Vasectomy and Risk of Prostate Cancer: How to Weigh Current Evidence
A partir des données de la cohorte européenne EPIC portant sur 84 753 participants âgés de 35 à 79 ans, cette étude de cohorte évalue l'association entre une vasectomie, le risque de développer un cancer de la prostate et la mortalité spécifique (durée moyenne de suivi : 15,4 ans ; 4 377 patients atteints d'un cancer de la prostate)
In the article accompanying this editorial, Byrne et al1 present results from the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. They followed 84,753 men for an average of 15 years and observed 4,377 incident cases of prostate cancer. They found that vasectomy was not associated with subsequent risk of overall, advanced, high-grade, or fatal prostate cancers.1 These findings contribute to a growing body of literature investigating whether vasectomy confers an increased risk of prostate cancer. Several studies have found no association,1-3 whereas others have observed small to moderate elevated risk.4-6 To facilitate this scientific debate, we aim to provide a contextual framework for evaluating the evidence. The first framework component is determining whether inconsistent findings may arise due to systematic bias. Detection bias due to differential prostate-specific antigen (PSA) screening is highly relevant to studies of vasectomy and prostate cancer, because PSA screening is often directly associated with use of medical care and maintenance of healthy behaviors. If men with vasectomy are more likely to be screened, as appears to be true in all populations that have been studied, detection bias may lead to a spurious association with disease diagnosis. An observed association between vasectomy and overall or localized prostate cancer, but not with advanced or lethal disease, may indicate the association is due to differential PSA screening, because less-advanced disease is more likely to be diagnosed through PSA screening. Conversely, an association with lethal but not localized disease is unlikely a result of detection bias, because more-advanced disease can be detected clinically as opposed to through screening alone. Moreover, because screening reduces the risk of lethal prostate cancer,7 a true association of vasectomy with potentially lethal prostate cancer could partially be masked due to higher screening rates in the vasectomy group. For these reasons, collection of adequate information and adjustment for PSA screening in epidemiologic studies is imperative. (...)