Faculty Bibliography

Loss of Paneth cells and their antimicrobial granules compromises the intestinal epithelial barrier and is associated with Crohn's disease, a major type of inflammatory bowel disease^1-7. Non-classical lymphoid cells, broadly referred to as intraepithelial lymphocytes (IELs), intercalate the intestinal epithelium^8,9. This anatomical position has implicated them as first-line defenders in resistance to infections, but their role in inflammatory disease pathogenesis requires clarification. The identification of mediators that coordinate crosstalk between specific IEL and epithelial subsets could provide insight into intestinal barrier mechanisms in health and disease. Here we show that the subset of IELs that express γ and δ T cell receptor subunits (γδ IELs) promotes the viability of Paneth cells deficient in the Crohn's disease susceptibility gene ATG16L1. Using an ex vivo lymphocyte-epithelium co-culture system, we identified apoptosis inhibitor 5 (API5) as a Paneth cell-protective factor secreted by γδ IELs. In the Atg16l1-mutant mouse model, viral infection induced a loss of Paneth cells and enhanced susceptibility to intestinal injury by inhibiting the secretion of API5 from γδ IELs. Therapeutic administration of recombinant API5 protected Paneth cells in vivo in mice and ex vivo in human organoids with the ATG16L1 risk allele. Thus, we identify API5 as a protective γδ IEL effector that masks genetic susceptibility to Paneth cell death.

Introduction: Infliximab (IFX) has been efficacious in reducing colectomy rates among patients with moderate-to severe ulcerative colitis, but predictors of colectomy within 30 days of IFX among patients with acute severe ulcerative colitis (ASUC) are less established.
Method(s): We performed a single-center retrospective analysis of patients who received at least one dose of IFX while admitted between 2011-2022. We assessed demographic, clinical and laboratory predictors of colectomy within 30 days of first IFX dose. Multivariable and time-to-event analysis using Kaplan-Meier with log-rank statistics were used to assess risk factors for colectomy within 30 days.
Result(s): A majority of the 172 patients hospitalized with ASUC who received IFX received 10 mg/kg (87.79%). Overall, 22/172 patients (12.79%) underwent colectomy within 30 days of first IFX dose. On univariable analysis, age, sex, race, ethnicity, BMI and smoking status were not associated with risk of colectomy. Higher initial CRP was significantly associated with 30-day risk of colectomy (106.17 vs. 65.10 mg/dL among patients who did not undergo colectomy; p< 0.01), as was a decrease of CRP <=50% prior to discharge (p< 0.01). Lower initial albumin [< 3 (36.36%), 3.0-3.5 (40.91%), >3.5 g/dL (22.73%)] was associated with our primary outcome (p=0.046), as was a higher number of bowel movements in a 24-hour period prior to discharge (5.6 vs. 3.9 among patients who did not undergo colectomy; p=0.0256). On multivariable analysis, higher initial CRP (aOR 1.01, 95% CI 1.00 - 1.02), <=50% change in CRP after first dose of IFX (aOR 9.00, 95% CI 2.43 - 33.29) and higher number of bowel movements in a 24-hour period prior to discharge (aOR 1.24, 95% CI 1.01- 1.52) remained significantly associated with risk of colectomy when adjusting for relevant covariables (Table). On Kaplan-Meier analysis, initial CRP >100 mg/ dL, albumin < 3 g/dL and change in CRP <=50% prior to discharge were significantly associated with decreased time to colectomy (Figure).
Conclusion(s): Among patients with ASUC, higher CRP, decrease of CRP <=50% and higher number of bowel movements prior to discharge were associated with increased risk of colectomy within 30-days of receiving IFX. Initial CRP >100 mg/dL, albumin < 3 g/dL and decrease of <=50% in CRP prior to discharge were associated with decreased time to colectomy. These results can identify patients at highest risk and impact clinical decision-making regarding need for and timing of colectomy in patients with ASUC receiving IFX. (Table Presented)

