The Effect of Testosterone Replacement on Prostate Cancer
The Effect of Testosterone Replacement on Prostate Cancer
Prostate growth is dependent on the presence of testosterones; conversely, antiandrogen and orchidectomy can decrease prostate volume in patients with BPH. There is abundant evidence that androgens influence the development of prostate cancer. Although the risk that testosterone treatment triggers prostate cancer was not fully recognized. Resolving this question will inform methods of treating LOH accompanied by prostatic problems.
The 22 placebo-controlled, randomized studies included in this meta-analysis provide direct comparasions of the indications of prostate cancer (PSA levels' change, abnormal PSA levels, prostate nodule, prostate biopsy) and the incidence of new prostate cancer, which are all relative sensitive indicators for prostate cancer in testosterone therapy and placebo. In all, only 12, 5, 9 and 8 of the included 22 RCTs reported the incidence of abnormal PSA levels, prostate nodule, prostate biopsy and new prostate cancer, respectively. Namely, less than half of the included 22 RCTs reported the incidence of the indications of prostate cancer or new prostate cancer. It showed that neither short-term nor long-term TRT increased the risk of prostate cancer.
In our analysis, short-term TRT delivered by transdermal application was more likely to increase PSA levels than treatment with a placebo. A section of urologists consider that testosterone has a linear effect on prostate growth, and prostate dysfunction as an adverse effect of this was the dominant theory in the past. A new paradigm has been put forward: the saturation model of testosterone and the prostate. The theory holds that testosterone's effect on the prostate reaches a saturation point well below the physiological testosterone levels encountered in the clinical setting, beyond which additional testosterone does not have an increased effect. The inclusion criteria for patients receiving transdermal treatment included PSA level <1.5 ng ml, which was much lower than normal (Table 1). This may be the reason leading to increased PSA levels in the short term. But as to the incidence of new prostate cancer, prostate biopsy, prostate nodule and abnormal PSA levels, there was no apparent difference between testosterone with placebo, especially in transdermal administration method. Our study also evaluated the safety of long-term TRT (12–36 months) for prostate cancer.
All 22 RCTs included in our analysis involved rigorous, periodic monitoring of patients, and treatment was withdrawn when there were indications suspicious for prostate cancer or other serious complications. In addition, for all 22 RCTs no patient had prostate enlargement at baseline, all patients had normal PSA levels at baseline and one RCT reported patients who underwent prostate biopsy at study entry and at the end of the study, which implicated that testosterone can not be a prostate cancer stimulus. Consequently, short-term TRT is safe in terms of prostate cancer under judicious and careful monitoring. The fact that all 22 RCTs included rigorous periodic monitoring of patients and withdrew patients when they suspected prostate cancer might be the reason why the meta-analysis failed to detect an increased hazards of prostate cancer among those treated with TRT. Our conclusion is based on the fact that TRT has no adverse effect on prostate cancer in patients without prostatic hyperplasia. Whether or not TRT will increase the risk for prostate cancer in patients with BPH needs further investigation in larger, high-quality studies.
Although difference resided in the baseline level of testosterone included in 22 RCTs, the cohorts of 17 of the 22 studies in our meta-analysis were diagnosed as LOH. Controversy surrounds setting a lower limit of normal testosterone. There is also no generally accepted lower limits of normal testosterone level in aging males. It was demonstrated that demographic differences in testosterone level within population of aging male exist. Although different methods and dosages were used in the selected RCTs, all patients in the 22 RCTs achieved normal testosterone levels and supraphysiological levels were avoided.
The 22 RCTs included in this meta-analysis were all double-blinded. According to the quality-assessment scale that we developed, the quality of the individual studies in the meta-analysis was high. The results of this analysis are important from a scientific standpoint, and they apply to everyday clinical practice, particularly because data were analyzed by method of TRT delivery. The results of the subgroup and sensitivity analyses are in accordance with our findings, indicating that our results are robust and reliable.
Nevertheless, there are some limitations to our analysis. Data in the studies covered by this meta-analysis are insufficient to determine whether more severe grades of prostate cancer are found with testosterone treatment versus without testosterone treatment. Major limitations include heterogeneity in the populations examined (Table 1), testosterone dosage used (Table 1) and baseline PSA levels. LOH is a clinical and biochemical syndrome that adversely affects the function of multiple organ systems. The presence of symptoms associated with LOH was not consistent across the included studies, and this may have contributed to the heterogeneous population. Although we conducted subgroup and sensitivity analyses to assess the quality of the studies, the problem of heterogeneity still could not be completely avoided. Although 21 of the 22 RCTs drew morning blood samples to measure testosterone levels, 17 of the 22 studies reported six measurement assay methods that included tandem mass spectrometry-liquid chromatography, mass spectroscopy, chemiluminescent immunoassay, radioimmunoassay, fluoroimmunoassay and electrochemiluminescence. It is generally known that the first method is the gold standard. Although the results of subgroup analysis by testosterone assay types (gold standard and not gold standard) matched our findings, this made inclusion of studies using different testosterone assays problematic and potentially influenced the results of the meta-analysis.
