A Comparison of Radiation Toxicity Outcomes in Patients With and Without Autoimmune Diagnoses on the REQUITE Study

Abstract

The REQUITE study, a multicenter prospective observational study which includes standardized radiotherapy data, non-genetic risk factor data and biobank data including common genetic variants, may provide important information about differences in radiation toxicity and predictive biomarkers amongst patients with autoimmune disease. The goal of this analysis was to determine the association of late toxicity outcomes with autoimmune disease phenotypes in the REQUITE study cohort, for use in determining autoimmune-related genomic predictors of such toxicities.We compared late radiation toxicity in prostate cancer patients between those with and without an autoimmune phenotype. Autoimmune diagnoses included systemic lupus erythematosus, collagen vascular disease, rheumatoid arthritis, inflammatory bowel disease, and other autoimmune disorders (e.g., multiple sclerosis). Probability of symptom worsening from baseline to 12-24 months was determined for proctitis, rectal bleeding, hematuria, urinary obstruction, incontinence, frequency, retention, erectile dysfunction, ejaculation disorder, and orgasmic dysfunction. Adjusted estimates for the association between autoimmune phenotype and late toxicity were modeled with multivariable logistic regression, adjusting for multiple comparisons (false discovery rate, t = 0.10). Models were adjusted for other toxicity and progressive disease risk factors, including comorbidities (heart disease, hypertension, diabetes, hemorrhoids), smoking, treatment modality, prior TURP/radical prostatectomy, hormone therapy, and age. Patients with maximal scores (3/4, 3/3, or 2/2) at baseline were excluded.Of 1441 prostate cancer patients treated with radiotherapy, 76 (5.3%) had an autoimmune phenotype reported, and were more likely to also have hypertension. Among patients with baseline and longitudinal scores, the most common toxicities included: sexual toxicity (erectile dysfunction, N = 475/945 (50%); ejaculation disorder, N = 373/787 (47%); orgasmic dysfunction, N = 324/760 (43%)), urinary frequency (N = 396/1311 (30%)) and incontinence (N = 282/1313 (21%)), and proctitis (N = 224/1331 (17%)) and rectal bleeding (N = 218/1329 (16%)). Adjusted estimates demonstrate an independent association of autoimmune disease with proctitis (OR 2.3, 95% CI: 1.30-3.88; P = 0.003, q = 0.02) and rectal bleeding (OR 2.0, CI: 1.11-3.48; P = 0.02, q = 007). For rectal bleeding, current smoking (OR .45, CI: .25-.79; P = 0.01, q = 0.05) and hypertension (OR 0.66, CI: .48-.89; P = 0.01, q = 0.05) were protective.Certain gastrointestinal radiation treatment-related toxicities are associated with autoimmune phenotype in patients treated for prostate cancer in this large, prospective cohort. Our next step includes performing an analysis of candidate genes in addition to a polygenic risk score predicting for toxicity.