Improved Glycemic Control in Diabetic Patients Treated with Recombinant Human Erythropoietin.

Recombinant human erythropoietin (rHuEPO) originally used to treat anemia has also been found to have several pleiotropic actions, e.g. neuroprotection, cardioprotection, and immune effects. Based on our preliminary data in mice, and on the studies by others of reduced insulin resistance in rHuEPO-treated hemodialysis patients, we have raised the question of a potential association between EPO and glycemic control. Here we report our observation of glucose control in four rHuEPO-treated diabetic patients. These diabetic patients were treated with rHuEPO for their anemia associated with chronic renal failure (CRF) or myelodysplastic syndrome (MDS). We retrospectively compared the available clinical parameters reflecting blood glucose control before and after initiation of EPO therapy. The parameters included fasting glucose levels, hemoglobin A1C (HbA1C), and medication requirements. The data are presented in the Table; where no number is provided, the data were unavailable or incomplete. The number of measurements available in each patient varied: For glucose-level, patient 1 had 10 measurements over the 3 months before (pre-rHuEPO) and 6 over the 3 months after rHuEPO initiation (post-rHuEPO). Patient 2 had more than 50 measurements, both over 3 years pre-rHuEPO and 2 years post-rHuEPO; patient 3 had more than 50 in 3 years pre-rHuEPO, but only 12 in 2 years post-rHuEPO, while patient 4 had more than 50 measurements, both over 3 months pre-rHuEPO and 3 months post-rHuEPO. HbA1C: Patient 2 had 10 measurements pre-rHuEPO, and 6 post-rHuEPO; Patient 3 had 10 pre-rHuEPO and 10 post-rHuEPO, patient 4 had none. Patient 1 had only 2 measurements of HbA1c (7.7 pre-rHuEPO; 6.3 post-rHuEPO). Statistical significance (p<0.05) was computed using the two-tailed unpaired student’s t test; a diamond (♦) represents a statistically significant difference. The Table shows that in all four patients, glucose control improved following the initiation of rHuEPO treatment. Of note, patient 1 was treated with unchanging doses of oral hypoglycemic therapy. Patient 2 was treated with increasing doses of insulin as is often seen in diabetes mellitus, a progressive disease. The increasing insulin requirement continued even after initiation of rHuEPO therapy, yet it appeared that this increase did not fully account for the significantly improved glycemic control. Note that the insulin requirements in patients 3 and 4 were reduced post-rHuEPO, yet due to the paucity of post-rHuEPO data, the lower glucose of patient 3 did not reach statistical significance. We conclude that rHuEPO may have an important clinical impact on the control of glucose in these patients. Future trials are currently underway to substantiate and unravel the molecular mechanism underlying this phenomenon. We are examining the effect of rHuEPO administration on glycemic control, and thus exploring the potential role of rHuEPO as part of the armamentarium in the treatment of diabetes.