Corticosteriods, Neurocognition and Stress In Adult Survivors of Childhood Leukemia

Abstract 738

While recent research has demonstrated comparable neurocognitive outcomes in survivors of childhood leukemia (ALL) treated with different types of corticosteroids during induction therapy, the long-term impact of prolonged corticosteroid therapy has not been investigated. We compared neurocognitive and physiologic outcomes in 37 adult survivors of childhood leukemia who were treated with only chemotherapy (i.e., no cranial radiation therapy) on one of two standard therapy protocols (21 treated with 20 g/m² HD-IV methotrexate and repeated cycles of prednisone for two years; 16 treated with either 20 or 24 g/m² HD-IV methotrexate and repeated cycles of dexamethasone for two years). Patients underwent neurocognitive evaluations, physical exams, and laboratory tests after overnight fasting. Morning cortisol was assessed at baseline and 12 h after 1 mg dexamethasone (standard dexamethasone suppression test). Groups were similar for male:female ratio (p=0.73), current age (p=0.90), and cumulative HD-IV methotrexate exposure (p=0.68). No differences were detected in final adult height (p=0.85), body mass index (p=0.43), HDL (p=0.86) or LDL (p=0.99) cholesterol, heart rate (p=0.49), triglycerides (p=0.59), serum glucose (p=0.84), insulin (p=0.63), HgA1C (p=0.62), free T4 (p=0.78), IGF-I (p=0.77), or basal morning cortisol (p=0.72). Survivors treated with dexamethasone displayed significantly higher systolic blood pressure compared to the prednisone exposure group (mean=128.5 ± SD 11.71 vs. 117.7±13.78; p=0.02), and a trend for higher diastolic BP (mean=75.1±9.33 vs. 68.7±10.55; p=0.08). Survivors originally treated with dexamethasone displayed significantly less suppression of cortisol following the dexamethasone suppression test (mean 3.82±6.41 μg/dL) compared to those originally treated with prednisone (0.94 ± 0.75 μg/dL; p=0.05). The dexamethasone group demonstrated significantly lower performance on multiple measures neurocognitive function (Tables 1), including mathematical problem solving (p=0.002), semantic verbal memory (p=0.006) and cognitive flexibility (p=0.02). They also reported more difficulty with emotional regulation (p=0.04). These results suggest that adult survivors of childhood leukemia treated with dexamethasone are at increased risk for neurocognitive impairment and poor emotional regulation. These problems may occur in the presence of symptoms of physical stress. Surprisingly, these survivors appear to display persistent late signs of increased reactivity within the hypothalamic-pituitary-adrenal axis.