Abstract
BACKGROUND: Stratification using cytogenetics (CG), either metaphase karyotyping or fluorescence in situ hybridization (FISH) is used to identify patients (pts) with multiple myeloma (MM) who are at higher risk of relapse and tend to have poorer survival. It is largely unknown if autologous stem cell transplantation (auto-HCT) after high dose melphalan (200 mg/m2) is able to modify the survival of some of these high-risk MM pts and to make it comparable to standard risk. METHODS: Pts were classified as high risk (HR) if either conventional cytogenetics or FISH demonstrated at least one of the following 1q+, 1p-, 17p-, 13q-, t(4;14), or t(14;16), realizing that the inclusion of 13q- by FISH alone and 1p- in the HR MM definition is controversial. Pts with normal chromosomes or those with trisomies and hyperdiploidy were considered standard-risk (SR). We compared progression-free survival (PFS) and overall survival (OS) via a retrospective analysis of pts at the University of Miami Sylvester Comprehensive Cancer Center who underwent auto-HCT between January 2014 and December 2017 for MM. Survival analyses were performed using the log-rank test, with significance at p-value < 0.05. RESULTS: Male pts comprised 56% of the population, and 40% of pts were of Hispanic ethnicity. Of 205 pts undergoing auto-HCT, 108 (53%) had at least one HR cytogenetic abnormality. Interestingly, the depth of response to pre-transplant induction was higher in pts classified as HR, with 71% (77 of 108) achieving at least a very good partial response (VGPR), while 24% (23 of 96) SR pts achieved VGPR. While OS remained largely unaffected in HR pts (34.0 m vs. 35.1 m, p = 0.27); HR pts had an inferior PFS compared to SR pts (21.9 m vs. 25.7 m, p = 0.041). The presence of trisomies did not negate the poorer PFS of HR pts. When we evaluated specific HR CG abnormalities, OS and PFS in patients with 1q+ or t(4;14) were surprisingly comparable to SR pts, indicating a significant benefit from auto-HCT. On the other hand, OS was significantly decreased in pts with 1p- when compared to standard risk (16.5 m vs. 35.1 m, p = 0.004) or other high-risk patients (16.5 m vs. 35.4 m, p = 0.01), implying that 1p- group derive no benefit from auto-HCT. Similarly, OS was shorter in pts with t(14;16) (16.5 m vs. 34.4 m, p = 0.025) and with 17p- (26.6 m vs. 35.1 m, p=0.01), however the PFS was not affected in these populations. In pts with 13q-, PFS was significantly shorter (20.3 m vs. 25.7 m, p=0.023) compared to SR pts without affecting OS. CONCLUSION: At our center in a retrospective analysis of 205 pts: a) patients with HR MM responded better and faster than SR pts to induction, b) those with 1p- did not derive any benefit from transplant and c) pts with 17p- and t(14;16) had some short term benefit (similar PFS to SR group) but at the end their OS remained inferior. Nevertheless, we consider a very important finding the fact that, by having auto-HCT, pts with 1q+ and t(4;14), equalized their PFS and OS to those of SR pts. Based on these findings, patients with 1q+ and t(4;14) should still have a transplant in CR1 even after optimal induction. Whether 17p- and t(14;16) pts can extend PFS benefit after transplant into OS benefit with maintenance regimens stronger than lenalidomide alone remains to be determined. Disclosures Hoffman: Celgene: Honoraria, Speakers Bureau; Loxo: Current equity holder in publicly-traded company; Seattle Genetics: Research Funding.
