Prognostic Significance of Karyotype in Octogenarian Patients (Pts) with Acute Myeloid Leukemia (AML)–An International Study

We asked whether a modified European LeukemiaNet (ELN) karyotype-based classification without molecular markers has prognostic significance in AML pts aged 80 to 89 years (y). We queried the German-Austrian AML Study Group (AMLSG; 7 pts), the Acute Leukemia French Association Group (ALFA; 17 pts), the German AML Cooperative Group (AMLCG; 35 pts) and the Cancer and Leukemia Group B (CALGB; 81 pts) for pts treated with intensive induction (7+3 or similar) on whom conventional karyotype analyses were completed and reviewed centrally. Of the 140 pts, including 82 (59%) men and 58 (41%) women, with a median age of 82 y (range, 80–89), 117 pts had de novo and 23 had secondary or therapy-related AML (s-AML/t-AML). There were 2 pts in the modified Favorable Group [t(8;21) and inv(16)], 67 pts in the modified Intermediate I Group [cytogenetically normal (CN-AML) pts], 44 pts in the ELN Intermediate II Group [t(9;11) and abnormalities (abn) classified as neither favorable nor adverse] and 27 pts in the ELN Adverse Group [inv(3) or t(3;3), t(6;9), t(v;11)(v;q23), -5 or del(5q), -7, abn(17p) and complex karyotype (≥3 abn)]. In order to assess the impact of karyotype on outcome, we eliminated early deaths. The complete remission (CR) rate for all 92 (66%) pts surviving beyond 30 days (d) was 46% and their median disease-free survival (DFS) was 6 months (mo); 35% were disease-free at 1 y and 12% at 3 y. Similarly, the median overall survival (OS) for pts surviving beyond 30 d was 6 mo, with 35% alive at 1 y and 11% at 3 y. There was no difference in DFS or OS based on AML type (de novo v s-AML/t-AML; Table 1). As shown in Table 2, pts in the modified Intermediate I Group had better OS than pts in the Adverse Group (P=0.03). Among CN-AML pts with molecular testing completed, 9/21 (43%) were NPM1-mutated (mut), 8/25 (32%) had FLT3-internal tandem duplication (ITD) and 1/12 (8%) was CEBPA-mut. Interestingly, FLT3-ITD did not have prognostic significance in the CN-AML cohort alive beyond 30 d (data not shown), but NPM1 mutation resulted in trends for higher CR rates (67% v 25%, P=.09) and longer OS (91 v 10 mo, P=0.07) in the CN-AML cohort alive beyond 30 d (Figure). Of note, there was no difference between the 2 larger cohorts (AMLCG and CALGB) in regards to pt characteristics or outcome in de novo pts; there were insufficient numbers of pts with s-AML/t-AML to compare the 2 cohorts. We conclude that there is a difference in accrual to clinical trials of octogenarian AML pts among the different cooperative groups. Further, CN-AML pts had better OS in octogenarian AML, and NPM1 mutation may be of prognostic significance among the CN-AML pts.