A seasonal pattern of presentation in younger patients with de novo acute myeloid leukemia (AML).

Abstract

6617 Background: Seasonal variation in AML has been studied in a wide variety of populations and locations, with primarily negative results. These investigations may have been undermined by the heterogeneity of the disease, as de novo and secondary AML can vary in disease presentation and trajectory, likely reflecting distinct pathogenesis.We investigated seasonal variation in the incidence of AML diagnosed at Massachusetts General Hospital (MGH) from 1992 through 2011, focusing on de novo disease among younger patients. Methods: We assembled a database of 511 biopsy-proven cases of adult-onset AML (age > 18 years), from the MGH electronic medical record, using a systematic search algorithm with IRB approval. We subdivided this database into three cohorts: (1) de novo AML diagnosed prior to age 50 (2) de novo AML diagnosed after age 50, and (3) secondary AML, preceded by chemotherapy, myelodysplasia or myeloproliferative disease. Diagnosis dates were grouped into quarters (Jan-Mar, Apr-Jun, Jul-Sep, Oct-Dec). Divergence of quarterly incidence from a null hypothesis of uniformity was evaluated by chi square analysis. Results: Among patients under 50-years-old with de novo AML, we found a 44% increase in incidence between October and December (Table, p=0.04). There was no significant variation throughout the rest of the calendar year. Furthermore, the incidence of both de novo AML diagnosed after age 50 and secondary AML conformed to a uniform quarterly distribution. As expected, the majority of AML patients under age 50 demonstrated intermediate-risk cytogenetics, most frequently normal karyotype. Conclusions: We found a significantly increased incidence of de novo AML in younger patients between the months of October and December, diagnosed at MGH from 1992 through 2011. This may reflect a local environmental or infectious exposure that is not relevant to the pathogenesis of AML in older patients or those in whom AML develops due to prior therapy or previous hematological disorder. Investigation of this exposure is ongoing. [Table: see text]