Jonathan Olivier

Le dépistage du cancer de la prostate, réalisé par dosage sérique du PSA et toucher rectal, permet une diminution de la mortalité spécifique de 21% à 13 ans. Le diagnostic de lésions de volume et de grade significatifs en cas de test de dépistage suspects a été amélioré par l’IRM prostatique qui depuis la fin des années 2000 a permis de mieux détecter les lésions malignes notamment de faible volume ou d’accès difficile pour les biopsier.Du fait d’une valeur prédictive négative (VPN) élevée entre 85% et 95% pour les cancers significatifs, l’IRM prostatique a été utilisée à la fin des années 2010 pour diminuer le sur-diagnostic de cancers cliniquement non significatifs de faibles volume et grade en évitant de biopsier les hommes en cas d’IRM non suspecte. Cette stratégie d’utilisation de l’IRM comme un test de tri entre le dépistage et la pratique des biopsies a été évaluée en association à des critères cliniques (facteurs de risque familiaux) et des marqueurs (densité du PSA) pour augmenter la VPN de l’IRM. Cette évaluation correspond au premier objectif de la thèse.Cependant au début des années 2010, des biopsies en cartographie étaient indiquées en cas d’IRM non suspecte et entrainaient la détection d’un cancer cliniquement non significatif de faible volume et grade. Dans ces cas, la stratégie de surveillance de ces cancers a été proposée pour éviter leur sur-traitement. Cette situation concernait environ 20% des cancers détectés. La surveillance permet de différer le traitement jusqu’à progression à un stade plus élevé. L’évaluation des critères de sélection des cas à surveiller et de leur suivi correspond au deuxième objectif de la thèse.Dans la première partie nous avons évalué les facteurs cliniques, biologiques et familiaux pouvant améliorer la VPN de l’IRM. A l’aide d’une revue systématique de la littérature avec méta-analyse, nous avons mis en évidence que la densité du PSA (PSA/volume prostatique) était le facteur prédictif améliorant la VPN de l’IRM le plus validé pour le diagnostic de cancer significatif.Nous avons aussi validé rétrospectivement le risque de ne pas diagnostiquer un cancer significatif en d’IRM non suspecte dans cohorte de 503 patients qui avaient été tous été biopsiés initialement et suivis sur une durée d’inclusion de 10 ans. L’analyse a montré que ce risque était de 9% au moment du diagnostic et de 4% supplémentaire lors du suivi. Ce risque initial diminuait de 9% à 2.4% en intégrant dans l’analyse au moment du diagnostic la densité de PSA, le toucher rectal ou les antécédents familiaux pour décider ou non d’une biopsie.Dans la deuxième partie, nous avons étudié, comme critère de sélection à une surveillance active l’intérêt de l’IRM en plus des critères de cancers cliniquement non significatifs. Une étude rétrospective multicentrique, comparant 1035 patients ayant une IRM non suspecte et 1084 une IRM suspecte à l’inclusion a montré que le risque de progression tumoral et de sortie de surveillance active était diminué en cas d’IRM non suspecte à l’inclusion.Nous avons aussi évalué les examens de suivi des patients en surveillance active avec IRM non suspecte et plus particulièrement la réalisation de biopsies de confirmation à un an. Une étude prospective incluant 2 cohortes successives de 78 et 71 patients avec et sans biopsies de confirmation à un an a comparé les taux de diagnostic de progression tumorale à 2 ans et le rôle de la cinétique du PSA pour cette prédire cette progression. La conclusion était qu’il n’y avait pas de différence de diagnostic de progression entre les 2 groupes et que la cinétique du PSA suspecte à elle seule pouvait faire indiquer une biopsie. Cette étude nous a permis de proposer de ne plus réaliser ces biopsies de confirmation en cas d’IRM non suspecte au diagnostic

Objectives : Partial nephrectomy can lead to complications such as urinary fistulas, with an incidence ranging from 1 to 17%. Currently, there are no formal guidelines for managing these urinary fistulas. The aim of this study is to assess the time to fistula resolution according to different treatment options. Materials and Methods: The U-Leak study is a retrospective, observational, and multicenter French cohort based on analyses from the UroCCR database. The inclusion period extends from 2009 to 2024, included patients who underwent partial nephrectomy with post-operative urinary fistula. We identified two groups: patients managed interventionally and patients managed without any drainage. The primary objective was to evaluate the time to resolution of urinary fistulas after partial nephrectomy. Results: We included 127 patients, predominantly male (71.7%), with a mean age of 59.4 years. The majority of partial nephrectomies (55.2%) were performed via transperitoneal robotic surgery. Interventional management was applied to 82.4% of patients, 39.5% received a primary double J stent, 17.7% of patients were treated with a bladder catheter combined with a double J and 25.2% received percutaneous drain, bladder catheter or association of two or three treatment. Twenty- five patients required a second intervention. The median time to fistula resolution was 43 days [13.5–81.5] after the first treatment and 48 days [21.5–67.5] after the second. Overall, 79.8% of fistulas resolved after the first intervention. The median time to resolution was 44 days [29.0–69.5] without interventional treatment and 50 days [23.5–87.0] with treatment, with no statistically significant difference between the two groups (p = 0.432). Conclusion: We did not observe a significant difference in the time to resolution of urinary fistulas after partial nephrectomy between patients managed interventionally and those managed without any drainage.

