3393 - Therapeutic Ratio of Seminal Vesicle Sparing vs. Inclusion in Prostate Stereotactic Body Radiation Therapy: A Propensity Score-Matched Competing Risks Analysis
Presenter(s)
V. T. Tran1, Z. Chen2, T. Komiyama2, T. Suzuki3, Z. Mochizuki3, M. Matsuda2, M. Saito2, R. Tozuka2, H. Nemoto2, and H. Onishi2; 1Radiotherapy Dept, Oncology Center, Military Hospital 103, Vietnam Military Medical University, 261 Phung Hung St, Ha Dong, Hanoi, Viet Nam, 2Department of Therapeutic Radiology, University of Yamanashi, Chuo, Japan, 3CyberKnife Center, Kasugai General Rehabilitation Hospital, 436 Kou, Kasugai, Fuefuki-city, Yamanashi, Japan
Purpose/Objective(s):
To evaluate the therapeutic ratio of seminal vesicle (SV) inclusion versus sparing in prostate stereotactic body radiation therapy (SBRT), characterizing the trade-off between biochemical control and toxicity without rectal spacers.Materials/Methods: We retrospectively analyzed 128 patients treated with robotic SBRT (36.25 Gy / 5 fractions). Patients were paired 1:1 into SV-included (base, proximal, or total) and SV-sparing cohorts via propensity score matching based on age, iPSA, Gleason score, T-stage and D’Amico risk. Biochemical failure (BCF) was estimated using Fine-Gray competing risks model to account for intercurrent mortality. Toxicity was graded per CTCAE version 5.0. A subgroup analysis of non-spacer patients isolated the intrinsic toxicity of SV irradiation.
Results:
At a median follow-up of 72.5 months in the matched cohort (n = 100), the SV-included arm exhibited higher 5-year BCF (5.2% vs 0.0% [two vs zero events]; P = .01). Gray’s test showed higher BCF with SV-inclusion overall (P = .011) and in intermediate-risk (P = .045) but not high-risk disease (P = .134). At a 118.4-month landmark, BCF CIF was 37.4% vs 19.0%. Adjusted Fine-Gray models showed higher BCF risk with SV-inclusion (sHR, 10.81; 95% CI, 1.75–66.64), though wide CIs reflected limited events and potential residual confounding. SV inclusion was associated with a significant shift in toxicity severity (P = .014), driven by a 3.5-fold increase in acute grade 2 toxicity (28.0% vs 8.0%; adjusted P = .052). In non-spacer patients, SV inclusion was associated with numerically higher rates of acute GI (24.0% vs 15.4%), and late grade = 2 toxicity (20.0% vs 15.3%), suggesting a volume-dependent rectal risk despite the limited sample size.Conclusion: Routine SV inclusion may represent a therapeutic trade-off in this cohort, significantly increasing moderate-grade toxicity without improving biochemical control. Observed failures likely reflect aggressive systemic biology and residual confounding rather than geographic misses. SV-sparing strategies should be considered to optimize the therapeutic ratio for favorable intermediate-risk patients, reserving SV irradiation for radiographically confirmed invasion.
Table. Oncologic Efficacy and Toxicity Outcomes in Propensity Score-Matched Cohort (n=100) *P for overall toxicity distribution. †Holm-adjusted. ‡Acute GI distribution comparison. §Late toxicity distribution comparison.| Outcome Measure | SV-Sparing (n=50) | SV-Included (n=50) | sHR (95% CI) | P |
| Biochemical Failure | ||||
| 5-Year Cumulative Incidence (CIF) | 0.0% | 5.2% | ||
| Fine-Gray Regression (Adjusted) | Reference | 10.81 (1.75–66.64) | .010 | |
| Acute Toxicity | ||||
| Overall Severity Distribution* | .014 | |||
| Grade 2 Toxicity Rate | 8.0% | 28.0% | .052† | |
| GI Toxicity (Any Grade, Non-Spacer) | 15.4% | 24.0% | .499‡ | |
| Late Toxicity | ||||
| Grade =2 Toxicity | 12.0% | 26.0% | .198§ |