Patient weight and event-free survival for children under 2 years of age at diagnosis with favorable histology Wilms tumor

Abstract

20002 Background: Over- and underweight have been associated with excess mortality in certain childhood cancers. The impact of the child’s weight at diagnosis on event-free survival (EFS) in favorable histology Wilms tumor (FH WT) is unknown. Methods: Patients with FH WT under 2 years of age at enrolment on NWTS-5 were included. This age group was analyzed by body weight in kilograms because body mass index (BMI) norms do not exist for individuals less than 2 years old. Outcomes by BMI for children older than 2 years of age with FH WT will be analyzed separately. CDC 2000 growth charts were used. Patients were stratified for risk based on stage and chemotherapy protocol [EE4A = vincristine/dactinomycin] [DD4A = vincristine/doxorubicin/ actinomycin]. A univariate analysis of the relationship of weight-for-age and EFS was calculated. A Cox proportional hazards model was fitted for EFS examining four subsets of weight-for-age by percentiles: a) less than 5%, b) 5–9.9%, c) 90–94.9% and d) more than 95% and adjusting for risk/treatment groups via stratification. Results: 594 patients met the study criteria. 567 had weights recorded. Median follow-up was 4.7 years. 10% of patients had a weight for age percentile of 5.6 or below and 10% had a weight percentile of 94.1 or above. A univariate analysis of the relationship of weight-for-age and EFS showed no relationship (p=0.40, log-rank test). A Cox proportional hazards model, stratified by risk/treatment groups, showed that low or high weight-for-age was not predictive of outcome (p=0.24). Conclusions: There was no evidence that low or high weight-for-age was predictive for EFS among patients less than 2 years old with FH WT. There were more patients with lower or higher weight than would be expected. [Table: see text] No significant financial relationships to disclose.