A 47-year-old Greek diabetic man presented with erectile dysfunction and a decrease in sexual desire. The patient had type 2 diabetes for the previous 8 years and was on treatment with rosiglitazone and metformin with strict glycemic control (HbA1c 5.8%). No symptoms or signs of neuropathy were present. Hypogonadotrophic hypogonadism was found.
His plasma testosterone level was very low (100 ng/dl [reference range 300–1,000]) and there was no luetinizing hormone response to luetinizing hormone–releasing hormone (LHRH) test. Further work-up with a magnetic resonance imaging scan and hypophyseal function tests did not reveal any space-occupying lesions of the hypothalamic pituitary site.
The process led to the diagnosis of idiopathic hypogonadotropic hypogonadism. On further work-up, the patient was found to have a prostatic carcinoma. There was no evidence of metastatic disease (his plasma prostate specific antigen [PSA] level was 1.9 ng/ml).
Six years earlier, the patient was treated with finasteride for benign prostatic hypertrophy. A radical prostatectomy was performed and a poorly differentiated adenocarcinoma was found (Gleason grade 10, T3N1Mx).
Postoperatively his plasma testosterone rose to normal levels (530 ng/dl), and there was no need for diabetes medication, given that his fasting plasma glucose values never exceeded 6 mmol/l while only on diet.
His homeostasis model assessment of insulin resistance (HOMA-IR) (fasting serum insulin [μU/ml] × fasting plasma glucose [mol · l−1/22.5]) was 1.8, and an oral glucose tolerance test performed with 75 g glucose was absolutely normal. His PSA value was 0.3 ng/ml.
After surgery there was very little change in body weight. His BMI before the operation was 26.6 kg/m2 and after the operation 26.5 kg/m2. This minimal change in body weight could not account for the observed euglycemia, nor could his diet given that it was virtually unchanged.
Combined androgen blockade therapy was initiated with goserelin acetate and bicalutimide. Three weeks later his plasma testosterone level fell well in the hypogonadotropic range (80 ng/ml) with a simultaneous, abrupt worsening of his glycemic control. Fasting plasma glucose ranged from 10 to 15 mmol/l and HOMA-IR rose to 15, indicating an insulin-resistant state. Again there was no significant change in his body weight; his BMI was 26.5 kg/m2.
Treatment with metformin and rosiglitazone was reinstituted with a significant euglycemic response (fasting plasma glucose 5.5 mmol/l).
The presence of diabetes in the preoperative hypogonadal state, the remission of it in the immediate postoperative eugonadal phase, and the reappearance of insulin resistance after the institution of androgen deprivation treatment indicate that in this patient, the effect of testosterone was insulin sensitizing.
Marked hyperglycemia in prostatic cancer patients, after initiation of androgen deprivation therapy, has been reported in the literature with good response to pioglitazone (1). In the present case, prostatic carcinoma presented as hypogonadotropic hypogonadism. Schaeffer and Walsh (2) eloquently suggested that adenocarcinoma of the prostate should be considered in the differential diagnosis of hypogonadism based on the suppression of the hypothalamic-pituitary-testicular axis occasionally caused by this carcinoma. Undifferentiated prostatic carcinomas may yield normal PSA values, and this should be kept in mind when considering testosterone replacement therapy in men with hypogonadism. Another point of significance is that our patient was receiving rosiglitazone and peroxisome proliferator–activated receptor-γ ligands, which may modify PSA levels (3).
Based on the above information, it is clear that in diabetic patients presenting with sexual dysfunction, prostatic carcinoma should be considered in the differential diagnosis, regardless of the PSA level. Further research is needed to examine the possible relationship between testosterone and insulin resistance and the possible role of hyperinsulinemia in the course of prostatic carcinoma disease, given that in recent years it has become clear that there are multiple androgen-independent routes.
