Introduction

Treating perianal Crohn’s disease (pCD) is challenging, with at least 60% of the patients not responding or relapsing after a year of 1^st line biological therapy. Data on 2^nd and 3^rd line biological treatment effectiveness in pCD is scarce.

Aims & Methods

We conducted a multicenter retrospective cohort study to describe the effectiveness of 2^nd and 3^rd line biological therapy in pCD. Data was collected from 22 hospitals from 13 countries worldwide. Patients with pCD that were treated with >1 biologic were included and activity of disease was defined by the clinician at the point of initiation of each treatment line. Primary outcome for patients with active pCD was defined as perianal response under 2nd or 3rd line treatment judged by the clinician. Primary outcome for patients with inactive pCD was defined as a perianal recurrence under 2nd or 3rd line treatment. The main secondary outcome was fistula healing on pelvic MRI or transrectal ultrasonography (TRUS). Study cohort was divided into four sub-groups according to treatment line (2^nd/3^rd) and disease status (active/inactive) and by treatment (2^nd/3^rd line): anti-TNF, vedolizumab (VDZ) and ustekinumab (UST). A propensity score matching was based on the following variables: patient age, gender, pCD duration, perianal fistula, surgical fistula repair, seton placement and performance of examination under anesthesia (EUA) before 1st line treatment.

Results

A total of 577 pCD patients who received >1 biological therapy were found; 267/577 patients (46%) received 3^rd line treatment. 416/577 patients (72%) and 199/267 patients (74%) had active pCD at the point of 2^nd and 3^rd line initiation, respectively. Out of the initial cohort, 385/577 (66%) and 172/199 (86%) patients were included in the propensity-match score in 2^nd and 3^rd line, respectively.
In 2^nd line, 101/416 patients (24%) with active disease under UST treatment were compared to 229/416 (55%) and 55/416 (13%) propensity-matched cases of anti-TNF and VDZ, respectively. UST-arm showed significantly higher rates of clinical response (aOR 7.2, 95%CI 3.23-16.06, p<0.001) and remission (aOR 2.2, 95%CI 1.1-4.6, p=0.04). In 3^rd line, 101/199 (24%) patients with active disease under UST treatment were compared to 18/199 (9%) and 53/199 (27%) propensity-matched cases of anti-TNF and VDZ, respectively. UST-arm showed significantly higher rates of clinical response (aOR 5.1, 95%CI 2.4-10.6, p<0.001) and remission (aOR 2.73, 95%CI 1.2-10.64, p<0.001).
For patients with inactive disease at 2^nd line initiation, 17/161 patients (11%) under UST treatment were compared to 83/161 (51%) and 9/161 (6%) cases of anti-TNF and VDZ, respectively; there was no significant difference in perianal disease recurrence (aOR 0.17, 95%CI 0.03-1.09, p=0.06). In 3^rd line, 30/64 patients (47%) under UST treatment were compared to 5/64 (8%) and 11/64 (17%) cases of anti-TNF and VDZ, respectively. UST-arm was found to have significantly lower rates of perianal disease recurrence (aOR 0.17, 95%CI 0.04-0.76, p=0.02).
As for radiological evidence of treatment effectiveness in 2^nd line, 47/74 patients (63%) under UST therapy had fistula improvement demonstrated in MRI or TRUS, compared to 13/35 (37%) and 75/128 (59%) patients under VDZ and anti-TNF treatment, respectively (p=0.01).

Conclusion

Perianal Crohn’s disease may respond to 2^nd and 3^rd line biological therapies. Our large multicentre cohort with propensity scored matching demonstrated a superior response to ustekinumab as 2^nd or 3^rd line treatment as compared to vedolizumab or anti-TNFs. Our results merit validation in a prospective study.

Disclosure

These authors disclose the following: Shomron Ben-Horin has received consulting and advisory board fees and/or research support from AbbVie, MSD, Janssen, Takeda, and CellTrion; Uri Kopylov has received speaker and advisory fees from AbbVie, Jannsen, Gilead, MSD, Medtronic, and Takeda, and research support from Jannsen, Medtronic, and Takeda; and Bella Ungar has received consultation fees from Neopharm, Takeda, Janssen, and AbbVie. The remaining authors disclose no conflicts.

