Abstract 2807
hypomethylating agents, especially AZA, have become the reference treatment of higher risk MDS, but the prognostic value of baseline cytogenetics on response to AZA, and the impact of cytogenetic response (CyR) on outcome in responders remain uncertain.
Between Jan 2005 and Nov 2011, we treated at our center155 consecutive MDS patients (pts), including FAB RAEB-T / WHO AML with 20–30% blasts, with AZA (75 mg/m2/d x7 d every 4 weeks, for a median of 6 cycles). Karyotype at onset of AZA was evaluable in 143 pts, and abnormal in 95 (66%) pts. Median age was 74 years and IPSS high: 51, int2: 58, int 1: 14, NA: 20. 65 (42%) pts achieved hematological IWG 2006 response, including 28 (18%) CR, 8 (5%) PR, 13 (8%) Marrow CR, 16 (10%) stable with HI. With a median follow up of 28 months, median OS was 16 months.
Of the 95 pts with abnormal karyotype, 47 had −7/del(7q) including 9 isolated −7/del (7q), 37 had del(5q)/-5 including only 3 isolated del(5q)/-5,26 had +8 including 9 isolated +8, 9 had abnormalities leading to del (17p)(6 of them had complex karyotype), 16 had del (20q)and 44 had complex karyotype (>= 3 abnormalities).
Response and OS according to cytogenetics are summarized in table 1. None of the cytogenetic abnormalities studied (complex, normal, del20q, 17p, del5q/-5,7/del (7q) or +8) had a significant impact on response to AZA. Presence of del (17p) (median 7 vs 18 mo, p= 0.0001), del5q/-5 (12.5 vs 20 mo, p= 0.0008), −7/del (7q) (9.7 vs 20 mo, p=0.02) or complex karyotype (12 v 20 mo, p=0.002) was associated with significantly shorter OS. Among pts with complex karyotype, there was a trend for shorter OS for pts when 17p abn (median 6.7 vs 12.5 mo, p=0.12) del (5q)/-5 (9 v 21 mo, p=0.16) or −7/del (7q) (7 v 17 mo, p=0.06) abnormality was part of the complex karyotype. By contrast, isolated −7/del (7q) (21 vs 16 mo, p=0.3) and +8 (all+8:20 vs 14 mo, p=0.48; isolated +8:23 vs 16 mo, p=0.92) had no significant impact on OS. According to IPSS cytogenetic risk, response rates and CR rates were similar across the 3 groups, but OS was significantly longer in the good risk category (p=0.04) (table 1). Cytogenetics could be reclassified using new IPSS-R cytogenetic groups in 138 pts (Shanz, JCO, 2011) in 1 very good, 52 (38%) good, 24 (17%) int, 30 (22%) poor and 31 (23%) very poor. According to this IPSS-R cytogenetic classification, response rates and CR rates were similar across the 4 main groups. Median OS was 20.6 mo, 23 mo, 14 mo and 12 mo in the good, int, poor and very poor risk groups respectively (p= 0.037).
66 of the pts with baseline cytogenetic abnormalities had cytogenetic analysis at treatment evaluation, after 4 to 6 cycles of AZA, of whom 32 had achieved hematological response. In those 32 pts, 34% achieved complete CyR(CCyR), none partial CyR, 37% had stable cytogenetics (and the remaining pts had cytogenetic failure). In those 32 pts, achievement of CCyR had no significant impact on OS (p=0.36) but the number of pts was relatively small. In a landmark analysis performed at D100 in pts with baseline cytogenetic abn, achieving CCyR was not associated with an OS advantage compared to stable cytogenetics (median 22 vs 17 mo, p=0.82). Of note, only 1 patient with baseline cytogenetic abn (+8) who did not achieve hematological response achieved CCyR, of 6 months duration, before disease progression.
Baseline cytogenetic findings were strong predictors of survival in patients with MDS treated with AZA, while impact on response was limited. In hematological responders with baseline cytogenetic abnormalities, achieving cytogenetic response was not was associated with an OS advantage, butthe number of patients analyzed may have been insufficient to conclude.
Gardin:celgene: Honoraria. Fenaux:Amgen: Honoraria, Research Funding; Celgene: Honoraria, Research Funding; Janssen: Honoraria, Research Funding; Roche: Honoraria, Research Funding; GSK: Honoraria, Research Funding; Novartis: Honoraria, Research Funding.