Discussion
Prostate growth is dependent on the presence of testosterones; conversely, antiandrogen and orchidectomy can decrease prostate volume in patients with BPH. There is abundant evidence that androgens influence the development of prostate cancer. Although the risk that testosterone treatment triggers prostate cancer was not fully recognized. Resolving this question will inform methods of treating LOH accompanied by prostatic problems.
The 22 placebo-controlled, randomized studies included in this meta-analysis provide direct comparasions of the indications of prostate cancer (PSA levels' change, abnormal PSA levels, prostate nodule, prostate biopsy) and the incidence of new prostate cancer, which are all relative sensitive indicators for prostate cancer in testosterone therapy and placebo. In all, only 12, 5, 9 and 8 of the included 22 RCTs reported the incidence of abnormal PSA levels, prostate nodule, prostate biopsy and new prostate cancer, respectively. Namely, less than half of the included 22 RCTs reported the incidence of the indications of prostate cancer or new prostate cancer. It showed that neither short-term nor long-term TRT increased the risk of prostate cancer.
In our analysis, short-term TRT delivered by transdermal application was more likely to increase PSA levels than treatment with a placebo. A section of urologists consider that testosterone has a linear effect on prostate growth, and prostate dysfunction as an adverse effect of this was the dominant theory in the past. A new paradigm has been put forward: the saturation model of testosterone and the prostate. The theory holds that testosterone's effect on the prostate reaches a saturation point well below the physiological testosterone levels encountered in the clinical setting, beyond which additional testosterone does not have an increased effect. The inclusion criteria for patients receiving transdermal treatment included PSA level <1.5 ng ml, which was much lower than normal (Table 1). This may be the reason leading to increased PSA levels in the short term. But as to the incidence of new prostate cancer, prostate biopsy, prostate nodule and abnormal PSA levels, there was no apparent difference between testosterone with placebo, especially in transdermal administration method. Our study also evaluated the safety of long-term TRT (12–36 months) for prostate cancer.
All 22 RCTs included in our analysis involved rigorous, periodic monitoring of patients, and treatment was withdrawn when there were indications suspicious for prostate cancer or other serious complications. In addition, for all 22 RCTs no patient had prostate enlargement at baseline, all patients had normal PSA levels at baseline and one RCT reported patients who underwent prostate biopsy at study entry and at the end of the study, which implicated that testosterone can not be a prostate cancer stimulus. Consequently, short-term TRT is safe in terms of prostate cancer under judicious and careful monitoring. The fact that all 22 RCTs included rigorous periodic monitoring of patients and withdrew patients when they suspected prostate cancer might be the reason why the meta-analysis failed to detect an increased hazards of prostate cancer among those treated with TRT. Our conclusion is based on the fact that TRT has no adverse effect on prostate cancer in patients without prostatic hyperplasia. Whether or not TRT will increase the risk for prostate cancer in patients with BPH needs further investigation in larger, high-quality studies.
Although difference resided in the baseline level of testosterone included in 22 RCTs, the cohorts of 17 of the 22 studies in our meta-analysis were diagnosed as LOH. Controversy surrounds setting a lower limit of normal testosterone. There is also no generally accepted lower limits of normal testosterone level in aging males. It was demonstrated that demographic differences in testosterone level within population of aging male exist. Although different methods and dosages were used in the selected RCTs, all patients in the 22 RCTs achieved normal testosterone levels and supraphysiological levels were avoided.
The 22 RCTs included in this meta-analysis were all double-blinded. According to the quality-assessment scale that we developed, the quality of the individual studies in the meta-analysis was high. The results of this analysis are important from a scientific standpoint, and they apply to everyday clinical practice, particularly because data were analyzed by method of TRT delivery. The results of the subgroup and sensitivity analyses are in accordance with our findings, indicating that our results are robust and reliable.
Nevertheless, there are some limitations to our analysis. Data in the studies covered by this meta-analysis are insufficient to determine whether more severe grades of prostate cancer are found with testosterone treatment versus without testosterone treatment. Major limitations include heterogeneity in the populations examined (Table 1), testosterone dosage used (Table 1) and baseline PSA levels. LOH is a clinical and biochemical syndrome that adversely affects the function of multiple organ systems. The presence of symptoms associated with LOH was not consistent across the included studies, and this may have contributed to the heterogeneous population. Although we conducted subgroup and sensitivity analyses to assess the quality of the studies, the problem of heterogeneity still could not be completely avoided. Although 21 of the 22 RCTs drew morning blood samples to measure testosterone levels, 17 of the 22 studies reported six measurement assay methods that included tandem mass spectrometry-liquid chromatography, mass spectroscopy, chemiluminescent immunoassay, radioimmunoassay, fluoroimmunoassay and electrochemiluminescence. It is generally known that the first method is the gold standard. Although the results of subgroup analysis by testosterone assay types (gold standard and not gold standard) matched our findings, this made inclusion of studies using different testosterone assays problematic and potentially influenced the results of the meta-analysis.
Source...