Extracellular vesicle glycan profiling for pre-biopsy detection of prostate cancer

Résumé fourni par la source

Biomarkers are needed to detect clinically significant prostate cancer (csPCa; Gleason grade group [GG] ≥ 2), as serum prostate-specific antigen (PSA) testing often overdiagnoses indolent GG1 disease and prompts unnecessary biopsies. Extracellular vesicles (EVs) in post-digital rectal examination (DRE) urine carry glycan cargoes reflective of underlying tumor biology, which may improve discrimination between aggressive and indolent disease. This study evaluated whether profiling glycan signatures on post-DRE urine EVs improves pre-biopsy detection of (i) csPCa (GG ≥ 2 versus GG1/benign) and (ii) very clinically significant prostate cancer, vcsPCa (GG ≥ 3 versus GG2/GG1/benign). EVs were enriched from post-DRE urine of 135 men undergoing prostate biopsy (benign, GG1-GG5) and analyzed using Glycoflow, a nanoflow cytometry assay quantifying six glycan biomarkers together with prostate-specific membrane antigen (PSMA) and CD63 on EVs. Predictive models integrating Glycoflow with PSA or clinical parameters (CP: PSA, age, family history, and DRE-findings) were trained on 120 patients and evaluated using cross-validation and bias-corrected performance estimation. Performance (area under the receiver operating characteristic curve [AUC]) was compared to PSA- and CP-alone models. A small validation cohort ( n = 15) assessed signal reproducibility and computational pipeline stability. Models combining Glycoflow with PSA or CP improved discrimination over PSA- or CP-alone models for csPCa (AUC 0.80 vs. 0.72; 0.80 vs. 0.71) and vcsPCa (0.82 vs. 0.72; 0.81 vs. 0.73). Bias-corrected out-of-sample AUCs were 0.72/0.72 for csPCa and 0.73/0.72 for vcsPCa, comparable to clinical models. Patient-level analyses demonstrated improved classification of csPCa and vcsPCa among individuals misclassified by clinical-alone models, with uncorrelated prediction scores between Glycoflow-alone and clinical-alone models, indicating complementary diagnostic value of EV glycan profiling. Glycoflow, a urine-based EV glycan assay, provides diagnostic information complementary to PSA and CP, improving pre-biopsy detection of csPCa and vcsPCa. With further validation in larger cohorts, Glycoflow may support biopsy decision-making and reduce overdiagnosis.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.