Azacitidine (AZA) After Failure of Lenalidomide (LEN) in Low/Int-1 Risk MDS with Del 5q

LEN is currently the reference treatment of low/int 1 (lower) risk MDS with del 5q, yielding RBC transfusion independence (TI) in two thirds of the cases, with a median RBC-TI duration of 2.2 years. (MDS 003 trial, List, NEJM, 2006). In case of primary or secondary failure of LEN, however combined analysis of MDS 003 and MDS 004 trial showed prognosis to be unfavorable and treatment approaches uncertain. We report results of treatment with AZA in 13 lower risk MDS with del 5q with primary or secondary failure of LEN.

The 13 cases were treated between November 2005 and June 2011 in 10 centers of the Groupe Francophone des Myelodysplasies (GFM). At onset of LEN, median age was 71 years, M/F 0.43, 7 patients (pts) had RAEB 1 and 6 RA, 11 had isolated del 5q and 2 had 1 additional chromosomal abnormality. IPSS was low in 6 and int 1 in 7 patients. The LEN dose was 5mg/d, 10mg/d 3 weeks/4weeks, 10mg/d in 8, 3 and 2 patients, respectively. 11 (85%) patients had achieved RBC-TI, for a median duration of 26.3m (range 5.8–57.3)(secondary failures), while 2 patients had not achieved RBC-TI (primary failures). At onset of AZA, 11 patients had progressed with a change in WHO category (RA to RAEB1, or RAEB 1 to RAEB2) in 8 patients, and/or acquisition of additional chromosomal abnormalities in 9 patients, 5 eventually having complex karyotype, of whom 2 had chromosome 17 abnormality leading to TP53 gene deletion (TP53 mutation analysis is in progress); IPSS was low (n=0), int 1 (n= 3), int 2 (n=2), high (n=8). Prior to onset of AZA, attempts had been made to increase the LEN dose from 5 to 10 mg/d in 2 patients, without success.

AZA was administered at the FDA/EMEA approved schedule (75mg/m2/dx7) in 12 pts, and reduced (5 day schedule) in 1 pt. The median number of cycles administered was 6 (range 1–23). Five patients (46%) achieved a response according to IWG 2006 criteria, including CR, PR, HI in 1, 0 and 4 pts respectively. All 5 responders achieved RBC –TI. 4 patients received less than 4 cycles of AZA: 2 stopped AZA after 1 cycle for hematological toxicity, 1 stopped after 3 cycles for progression, 1 pt died from septic shock after 3 cycles. Response duration was 2.9+, 4.4+, 6.2+, 13.8, and 24.6 months, respectively. No responder was allografted, but the age of responders was 68, 70, 78, 78 and 86 years, respectively. Median survival from onset of AZA was 8.8 m (range 0.8–24.9). The 5 responders were still alive 3+, 4.4+, 6.2+, 13.9+ and 24.8+ m after onset of AZA, while median survival in non-responders was 8.7 months. Among the 5 patients who responded to AZA, 4 had at onset of AZA a high IPSS, 3 had complex karyotype and 2 one additional chromosomal abn, while 4/5 had previously achieved RBC-TI with LEN (with a median duration of 26.1 months). Neither of the 2 patients with TP53 deletion responded to AZA. Finally, among the 2 primary failures to Lenalidomide, one achieved HI with AZA (of 4.4+ months duration) and the other died within 2 months.

In this relatively small series almost 40% of the patients with lower risk MDS and del 5q who progressed under LEN could be salvaged by AZA, and might expect survival prolongation. The appearance of novel chromosomal abnormalities or complex karyotype at progression did not predict poor response to AZA, while we are currently analyzing the prognostic value of TP 53 mutations, which occur frequently at th at disease stage (Jadersten, JCO 2011), on response to AZA.

Fenaux:Celgene: Honoraria, Research Funding.