Reviewed research

Authors Potosky AL, Haque R, Cassidy-Bushrow AE, et al.

Review Date May 2014

Citation J Clinical Oncol 2014; 32(13): 1324-30



Androgen deprivation therapy (ADT) is an effective palliative treatment for metastatic prostate cancer and is a useful treatment in certain clinical settings. However, despite not being recommended in professional guidelines, ADT is often used (in the USA) to treat clinically localised prostate cancer. The effect of ADT on cause-specific and overall mortality has not been well established while evidence of adverse side effects of ADT has been accruing in recent years. Accurate mortality data can help to inform treatment decisions.



Given the increasing use of primary ADT and the risks of serious adverse effects, the aim of this study was to provide accurate mortality data for men diagnosed with clinically localised prostate cancer to assess the all-cause mortality effect of primary treatment with ADT. The authors noted that other studies addressing similar questions have had various shortcomings.



This retrospective cohort study used utilisation and cancer registry data from three integrated health care plans in the USA that collect comprehensive data on diagnoses, clinical encounters, treatments and tumour-registry data. Men who were diagnosed between 1995 and 2008, were not treated with curative intent therapy, and received follow-up until December 2010 were included in the study (n = 15,170; aged 35 years and older). Primary ADT was defined as receipt of medical ADT for localised prostate cancer within the first 12 months of the diagnosis and without radiation or radical prostatectomy. The main outcomes were all-cause and prostate cancer-specific mortality. Data analysis included Cox proportional hazards models with and without propensity score analysis. For each outcome, adjustment was made for socio-demographic and clinical prognostic factors and 34 comorbidities. Sub-group analyses were conducted to see if associations of ADT with mortality differed according to clinical subgroups defined by baseline PSA/Gleason score/clinical stage or age at diagnosis.



3,435 of the 15,170 men received ADT in the first 12 months; this group had worse prognostic factors but after adjusting for propensity scores, balance was achieved between groups. There were 4,921 deaths in the study, 32% from prostate cancer. Overall, primary ADT was associated with neither all-cause mortality (hazard ratio [HR], 1.04; 95% CI, 0.97 to 1.11) nor prostate-cancer-specific mortality (HR, 1.03; 95% CI, 0.89 to 1.19) after adjusting for all sociodemographic and clinical characteristics. Primary ADT was associated with decreased risk of all-cause mortality only among the subgroup of men with a high risk of cancer progression (HR, 0.88; 95% CI, 0.78 to 0.97) and with increased risk of overall mortality in men with low-risk prostate cancer (HR, 1.41; 95% CI, 0.99 to 1.82).



The study did not identify an all-cause or prostate cancer-specific mortality benefit from primary ADT compared with no primary ADT for men with clinically localised prostate cancer who did not receive curative intent therapy. Men with higher-risk disease may derive a small clinical benefit from primary ADT. The authors claim their study provides the best available contemporary evidence on the lack of survival benefit from primary ADT for men with clinically localised prostate cancer and that clinical trials to answer this question are unlikely. Given the risk of adverse effects from ADT, there is no clinical rationale for using ADT in this context.


Points to Note
  1. The use of ADT in treating prostate cancer has increased in the past 10 years and it appears that ADT is being used in men for whom it is not indicated such as those with localised prostate cancer with no curative therapy.
  2. The adverse effects of ADT are well documented including a possible increased risk of cardiovascular disease and diabetes, such that risks need to be weighed against benefits before implementing ADT.
  3. One strength of the study was the large number of clinical prognostic factors available with the use of propensity score analysis.
  4. However, the study limitations include its retrospective design and use of databases/registries where unmeasured confounding could be having an effect; it is possible that the men treated with ADT had clinical factors that were associated both with the decision to use ADT as a primary therapy and with the risk of death, but were not captured in the data.
  5. The finding of a reduced mortality risk in the high-risk sub-group may be due to chance given no adjustment for the multiple comparisons made in the study.
  6. These data are from the USA; the indications for use of ADT in Australia may not be the same as presented in this study, although a significant increase in use of ADT in Australia in the past decade or so has been documented.


Website: http://www.ncbi.nlm.nih.gov/pubmed/24638009

Reviewed by
Updated on