Forty-four studies were included.
Men without a history of prostate cancer were assessed in 11 placebo-controlled RCTs (n=893) and 29 non-placebo controlled studies (n=1,732, which comprised: one case series n=20; 15 prospective studies n=1,354; five retrospective studies n=349; and eight case reports n=9).
Men with a history of prostate cancer were assessed in four studies: one prospective (n=5) and three retrospective case series (n=48).
Men without a history of prostate cancer:
Randomised-placebo controlled trials (11 studies, nine with follow up of one year or less): rates of prostate cancer were similar for testosterone and placebo–treated groups (1.3% versus 1.5%; overall seven out of 542 testosterone-treated men developed prostate cancer). The incidence across studies ranged from 0% in both groups to 9.5% in a testosterone-group versus 21% for a placebo group in one study; this study was the only one that routinely performed end-of-study prostate biopsies.
Non-placebo controlled studies (21 studies): seven studies reported a total of 12 cases of prostate cancer; the incidence per study ranged from 1.2% to 4.5%. One retrospective study examined 20 cases of prostate cancer in men who received testosterone; 11 cancers were detected in the first two years of testosterone therapy and nine were detected 28 months to eight years after the start of treatment.
Men with a history of prostate cancer:
The four studies (n=53) found no evidence of a recurrence of prostate cancer over follow up that ranged from 0.5 to 12 years. Following treatment, prostate-specific antigen levels were less than 1ng/mL after 8.5 years in one retrospective study and undetectable for up to 12 years in two other retrospective studies.