Predictive factors for latency period and a prognostic model for survival in patients with therapy-related acute myeloid leukemia.

Therapy-related acute myeloid leukemia (t-AML) is an increasingly recognized sequela in patients receiving chemotherapy or radiotherapy for a primary malignancy or autoimmune disease. This study assessed factors related to the latency period (LP) between the antecedent disorder (AD) and t-AML diagnosis and developed a comprehensive prognostic model to predict overall survival (OS). We evaluated a cohort of newly diagnosed t-AML patients treated with cytarabine-based induction therapy from 2001 to 2011. Multivariable linear and proportional hazards models were used to assess the impact of different classes of chemotherapy on the LP and to identify independent prognostic factors for OS. Of 730 treated AML patients, 58 (7.9%) had t-AML. Median LP to t-AML was 5.6 years (range, 0.5-38.4). 64% of patients achieved CR and median OS was 10.7 months. Independent prognostic factors of short LP were age at AD (P < 0.0001) and prior treatment with mitotic inhibitors (P = 0.05). Unfavorable cytogenetics (P = 0.004), antecedent hematologic or autoimmune disease (P = 0.01), age >60 (P = 0.03), and platelet count <30,000 μL (P = 0.04) at the time of t-AML diagnosis were prognostic for inferior OS. A prognostic model using these factors was developed that risk stratified t-AML patients into two groups: favorable and unfavorable. Patients in the favorable group had a median OS of 37.6 months compared with 6.4 months in patients comprising the unfavorable group (P < 0.0001). Multicomponent prognostic models integrating disease or treatment-related covariates can help better understand how t-AML evolves; and can be clinically useful in risk stratifying t-AML patients undergoing induction therapy.