Enhanced risk stratification in young and elderly patients treated with induction chemotherapy as well as in frail patients with AML.

Abstract

6582 Background: Cytogenetic aberrations represent the main risk factor in AML. The recently published Muenster AML score ameliorated risk stratification in elderly patients treated with induction chemotherapy in a large trial using additional clinical factors. A second score for prediction of complete remission (CR), overall survival (OS) and early death was established for cases, where cytogenetic information is not available. Methods: We first retrospectively applied these two scores in 59 elderly patients with newly diagnosed AML treated with induction chemotherapy. Although the score was not designed to, we then tested these scores in a cohort of young (n=73) patients who received induction therapy and additionally in frail (n=118) patients with cytoreductive therapy or best-supportive care only. Results: For the proposed patient-cohort of elderly patients treated with induction chemotherapy, the AML Score including cytogenetic information was, as expected, highly predictive for achieving CR (100% vs. 42.9%, p<0.001) and for OS (44.6 vs. 2.6 months, p<0.001). Using the score without cytogenetic information, prediction of OS was also possible (16.2 vs. 5.9 months, p=0.02) whereas prediction of CR lost statistical significance. Interestingly, in the not intended cohort of young patients, both scores achieved a very good discrimination of risk groups, even without using cytogenetic information (OS: 88.6 vs. 24.4 months, p<0.01; achieving CR: 94.4% vs. 59.1%, p=0.02). In our cohort of frail patients not suitable for induction therapy, using the score with cytogenetics also clearly discriminated risk groups for short OS (20.7 vs. 3.8 months, p<0.001). Discrimination using the score without cytogenetic information was not statistically significant (15.0 vs. 5.2 months, n.s.). However, prediction of early death showed no statistically significance in all of our three patient cohorts. Conclusions: The Muenster AML Scores may enhance selective risk stratification in young and elderly patients treated with induction chemotherapy as well as in frail patients without intensive treatment also in cases, where cytogenetic information is not available.