OBJECTIVES: To assess the predictive value of patient-reported stool frequency and rectal bleeding for mucosal healing based on endoscopic evaluation in patients with active UC. METHODS: Patients with active UC defined by a Mayo score of 6-12 including an endoscopy subscore of ≥2 were randomized to receive placebo or golimumab (GLM) 200mg/100mg or 400mg/200mg at weeks 0 and 2, respectively. Patient-reported stool frequency and rectal bleeding from weeks 0 through week 6 were collected from the most recent consecutive 3-day period within the 2 weeks prior to the visit. Responder was defined as normalized stool frequency or no rectal bleeding. Mucosal healing is defined as an endoscopy subscore of the Mayo score of 0 or 1. Odds Ratio (OR), positive predictive value (PPV) and likelihood Ratio Positive (LR+) were estimated for predicting mucosal healing using normalized stool frequency or no rectal bleeding. Receiver Operating Characteristic (ROC) curve was developed based on different cut-off of combined stool frequency and rectal bleeding scores (0-6). RESULTS: At week 0, 99% of patients reported abnormal stool frequency and 86.9% reported rectal bleeding.  Among patients who reported no rectal bleeding at week 6, 66.5% achieved mucosal healing compared to 16.0% of patients with rectal bleeding (OR=9.9, p<0.001, PPV=0.65, LR=3.0). Similarly, among patients with normalized stool frequency at week 6, 77.1% achieved mucosal healing vs. 32.8% % of patients with abnormal stool frequency at week 6 (OR=6.9, p<0.001, PPV=0.77, LR=5.3). On ROC curve, a cut-off of ≤2 on a combined stool frequency and rectal bleeding score resulted in a sensitivity of 85% and specificity of 70.6% for predicting mucosal healing at Week 6. CONCLUSIONS: Normalized stool frequency and no rectal bleeding are predictors for mucosal healing in patients with UC.