Prognostic and Predictive Performance of a 24-Gene Post-Operative Radiation Therapy Outcomes Score (PORTOS) in a Phase 3 Randomized Trial of Dose-Intensified Salvage Radiotherapy after Radical Prostatectomy (SAKK 09/10)

Abstract

<h3>Purpose/Objective(s)</h3> The SAKK 09/10 phase 3 trial showed that dose-intensified salvage radiotherapy (SRT) was not superior to conventional dose SRT for biochemically recurrent prostate cancer (PCa) after radical prostatectomy (RP). We previously reported on the validation of the 22-gene genomic classifier as prognostic in this trial but did not observe an interaction with RT dose. Herein, we examine the PORTOS classifier, an expression signature of 24 DNA damage repair and immune pathway genes, previously shown to predict benefit from postoperative radiotherapy, to test its predictive ability to SRT dose. <h3>Materials/Methods</h3> The SAKK 09/10 trial (NCT01272050) was a multicenter, randomized phase 3 trial that included PCa patients with biochemical progression after RP (PSA > 0.1 to 2 ng/mL at randomization). Patients were randomized to 64Gy vs 70Gy to the prostate bed. No patients received concurrent androgen suppression or pelvic nodal radiotherapy. After central review, the highest-grade tumors were profiled using the Decipher whole-transcriptome assay (Veracyte, San Diego, CA) to generate PORTOS. The independent prognostic ability and interaction effects were examined for the biochemical progression, clinical progression-free survival (CPFS), receipt of salvage hormone therapy, and metastasis-free survival endpoints with Cox multivariable and interaction analyses (MVA). Categorical analyses used the published threshold for PORTOS radiotherapy response (high vs low). <h3>Results</h3> Of the 226 trial patients with transcriptome data, 16% had high PORTOS radiation response compared to 84% with lower PORTOS. Patients with high PORTOS were well balanced between the randomization arms. PORTOS was not prognostic for the endpoints in MVA with the trial stratification factors. A high PORTOS, however, was predictive for improved outcomes in patients receiving dose-intensified SRT. Patients with high PORTOS that received 70Gy dose had 5-year CPFS of 94% compared to 49% with the 64Gy dose (log-rank p=0.006). In interaction analysis, PORTOS had a statistically significant interaction with RT dose for multiple endpoints: freedom from biochemical progression (interaction p=0.04), clinical progression-free survival (interaction p=0.003) and metastasis-free survival (interaction p=0.037). <h3>Conclusion</h3> In a phase 3 trial testing SRT dose intensification after RP, patients with high PORTOS had significantly better outcomes with dose-intensified treatment, suggesting that higher doses should be considered in this subgroup. Further studies are warranted to validate PORTOS as the first predictive biomarker to help personalize dose prescription in the postoperative setting.