Introduction: Ileocecal resection (ICR) often leads to remission of Crohn's Disease (CD), but relapse is common. Guidelines suggest postoperative biologic prophylaxis in high-risk patients and colonoscopy within 6-12 months of surgery to assess for post-operative recurrence (POR). Guidance on adjunctive disease monitoring modalities such as biomarkers and cross-sectional imaging is lacking. We aimed to describe the real-world surveillance approach for CD patients after ICR in relation to evidence-based guidelines.
Method(s): This was a dual center retrospective study of CD patients who underwent ICR with >=1 year of follow-up. We grouped patients into high- (HR) and low-risk (LR) for POR per guidelines and assessed the use of biomarkers, imaging, and colonoscopy postoperatively. Approaches and recurrence rates in patients who received resection prior to or after 2015, accounting for changing practices with guidelines, were compared. Biomarker, radiographic, and endoscopic POR were defined as high CRP/fecal calprotectin (FC), active inflammation on CT/MRE, and modified Rutgeerts >=i2b, respectively. P-values were calculated using Wilcoxon and Chi squared tests.
Result(s): Of 1026 CD patients who underwent ICR, 798 were HR. For LR patients, median time to first CRP was 244 days (d), FC was 267d, imaging was 579d, and colonoscopy was 392d. For HR patients, median time to first CRP was 183d, FC was 241d, imaging was 460d, and colonoscopy was 352d. 72% of HR patients had at least 1 modality within 1 year compared to 59% of LR. Compared to pre-2015, patients who underwent an ICR in 2015 or later had significantly earlier imaging (543d vs. 379d, p< 0.001) and colonoscopy (404d vs. 292d, p< 0.001). There was no difference in time to first CRP or FC. Timing of ICR was significantly associated with postoperative biologic use and the detection of POR by all methods (p< 0.001).
Conclusion(s): 30% of HR CD patients did not undergo any monitoring within the first year after ICR. Evolving practice patterns suggest earlier disease monitoring with imaging and colonoscopy in more recent years whereas utilization of biomarkers was not changed. These data suggest that while guidelines have changed practice, allowing for the earlier identification of POR and initiation of therapy, many patients remain under-monitored. As earlier monitoring may improve long-term clinical outcomes, additional studies are required to further guide optimal surveillance intervals and use of biomarkers

Introduction: Clostridium difficile infection (CDI) is one of the most common hospital-acquired infections leading to prolonged hospitalization and significant morbidity. Only a few prior studies have developed predictive risk models for CDI and all but one have utilized logistic regression (LR) models to identify risk factors. Automated machine learning (AutoML) programs consistently outperform standard LR models in non-medical contexts. This study aims to investigate the utility of AutoML methods in developing a model for post-operative CDI prediction.
Method(s): We used an AutoML system developed by Amazon, Autogluon v0.3.1, to evaluate the prediction accuracy of post-surgical CDI using the 2016-2018 ACS NSQIP database. A total of A total of 3,049,617 patients and 79 pre-operative features were included in the model. Post-operative CDI was defined as CDI within 30 days of surgery. Models were trained for 4 hours to optimize performance on the Brier score, with lower being better. Validation of all performance metrics was done using the 2019 NSQIP database.
Result(s): 0.36% of the patients (n = 11,001) developed post-operative CDI. Brier scores were calculated for each model with the top performing model being an ensembled neural net model having a Brier score of 0.0027 on the test set. The corresponding AUROC and AUC-PR was 0.840 and 0.015 respectively (Figure).
Conclusion(s): The models generated via AutoML to predict post-operative CDI had discriminatory characteristics greater than or equal to those models reported in the literature. Future post-operative CDI models may benefit from automated machine learning techniques

