ESPE Abstracts

Background: Pediatric acute lymphoblastic leukemia survivors (pALL-S) carry a risk for developing metabolic abnormalities, including obesity, abnormal glucose metabolism (AGM) and dyslipidemia. Previous studies showed conflicting data regarding the pathophysiology of AGM in those survivors. Additionally, there has never been a study comparing glucose metabolism between pALL-S and simple obese children (Ob-C). This study aimed to assess glucose metabolism in pALL-S in comparison with Ob-C.

Methods: pALL-S and age-matched Ob-C were enrolled. All of them underwent oral glucose tolerance test (OGTT) and body composition analysis. Data of Ob-C were derived from our previous study. Abnormal OGTT results i.e. AGM were defined as hyperinsulinemia, impaired fasting glycemia, impaired glucose tolerance or diabetes mellitus. Insulin sensitivity and secretion indices were calculated from the serum glucose and insulin levels derived from the OGTT. Comparisons of these parameters between pALL-S and Ob-C were performed.

Results: There were 91 pALL-S and 100 Ob-C included. Twenty-seven (30%) out of 91 pALL-S were overweight or obese. Median (IQR) ages of pALL-S and Ob-C were 14.8 (12.4, 17.1) and 14.4 (13.8, 15.4) years, respectively. Median (IQR) BMI Z-scores were 0.2 (-0.6, 1.8) and 2.7 (2.4, 3.0) in pALL-S and Ob-C groups, respectively. AGM were demonstrated in 40 (44%) of pALL-S and in 76 (76%) of Ob-C. Comparing between pALL-S and Ob-C with AGM, fat mass index (FMI) was lower in pALL-S [7.6 (5.4, 10.5) vs 13.8 (11.7, 16.1) kg/m2, p <0.001]. However, FMI of pALL-S was slightly greater than that reported in age-matched lean children of 5.4 kg/m2. As expected, Ob-C with AGM had greater insulin resistance as demonstrated by having greater homeostasis model assessment of insulin resistance [4.4 (3.5, 6.9) vs 1.9 (1.1, 2.8), p <0.001] and lower whole body insulin sensitivity index [1.6 (1.2, 2.5) vs 3.7 (2.6, 5.3), p <0.001]. Interestingly, pALL-S with AGM had lower insulin secretion index, homeostasis model assessment of beta-cell function (HOMA-β) [175 (117, 301) vs 474 (270, 808), p <0.001]. Additionally, HOMA-β of pALL-S was lower than that reported in age-matched lean children of 213. The findings indicated that impaired insulin secretion was the main pathophysiology underlying AGM in pALL-S.

Conclusions: Despite not being obese, a number of pALL-S had AGM (44%). AGM could occur in pALL-S at lower BMI and FMI as compared with Ob-C. The pathophysiology underlying AGM in pALL-S was impaired insulin secretion rather than insulin resistance which was in contrast with Ob-C.