OP12 Does early initiation of biologics change the natural history of IBD? a nationwide study from the epi-IIRN

Abstract Background Studies have reported conflicting findings regarding the effect of timing of biologics initiation on disease course, often affected by selection and referral biases. In this nationwide cohort study we aimed to explore whether early initiation of biologics is associated with better disease outcomes in CD and UC. Methods Data of patients diagnosed with IBD in the epi-IIRN cohort from 2005 to 2020 were retrieved from 4 Israeli Health-Maintenance-Organizations covering 98% of the population.The primary analysis was time to steroid-dependency and IBD-related surgery. To control for confounding by indication in which the severity of disease could account both for the timing of biologics and risk of disease outcomes,we compared treatment strategies using the cloning,censoring and weighting method to emulate a target trial. Inverse probability of treatment weights were fitted per patient to adjust by weighting for time varying confounding and selection bias.Survival probabilities(POS) were estimated and compared at 10 years post diagnosis for each strategy through weighted non-parametric Kaplan-Meier estimators. The 95%CIs for survival probabilities were calculated using non-parametric bootstrap with 500 replicates. Results Of the 34,375 patients diagnosed with IBD, 7,452(39%) CD patients and 2,235(15%) UC patients received biologics and included in the analysis. In total,1,139 CD and 206 UC patients received biologics during the first three months after diagnosis,2,108 CD and 450 UC during the 3-12 month period,1,161 CD and 396 UC during 1-2 year period,and 719 CD and 271 UC during 2-3 years post diagnosis. Patients who initiated biologics after 3 years post diagnosis were excluded as they were not considered early biologics initiation.In CD,at 10 years post diagnosis there was a significant but modest decrease in IBD-related surgeries when initiating biologics prior to 3 months compared to the other periods (POS:0-3 months 0.75[CI 0.71- 0.79] vs 2-3 year 0.7[CI 0.66-0.73] ;p<0.05).The same observation with a larger effect, was apparent for steroid-dependency (POS:0-3 months 0.78[CI 0.75-0.82] vs 2-3 year 0.69[CI 0.66-0.71]; p<0.001)(Fig 1). In UC, no significant differences in colectomy rate at 10 years was observed between the treatment periods (POS:0-3 months 0.92[CI 0.87-0.97] vs. 2-3 years 0.94[CI 0.91-0.98];p=0.42). On the other hand, a significant decrease was apparent for steroid-dependency (POS:0-3 months 0.78[CI 0.75-0.82] vs 2-3 year 0.69[CI 0.66-0.71];p<0.001)(Fig 2). Conclusion Our study showed a slight advantage to earlier initiation of biologics,as reflected by longer time to IBD-related surgery and steroid-dependency in CD.In UC, the effect of early biologics was minor and was only apparent for steroid dependency in UC.