Introduction: Inflammatory bowel disease (IBD), comprised of Ulcerative Colitis (UC) and Crohn's Disease (CD), is caused by a combination of environmental factors, immune dysregulation, and genetic susceptibility. Other immune-mediated phenomena, like hypothyroidism, have also been observed in this population. Thus, we sought to explore clinical characteristics and outcomes among IBD patients with hypothyroidism compared to IBD patients without hypothyroidism.
Method(s): In a retrospective chart review from a large, tertiary, academic medical center, baseline demographics and clinical data were extracted for patients diagnosed with either UC or CD and having at least one thyroid stimulating hormone (TSH) measurement from prior to 2016. Based on the presence of a documented hypothyroidism ICD-10 code, patients were then divided into two groups, those with IBD alone and those with both IBD and hypothyroidism, as described in Figure. Individual charts were then further examined for disease characteristics, biomarkers, healthcare utilization, medication use, and other comorbidities from 2016 to 2022. Demographic and clinical variables were then compared between the two groups, as seen in Table.
Result(s): We identified 166 adult IBD patients (CD 53%, UC 47%). The mean age was 62.9 years. Among these patients, 116 patients (69.9%) had IBD and hypothyroidism. The most common causes of hypothyroidism were Hashimoto, subclinical, and acquired hypothyroidism. No differences were noted in race, smoking status, or BMI. IBD disease location, behavior, and prevalence of extra-intestinal manifestations did not significantly differ between the two study groups. Both groups had similar number of colonoscopies, hospitalizations, as well as comparable medication use (SSRI/SNRI, steroids, 5-ASA, immunomodulators, biologics). However, patients with IBD and hypothyroidism had higher rates of anemia (p=0.03), hypoalbuminemia (p=0.007), and CRP elevations (p=0.002). Furthermore, patients with both IBD and hypothyroidism had a greater median number of emergency department visits (p=0.039) and axial radiography (p=0.002).
Conclusion(s): IBD patients with hypothyroidism experience a more severe disease course with higher biomarkers of inflammation and healthcare utilization than those without hypothyroidism despite similar IBD phenotype and therapy exposures. This highlights a potential subgroup of IBD patients who may be at risk for increased disease severity and associated poor outcomes. (Table Presented)

Introduction: Patients with IBD have a high degree of sexual dysfunction (SD) which has been correlated with depression, disease activity, and past medication use such as steroids and biologic therapy. We aimed to track SD longitudinally and assess the impact of biologic therapy, using IBD-specific scales.
Method(s): Patients with Crohn's disease (CD) and ulcerative colitis (UC) starting a new biologic therapy (anti-TNF, anti-integrin, anti-IL12/23, JAK inhibitor) were surveyed at start of induction therapy and at 6-months. Surveys included the IBD- FSDS and MSDS, PROMIS Brief Sexual Function and Satisfaction Profile, clinical disease activity indices [Harvey-Bradshaw index (HBI), partial Mayo (pMayo) score] and scales that assessed depression [Patient Health Questionnaire-9 (PHQ-9)], and quality of life [Short IBD Questionnaire (SIBDQ)]. Clinical data included inflammatory markers and prior IBD therapies. Therapy response was defined as a reduction in HBI, pMayo, SCCAI >=3 or total HBI <= 4, pMayo < 2, SCCAI <=2 at 6 months.
Result(s): 158 patients (86 males and 72 females) completed surveys at induction, and 101 completed at 6 months. The median age was 31 years, 58% had CD, 42% had UC, and 32% were non-white. At induction, the median MSDS score was 5.5 out of 40 (IQR 2-13) and FSDS was 12 out of 60 (3-27; Table). SD correlated with the SIBDQ (r=0.56, p< 0.001), and PHQ-9 (r=0.51, p< 0.001). MSDS and FSDS scores strongly correlated with PROMIS scores (r= 0.70, p< 0.001), and moderately correlated with the HBI (r=0.49, p=0.002), pMayo and SCCAI score (0.44, p=0.02). SD did not correlate with markers of inflammation. MSDS scores significantly improved at 6 months among all participants (p= 0.048). FSDS and PROMIS scores numerically improved among all participants, but did not reach significance. Both MSDS and FSDS scores significantly improved among therapy responders (p=0.004 and p= 0.042, respectively) as did PROMIS scores. Both patients with prior biologic use and biologic naive patients experienced improvement in sexual function among therapy responders (p=0.02, 0.04).
Conclusion(s): There was a strong correlation between SD, disease activity, depression, and quality of life indices. Biologic therapy improves sexual function in therapy responders, which is again evidenced in this updated cohort. Despite prior data correlating prior biologic use with SD, our new findings in this longitudinal study show improvement in SD in patients who are both biologic naive and those with prior use. (Table Presented)

