G. Rajeev-Kumar1,2, Y. Che3, C. J. Stepaniak4, and S. Liauw2; 1University of Chicago, Chicago, IL, 2Department of Radiation and Cellular Oncology, University of Chicago, Chicago, IL, 3University of Chicago Department of Public Health Sciences, Chicago, IL, 4University of Chicago, Department of Radiation and Cellular Oncology, Chicago, IL
Purpose/Objective(s): Circadian rhythms regulate cellular mechanisms related to radiation response, including cell death and repair. We explored whether time of day was associated with disease outcome and recovery in a large cohort of men treated with radiation therapy (RT) for prostate cancer. Materials/
Methods: This retrospective study included 336 men treated with curative intent with external beam RT between 2010 and 2019. The median age was 69 years, 69% were black, median PSA was 11.3, and 40% had Gleason 8+. Median dose was 78 Gy/39 fractions; 27% received concurrent hormonal therapy. Average treatment time was calculated according to the average time of each patient’s daily treatments. For analysis, average time of treatment was separated into quartiles and by median, based on distribution. Disease outcomes including freedom from biochemical failure (FFBF) and distant metastasis (FFDM), RTOG and CTC gastrointestinal (GI) and genitourinary (GU) toxicity, and patient reported quality of life (QOL, EPIC-26) were analyzed on univariate and multivariable analysis (MVA). Median followup was 55 months. Results: The median average treatment time was 10:47AM (IQR, 9:12AM - 12:42PM, range 6:37AM-5:12PM). There were no imbalances in NCCN risk category or race across treatment times. There were no overall differences observed in FFBF or FFDM by treatment time. However, on subset analysis by race, white men treated in the first half of the day had better FFBF (89% vs 67%, p=0.0139) and FFDM (93% vs 72%, p=0.0268) than those treated in the second; this difference was not observed for black men. The crude rate of BF in white men by quartile of average treatment time was 5%, 10%, 28%, and 32%, respectively (p=0.04), whereas it was 14%, 17%, 18%, and 21% in black men (p=0.81). On MVA, treatment time was not associated with FFBF or FFDM when included in a model with NCCN risk category, or a model including NCCN risk category and race. On an MVA of white men only, treatment time was associated with FFBF (HR 2.8, p=0.06) in a model also including risk category (p=0.21); a model for FFDM indicated an association for risk category (p<0.01) but not treatment time (p=0.18). No significant differences were observed for grade 2-3+ toxicity and treatment time. A trend for worse grade 2+ GU toxicity (by median time, p=0.2) and grade 3+ GU toxicity (median, p=0.1) were observed for later treatment times. There were trends (p<0.2) observed in QOL for the urinary incontinence (quartile 4, p=0.16), urinary obstruction (quartile 3, p=0.09), and sexual function (median, p=0.16) for later treatment time, with stronger associations for white race than black race. Conclusion: In this exploratory analysis, treatment time of prostate RT may be associated with disease outcome depending on race, and later treatment time was weakly associated with GU toxicity and QOL. Further study is warranted to confirm these findings and understand the causality of these relationships.
