S193
Clinical - Biomarkers of clinical response
ESTRO 2026
3297 Prognostic Significance and Utility of the PCA3 Score in Post-Treatment Surveillance of Prostate Cancer Patients Treated with Radiotherapy Pantea Bayatfard 1 , Oytun Portakal Akcin 2 , Gokhan Ozyigit 1 1 Radiation Oncology, Hacettepe University, Ankara, Turkey. 2 Biochemistry, Hacettepe University, Ankara, Turkey Purpose/Objective: Prostate cancer antigen-3 (PCA3) mRNA is a well- established urinary biomarker for prostate cancer (PCa); however, its expression dynamic after radiotherapy (RT) remains unclear. Herein we prospectively evaluated PCA3 mRNA over-expression levels in high-risk PCa patients and investigated how these levels changed following stereotactic body radiotherapy (SBRT) in combination with androgen deprivation therapy (ADT). Material/Methods: Twenty-two patients, who had biopsy-proven high-risk PCa (PSA ≤ 20 ng/ml or Gleason score >7 or T2c-3b), were included in the study.Urine sample was collected according to instructions provided with the PROGENSA® assay. Cycle thresholds for PSAmRNA and PCA3mRNA were measured using two-step Real- Time Polymerase Chain Reaction to calculate PCA3 scores (PCA3-s) using the formula: [PCA3 mRNA/PSA mRNA] × 1000.A PCA3-s cut-off value of 35 was accounted for PCa positivity, per the PROGENSA® assay guideline. All patients received a total dose of 36.5 Gy SBRT in 5 consecutive fractions, along with 22.5 mg of leuprolide acetate up to 2 years. Patients with lymph node positivity also received whole pelvic RT. Follow-up PCA3 scores were evaluated at the 3rd month post-SBRT. All statistical analysis conducted via SPSS v23. Results: Patient and tumor characteristics are presented in Table-1. The median PCA3-s prior to SBRT was 10 (range:4-150), with only two patients (9%) exhibiting PCA3-s above the 35 cut-off. Eighteen patients (81%) began ADT concurrently with SBRT, while the remaining 4 (19%) had initiated prior to SBRT referral. Notably, PCA3 over-expression was not detected in the urine of patients who had received neoadjuvant ADT. Four patients had (19%) pelvic nodal involvement, 2 had received neoadjuvant ADT with suppressed PCA3- s, while the other 2 patients had PCA3-s of 13 and 15— both below the cut-off value. No significant association were found between pre-SBRT PCA3-s and pathological risk factors, including gleason score, perineural invasion, cribriform pattern, comedonecrosis, Suvmax value on 68Ga-PSMA and T stage (p>0.05). All patients successfully completed their planned SBRT, and follow-up urine samples were
Among 147 patients, 78 developed DM and 69 did not. In the independent test cohort, the model achieved an Area Under the Receiver Operating Characteristic Curve (AUC) of 0.932 (95% CI: 0.823–1.000) for DM prediction. For organ-specific metastasis prediction, the macro-AUC and micro-AUC were 0.7342 (95% CI: 0.619–0.839) and micro-AUC was 0.6374 (95% CI: 0.517–0.761), respectively. Risk stratification based on model-derived scores significantly distinguished patient outcomes (HR = 0.05, 95% CI: 0.01–0.40; log- rank P < 0.001), indicating strong prognostic discrimination. Conclusion: A multitask DL model using H&E-stained WSIs accurately predictDM risk and organ-specific metastasis in LANPC. This noninvasive tool may support personalized risk assessment and treatment planning. References: immune checkpoint-based signature to predict prognosis in nasopharyngeal carcinoma using computational pathology analysis. J Immunother Cancer 2019;7:298. https://doi.org/10.1186/s40425- 019-0752-4.[2] Lu MY, Williamson DFK, Chen TY, Chen RJ, Barbieri M, Mahmood F. Data-efficient and weakly supervised computational pathology on whole-
slide images. Nat Biomed Eng 2021;5:555–70. https://doi.org/10.1038/s41551-020-00682-
w.[3] Loeffler CML, Bando H, Sainath S, Muti HS, Jiang X, van Treeck M, et al. HIBRID: histology-based risk- stratification with deep learning and ctDNA in colorectal cancer. Nat Commun 2025;16:7561. https://doi.org/10.1038/s41467-025-62910-8. Keywords: Pathology, Nasopharyngeal carcinoma, Metastasis
Digital Poster
Made with FlippingBook - Share PDF online