Introduction: Obesity is associated with an increased risk of colorectal neoplasia, but this relationship has not been studied in patients with inflammatory bowel disease (IBD). Both IBD and obesity induce a chronic inflammatory state, so the combination of the two could have an additive or synergistic effect on risk of colorectal neoplasia. Given the increased baseline incidence of dysplasia among IBD patients, identifying modifiable risk factors, such as obesity, could have a significant impact on long term cancer-related outcomes.
Method(s): We performed a retrospective case-control study of IBD colitis patients at an academic IBD Center between January 2006 and February 2022. Demographic and disease-related data, known risk factors for dysplasia, and median BMI during the follow-up period were obtained. Only patients with at least 5 years of colonoscopy reports were included. A case was defined as any patient with biopsy proven dysplasia-indefinite, low-grade, or high-grade-during the study period. A control was defined as any patient with absence of biopsy-proven dysplasia. Obesity was defined as BMI of 30 or greater. Univariate analysis was performed using T-test for continuous variables and chi-square for categorical variables. Multivariate analysis was performed using logistic regression to model dysplasia risk.
Result(s): 106 cases had biopsy-proven colorectal dysplasia (64 IND, 36 LGD, 10 HGD); 125 controls had no dysplasia. Number of colonoscopies (p < 0.001) IBD subtype ulcerative colitis (p = 0.016), maximum histologic severity (p = 0.127), pseudopolyps (p = 0.162), IBD duration (p = 0.098), sex (p = 0.18), age (p < 0.001), smoking history (p = 0.048), prior dysplasia (p < 0.001), and obesity (p < 0.001) were associated with dysplasia on univariate analysis. On multivariable regression, number of colonoscopies (OR 1.26, 95% CI 1.08 - 1.48, p = 0.004), prior dysplasia (OR 3.98, 95% CI 1.23 - 12.86, p = 0.021), and obesity (OR 2.90, 95% CI 1.21 - 6.95, p = 0.017) were each independently associated with increased dysplasia risk. (Figure)
Conclusion(s): Patients with IBD have an increased risk of colorectal neoplasia, but a variety of comorbid states may exacerbate this risk. Notably, we identified obesity as an independent risk factor for dysplasia. Further research is needed to determine whether this risk functions synergistically with IBD or just as an independent risk factor. Furthermore, targeted weight-loss interventions may reduce the incidence of dysplasia among patients with IBD. (Table Presented)

