To use a clinical model of androgen-deprivation therapy (ADT) followed by radical prostatectomy (RP) to test the hypothesis that prostatic intraepithelial neoplasia (PIN, a premalignant lesion of the prostate causally linked to prostate cancer) is heterogeneous for hormone responsiveness, which might explain aspects of the heterogeneity of the natural history of prostate cancer, for although ADT has been used to reduce prostate cancer, there are controversial data on the effect of ADT on PIN.
PATIENTS AND METHODS
We assessed retrospectively patients with biopsy-confirmed prostate cancer who had RP; some patients had received ≥3 months of ADT at the discretion of their surgeons, and patients from the same cohort who did not have ADT were used as controls. Patients were sequentially selected from the database and their pathology slides were reviewed by a pathologist unaware of the initial presence of PIN (assessed by an independent observer). Fisher’s exact test was used to compare the proportions of patients who had residual PIN in the study and control groups. Exact logistic regression was used to evaluate the duration of ADT on PIN regression.
Eighteen patients initially diagnosed with PIN who had no ADT were identified, and 28 with PIN who had ADT were also assessed. All patients who had had no ADT had residual PIN, whereas seven of 28 receiving ADT had no residual PIN (P = 0.043). The evaluation of ADT between responders and nonresponders showed a statistically significant association between PIN regression and the duration of ADT (P <0.001). However, the response of PIN to ADT was not uniform, as 16% of patients on ADT for >6 months had residual PIN, suggesting variable sensitivity of PIN to ADT.
These results show that ADT causes PIN to regress, and that there is heterogeneity in this effect with the duration of ADT. We propose future prospective, multicentre, randomized trials in which the effect of ADT on PIN is characterized further.