Objectives: To determine if quantifying Gleason pattern 4 (GP4) is superior to Grade Group (GG) in predicting any and distant metastatic recurrence after radical prostatectomy (RP) for prostate cancer, and compare various GP4 quantification methods for predicting metastatic recurrence using preoperative targeted biopsy and magnetic resonance imaging (MRI) data. Patients and Methods: We conducted a retrospective study of patients who underwent MRI‐guided biopsies and RP from 2009 to 2018. Patients with GG 2–4 without GP5 disease on biopsy and/or RP specimen were included. The predictors compared were biopsy GG, percentage of GP4 in biopsy cores, millimetres of GP4 in biopsy cores, and volume of GP4 based on MRI lesion volume. These methods were also compared to the Cancer of the Prostate Risk Assessment (CAPRA), International Staging Collaboration for Cancer of the Prostate (STAR‐CAP), European Association of Urology (EAU), and National Comprehensive Cancer Network (NCCN) risk classifications. The C‐index for each model was calculated to evaluate discrimination performance. Results: A total of 446 patients were analysed, with a median follow‐up of 6.9 years for patients without an event; 46 patients developed any metastasis. For any metastatic recurrence based on biopsy findings, the CAPRA score (C‐Index = 0.72) showed the highest discrimination among risk scores, surpassing biopsy GG (C‐Index = 0.70), but was outperformed by percentage GP4 (C‐Index = 0.74), millimetres GP4 (C‐Index = 0.77), and volume of GP4 (C‐Index = 0.80). Conclusion: For patients with GG 2–4 prostate cancer containing GP4, preoperative GP4 volume estimation using MRI and targeted biopsy outperforms Gleason scoring classification and standard risk scores in predicting any and distant metastatic recurrence. Further research is warranted on the best methods to quantify GP4 before incorporation in treatment decision‐making.

Background/Objectives: Focal therapy (FT) for prostate cancer (PCa) is an alternative to radical treatments that aims to balance cancer control and quality of life preservation in well-selected patients. Understanding its general principles and outcomes is key for its widespread adoption and proper implementation. Methods: The International Consultation on Urological Diseases nominated a committee to review the literature on FT for PCa. A comprehensive PubMed search was conducted to identify articles focused on the different aspects of FT, including patient selection, imaging techniques, treatment modalities, cancer control and safety outcomes, integration with other approaches and future perspectives. Results: FT for PCa was introduced in the 1990s with cryotherapy and high-intensity focused ultrasound (HIFU) as pioneering modalities. Though initially guided by transrectal ultrasound (TRUS) and large biopsy templates, FT implementation expanded significantly with the advent of multiparametric magnetic resonance imaging (MRI) and the validation of the index lesion concept. Appropriate patient selection is key for FT and relies on prostate-specific antigen (PSA) metrics, MRI findings and targeted biopsy information. Multiple energy sources are now available, each with specific technical characteristics. Cancer control rates vary by energy modality, tumor characteristics, and institutional experience, demonstrating comparable outcomes to radical treatments in well-selected patients. The safety profile is excellent, with high rates of urinary continence and sexual function preservation. Post-treatment surveillance integrates PSA measurements, imaging, and histological assessment. Future directions for further FT adoption include the availability of long-term data, protocol standardization and technological improvements to enhance patient selection and treatment planning and delivery. Conclusions: FT is a valuable therapeutic option for selected patients with localized PCa, demonstrating promising oncological outcomes and better functional preservation compared to radical treatments. Understanding its principles and technical aspects is essential for offering comprehensive PCa care.

Importance: Magnetic Resonance Imaging (MRI) coupled with Prostate Imaging-Reporting and Data System (PI-RADS) provides a standardized scoring system for assessing the clinical significance of prostate cancer (PCa). The association between PI-RADS scores and key clinical end-points remains underexplored due to limited follow-up data. Objective: To evaluate the association of PI-RADS with prostate cancer-specific mortality (PCSM), overall survival (OS), metastasis-free survival (MFS), and biochemical recurrence (BCR) across three cohorts. Design, Setting, Participants: A retrospective registry cohort, Helsinki University Hospital (HUS) for men with clinical suspicion of PCa, included 4,674 men who underwent diagnostic MRI during the PI-RADS version-2 era (2015-2019, median follow-up [mFU] 5.0 years). Validation cohorts were: Tampere University Hospital (n=1,474, 2016-2021, mFU 2.2 years) and Lille University Hospital (N=301, 2016-2024, mFU 6.2 years). Exposures: Diagnostic MRI imaging was scored using PI-RADS v2. Main outcomes and measures: Primary outcomes were PCSM, OS, MFS, and BCR. Secondary measures included clinical confounders, such as age, Grade Group (GG), and Charlson’s Comorbidity Index (CCI). Results: 84 of 4,674 HUS-cohort patients (1.7%) died of PCa, with 79 patients (94%) presenting with PI-RADS score 5. Multivariable Cox regression adjusted for clinical confounders found PI-RADS score 5 was significantly associated with PCSM (Hazard Ratio (HR) 18.4, 95% CI [6.62-51.1], p<0.001), along with biopsy GG 5 (HR 5.45 [1.82-16.3], p=0.002), CCI (HR 1.53 [1.42-1.7], p<0.001), and log2-PSA (HR 1.28 [1.11-1.5], p<0.001). Kaplan-Meier curves demonstrated negligible PCSM after negative MRI (nMRI; PI-RADS score 0-2) (log-rank p<0.001). PI-RADS score 5 was associated with OS in the Helsinki and Tampere cohorts (p<0.001) and MFS and BCR in the Lille cohort (p<0.001). Conclusion and relevance: PI-RADS score 5 is strongly associated with an increased risk of PCSM and other clinically relevant outcomes. A non-suspicious MRI associates with a negligible risk of adverse outcomes. These findings support the use of MRI for risk stratification in PCa.