Introduction: As the inflammatory bowel disease (IBD) patient population ages, there will be an increasing number of individuals requiring advanced therapies. Although older age is thought to be associated with immunosenescence, there are data suggesting that older adults may be at higher risk for antibody development as the result of biologic use.
Method(s): Using a large commercial laboratory database (Prometheus Laboratories), we extracted infliximab (IFX) dosing as well as antibody to infliximab (ATI) levels for all individuals using this assay from 2015-2021. Our primary outcome was the presence of ATI (titer >3.1 U/mL). Frequencies were recorded as categorical variables with chi-square analysis used, and multivariable logistic regression was employed to assess the impact of IFX dose, age (< 60 years-old v. >=60 years-old), and IBD subtype on the development of ATI.
Result(s): Overall, there were 22,197 unique specimens, with 3,028 (13.6%) having ATI. When stratified by age, individuals >=60 years-old developed ATI 18.1% (473/2,612) of the time as compared to 15.0% (2,555/17,030) for individuals < 60 years of age (p< 0.01, Figure). Among all individuals with IFX dose < 10mg q8 weeks, older adults (>=60 years of age) were more likely to develop ATI as compared to younger adults (22.8% vs. 16.2%, respectively, p< 0.01); however, when IFX dose was >=10mg/kg q8 weeks, age >= 60 years-old was no longer significantly associated with the development of ATI (9.9% if < 60 years-old vs. 10.6% if >=60 years-old) on univariable analysis. Overall, older adults were less likely to receive IFX doses >=10mg/kg q8 weeks (38.4% in older adults vs. 49.7% in younger adults; p< 0.01). On multivariable analysis, age >=60 years-old (adjOR 1.35, 95%CI 1.20-1.51), IFX dose >= 10mg/kg q8 weeks (adjOR 0.53, 95%CI 0.49-0.57) and having ulcerative colitis as compared to Crohn's disease (adjOR 1.44, 95%CI 1.33-1.57) were independently associated with the development of ATI.
Conclusion(s): Older adults with IBD develop ATI more frequently than younger adults when adjusting for IFX dose and IBD subtype. However, when IFX dose >=10mg/kg q8 weeks, ATI was significantly less likely to develop among older adults, and occurred in a similar proportion of younger individuals. Further education is needed, highlighting that older adults with IBD are more likely to develop ATI as compared to younger adults, particularly when using lower doses of IFX, and that higher doses may decrease this likelihood. (Figure Presented)

Inflammatory bowel diseases (IBD), including Crohn disease and ulcerative colitis, are chronic inflammatory conditions of the gastrointestinal tract. Individuals with IBD are at increased risk for several malignancies originating in the intestine, such as colorectal cancer, small bowel adenocarcinoma, intestinal lymphoma, and anal cancer. There are also several extraintestinal malignancies associated with IBD and IBD therapies, including cholangiocarcinoma, skin cancer, hematologic malignancies, genitourinary cancer, cervical cancer, and prostate cancer. The authors summarize the risk of cancer in patients with IBD, diagnosis and management of colorectal neoplasia in IBD, and management of patients with IBD and active or recent cancer.

BACKGROUND:Colorectal strictures have been considered independent risk factors for neoplasia in patients with inflammatory bowel disease (IBD). We examined the association between colorectal stricture and subsequent risk of colorectal neoplasia (CRN) in patients with IBD colitis undergoing colonoscopic surveillance. METHODS:We conducted a retrospective cohort analysis of patients with IBD colitis enrolled in colonoscopic surveillance for CRN at an academic medical center between 2005 and 2017. Inclusion criteria were IBD involving the colon for ≥8 years (or any duration with primary sclerosing cholangitis [PSC]) undergoing surveillance. Exclusion criteria were advanced CRN (ACRN; colorectal cancer [CRC] or high-grade dysplasia [HGD]) prior to or at enrollment, prior colectomy, or limited (<30%) disease extent or proctitis. Multivariable logistic and Cox regression analysis estimated the association between colorectal stricture on the index colonoscopy and ACRN, CRN (indefinite dysplasia, low-grade dysplasia, HGD, CRC), or colectomy. RESULTS:Among 789 patients with IBD undergoing CRC surveillance, 72 (9%; 70 with Crohn's colitis) had a colorectal stricture on index colonoscopy. There was no significant difference in the frequency of ACRN or requirement for colectomy between patients with vs without a colorectal stricture (P > .05). Colorectal stricture was not associated with subsequent ACRN (adjusted odds ratio [aOR], 1.41; 95% CI, 0.49-4.07), CRN (aOR, 1.15; 95% CI, 0.51-2.58), or colectomy (aOR, 1.10; 95% CI, 0.65-1.84). CONCLUSIONS:In this analysis of patients with IBD colitis undergoing CRN surveillance, the presence of a colorectal stricture was not independently associated with risk of ACRN or colectomy. Multicenter, prospective studies are needed to confirm these findings, particularly in patients with ulcerative colitis-associated colorectal stricture.