Prognostic Impact of Genetic Characterization in the GIMEMA LAM99P Study for Newly Diagnosed Adult AML. Relevance of Combined Analysis of Conventional Karyotyping, FLT3 and NPM Mutational Status.

Between 1998 and 2002, 509 patients with AML (median age 46 yrs, range 15–60) were enrolled in the multicenter LAM99P study of the Italian GIMEMA group. To better evaluate the clinical impact of genetic characterization, all patients received a uniform protocol and diagnostic samples were centralised for cytogenetic and molecular studies. Therapy consisted of HU pre-treatment (2g/m² for 5 days) followed by induction with DNR (50 mg/m2 d 1, 3, 5), cytarabine (100 mg/m2 d 1–10) and etoposide (100 mg/m2 d 1–5) and consolidation with cytarabine (500 mg/m2/q12 hrs d 1–6) and DNR (50 mg/m2 d 4–6). After consolidation, eligible patients with an identical HLA donor were to receive allogeneic SCT and the remaining peripheral blood autologous SCT. Cytogenetic and molecular genetic characterization (including analysis of major fusion genes, FLT3 and NPM status) was available in 397 (78%) patients. Compared to previous GIMEMA studies, the possibility to collect samples during the 5d of HU pretreatment considerably improved genetic characterization and in particular centralised karyotyping by overcoming the problem of sampling and shipment over the w-end. After induction, 269/397 (68%) patients achieved CR. For induction response, conventional K identified 3 distinct risk groups as follows: low risk (inv. 16 and t8;21), intermediate (normal K and other anomalies not comprised in the high risk group) and high risk (t3;3, inv.3, t9;22, 11q23, 5/7 abnormalities complex K,) with CR rates of 92%, 67% and 39%, respectively (P<.0001). NPM mutations were significantly associated with older age, higher WBC, normal K and FLT3-ITD. CR rates in NPM+ (mutated) vs. NPM- (wildtype) groups were 76% vs. 60% for the whole population and 81% vs, 61% for patients in the normal K group (P<.001 for both comparisons). Multivariate analysis for CR indicated that low risk K and NPM+ were independent factors favorably affecting CR achievement while FLT3 status had no significant impact on CR. The analysis of prognostic factors for DFS and OS was carried out in 269 patients in CR (median follow-up of 39 mos.) and multivariate analysis performed after adjusting for unfavorable factors (WBC count). Multivariate analysis of variables influencing OS showed the following: low vs intermediate K, P=.0005; high vs intermediate K, P<.0001; FLT3+ vs. FLT3−, P=.06. Multivariate analysis for DFS showed: low risk vs. intermediate risk K, p=.01; high risk vs. intermediate risk K, p= .03; FLT3+ vs. FLT3-, p=.0002. NPM status did not significantly influence DFS in either the whole population or in the normal K group. In particular, there was no difference in the DFS rates among patients NPM+ and NPM- in the normal K/FLT3- group while in the normal K/FLT3+ group there was a trend (p=.06) for lower relapse rate for NPM+ patients as compared to NPM- ones. These results highlight the relevance of combining cytogenetic and molecular studies in the diagnostic work up of AML and confirm the impact of karyotype on all outcome estimates as well as of FLT3 status on DFS. As to NPM mutations, these appear a favorable predictor of CR achievement. Further investigations in large clinical trials are needed to assess the prognostic value of NPM mutations on outcome in AML with normal karyotype.