Mortality in men with advanced prostate cancer appears to be reduced with radical treatment compared to androgen deprivation alone

Sooriakumaran P.l, Nyberg T.2, Akre 0.3, Carlsson S.3, Haendler L.3, Steineck G.2, Wiklund P.3

lUniversity of Oxford, Nuffield Dept, of Surgical Sciences, Oxford, United Kingdom, 2Karolinska Institutet, Dept, of Clinical Cancer Epidemiology, Stockholm, Sweden, 3Karolinska Institutet, Dept, of Molecular Medicine & Surgery, Stockholm, Sweden

INTRODUCTION & OBJECTIVES: Standard management for men with advanced prostate cancer is androgen deprivation therapy (ADT) alone. In other malignancies, such as renal and ovarian cancer, there is evidence that treatment of the primary tumour might reduce progression of metastatic disease and lead to improved survival. Our objective was to evaluate whether radical treatment of the primary tumour in men with advanced prostate cancer improved survival over those managed with ADT alone.

MATERIAL & METHODS: The National Prostate Cancer Registry of Sweden is linked to eight national datasets to form PCBaSe and provides >98% complete information on virtually all patient and tumour covariates of all prostate cancer cases in Sweden diagnosed from 1996. We abstracted data from PCBaSe on men with PSA>50 or M+ or T4 disease, and matched them exactly for grade, T-stage, M-stage, and Charlson comorbidity index (CCI). We produced cumulative incidence curves (CICs) as well as performed statistical adjustments via classical regression techniques and propensity scoring, to compare prostate cancer mortality (PCM) and other cause mortality (OCM) in those men treated with ADT as primary therapy versus those treated initially with radical therapy (surgery or radiotherapy).

RESULTS: The matched sample was 699 cases per group. CICs demonstrated lower PCM and OCM in the radical treatment group compared to the ADT group up to 14 years follow-up there were 231/699 (33.0%) deaths from prostate cancer in the ADT group versus 93/699 (13.3%) in the radical treatment arm, and 90/699 (12.9%) deaths from other causes in the ADT cohort versus 62/699 (8.9%) in the radical treatment group. Propensity score-adjusted subdistribution hazard ratios (sHR) for ADT versus radical treatment were 2.89 (95% Cl, 2.25-3.71) for PCM and 1.41 (95% Cl, 1.01-1.98) for OCM. A post-hoc sensitivity analysis found that a residual confounder differentially prevalent in 90% versus 10% in the ADT and radical treatment groups respectively would need a sHR of 4.09 to account for the above PCM point estimate, with still higher confounder sHR for less skewed confounder prevalences.

CONCLUSIONS: This large, observational study from a comprehensive dataset suggests that men with advanced prostate cancer might benefit in survival terms from being managed with radical therapy as their primary treatment rather than ADT alone. Confounding is highly unlikely to account for these results, and this study provides an epidemiologic rationale to consider a future RCT in this setting.

Press release note: the original abstract contained a figure, which has been removed from this press release for possible use in subsequent publication.