant therapy. HRM was associated with shorter PFS and further shortened post-relapse survival. Patients with chromosome 1 abnormalities or del 17p alone appeared to have similar outcomes to those with NHRM while those with t(4;14) and those with more than 1 high-risk marker had the least favorable outcomes. In addition to HRM, obtaining less than complete response prior to transplant and the lack of post-transplant therapy were associated with worse PFS and OS. Finally, African-American race and higher stage at diagnosis were also associated with lower OS in our study.

Figure 1.

View largeDownload slide

Figure 1.

View largeDownload slide

Close modal