Prediagnosis Prostate-Specific Antigen Testing History in Patients With Incident Prostate Cancer
Menée en France à partir de données 2011-2022 du "Système National des Données de Santé" portant sur des patients atteints d'un cancer de la prostate diagnostiqué entre 2016 et 2022, cette étude rétrospective examine l'association entre l'utilisation d'un ou plusieurs tests PSA dans les 5 ans qui précèdent la détection de la maladie et le stade tumoral au diagnostic
Introduction : Although organized prostate-specific antigen (PSA) screening for prostate cancer (PCa) is not recommended by most health authorities, including those in France, a high frequency of PSA testing is observed in many countries with significant association with incidence but not PCa-specific mortality.1-3 However, ecological analysis in the US showed that the incidence of metastatic PCa, which had decreased during the first phase of intense PSA screening, has slightly increased since 2012.4
Little is known of the association between opportunistic PSA screening and incidence of late-stage PCa at the individual level. This study aimed to describe the trend in PSA testing history according to PCa stage at diagnosis.
Methods : In this population-based nationwide cohort study based on the French National Health Data System (SNDS),5 data relating to health care reimbursements or registration for long-term disease over the period from 2011 to 2022 in men older than 40 years were analyzed to identify 2016 to 2022 incident PCa with no other cancer, using a 5-year washout period to exclude history of PCa. Since clinical stage was not reported in SNDS, the stage at diagnosis (localized, late-stage PCa [advanced or metastatic PCa]) was established based on the first-line treatment (eAppendix in Supplement 1). This report follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. Caisse Nationale d’Assurance Maladie has permanent access to the SNDS by decree No. 2016-1871, so no patient consent or specific ethics committee approval were needed for this study. Outpatient PSA test reimbursements were recorded for the 5 years preceding PCa diagnosis, excluding tests from the year immediately prior, as they may be directly related to the diagnosis. A history of 5 to 2 years of PSA testing was also recorded in men without PCa during the same period. The average annual percentage changes (AAPC) for the age-adjusted proportions of 2- to 5-year prediagnosis PSA testing (≥1 test) stratified on stage at diagnosis were derived from logistic regression models adjusted on 5-year age groups. t Tests were used, and a 2-sided P < .05 was considered statistically significant. Data were analyzed in SAS Enterprise Guide version 8.6 (SAS Institute) from July 2024 to July 2025.
Results : Among the 359 741 individuals with incident PCa diagnosis with a median (IQR) age of 70 (64-76) years, 240 605 (66.9%) had localized disease at diagnosis, 103 859 (28.9%) had late-stage disease, and 15 577 (5.3%) had disease that remained unstaged. Localized PCa was detected at a younger age than late-stage PCa (Figure).
JAMA Network Open , article en libre accès, 2025