P769 Impact of discontinuation of anti-TNF therapy: A long term follow-up data of risks and predictors relapse in a large Indian cohort

Abstract

Abstract Background Discontinuation of anti-TNF therapy in IBD is an important consideration in TB endemic low resource settings of the developing world. There are no established guidelines on withdrawal of biologics. A proper exit strategy on when and how biologics can be stopped is therefore warranted. We aimed to identify risk factors and rates of relapse post withdrawal in a large real-life cohort of patients from India. Methods Prospectively collected data of consecutive patients treated with anti-TNFs from the IBD registry of a large tertiary care centre was analyzed. Demographics,baseline disease characteristics,reasons for discontinuation, response to anti-TNF,and relapse rate after withdrawal were analysed. Time to event analysis was done to evaluate relapse rates over time. Cox proportional hazard analysis and non-parametric tests were performed to identify risk factors of relapse. Results Of 8300 patients, only 725(8.7%),(Median age-36y,IQR:26-47,61.2% male,61.2%CD, 32%UC, IBD-U 1.8%) received anti-TNFs. 445 (61.3%) discontinued with median follow up of 54 m(IQR:31-81). Reasons for discontinuation were primary non-response(64,14.3%);secondary loss of response (44,9.8%);financial constraints(27,6.1%);TB reactivation (18,4%), side effects(17,3.2%); infusion reactions(20,4.4%);non-compliance(4,0.9%), and COVID lockdown (6,1.3%). Overall, 232/445 (52%) were in clinical and endoscopic remission while stopping anti-TNFs with median follow up of 41 months(IQR:24.5-67). Relapse rates at 1,2,3,4, and 5 years after discontinuation were 16%,5.6%,5.2%,2.6% and 1.3% respectively. Overall,161/232(69%) patients remained in remission(Fig1A). Patients who stopped after clinical and endoscopic remission had higher likelihood of maintaining remission after discontinuation of biologics (Log-rank test,p=0.001). On multivariate analysis, fistulizing disease(p=0.03), prior history of bowel surgery (p=0.019), immunomodulator usage (p=0.001) and shorter duration of therapy (p=0.017) were associated with significantly higher rate of relapse(Table1). Cumulative time to relapse was significantly longer in those with or without endoscopic remission (median 17.3 vs 7 months;p=0.04)(Fig1B). On the other hand, shorter disease duration (p=0.001) and absence of extra-intestinal manifestations (p=0.009) were associated with lower risks of relapse. Conclusion 60% of patients discontinued anti-TNF therapy in this real world cohort. However, long-term cumulative relapse rates were low(30%), particularly in non-fistulizing disease, with no history of bowel surgery or extra intestinal manifestations and in endoscopic remission. A definite exit strategy appears feasible and would be a practical and affordable management approach.