Review question
Men with advanced prostate cancer get hormonal treatment that lowers the level of the male sex hormones. This does not cure men from cancer but can stop the cancer from growing and help men live longer. However, it is not clear whether it is better to start these hormone treatments early on or later, when there are x-ray or laboratory findings showing that the cancer is growing or when men start having symptoms from the prostate cancer. We did this study to compare starting treatment early versus late.
Background
Prostate cancer can be cured if the disease is only in the prostate gland. These men can have radiation or surgery to remove their prostate. If the cancer has spread outside the prostate, for example to the lymph nodes or the bones, there is no cure. Hormonal treatment that lowers the level of the male sex hormones can slow down cancer growth and prevent it from causing problems. This treatment can be started straight after the diagnosis is made (early) or when the cancer has been shown to grow (late) based on x-ray or laboratory findings or when it has started causing problems (also late).
Study characteristics
We considered only studies in which chance decided whether men with prostate cancer got early or late hormonal treatment.
Key results
We found 10 studies that matched our question. We found that early hormonal treatment probably lowers the risk of dying from any cause. The risk of serious unwanted effects may be similar to that of late treatment.
Early hormonal treatment probably lowers the risk of dying from prostate cancer and slightly lowers the risk of problems related to cancer spreading to the bones.
Men getting early treatment may be more likely to feel tired and develop heart weakness.
Overall quality of life is probably unaffected (or only slightly affected) by early treatment.
The certainty of evidence was either moderate, which means that the true results are likely close to what we found; or low, in which case our concern is that the true results could be quite different to what we found.
Early AST probably extends time to death of any cause and time to death from prostate cancer. It may slightly decrease the rate of skeletal events. Rates of serious adverse events and quality of life may be similar. It may increase fatigue and may increase the risk of heart failure. Better quality trials would be particularly important to better understand the outcomes related to possible treatment-related harm, for which we only found low-certainty evidence.
Standard androgen suppression therapy (AST) using surgical or medical castration is considered a mainstay of advanced hormone-sensitive prostate cancer treatment. AST can be initiated early when disease is asymptomatic or deferred when patients suffer symptoms of disseminated prostate cancer.
To assess the effects of early versus deferred standard AST for advanced hormone-sensitive prostate cancer.
For this Cochrane Review update, we performed a comprehensive search of multiple databases (CENTRAL, MEDLINE, Embase, Web of Science; last searched November 2018) and two clinical trial registers, with no restrictions on the language of publication or publication status. We also searched bibliographies of included studies and conference proceedings (last searched January 2019).
We included all randomised controlled trials (RCTs) with a direct comparison of early versus deferred standard AST. We excluded all other study designs. Participants included had advanced hormone-sensitive prostate cancer receiving surgical or medical castration.
Two review authors independently classified studies and abstracted data. The primary outcomes were time to death of any cause and serious adverse events. Secondary outcomes were time to disease progression, time to death from prostate cancer, adverse events and quality of life. We performed statistical analyses using a random-effects model and assessed the certainty of evidence according to GRADE. We performed subgroup analyses for advanced but non-metastatic disease (T2-4/N+ M0), metastatic disease (M1), and prostate-specific antigen (PSA) relapse.
We identified seven new RCTs since publication of the original review in 2002. In total, we included 10 RCTs.
Primary outcomes
Early AST probably reduces the risk of death from any cause over time (hazard ratio (HR) 0.82, 95% confidence interval (CI) 0.75 to 0.90; moderate-certainty evidence; 4767 participants). This corresponds to 57 fewer deaths (95% CI 80 fewer to 31 fewer) per 1000 participants at 5 years for the moderate risk group and 23 fewer deaths (95% CI 32 fewer to 13 fewer) per 1000 participants at 5 years in the low risk group. We downgraded for study limitations. Early versus deferred AST may have little or no effect on serious adverse events (risk ratio (RR) 1.05, 95% CI 0.95 to 1.16; low-certainty evidence; 10,575 participants) which corresponds to 6 more serious adverse events (6 fewer to 18 more) per 1000 participants. We downgraded the certainty of evidence for study limitations and selective reporting.
Secondary outcomes
Early AST probably reduces the risk of death from prostate cancer over time (HR 0.69, 95% CI 0.57 to 0.84; moderate-certainty evidence). This corresponds to 62 fewer prostate cancer deaths per 1000 (95% CI 87 fewer to 31 fewer) at 5 years for the moderate risk group and 24 fewer death from prostate cancer (95% CI 34 fewer to 12 fewer) per 1000 men at 5 years in the low risk group. We downgraded the certainty of evidence for study limitations.
Early AST may decrease the rate of skeletal events (RR 0.37, 95% CI 0.17 to 0.80; low-certainty evidence) corresponding to 23 fewer skeletal events per 1000 (95% CI 31 fewer to 7 fewer). We downgraded for study limitations and imprecision. It may also increase fatigue (RR 1.41, 95% CI 1.23 to 1.62; low-certainty evidence), corresponding to 31 more men with this complaint per 1000 (95% CI 18 more to 48 more). We downgraded for study limitations and imprecision. It may increase the risk of heart failure (RR 1.90, 95% CI 1.09 to 3.33; low-certainty evidence) corresponding to 27 more events per 1000 (95% CI 3 more to 69 more). We downgraded the certainty of evidence for study limitations and imprecision.
Global quality of life is probably similar after two years as assessed with the EORTC QLQ-C30 (version 3.0) questionnaire (mean difference −1.56, 95% CI −4.50 to 1.38; moderate-certainty evidence) with higher scores reflecting better quality of life. We downgraded the certainty of evidence for study limitations.