Impact of Treatment on Overall Survival (OS) in Higher-Risk Myelodysplastic Syndromes (MDS): A Report from the Erasme Study

Introduction: Treatment of MDS has undergone a dramatic change in recent years with the emergence of demethylating agents; however, there are limited data on the impact of these agents on OS in the "real-world" setting. The aim of this study is to evaluate the use and outcomes of different therapies in patients (pts) with International Prognostic Scoring System (IPSS)-defined Intermediate-2- and High-risk (higher-risk) MDS.

Methods: The ERASME study (CEL-SMD-2012-01) is an observational, prospective study of pts with either MDS or chronic myelomonocytic leukemia (CMML); pts were defined using the 2008 World Health Organization (WHO) classification system. Initial pt management strategy was classified into 3 groups: active therapy (AT), such as chemotherapy or treatment with azacitidine (AZA), lenalidomide (LEN), etc.; allogeneic hematopoietic cell transplantation (HCT), which included those pts receiving other therapies before transplantation; and observation and support (OB&SP), which included red blood cell and platelet transfusions, and growth factors. OS rates of higher-risk pts are presented using the Kaplan-Meier method.

Results: A total of 160 pts with higher-risk MDS were recruited between Jan 2013 and Feb 2015. Median follow-up was 8.6 months (interquartile range [IQR] 4.0-11.8). Pt characteristics are described in the Table. AT was the first therapeutic decision in 83 (52%) pts, HCT in 43 (27%), and OB&SP in 34 (21%). Within the AT group, 75 (90%) pts received AZA alone, 5 (6%) AZA plus chemotherapy, and 3 (4%) received other options; most pts (n = 78; 94%) were treated with a 7-day regimen. Among the 43 potential HCT candidates, 40 (93%) had received prior therapy: 25 (58%) with AZA alone, 8 (19%) with chemotherapy, 6 (14%) with AZA plus chemotherapy, and 1 (2%) with LEN; 3 (7%) had not received prior therapy. Of the potential HCT candidates, 20 had undergone transplantation within a median of 8 months (IQR 4.1-13.3); 7 (16%) pts died before transplantation and 16 (37%) are still awaiting transplantation. The main reasons for OB&SP being the initial pt strategy were disease risk (100%), age (95%), comorbidities (60%), and symptomatology (44%). At last follow-up, 70 of 160 pts (44%) had died after a median of 6 months (IQR 2.9-11.8): 34 (41%), 14 (32%), and 22 (65%) pts undergoing AT, HCT, and OB&SP, respectively (log-rank 6.9; P = 0.032). Median OS in pts who received AT, HCT, and OB&SP was 18.60 months (95% confidence interval [CI] 13.1-21.8), 14.92 months (95% CI 11.6-not reached), and 8.44 months (95% CI 4.3-13.4), respectively. Median OS was significantly longer in the AT group compared with the OB&SP group (hazard ratio [HR] 1.9; 95% CI 1.1-3.3; P = 0.0197). No statistically significant difference was observed in OS between pts treated with AT and those considered potential HCT candidates (HR 0.9; 95% CI 0.5-17.7; P = 0.8).

Conclusions: The preliminary data from this observational, prospective study indicate that AT significantly prolongs OS compared with OB&SP when treating pts with higher-risk MDS in a real-world setting. AZA was the most common treatment in the AT group.

Abstract presented on behalf of the ERASME Study Investigators Group.