Introduction

Colorectal endoscopic mucosal resection (EMR) is the treatment of choice for large colorectal lesions, but it has an important recurrence risk, requiring endoscopic surveillance. This study aims to identify factors associated with relapse and complications, by analysing the 5-year follow-up.

Aims & Methods

Prospective cohort study conducted in a single centre, including all patients submitted to colorectal pEMR between 2009 and 2021 with follow-up ≥3 months. We analysed lesion characteristics, calculated Sydney EMR recurrence tool (SERT) and SMSA (size, morphology, site, access) scores and compared complication and recurrence rates and respective risk factors.

Results

Included 568 mucosal resections, median age 69.5 years (IQR 62.3-76.0), male sex 58.1%. From the lesions included: 32.6% ≥40mm; 51.6% plane and 48.4% sessile; 35.9% in the right colon, 27.6% in the rectum, 23.2% in the transverse colon and 13.2% in the left colon. The median SMSA score was 4 (IQR 3-4) and median SERT score 1 (IQR 0-2), complication rate of 7.0% (bleeding, n=33; perforation, n=4; post-polypectomy syndrome, n=3) and overall recurrence rate of 20.3%: 15.1% in 3 months, 6.3% in 12 months, 2.5% in 36 months and 0% in 60 months.
The variables associated with recurrence were size ≥40mm (35.0% vs. 14.6%, p<0.001), anal location (35.1% vs. 19.7%, p<0.01), SERT score>0 (27.5% vs. 11.8%, p<0.001) and SMSA score=4 (30.6% vs. 11.5%, p<0.001), but not argon plasma therapy (22.8% vs. 18.7%, p=0.26). By logistic regression, size ≥40mm was associated with higher recurrence, but not statistically significant (OR 3.19, CI95% 0.69-1.63).
Use of anticoagulant/antiplatelet drugs was associated with higher risk of complications (12.7% vs. 4.7%, p<0.001), particularly bleeding (11.4% vs. 4.2%, p<0.01), with no difference for size ≥40mm (9.2% vs. 6.0%, p=0.17) or SMSA score=4 (7.6% vs. 6.5%, p=0.61).

Conclusion

The 5-year surveillance proved colorectal EMR’s efficacy. Recurrence is higher in lesions ≥40mm, with anal location, SERT score>0 and SMSA score=4, whereas bleeding is more frequent with the use of anticoagulant/antiplatelet drugs.

Introduction

Patients with ulcerative colitis (UC) are at an increased risk of cytomegalovirus (CMV) infection. CMV infection in acute severe ulcerative colitis (ASUC) may worsen disease severity and clinical outcomes. Current guidelines recommend colonic biopsies for the exclusion of CMV in all patients with ASUC. One study found that CMV positivity in ASUC was associated with greater need for rescue therapy, longer hospital length of stay (LOS) and higher readmission rates but found no difference in rate of colectomy.¹ We aimed to investigate the impact of positive CMV result on clinical outcomes in patients presenting with ASUC.

Aims & Methods

A retrospective review included all ASUC presentations to a quaternary hospital over a 10-year period based on modified Truelove and Witts’ criteria. ASUC presentations were grouped based on CMV positivity as determined by colonic biopsy. Data on pre-admission medications was collected. Data was collected on rates of rescue therapy, referral to colorectal surgery, colectomy, total parenteral nutrition (TPN). Outcomes assessed included intensive care unit (ICU) admission, 30- and 90-day readmission rates, 30- and 90-day mortality rates and mean LOS.

