Drug-Induced Diabetes during Treatment of ALL in Children: Who Is at Risk and When?

Hyperglycemia is a prevalent metabolic disturbance during ALL treatment, with a reported incidence of 8-35% in children with ALL. It is most commonly related to high-dose glucocorticoids and asparaginase used in treatment protocols.

The study was designed to examine the effects of different doses of glucocorticoids and asparaginase on the development of diabetes during ALL treatment; Answers were sought to the questions of who is more likely to get diabetes and when?

278 children who received ALL treatment between 2008 and 2023 were included in the study. 173 children received the Modified St Jude Total XV (St Jude) protocol between 2008 and 2016, while 105 received the ALL IC BFM 2009 (BFM) protocol between 2016 and 2023. The median age of 278 children (115 girls, 41.4%) was 5 years. The main differences between the two protocols are the dosage of glucocorticoids and asparaginase throughout the remission induction period. The average cumulative glucocorticoid dose was 26600 mg/m² hydrocortisone equivalent of methylprednisolone and 75000 U/m² L-asparaginase in St. Jude protocol, whereas 7170 mg/m² hydrocortisone equivalents of prednisolone and 40000 U/m² L-asparaginase in BFM protocol. Clinical and laboratory determinants for the ALL risk group and data related to hyperglycemia were gathered from the file records. A blood glucose level above 200 mg/dl in a single measurement or for less than 24 hours was defined as short-term hyperglycemia, while one lasting more than 24 hours was defined as diabetes. There were no significant variations in the initial presentation (age, gender, WBC count, immunophenotype, and risk group) of patients treated with two different protocols. The rate of short-term hyperglycemia and diabetes were similar between the two protocols (p = 0.617 and p = 0.180, respectively). Therefore, both groups were evaluated together. When consider the whole treatment period, 60 out of 278 (21.5%) experienced short-term hyperglycemia and 45 out of 278 (16.2%) developed diabetes. Short-term hyperglycemia and diabetes occurred more frequently during the remission induction period (%15.4, %9.7, respectively) and during the relapse (%7.4, %26, respectively). Since the rate of diabetes is highest and the difference between the two protocols is minimal during the remission induction period, risk factors for diabetes were determined for this period. Advanced age, female gender, and central nervous system (CNS) involvement were found to be independent predictor for the development of diabetes in multivariate analyses. Diabetes risk increases by 1.39 times each year of age, 2.87 times for female sex, 4.16 times for CNS involvement, and 3.85 times for traumatic LP. 25 (55.5%) of the 45 patients with diabetes received insulin therapy. Insulin administration lasted for a median of 24.7 days. Children diagnosed with diabetes and who used insulin for more than four days had higher HbA1c levels and a higher glucose/insulin ratio (p = 0.002 and p = 0.009, respectively) before the initiation of insulin therapy, suggesting that patients who used insulin for a longer period of time had an inadequate insulin response to high glucose levels. The mortality rate was higher in patients with diabetes compared to patients without diabetes (36% vs. 3.4%, p<0.001).

In children with ALL, patient-related factors play a more important role in the development of diabetes rather than glucocorticoid or L-asparaginase doses used in ALL treatment. Advanced age, female gender, and CNS involvement were the most important predictors for the development of diabetes throughout the remission period. Due to the high rate of diabetes (¼ of cases) after relapse, monitoring blood glucose is required in children who relapse.

No relevant conflicts of interest to declare.