Studies on Hormone Replacement Therapy (HRT) and Cancer Risk For Men
Studies are now plentiful that show the benefits of hormone replacement, not just for health benefits, but also for reduced cancer risk. My mentor in the field of Hormone Replacement Therapy (HRT), Neal Rouzier MD, gave a superb summary of the literature that I have summarized below with regards to Testosterone (and growth hormone) replacement for men and the risk of prostate cancer.
The study by Fuhrman Et. Al., “Basal Growth Hormone Concentrations in Blood and The Risk for Prostate Cancer: A Case-Control Study”, found a statistically significant trend of decreasing prostate cancer risk across increasing Growth Hormone (GH) quintiles, and that lower levels of Growth Hormone in the serum are associated with increased prostate cancer risk. This fits with the experience that children on GH do not have increased risk for cancers. 
The study by Devi et. al., “Insulin-like Growth Factor Binding Protein-3 Induces Early Apoptosis in Malignant Prostate Cancer Cells and Inhibits Tumor Formation in Vivo”, suggests that IGFBP-3 induces early apoptosis and has potential tumor suppressive effect in prostate cancer. The fact is that reducing visceral fat reduces cancer risk. IGFBP-3 protects from that risk and the only way to increases IGFBP-3 is by giving GH (not currently possible for regulatory reasons). 
The study by Mehmed, “Acromegaly and Cancer: Not a problem”, showed that elevated IGF-1 bioactivity and activation of IGF-1 receptor are associated with cell proliferation and growth advantage, whereas IGFBP-3 bioactivity promotes an apoptotic advantage. IGHBP-3 inhibits IGF-1 induced prostate cancer cell growth and breast cancer cells are diverted into an apoptotic phase by IGFBP-3. So, while IGF-1 is mitotic and stimulates growth of cancer cells, this is halted by IGFBP-3. 
The study by Chen et. al., “Prostate Carcinoma Incidence in Relation to Prediagnostic Circulation Levels of Insulin-Like Growth Factor 1, Insulin-Like Growth Factor Binding Protein 3 and Insulin”, showed that those with the highest level of IGF-1 were associated with risk of prostate cancer in range of 0.67 to 0.73 (protective) compared with those with the lowest levels of IGF-1. An increase in IGFBP-3 was associated with a 0.4 relative risk of developing prostate cancer. It is known that estrogen is protective against prostate cancer (a reason not to lower your estrogen levels intentionally), and this study shows IGFBP-3 is associated with decreased risk of developing prostate cancer. The way one would raise their IGHBP-3 would be to normalize GH levels for those deficient (currently difficult due to regulatory issues). 
The study by Carter et. al., “Longitudinal Evaluation of Serum Androgen Levels in Men With and Without Prostate Cancer”, showed there was no measurable difference in serum testosterone levels among men who were destined to develop prostate cancer and those without. 
The study by Morgentaler et. al., “Prevalence of Prostate Cancer Among Hypogonadal Men With Prostate-Specific Antigen Levels of 4.0 ng/ml or less”, found that prostate cancer was present in more than 1 of 7 hypogonadal men with PSA of 4.0 ng/ml or less. An increased risk of prostate cancer was associated with more severe reductions in testosterone. If testosterone caused prostate cancer to grow, we would see it in studies like this. Clearly, taking testosterone if your levels are low is protective. 
The study by Scherr et. al., “The Nonsteroidal Effects of Diethylstilbestrol: The Rationale for Androgen Deprivation Therapy Without Estrogen Deprivation in The Treatment of Prostate Cancer”, reminds us that estrogen has been used for decades to treat prostate cancer, as it is protective or can treat prostate cancer. This study used DES (not ideal as it is thrombotic) but the study did show benefits. Likely, the higher the testosterone the higher the estradiol which is protective. This gives further support of not lowering estrogen levels in men who are on testosterone replacement therapy and as a result have higher estrogen levels. 
The study by Eaton et. al., “Endogenous Sex Hormones and Prostate Cancer: A Quantitative Review of Prospective Studies”, finds there is no correlation between serum levels of estrogen or testosterone and prostate cancer. 
The study by Shabsign et. al., “Testosterone Therapy in Hypogonadal Men and Potential Prostate Cancer Risk: A Systematic Review Testosterone Therapy and Prostate Cancer Risk”, showed that none of the studies where testosterone therapy was used for hypogonadism resulted in increased prostate cancer risk or increased Gleason grade of cancer detected in treated vs untreated men. 
The study by Savage et. al., “Sex Hormones Unrelated to Prostate Cancer Risk”, looked at data from 18 studies to determine the relationship between blood levels of sex hormones and prostate cancer. The authors write: “The results of this collaborative analysis of the existing worldwide data on the associations between endogenous hormone concentrations and prostate cancer risk indicate that circulating concentrations of androgens [and estrogens] do not appear to be associated with the risk of prostate cancer”. 
The study by Heikkila et. al., “Serum Testosterone and Sex-Hormone-Binding Globulin Concentrations and the Risk of Prostate Carcinoma”, concluded that the results of their study do not corroborate the hypothesis that serum testosterone, SHBG or andro stenedione are determinants of subsequent occurrence of prostate carcinoma.