Purpose: Laboratory and epidemiological studies suggest that high circulating insulin-like growth factor (IGF)-1 and low IGF binding protein-3 are associated with increased prostate cancer risk. However, the usefulness of serum IGF-1 or IGF binding protein-3 for predicting pathology results in men undergoing prostate biopsy is unclear. We examined the relationships of serum IGF-1, IGF binding protein-3 and the results of prostate biopsy.
Materials and methods: A total of 652 consecutive patients with elevated serum prostate specific antigen (PSA) or abnormal digital rectal examination who were referred for transrectal ultrasound sextant prostate needle biopsy underwent blood sampling before biopsy. PSA, free PSA, IGF-1 and IGF binding protein-3 were measured. There were 244 men (37.4%) with cancer and 408 controls with benign conditions.
Results: Mean IGF-1 plus or minus SD in the cancer and control groups was 176.1 +/- 58.3 and 178.7 +/- 54.7 ng./ml., respectively (p = 0.57). Mean IGF binding protein-3 in the cancer and control groups was 2,724 +/- 647 and 2,673 +/- 589 ng./ml., respectively (p = 0.3). Adjustment for age and PSA showed significantly lower IGF-1 in cancer cases, while IGF binding protein-3 was not significant. ROC values were significantly higher for free-to-total PSA and PSA than for crude and age adjusted IGF-1 and IGF binding protein-3.
Conclusions: Our data indicate that serum IGF-1 or IGF binding protein-3 does not predict the results of prostate biopsy in men with elevated PSA or abnormal digital rectal examination. This finding implies that while there is evidence that the IGF-1 level is a risk factor for prostate cancer, neither IGF-1 nor IGF binding protein-3 can be used as a tumor marker for this disease.