P1052Perianal Lesion Index as a Predictor of Advanced Therapy Maintenance in Perianal Crohn’s Disease

Background Advanced therapies (AT) have transformed perianal Crohn’s disease (pCD) management, yet predicting treatment durability remains challenging. The Perianal Lesion Index (PLI)—a novel endoscopy-based tool adapted from SES-CD for the distal rectum and anal canal, incorporating drainage and seton presence (score 0-16)—may stratify patients by treatment response. We evaluated whether baseline PLI predicts AT maintenance rates and characterised discontinuation patterns. Methods We retrospectively analysed 42 patients with active pCD initiating AT between 2011-2024 (49 episodes). PLI assessed ulcer size, ulcerated surface, affected segment, stricture, drainage, and seton requirement. Patients were stratified as PLI<10 (n = 38) versus PLI≥10 (n = 11). Primary endpoint was AT maintenance at 1-2 years by Kaplan-Meier analysis. AT included infliximab (IFX, n = 17), adalimumab (ADA, n = 5), ustekinumab (UST, n = 17), vedolizumab (VDZ, n = 7), upadacitinib (n = 2), risankizumab (n = 1). Secondary outcomes included discontinuation rates and reasons by PLI and drug class. Results Median age at AT initiation was 30 years (14-59); 84% male. Mean baseline PLI was 7.3±2.6 (77.6% PLI<10, 22.4% PLI≥10). Prior biologic exposure: 61% naïve, 20% one agent, 18% ≥two agents. Median follow-up 3.7 years.One-year AT maintenance rates were significantly higher in PLI<10 versus PLI≥10 groups (97.4% vs 63.6%, p = 0.013; Figure 1). Two-year rates were 86.8% and 45.5%, respectively. Overall discontinuation was 43% (PLI≥10: 73% vs PLI<10: 34%).TNFα inhibitors (n = 22, mean PLI 6.5) showed 32% discontinuation, primarily from loss of response (71%; Table 1). Non-TNFα biologics (n = 24, mean PLI 7.9) had 54% discontinuation, mainly primary non-response (62%; Table 1). Drug-specific 1-year/2-year maintenance: IFX 100%/65%, ADA 100%/60%, UST 94%/77%, VDZ 57%/43%.TNFα inhibitors were used in less severe disease (mean PLI 6.5), whilst non-TNFα agents were reserved for refractory cases (mean PLI 7.9). PLI≥10 predicted treatment failure regardless of agent class. Conclusion Baseline PLI significantly predicts AT maintenance in pCD, with PLI≥10 independently associated with treatment failure (p = 0.013). PLI-based assessment may help guide treatment selection and timing. Early intervention before disease progression may improve outcomes, though further validation is warranted. Regular PLI assessment warrants consideration for incorporation into clinical practice. References: 1. Yokoo T, et al. J Crohns Colitis. 2024;18(Suppl 1):i1799-i1801. 2. Schwartz DA, et al. Inflamm Bowel Dis. 2021;27(7):1065-1082. 3. Gecse KB, et al. J Crohns Colitis. 2021;15(9):1416-1431. 4. Panés J, et al. Lancet. 2016;387(10032):1981-1990. Conflict of interest: Dr. Yokoo, Takashi: Personal Fees: Honoraria for lectures from AbbVie, Bristol Myers Squibb, Chugai, EA Pharma, Gilead Sciences, Johnson & Johnson, Kaken, Kyorin, Mitsubishi Tanabe Pharma, Mochida, and Takeda (past 36 months). Yoshikawa, Shusaku: No conflict of interest Masuda, Tsutomu: No conflict of interest Terauchi, Seiji: No conflict of interest Uchida, Hideki: No conflict of interest Nakao, Takeshi: No conflict of interest Inagaki, Mizumi: No conflict of interest Tani, Takafumi: No conflict of interest Okamoto, Kouhei: No conflict of interest Shibata, Yusuke: No conflict of interest Inatsugi, Naoki: No conflict of interest