3197 - Radiation Therapy for Prostate Cancer: Baseline Platelet Count and Anti-Platelet and Anticoagulant Use Are Associated with Disease Outcomes in a Cohort of 1000 Men

M. Ji¹, Y. Che², S. Gutiontov³, A. A. Solanki⁴, and S. Liauw⁵; ¹University of Chicago, Chicago, IL, ²University of Chicago Department of Public Health Sciences, Chicago, IL, ³Aurora Health, Milwaukee, WI, ⁴Department of Radiation Oncology, Stritch School of Medicine, Cardinal Bernardin Cancer Center, Loyola University Chicago, Maywood, IL, ⁵Department of Radiation and Cellular Oncology, University of Chicago, Chicago, IL

Purpose/Objective(s): Prior studies have demonstrated antiplatelet/anticoagulant (AP/AC) medications to be associated with improved biochemical outcomes and cause-specific survival in men with prostate cancer treated with radiation therapy (RT). We explored this association and the proposed link to platelet count in a large, single-institution database. Materials/

Methods: 1000 men with localized prostate cancer received definitive RT from 1989-2018, including 860 RT alone, 83 brachytherapy alone, and 57 in combination. Median age was 68 y. NCCN risk category was 28/41/31% for low/intermediate/high risk, respectively. Median dose was 76 Gy/38 fx, and 52% received androgen deprivation therapy (ADT). Medication use at consultation including AP (n=410, e.g. aspirin, clopidogrel) and AC (n=65, e.g. warfarin) medications and platelet count within 6 mo of RT start were retrospectively recorded. The effects of AC/AP medications and platelet counts were analyzed with respect to freedom from biochemical failure (FFBF, nadir+2), distant metastasis (FFDM), and cause-specific survival (CSS) on univariate (UVA) and multivariate analysis (MVA).
Results: Median follow-up was 78 mo. 7-y FFBF was 88%, 82%, 71%, and 69% for NCCN low, favorable intermediate, unfavorable intermediate, and high-risk disease (p<0.01). On UVA, AC/AP use was associated with increased FFBF (7-y, 82% vs 75%, p=0.02), FFDM (p=0.02), and CSS (p<0.01). Relationships were also observed between the lowest quartile (<181) platelet count and FFBF (5-y 81% vs 70%,p<0.01) but not FFDM (p=0.07), or CSS (p=0.84). MVA models including AP/AC use, platelet count < 181, PSA (continuous), Gleason score (6, 7, or 8+), T2b stage, and ADT, showed an association with disease outcomes for low platelet count (FFBF HR 1.7, p<0.01, FFDM HR 1.61, p=0.04) and AP/AC use (CSS HR 0.40, p=0.02). The potential interaction between AP/AC use and low platelets was further explored by analyzing FFBF, FFDM, and CSS in separate subgroups. The association between low platelets with inferior FFBF was observed in AP/AC non-users (p<0.01) but not in AP/AC users (p=0.54), and in men not receiving ADT (p<0.01) rather than in men on ADT (p=0.31). MVA in AP/AC non-users including covariates as above demonstrated a lower FFBF for low platelets (HR: 2.09, p=0.0001).
Conclusion: Low platelet count and AP/AC use demonstrated associations with FFBF, FFDM, and CSS on either UVA or MVA in prostate cancer patients receiving RT. The negative prognostic association of low platelet count was strongest in AP/AC non-users. The potential interaction between AP/AC and low platelet count should be further verified and investigated for possible mechanisms, as it may present an opportunity to offer therapeutic benefits in select men.