Results

A total of 172 ASUC presentations were identified. Of these, 133 had available colonic biopsy CMV results. 11 patients (8%) had positive CMV. Positive CMV result was significantly associated with increased need for TPN (36% vs 11%, p=0.02), ICU admission (27% vs 5%, p=0.004) and referral to colorectal surgery (82% vs 43%, p=0.01). Patients with CMV had longer LOS (27.1±14.0 vs 11.0±7.4, p=0.003). There were no significant differences in rates of rescue therapy, inpatient colectomy, readmission or mortality rates. 9 of the patients (82%) were treated with ganciclovir or valganciclovir. Patients on maintenance advanced UC therapies, including biologics and small molecule inhibitors, had significantly higher rate of CMV (22% vs 5%, p=0.01). Biologic therapy alone was associated with higher rate of CMV (24% vs 7%, p=0.02), whereas small molecule inhibitor alone was not (p=0.6). Pre-admission use of 5-aminosalicylates, high dose steroids, thiopurines or combination therapies were not associated with CMV positivity.

Table 1: Comparison data of clinical outcomes in ASUC between patients with and without CMV positivity on colonic biopsy

Conclusion

Patients with CMV positive colonic biopsies tend to require more inpatient interventions, including increased need for TPN, ICU admission, referral to colorectal surgery and longer hospital length of stay. Despite more complex admission course of ASUC patients with concurrent CMV infection, these patients were not observed to have worse outcomes than those without CMV diagnosis. Maintenance biologic and small molecule inhibitor use was associated with higher rates of CMV, reflecting the immunosuppressant effects of advanced UC therapies. CMV detection needs to be considered as a potential exacerbating factor in disease severity.

References

1. Huang D, Rennie M, Krasovec A, Nagubandi S, Liu S, Ge E, Khehra B, Au M, Sivagnanam S, Kwan V, Rogge C, Mitrev N, Kariyawasam V. Impact of cytomegalovirus on outcomes in acute severe ulcerative colitis: a retrospective observational study. Ther Adv Chronic Dis. 2024 Mar 28;15:20406223241233203. doi: 10.1177/20406223241233203. PMID: 38560721; PMCID: PMC10981253.

Introduction

Vedolizumab (VDZ), a monoclonal antibody therapy approved for moderate-to-severe ulcerative colitis (UC) and Crohn’s disease (CD) in China in 2020¹ has a unique, gut-selective, anti-lymphocyte trafficking (GSALT) mechanism of action.² However, there is limited data on VDZ's real-world effectiveness and safety in Chinese IBD patients.³ Interim results from the VALUE study that enrolled patients with UC have been published ^3,4 and final results are reported here.

Aims & Methods

VALUE (NCT04872491), a multicenter, single-arm, prospective, observational study evaluated the safety and effectiveness of VDZ in adult Chinese patients with IBD (N = 500) over 72 weeks. Clinical response, clinical remission, and endoscopic remission were assessed at Weeks 14, 30, and 54. Partial Mayo and Mayo endoscopic subscores at baseline and Weeks 14, 30, and 54 were calculated. Clinical response was defined as ≥2-point reduction in Partial Mayo Clinic score and ≥25% decrease from baseline score accompanied with ≥1-point decrease in the rectal bleeding or absolute rectal bleeding subscore ≤1. Clinical remission was defined as Partial Mayo Clinic score ≤2 with no subscore >1, endoscopic remission as Mayo endoscopic subscore ≤1, and complete endoscopic remission as Mayo endoscopic subscore 0. Safety results and CD patients’ data are presented in separate VALUE abstracts (IDs: AS-UEG-2025-01329 and AS-UEG-2025-01324).

