Allergic reactions to insulin are rare now that recombinant human insulin is clinically available. However, certain patients still suffer from insulin allergy, even with human insulins. We report herein the case of a 45-year-old type 1 diabetic Japanese man (weight 62 kg, BMI 19.3 kg/m2) with generalized allergy to human insulin, who was successfully treated with human analog (GlyA21, ArgB31, ArgB32) insulin glargine (HOE901). He was diagnosed with type 1 diabetes because of the high titer of anti-GAD antibody (65 U/ml, normal range <5) and poor C-peptide responses after glucagon loading (0.43 ng/ml).

In 1985, his first insulin therapy was started. Although he had no history of past illness, including any type of allergy or atopic background, the first subcutaneous injection with pork insulin (semi-lente type) induced allergic reactions, including transient skin rash, itching all over the body, and facial edema. Several days later, injection was restarted once daily with beef insulin (semi-lente type). However, beef insulin also induced generalized allergic reactions. Fortunately, the allergic reactions to beef insulin were far less intense than those to pork insulin. In 1989, the insulin therapy was switched from beef to human insulin. However, the allergic reactions presented yet again. An antihistaminic drug reduced the allergy, but did not eliminate the reactions.

A blood examination did not show any evidence of allergic reaction to insulins. The percentage of eosinophils in the peripheral white blood cell count was 3.1% (normal range <7). Total IgE was 142 U/ml (5-419), and insulin-specific IgE and IgG were not detected. Drug-induced lymphocyte stimulation tests for regular and NPH human insulin were negative. However,intradermal tests showed positive skin reactions to human regular insulin, NPH insulin, and crystalline insulin zinc. The patient did not show any positive skin reaction to concomitant material or any preservatives.

Insulin glargine was recently developed as a new human insulin analog with protracted action, and it may be an ideal preparation for maintaining basal insulin substitution (1). As the prick test of insulin glargine was negative, we initiated treatment with insulin glargine. The patient tolerated subcutaneous insulin glargine without any evidence of allergy. Subsequently, we started an intensive insulin therapy with regular insulin before each meal and insulin glargine at a bedtime. Intensive insulin therapy using 8 U/day of insulin glargine was administered for >6 months. During the treatment, the HbAlc levels decreased from 9 to 7%. No other allergy was noted, even with regular insulin;therefore, antihistaminic drugs were no longer required.

His allergic epitopes appear to contain both human insulin A-chain and animal insulin (AlaB30), because all of the pork, beef, and human insulins induced allergic reactions, and both epitopes are modified in insulin glargine. The tolerance of insulin glargine appeared to suppress the allergy to human regular insulin. The epitope spreading of the tolerance by peptide therapy in autoimmune diabetic mice has recently been reported(2). Similar anergic epitope spreading may have occurred in the insulin allergy. Although the precise mechanism to induce the tolerance across other insulins remains unclear, this is the first report to demonstrate that insulin glargine allows generalized insulin allergy to be overcome.

Pieber TR, Eugène-Jolchine I,Derobert E, The European Study Group of HOE 901 in Type 1 Diabetes: Efficacy and safety of HOE 901 versus NPH insulin in patients with type 1 diabetes.
Diabetes Care
Tian J, Lehmann PV, Kaufman DL: Determinant spreading of T helper cell 2 (Th2) response to pancreatic islet autoantigens.
J Exp Med