Gene signature predictor of dose-response to prostate radiation: Validation of PORTOS in phase III trials.

• DOI: 10.1200/jco.2025.43.5_suppl.308
• Publication Date: 2025-02-18T14:38:27Z
• Authors: Shuang Zhao; Hyunnam Monica Ryu; James A. Proudfoot; Elai Davicioni; Jeff M. Michalski; Daniel Eidelberg ...; Stefanie Hayoz; Jeffry Simko; Howard M. Sandler; Alan Pollack; Matthew Parliament; Ian S. Dayes; Rohann Jonathan M...; Theodore Karrison; William Adrian Hall; Daniel M. Aebersold; Felix Y Feng; Pirus Ghadjar; Phuoc T. Tran; Alan Dal Pra
• Keywords: Gene signature; Signature (topology)

308 Background: NRG/RTOG 0126 and SAKK 09/10 were phase III randomized trials examining whether higher dose resulted in better response/outcomes PCa patients following definitive post-operative RT, respectively. RTOG showed a benefit for RT escalation (DE) from 70.2Gy to 79.2Gy though did not show 64Gy 70Gy. We hypothesized that previously developed 24-gene prostate cancer gene expression score (PORTOS) could distinguish who benefited DE both trials. Methods: PORTOS scores calculated on biopsy samples prostatectomy as published. Since the original cutoffs post-op setting, we utilized tertile groups 0126, whereas published used 09/10. The primary objective was evaluate predictive biomarker of biochemical failure (BF) via Phoenix criteria (N=215) clinical progression-free survival (CFPS) (n=226). In addition, also investigated molecular correlates large real-world datasets 31,107 42,407 radical samples. Results: with lower scores, there no difference BF (sHR 1.14 [0.54-2.40], P=0.73). However, middle range, significant (middle PORTOS: sHR 0.45 [0.22-0.90], P=0.02; 0.30 [0.12-0.75], P=0.009). An interaction test indicated between (P=0.048). Similarly trial, only group (CPFS HR 0.19 [0.05-0.70]; P=0.01), biomarker-treatment vs. treatment arm (P=0.003). Interestingly, consistently associated clinicopathologic variables either trial or datasets. Biologically, datasets, modestly hypoxia signatures consistent its role radio-resistance, strongly immune subtypes. Conclusions: two trials, have validated can identify well those do localized provides first evidence any be able predict response. personalize radiation clinically, allow selection most likely while sparing others potential increased risk toxicity.