Azacitidine Treatment for Lenalidomide (LEN)-Resistant Myelodysplastic Syndrome (MDS) with Del 5q

LEN is a highly effective treatment for lower risk transfusion dependent (TD) MDS patients with chromosome 5q deletion [del(5q)]. Effective treatment alternatives after LEN treatment failure are limited and outcome reported to date is poor. With decitabine, Lubbert et al reported responses among 4/5 patients with del (5q) and > 5% myeloblasts (one patient had isolated del (5q)) and Kantarjian et al in 2/16 higher risk del (5q) MDS patients. We assessed the response to treatment and outcome with azacitidine (AZA) in del(5q) MDS patients after LEN treatment failure.

Patients with del(5q) MDS who were treated with AZA after LEN failure were identified through the Moffitt Cancer Center (MCC) MDS and the Groupe Francophone des Myelodysplasies(GFM) database. Individual charts were reviewed. Data collected included baseline characteristics, duration of response to LEN, responses to AZA by IWG 2006 criteria, and overall survival (OS). Descriptive statistics were used for analysis and Kaplan Meier estimates were used for OS.

Thirty six del(5q) MDS patients treated with AZA after LEN failure were identified. The median age was 68 years (48–93), with M/F :20/16. The median duration of follow up from date of AZA initiation was 848 days. Table-1 summarizes baseline characteristics. The median duration of LEN therapy prior to AZA was 13 mo (1–58 mo), The median time between end of LEN treatment and start of AZA treatment was 2.5 mo (0–30 mo). The median duration on AZA treatment was 7 month (1–25 mo). The best response to AZA by IWG 2006 criteria was CR in 2 patients (6%), 3 (8%) marrow CR (mCR), 2 (6%) partial remission (PR),11 (31%) stable disease with hematological improvement (HI), 9 (25%) had stable disease without HI (SD) and 7 (19%) had progressive disease (PD); response data was not available for 2 (5%) patients. Overall response rate was 50%. HI lineage responses included erythroid (HI-E) 12 of 28 (46%), HI-P (Platelets) 8 of 23 (35%), and HI-N (neutrophil) 4 of 19 (21%). The median duration response (HI or better) was 12 mo (95%CI 5–19 mo). Median OS was 22 mo (95%CI 13–31 mo) from the time of AZA initiation, whereas median OS among patients with HI or better response to AZA was 32 mo (95%CI 20–44 mo) vs. 13 mo (95% CI 4–22 mo) in those with SD/PD. (p=0.001). Patients with blast <5% at time of starting azacitidine therapy had better median OS (28 mo versus 17 mo) (p=0.04) compared to those with ≥ 5%. Patients with isolated del(5q) or int-1 risk IPSS prior to AZA had a trend for better OS (Table-2). Only 3 patients proceeded to allogeneic stem cell transplant with median OS 24 mo. Among all patients, 33% (12/36 patients) progressed to AML.

To our knowledge this is the largest cohort evaluating the activity of AZA in patients with del(5q) MDS after LEN treatment failure. Response rates are similar to those reported in non-del (5q) patients, indicating that AZA is an effective option for salvage treatment. Given the poor outcome among AZA non-responders patients should be considered for allogeneic stem cell transplantation.

Komrokji:Celgene: Honoraria, Speakers Bureau. List:Celgene: Consultancy. Fenaux:Celgene: Honoraria, Research Funding.