A 70-year-old patient with refractory multiple myeloma, renal failure, and colon cancer was admitted with severe hyperglycemia (612 mg/dl). During the previous 3 weeks he had developed polyuria, polydipsia, fatigue, and a 2-kg weight loss. He had no personal or family history of diabetes. He had developed stress-related hyperglycemia during hospitalization for colon cancer surgery 4 years earlier. He was treated with insulin in the hospital and discharged on medical nutrition therapy. He was diagnosed with multiple myeloma 3 months later and was treated with chemotherapy and steroids for 12 months. His plasma glucose levels ranged from 98 to 114 mg/dl during this period. He was not treated with glucocorticoids or antidiabetic agents for the next 3 years, and during this period plasma glucose values were <125 mg/dl. Four weeks prior to this hospitalization, thalidomide 400 mg/day was started for treatment of refractory multiple myeloma. Physical examination revealed an afebrile thin patient (BMI 21 kg/m2), and no precipitating factors for diabetic ketoacidosis were identified. Laboratory investigations showed no acidosis, ketonuria, or leukocytosis. A1C was elevated at 9.9%. He received 10 units of regular insulin subcutaneously on admission. His subsequent capillary blood glucose values (with corresponding insulin doses in parentheses) were 512 (10 units), 247 (2 units), 282 (4 units), 126 (0 units), and 94 mg/dl (0 units). He received a total of 26 units of regular insulin over a period of 36 h. He refused insulin therapy on a long-term basis and was treated with glipizide GITS 5 mg/day. Three weeks later his symptoms were resolved. His fasting plasma glucose values 4, 8, 12, and 16 weeks after admission were 83, 151, 141, and 80 mg/dl, respectively. A1C after 20 weeks of glipizide GITS therapy was 5.4%. Thalidomide therapy was continued for treatment of his myeloma.
Thalidomide, withdrawn for teratogenicity, was reintroduced in 1997 as an immunomodulator to treat erythema nodosum leprosum. Its mechanism of action is thought to involve the inhibition of tumor necrosis factor (TNF)-α–mediated angiogenesis of the lesion. To our knowledge, thalidomide treatment has not previously been reported to cause or to worsen diabetes. We believe this to be the first report of extreme hyperglycemia occurring after initiation of thalidomide therapy. Iqbal et al. (1) investigated the role of thalidomide as a TNF-α antagonist on six patients with diabetes. They administered placebo or 150 mg of thalidomide for 3 weeks in a crossover design and performed isoglycemic-hyperinsulinemic clamps before and after therapy. They reported that thalidomide decreased insulin-stimulated peripheral glucose uptake by 31% (increased insulin resistance) and decreased glycogen synthesis by 48%. Wilson et al. (2) studied insulin antagonism by using a bioassay (rat diaphragm assay) in mothers giving birth to children with congenital malformations in 1966. They observed that antagonism to insulin was present in 5 of 6 (83%) mothers exposed to thalidomide in their first trimester compared with 14 of 50 (28%) mothers in the control group.
In a prostate cancer study, Figg et al. (3) observed that decreasing the dose of thalidomide improved hyperglycemia, suggesting that thalidomide may have contributed to the hyperglycemia. Our patient developed diabetes shortly after initiation of thalidomide. We believe this to be the first report of extreme hyperglycemia associated with thalidomide therapy.
This case should prompt additional studies to evaluate hyperglycemia in patients treated with thalidomide. Until then, we recommend screening for diabetes before thalidomide treatment and regular follow-up of plasma glucose levels.
References
Address correspondence to Kandaswamy Jayaraj, MD, Endocrinology, Ochsner Clinic Foundation, 1514 Jefferson Highway, New Orleans, LA 70121. E-mail: [email protected].
