Association of abnormal cytogenetics at date of morphologic complete remission (CR) with overall (OS), disease-free survival (DFS) and higher relapse rate in acute myeloid leukemia (AML): Results from Cancer and Leukemia Group B (CALGB) 8461

Abstract

6514 Background: As most AML patients (pts) with morphologic CR ultimately relapse, better predictors for outcome are needed. Recently, Cheson et al. (JCO 2003;21:4642) proposed cytogenetic remission as part of the criteria for CR (CRc). This is the 1st large study of usefulness of CRc. Methods: Karyotypes at diagnosis (dx) and at date of 1st CR from pts treated on CALGB front-line studies were centrally reviewed. Pts with abnormal cytogenetics at dx and normal cytogenetics at CR (NCR; n=103) were compared to pts with abnormal cytogenetics both at dx and CR (ACR; n=16) for OS, DFS and cumulative incidence of relapse (CIR). Cox proportional hazards models assessed the prognostic impact of cytogenetics at CR, adjusting for other covariates. Results: Clinical features were similar for both groups, except favorable cytogenetics [t(8;21), inv(16), t(15;17)] at dx was present in 60 (58%) NCR vs 4 (25%) ACR pts (P=0.02) and NCRs had higher % marrow blasts (P=0.03). Median follow-up was 3.1 years (range, 1.0–11.4). ACRs had shorter OS (P=0.003) and DFS (P<0.0001) and higher CIR (P<0.0001). At 3 and 5 years, the rate of relapse or death was worse for ACRs. Similar trends were observed when only pts without favorable cytogenetics were analyzed. In multivariable models, the NCR/ACR groups were significant predictors for OS (P=0.02), DFS (P=0.01) and CIR (P=0.03). The relative risk of relapse or death for ACRs was 2.0 - 2.2 times that of NCRs (95%CI: 1.1–1.2 to 3.8–4.3 depending on endpoint). Conclusions: Our data suggest that reverting to a normal karyotype at time of 1st CR is an important prognostic factor and support the use of cytogenetic remission (CRc) as a criterion for CR in AML. No significant financial relationships to disclose.