L-Asparaginase-Induced Continuous Hyperglycemia With Type 1 Diabetes-Related Antibodies and HLA Genotypes: A Case Study

A 19-year-old male presented with fatigue and dyspnea on exertion. He was diagnosed with acute T-cell lymphoblastic leukemia. After following the Group for Research on Adult Acute Lymphoblastic Leukemia (GRAALL) 2003 protocol that incorporates L-asparaginase (L-Asp) treatment, blood glucose levels became elevated for more than one year and insulin secretion was depleted. Anti-glutamic acid decarboxylase (GAD) and anti-islet antigen 2 (IA-2) antibody levels were both positive, which is rare. The patient’s HLA genotype was sensitive for type 1 diabetes. L-Asp can cause transient hyperglycemia as a side effect. However, cases with the anti-GAD antibody have not been reported in L-Asp-induced diabetes. In summary, L-Asp-induced continuous hyperglycemia might be associated with a type 1 diabetes-related HLA genotype through elevations of anti-GAD and anti-IA-2 antibodies.

Type 1 diabetes is caused by the progressive autoimmune destruction of pancreatic beta cells producing insulin. The anti-glutamic acid decarboxylase (GAD) antibody is present in 70-80% of patients with newonset type 1 diabetes [11].
No cases with the anti-GAD antibody have been reported in L-Asp-induced diabetes. Here, we report the first case of L-Asp-induced diabetes, together with the transient appearance of anti-GAD antibody, associated with the type 1 diabetes susceptibility HLA alleles, DRB1*04:05/*08:02.

Case Presentation
A 19-year-old male experienced fatigue and dyspnea on exertion for two months. His blood count revealed leukocytosis, erythrocytopenia, and thrombocytopenia. As a result of a bone marrow biopsy, he was diagnosed with acute T-cell lymphoblastic leukemia.
The Group for Research on Adult Acute Lymphoblastic Leukemia (GRAALL) 2003 protocol is outlined in Table 1 [14]. Remission induction therapy, including L-Asp and a large amount of corticosteroid, was initiated. L-asparaginase was only used during the first remission induction therapy. After five weeks, a bone marrow test showed hematological remission. 1 1, 2, 3 1 1 1  Consolidation therapy was then started. The fasting plasma glucose level increased to over 400 mg/dL. Selected biochemical test results are shown in Table 2. Plasma levels of insulin and C-peptide were 2.0 μU/mL and 0.25 ng/mL, respectively. Anti-GAD and anti-islet antigen 2 (IA-2) antibody levels were measured at 13.4 U/mL and 3.1 U/mL, respectively, and samples were therefore considered positive for these two antibodies.   Intensive insulin therapy was started on a sliding scale. The total amount of insulin given was approximately 60 units per day. An insulin secretory defect was thought to cause hyperglycemia. Lasparaginase induces diabetes mellitus as a side effect [14], and both anti-GAD and anti-IA-2 antibody levels were found to be positive. Since this was not common, we regularly measured C-peptide and anti-GAD and anti-IA-2 antibody levels during the intensive insulin therapy.

Remission induction therapy protocol
The time courses for measured glucose, insulin, and C-peptide levels are shown in Table 3. The fasting Cpeptide level was found to gradually recover. The total amount of insulin administered decreased from approximately 60 units to about two units per day.    In order to evaluate insulin secretion 11 months after remission induction therapy was started, a glucagon test was performed and plasma glucose/insulin/C-peptide levels before and after breakfast, as well as urinary C-peptide, were assessed. These findings are shown in Table 4. Insulin secretion did not become depleted.  The patient was administered because of a relapse of acute T-cell lymphoblastic leukemia. He felt a loss of appetite and fatigue; such stresses subsequently induced hyperglycemia. A computed tomography (CT) scan showed atrophy of the pancreas (Figure 2). Therefore, the pancreatic exocrine function was evaluated. A fecal digestion status test showed fatty stool. A low pancreatic function diagnostic test score of 20.6% (standard: >73.4%) was obtained.