Results

Among 409 UC patients enrolled in the final analysis, 91.9% (376/409) were included in the effectiveness analysis set. Mean age (±standard deviation [SD]) of patients was 46.3 (±15.1) years, and 59.4% (n=243) were male. At baseline, 30.8% (126/409) of patients had moderate UC, and 29.8% (122/409) had severe UC. The median duration of UC was 2.5 years, concomitant steroids were prescribed in 18.9% (71/376). Mean (±SD) partial Mayo score was 5.1 (2.1) at baseline, which declined to 1.7 (1.7), 1.5 (1.7) and 1.3 (1.5) by Weeks 14, 30 and 54, respectively, post-VDZ treatment. Patients achieving clinical response, clinical remission, endoscopic remission, and complete endoscopic remission were 76.9% (220/286), 71.1% (207/291), 63.6% (35/55), and 21.8% (12/55) at Week 14; 78.1% (193/247), 76.3% (190/249), 63.6% (28/44), and 27.3% (12/44) at Week 30; and 80.3% (155/193), 83.3% (165/198), 71.4% (20/28), and 50.0% (14/28) at Week 54, respectively. Patients with moderate and severe UC; biologic naïve patients and patients with prior exposure to biologics at baseline also achieved clinical response, clinical remission, and endoscopic remission (Table). Compared with baseline, mean (±SD) reduction of C-reactive protein (mg/L) at Weeks 14, 30, and 54 were 7.8 (20.2), 6.5 (22.2), and 3.4 (24.8), respectively; mean (±SD) reduction of fecal calprotectin (mu g/g) was 42.7 (150.7) at Week 14 and 212.6 (480.1) at Week 30. The proportion of patients achieving clinical remission without steroid use by Week 14 was 37.7% (23/61), which increased to 60.6% (20/33) by Week 54.

Conclusion

This study indicated that VDZ demonstrated a good effectiveness profile in Chinese UC patients in the real-world setting.

References

1. Yan J et al., Medicine (Baltimore). 2024; 103: e38759.
2. Soler D et al. J Pharmacol Exp Ther. 2009; 330: 864-875.
3. Baili C et al. Am J Gastroenterol. 2023; 118: S213-S214.
4. Yan C et al. Am J Gastroenterol. 2023; 119: S894-S895.

Disclosure

This study was funded by Takeda (China) International Trading Co. Ltd. Xiuli Zuo, Ye Chen, and Yingchao Li have no conflicts of interest. Lifei Gu and Fang Zhou are Takeda employees and hold Takeda stock options. Minhu Chen received speaker honorariums from Takeda China, AstraZeneca China, Xian Janssen, and Eisai China.
Medical writing support for the development of this abstract under the direction of the authors was provided by Vibhuti Singh, Ph.D., of Cactus Life Sciences (part of Cactus Communications), funded by Takeda (China) International Trading Co., Ltd.

Introduction

Autoimmune atrophic gastritis (AAG) is an immune-mediated disorder characterized by corpus oxyntic mucosa atrophy and impaired gastric acid secretion(1,2). The increased gastric cancer (GC) risk in AAG is still debated. One mechanism favoring gastric carcinogenesis may be increased intragastric pH leading to intragastric microenvironment changes and facilitating bacterial dysbiosis(3-5). Oral microbiota and periodontal-dental disorders alterations have been linked to increased GC risk as well(6-8). In AAG, the increased intragastric pH may lead to survival of swallowed bacteria, possibly further contributing to the increased GC risk. The oral and gastric microbiota composition together have not been investigated so far in AAG patients.

Aims & Methods

The aim of the study was to investigate the oral and gastric microbiota composition in AAGpts in order to clarify the role of oral bacteria in gastric dysbiosis.
We conducted a case-control study: n=20 histologically diagnosed AAGpts(F75%,age 64 yrs) and n=20 ctrs with healthy stomach(HS)(F45%,age 56 yrs). From all pts and ctrs a sputum sample was obtained in the morning fasting without tooth brushing. Within 1 hour after sputum collection, pts and ctrs underwent gastroscopy with biopsies (updated Sydney system) for histopathological assessment and one further corpus biopsy for DNA extraction. Sputum(n=40) and biopsy(n=40) samples were collected in vials with DNA shield and frozen at -20°C since processing. DNA was extracted (QIAmp DNA kit) and prepared (Illumina Nextera XT DNA Library Preparation Kit). Paired-end sequencing was performed (Illumina MiSeq sequencer,MiSeq Reagent Kit v3). Fastq files were filtered to remove low-quality reads; filtered reads were taxonomically classified and subjected to Homo sapiens filtering (METAnnotatorX2 pipeline, up-to-date RefSeq database, NCBI).

