CAG polymorphic repeat length in androgen receptor gene combined with pretreatment serum testosterone level as prognostic metastatic prostate factor in patients with cancer

Objective: Androgen ablation has been the initial treatment of choice for men with metastatic prostate cancer, but the disease generally relapses to an androgen-independent state thereafter. To understand which groups respond well or poorly to endocrine therapy is thus important. Several studies have shown that pretreatment serum testosterone (T) levels and the length of the CAG repeat at the N-terminal region of the androgen receptor are significant. However, the relevance of a combination of these factors has not been reported. We therefore investigated the clinical significance of CAG repeat length and pretreatment serum T levels among Japanese patients with metastatic prostate cancer (TxNxM1), and analyzed their relevance to survival. Methods: Fifty-two Japanese patients with metastatic prostate cancer were enrolled in this study. We determined the length of the CAG repeat by both PCR sequencing and fragment analysis. Pretreatment serum T levels were measured using a radioimmunoassay. We examined the clinical significance of the CAG repeats and T levels individually and in combination with respect to several clinical factors. Results: The pretreatment T level in the responder group was significantly higher than that in the nonresponders (p = 0.009) and the mean was 4.33 +/- 2.12 ng/ml. Kaplan-Meier analyses revealed that cause-specific survival was significantly enhanced in patients with higher levels of T (p = 0.0489). The length of the CAG repeat was positively associated with age at diagnosis (p = 0.032). The mean CAG repeat length was 22.5 +/- 3.0 and this value was significantly shorter in patients with poorly differentiated, than with well and moderately differentiated tumors (p = 0.019). Kaplan-Meier analyses revealed a significantly better cause-specific survival rate as well as progression-free survival rate in patients with longer CAG repeats. Cause-specific survival curves were better in patients with higher T levels and longer CAG repeats than with lower T levels and shorter CAG repeats (p = 0.0066). A multivariate analysis showed that the most significant prognostic factor was histological grade, followed by EOD grade, marker response and the combination of T and CAG. Conclusion: Pretreatment serum T levels together with the length of the N-terminal CAG repeat of the androgen receptor gene can distinguish responders from non-responders to androgen ablation. These parameters appear to be clinically useful, in that therapies appropriate to individual patients could be selected. Further studies are necessary to confirm these results. (c) 2004 Elsevier B.V. All rights reserved.