Outcome of Patients with Lower-Risk Myelodysplastic Syndrome (MDS) After Azacitidine (AZA) Treatment Failure.

Abstract

2815 Background: Azacitidine has emerged as the standard of care for treatment of higher risk MDS based upon results of the AZA-001 study. Several groups reported poor outcomes after AZA failure in patients with int-2 or high risk International Prognostic Scoring System (IPSS) risk groups with a median overall survival (OS) ranging from 4–8 months (mo). In the USA, AZA is approved for all FAB types and risk groups and is as first or second line therapy for anemia after erythroid stimulating agents in low /int-1 risk non-deletion 5q MDS and is the treatment of choice for thrombocytopenia. The outcome of patients with lower risk myelodysplastic syndrome (MDS) after AZA failure has not been characterized. We report our experience in a large cohort of low/int-1 (lower risk) MDS patients after AZA failure. Methods: Patients were identified through the Moffitt Cancer Center (MCC) MDS database. Individual charts were reviewed and relevant clinical data was extracted. Patients with low or intermediate-1 (int-1) risk disease as defined by IPSS who had received AZA treatment were identified. These patients were also risk stratified based on Global MD Anderson Score (MDAS). The primary objective was to estimate OS in these patients after AZA failure. AZA failure was defined as failure to respond after 4 or more treatment courses, loss of response, or disease progression while on therapy. All responses were defined according to the International Working Group (IWG) 2006 criteria. The Kaplan–Meier method was used to estimate median overall survival. Results: Two hundred eighty MDS patients with low/int-1 IPSS risk who had received AZA treatment were identified. Most patients (81%) were greater than 60 years of age (median, 69 years), and 90% of AZA treated patients were RBC transfusion dependent. Refractory cytopenia with multilineage dysplasia (RCMD) was the most common WHO subtype (44%), and 81% of patients had good risk cytogenetics (Table-1). The median time from MDS diagnosis to AZA treatment was 12.3 months; median number of AZA cycles received was six. At the time of AZA treatment, 241 patients (86 %) were risk stratified as int-1 versus 39 patients (14 %) who were stratified as low risk IPSS. The IWG 2006 responses to AZA treatment included 4% CR (n=10 ), 1% marrow CR (n=2), 4% PR (n=10), 27% Hematological improvement (HI) (n=75), whereas 52% (n=146) had stable disease with no HI (n=146) and 10% had progressive disease (n=10); 6 patients (2%) died on therapy, and responses were missing in 2 patients (<1%). The overall best response (HI or better) was 36%. The median OS for the entire cohort after AZA failure was 18.5 months (95% CI [13.5–23.5 mo], Figure 1A). The median OS for patients with low risk IPSS disease from time of AZA failure was 46 months versus 15 mo for int-1 patients (p<0.005, Figure 1B). When utilizing MDAS, median OS was 33.3 months for low risk patients, 21 months for int-1, 11 months for int-2, and 7.5 months for poor risk patients (p=0.005). Conclusions: To our knowledge this is the first report describing the outcome of lower risk MDS patients after AZA treatment failure. Outcome is particularly poor for those patients with int-1 risk MDS, with a median OS of 15 mo. Global MDAS identified patients upstaged to int-2 or high risk with less than one year OS. There is unmet need for effective novel therapies for lower risk MDS patients after AZA failure. Disclosures: List: Celgene: Consultancy. Komrokji:Celgene: Speakers Bureau.