Results

AAGpts and ctrs were similar with regard to clinical features, oral hygiene and paraodontal-dental disorders. Bacterial abundance in biopsies was higher in AAGpts than in ctrs (403 vs 223,p=0.06), but not different in sputum (27,728 vs 31,544,p=0.54). On biopsies, AAGpts showed a lower Shannon Index than ctrs (2.553 vs 2.617, p=0.001). On sputum, AAGpts showed a higher Shannon Index than ctrs (2.828 vs 2.738, p=0.0). Beta-diversity (Bray-Curtis dissimilarity) was not different between pts and ctrs, neither in sputum nor biopsies.
At the taxa level, in biopsies Streptococcus oralis, Fusobacterium pseudoperiodonticum, Veillonella rogosae, and Gemella sanguinis were significantly associated with pts(p<0.03); while in sputum the most significant associations were G.sanguinis, Prevotella multisaccharivorax, F.nucleatum, and F.peridonticum(p<0.03). The most abundantly shared taxa between sputum and biopsy were S.oralis and P.histicola, while G.sanguinis, F.pseudoperidonticum, and V.rogosae were shared in pts only.

Conclusion

This study confirmed gastric dysbiosis in AAG as shown by lower biodiversity and evenness. Gastric bacterial abundance was higher in AAG, probably due to non-acidic gastric pH permitting overgrowth of gastric bacteria. Oral bacterial abundance was similar in AAG and ctr with higher biodiversity-evenness in AAG pts, probably due to upstream passage of unactivated bacteria from stomach to mouth. Oral taxa related to increased GC risk were more commonly associated to AAG and shared between mouth and stomach. In AAG these findings show the role of oral microbiota in gastric dysbiosis, possibly contributing to increased GC risk.

References

1) Lenti MV,Rugge M,Lahner E,Miceli E,Toh BH,Genta RM,De Block C,Hershko C,Di Sabatino A. “Autoimmune gastritis” Nat Rev Dis Primers. 2020 Jul 9;6(1):56.
2) Rugge M,Genta RM,Malfertheiner P,Dinis-Ribeiro M,El-Serag H,Graham DY,Kuipers EJ,Leung WK,Park JY,Rokkas T,Schulz C,El-Omar EM;RE.GA.IN;the Real-world Gastritis Initiative-updating the updates. Gut. 2024 Feb 23;73(3):407-441. doi: 10.1136/gutjnl-2023-331164. PMID: 38383142.
3) Conti L, Annibale B, Lahner E. Autoimmune Gastritis and Gastric Microbiota. Microorganisms. 2020;8(11):1827. Published 2020 Nov 19. doi:10.3390/microorganisms8111827
4) Lin L, Li W, Yan L, et al. Comparison of gastric microbiota in patients with different gastric lesions in high and low risk areas of gastric cancer. BMC Microbiol. 2025;25(1):202. Published 2025 Apr 9. doi:10.1186/s12866-025-03926-4
5) Conti L, Borro M, Milani C, et al. Gastric microbiota composition in patients with corpus atrophic gastritis. Dig Liver Dis. 2021;53(12):1580-1587. doi:10.1016/j.dld.2021.05.005
6) Huang C, Li Y, He C. Oral microbes and gastrointestinal cancers: new strategies and insights. Clin Transl Oncol. Published online April 5, 2025. doi:10.1007/s12094-025-03891-4
7) Xia R, Jiang Z, Zhou Y, et al. Oral microbiota and gastric cancer: recent highlights and knowledge gaps. J Oral Microbiol. 2024;16(1):2391640. Published 2024 Aug 16. doi:10.1080/20002297.2024.2391640
8) Aguiar FJN, Menezes FDS, Fagundes MA, et al. Gastric adenocarcinoma and periodontal disease: A systematic review and meta-analysis. Clinics (Sao Paulo). 2024;79:100321. Published 2024 Jan 31. doi:10.1016/j.clinsp.2023.100321