FIGURE 2: Abdominal plain computed tomography scan
The yellow arrow indicates an atrophic pancreas.
Furthermore, an HLA gene test was performed for a marrow transplant. Incidentally, the patient's HLA-DRB1, consisting of 04:05 and 08:02, was found to be a susceptible gene for type 1 diabetes [15].

Discussion
L-asparaginase is an enzyme that hydrolyses asparagine to aspartic acid and ammonia, causing a depletion of asparagine in cells. Insulin incorporates three molecules of asparagine; therefore, L-Asp can inhibit its synthesis in pancreatic beta cells to cause transient hyperglycemia and diabetes mellitus [16,17]. The percentage of cases with acute lymphoblastic leukemia in whom hyperglycemia occurs due to L-Asp with steroid therapy is 2.5-23% [18]. Hyperglycemia occurs within 5-10 days after the initiation of L-Asp therapy. Although patients require transient insulin therapy, hyperglycemia improves in almost all cases.
In addition to transient glucose intolerance, some rare cases develop pancreatic diabetes with L-Asp-related pancreatitis. Twenty-one percent of patients with L-Asp-related pancreatitis require insulin therapy in the acute phase and 6% of patients continue insulin therapy for one year [19,20]. However, no L-Asp-related diabetes mellitus cases have been positive for both anti-GAD and anti-IA-2 antibodies.
When we explored a mechanism for how the antibodies had appeared, i t became clear the patient had specific HLA genes that made him susceptible to type 1 diabetes mellitus. The relative risk of this genetic combination was estimated to be 15 times [15]. It may be that the patient's condition made him more susceptible to developing type 1 diabetes mellitus. L-asparaginase subsequently induced pancreatic beta-cell impairment, leading to the leakage of intracellular substances into the blood. As a result of antigen presentation, anti-GAD and anti-IA-2 antibody levels became transiently positive.
Next, we considered how the increase in these antibodies affected the decrease in insulin due to the impairment of pancreatic beta cells.
In Table 3, C-peptide levels seemed to be recovering gradually. The recovery might be mainly due to the resolution of glucotoxicity and functional beta cell impairment induced by L-Asp. However, the results of the glucagon test and assessment of plasma glucose/insulin/C-peptide levels before and after breakfast indicated hyperglycemia and insufficient C-peptide. In spite of the hyperglycemia, urinary C-peptide was only 36 μg/day ( Table 4). To assess these results comprehensively, the significantly decreased insulin level after L-Asp administration gradually recovered. However, the plasma glucose level was not normal. We considered that insulin secretion had decreased compared to before L-Asp administration.
We discussed whether the limited decrease of insulin depended on the L-Asp effect or autoimmune destruction of pancreatic beta cells due to type 1 diabetes mellitus. Compared to a typical course of L-Aspinduced diabetes, it was uncommon that pancreatitis did not occur clinically and insulin therapy was necessary after a year. We, therefore, speculated that the mechanism involved was due to type 1 diabetes mellitus. However, we thought that the speculation was unlikely because type 1 diabetes mellitus usually causes extensive destruction of beta cells and needs the administration of massive insulin permanently. On the other hand, the limited decrease in insulin was explainable if L-Asp had caused pancreatitis. The atrophic pancreas in the CT scan and the decrease in pancreatic exocrine function were consistent with post-pancreatitis changes. We might have missed an acute phase of pancreatitis.

Conclusions
We describe a unique case of L-asparaginase-induced continuous hyperglycemia with type 1 diabetesrelated antibodies. Although diabetes mellitus occurred after the administration of L-Asp, it was found that anti-GAD and anti-IA-2 antibodies, which are indicators of type 1 diabetes mellitus, were transiently high. This could be explained by the presence of a type 1 diabetes mellitus-sensitive HLA gene.
It is our firm belief that HLA genotype testing should be performed when the anti-GAD antibody is detected and hyperglycemia persists after L-Asp administration.

Additional Information Disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.