SIU Academy® - Official eLearning Portal of SIU (Société Internationale d'Urologie)

We were glad to read the article from Sooriakumaran et al. on comparative effectiveness of radical prostatectomy (RP) and radiotherapy (RT) in prostate cancer (PCa). This paper presents high-quality evidence by comparing the most widely used therapeutic alternatives to date for the treatment of PCa. Although previous studies have evaluated the comparative effectiveness of treatments, most such reports were arguably biased by lack of complete and accurate data that would help predict mortality.
The management of prostate cancer (PCa) remains controversial because of its variable natural history, the diversity of available treatments, and the lack of randomized controlled trials (RCTs) comparing the different treatment approaches. The choice of an adequate therapeutic option for PCa depends on several factors, including tumour stage, prostate-specific antigen (PSA) value, Gleason score, patient's age, comorbidity, life expectancy, and, ultimately, patient's preferences.
The current gold standard management of localized PCa is radical therapy, either as surgery or radiation therapy. The literature has described radiation therapy to be associated with an increased mortality compared with surgery [1]. A recently published study suggested that surgery offers a better survival benefit for men with localized Pca [2]. The large, collaborative, Swedish-Dutch, observational, multicentre study led by Prasanna Sooriakumaran compared the oncologic effectiveness of RP and RT in PCa, and analyzed the mortality outcomes in 34,515 patients treated primarily with each therapeutic alternative with up to 15 years of follow-up. The authors found that RP was associated with better cancer-specific survival (CSS) than RT among men with non-metastatic PCa, being likely superior to radiation, particularly for the younger age group and those with no or few comorbidities. Data from Sweden's National Prostate Cancer Registry (PCBaSe) showed that men were treated for PCa throughout Sweden with either surgery (n=21,533) or radiotherapy (n=12,982) as their first treatment option and constituted the study cohort. Patients were categorized by risk group (localized low risk, localized intermediate risk, localized high risk, and non-localized, with the patients having any T3-4, N+, M+, PSA >50), age (younger than 65 years, 65 to 74 years, or 75 years and older), and Charlson comorbidity score (0, 1, or ≥2). The primary outcome of interest was death from PCa, having defined survival time as the interval between the date of diagnosis of PCa and the date of death, emigration, or end of follow-up.
In their results, the researchers found that patients who received RT were associated with higher Gleason scores and clinical stages, were older, and had higher PSA levels than patients who underwent RP (p<.0001 for all comparisons). In addition, PCa mortality became a larger proportion of overall mortality as risk group increased for both the RP and RT cohorts. The study also showed that survival outcomes favoured RP for patients with localized PCa (risk groups 1 to 3), and that treatment results were similar for patients with locally advanced/metastatic disease. As for men with advanced PCa, both modalities appeared equivalent, and thus they concluded that the conventional view that surgery is not indicated in this group may be incorrect.
From our perspective, the study has several strengths. On one hand, this is reflected by its large population-based dataset obtained by the highly reliable Swedish registry, a complete data collection of PCa cases at the time of diagnosis and during follow-up, with their important and highly validated patient-tumour covariates including comorbidity status. On the other hand, the study's strengths are reflected by the application of a welldescribed statistical methodology to adjust for differences in the distribution of covariates between surgical and radiotherapy cohorts and to account for competing risks for mortality.
However, even sophisticated statistical techniques cannot completely eliminate the biases associated with observational studies, such as confounding by indication and the lack of information regarding secondary treatments applied, which might have showed a greater proportion of men in the RP arm receiving RT after surgery, thus having biased the authors' findings in favour of surgery. In addition, data was obtained from a Swedish Caucasian majority of a non-screen diagnosed population; hence, it is uncertain how the findings may relate to screen diagnosed multi-ethnic populations.
One could argue about the observational nature of this study. In this context, results of RCTs throughout the literature may have limited the generalizability due to the existent heterogeneity among the enrolled population and community populations; thus Sooriakumaran et al. highlighted the importance of using observational data from actual medical practice in comparative effectiveness studies to complement the evidence from RCTs. On the other hand, particularly in PCa, most such studies have evaluated biochemical recurrence as the endpoint, varying among RP and RT cohorts, thus having shown conflicting results [3,4]. Hence, they support the fact that death remains the most valid endpoint for comparative studies in PCa.
The take-home message from this paper is the clear idea that surgery is likely superior to radiation for the majority of men who have localized PCa, particularly for the younger age group and those with no or few comorbidities. From our perspective, even when largescale RCTs are still needed to confirm the true superiority of RP over RT for localized PCa, we strongly believe that results extracted from these large population-based studies should bear influence not only in the decisionmaking process as urologists, but also in the entire health care system.