Insulin and glucose homeostasis in childhood cancer survivors treated with abdominal radiation: A pilot study

108 Background: Previous reports have suggested an increased risk of Type I and Type II diabetes mellitus (DM) in childhood cancer survivors exposed to abdominal radiotherapy (RT). The mechanisms leading to DM in this population, however, remain unknown. We sought to clarify the pathophysiology leading to these derangements by performing dynamic testing of glucose and insulin in survivors previously treated with abdominal RT. Methods: Cross-sectional pilot study of 2-year survivors of childhood cancer treated with abdominal RT at Memorial Sloan Kettering between 1975 – 2009. Eligible participants were < 21 years of age at exposure to abdominal RT; those with a known diagnosis of DM or prior exposure to brain or total body RT were excluded. Survivors underwent formal 2-hour glucose tolerance testing; auto-antibodies (insulin auto-antibodies, islet cell autoantibody, glutamic acid decarboxylase) typically present in patients with Type I DM and hemoglobin A1c levels were assessed. Insulin resistance was calculated by the homeostasis model assessment of insulin resistance (HOMA-IR) and Matsuda index. Results: 21 survivors were enrolled (male: 47.6%; median age at RT: 3.3 years; median age at study: 14.4 years [range: 8.3 – 46.9]; median time from abdominal RT: 10.7 years). Primary diagnoses included neuroblastoma (n = 15), rhabdomyosarcoma (n = 3), Wilms (n = 1), Hodgkin lymphoma (n = 1), rhabdoid tumor (n = 1). None of the participants were obese (body mass index [BMI] range: 14.7 – 23.2 kg/m2). Five participants (23.8%, 95% confidence interval: 8% – 47%) had glucose derangements at a median of 8.4 years after RT (one with impaired fasting glucose [fasting glucose ≥ 100 mg/dl) and four with impaired glucose tolerance [2-hour glucose 140-199 mg/dl]). Two additional participants with normal glucose tolerance had impaired insulin sensitivity based on an abnormal Matsuda Index and HOMA-IR. None of the participants had abnormal autoantibodies, insulinopenia, or hemoglobin A1c levels. Conclusions: These findings suggest that nonobese childhood cancer survivors treated with abdominal RT may be at high-risk for subclinical derangements of glucose and insulin. Further study is warranted in larger survivor cohorts. Clinical trial information: NCT02248779.