Artificial Intelligence Enabled Histological Prediction of Remission or Activity and Clinical Outcomes in Ulcerative Colitis.

BACKGROUND & AIMS: Microscopic inflammation has significant prognostic value in ulcerative colitis (UC); however, its assessment is complex with high interobserver variability. We aimed to develop and validate an artificial intelligence (AI) computer-aided diagnosis system to evaluate UC biopsies and predict prognosis. METHODS: A total of 535 digitalized biopsies (273 patients) were graded according to the PICaSSO Histologic Remission Index (PHRI), Robarts, and Nancy Histological Index. A convolutional neural network classifier was trained to distinguish remission from activity on a subset of 118 biopsies, calibrated on 42 and tested on 375. The model was additionally tested to predict the corresponding endoscopic assessment and occurrence of flares at 12 months. The system output was compared with human assessment. Diagnostic performance was reported as sensitivity, specificity, prognostic prediction through Kaplan-Meier, and hazard ratios of flares between active and remission groups. We externally validated the model in 154 biopsies (58 patients) with similar characteristics but more histologically active patients. RESULTS: The system distinguished histological activity/remission with sensitivity and specificity of 89% and 85% (PHRI), 94% and 76% (Robarts Histological Index), and 89% and 79% (Nancy Histological Index). The model predicted the corresponding endoscopic remission/activity with 79% and 82% accuracy for UC endoscopic index of severity and Paddington International virtual ChromoendoScopy ScOre, respectively. The hazard ratio for disease flare-up between histological activity/remission groups according to pathologist-assessed PHRI was 3.56, and 4.64 for AI-assessed PHRI. Both histology and outcome prediction were confirmed in the external validation cohort. CONCLUSION: We developed and validated an AI model that distinguishes histologic remission/activity in biopsies of UC and predicts flare-ups. This can expedite, standardize, and enhance histologic assessment in practice and trials.

Ulcer-associated cell lineage expresses genes involved in regeneration and is hallmarked by high neutrophil gelatinase-associated lipocalin (NGAL) levels.

Neutrophil gelatinase-associated lipocalin (NGAL), also known as Lipocalin 2, is an antimicrobial protein, encoded by the gene LCN2, strongly upregulated in inflammatory bowel disease (IBD) and a promising biomarker for IBD. Here we demonstrate that NGAL is highly expressed in all parts of pyloric metaplasia, also known as the ulcer-associated cell lineage (UACL), a metaplastic cell lineage suggested to play a role in wound healing in Crohn's disease (CD). We further show NGAL expression in regenerative intestinal crypts and in undifferentiated patient-derived colonoids. This indicates that NGAL is important in the tissue regeneration process. The remarkable overexpression of NGAL in UACL led us to explore the pathobiology of these cells by transcriptome-wide RNA sequencing. This study is, to our knowledge, the first to characterize the UACL at this level. Biopsies with UACL and inflamed non-UACL epithelium from the terminal ileum of CD patients and epithelium from healthy controls were laser capture microdissected for RNA sequencing. Among the 180 genes differentially expressed between UACL and control epithelium, the ten most-upregulated genes specific for UACL were MUC5AC, PGC, MUC6, MUC5B, LCN2, POU2AF1, MUC1, SDC3, IGFBP5, and SLC7A5. PDX1 was among the most upregulated in both UACL and inflamed non-UACL epithelium. Immunohistochemistry and iDisco 3D visualization was used to characterize UACL histo-morphologically, and to validate protein expression of 11 selected differentially expressed genes. Among these genes, LCN2, NOTCH2, PHLDA1, IGFBP5, SDC3, BPIFB1, and RCN1 have previously not been linked to UACL. Gene expression results were analyzed for functional implications using MetaCore, showing that differentially expressed genes are enriched for genes involved in cell migration and motility, and for biomarkers of gastrointestinal neoplasia. These results support a role for UACL as part of the reepithelialization process during and after destructive intestinal inflammation. © 2019 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of Pathological Society of Great Britain and Ireland.

The ileal fungal microbiota is altered in Crohn's disease and is associated with the disease course.

Introduction: Fungal microbiota's involvement in the pathogenesis of Crohn's disease (CD) is incompletely understood. The terminal ileum is a predilection site both for primary involvement and recurrences of CD. We, therefore, assessed the mucosa-associated mycobiota in the inflamed and non-inflamed ileum in patients with CD. Methods: The mucosa-associated mycobiota was assessed by ITS2 sequencing in a total of 168 biopsies sampled 5 and 15 cm proximal of the ileocecal valve or ileocolic anastomosis in 44 CD patients and 40 healthy controls (HC). CD patients with terminal ileitis, with endoscopic inflammation at 5 cm and normal mucosa at 15 cm and no history of upper CD involvement, were analyzed separately. The need for additional CD treatment the year following biopsy collection was recorded. Results: CD patients had reduced mycobiota evenness, increased Basidiomycota/Ascomycota ratio, and reduced abundance of Chytridiomycota compared to HC. The mycobiota of CD patients were characterized by an expansion of Malassezia and a depletion of Saccharomyces, along with increased abundances of Candida albicans and Malassezia restricta. Malassezia was associated with the need for treatment escalation during follow-up. Current anti-TNF treatment was associated with lower abundances of Basidiomycota. The alpha diversity of the inflamed and proximal non-inflamed mucosa within the same patients was similar. However, the inflamed mucosa had a more dysbiotic composition with increased abundances of Candida sake and reduced abundances of Exophiala equina and Debaryomyces hansenii. Conclusions: The ileal mucosa-associated mycobiota in CD patients is altered compared to HC. The mycobiota in the inflamed and proximal non-inflamed ileum within the same patients harbor structural differences which may play a role in the CD pathogenesis. Increased abundance of Malassezia was associated with an unfavorable disease course.

Establishment of a Patient-Derived Xenograft Model of Colorectal Cancer in CIEA NOG Mice and Exploring Smartfish Liquid Diet as a Source of Omega-3 Fatty Acids.

Cancer patient-derived xenografts (PDXs) better preserve tumor characteristics and microenvironment than traditional cancer cell line derived xenografts and are becoming a valuable model in translational cancer research and personalized medicine. We have established a PDX model for colorectal cancer (CRC) in CIEA NOG mice with a 50% engraftment rate. Tumor fragments from patients with CRC (n = 5) were engrafted in four mice per tumor (n = 20). Mice with established PDXs received a liquid diet enriched with fish oil or placebo, and fatty acid profiling was performed to measure fatty acid content in whole blood. Moreover, a biobank consisting of tissue and blood samples from patients was established. Histology, immunohistochemistry and in situ hybridization procedures were used for staining of tumor and xenograft tissue slides. Results demonstrate that key histological characteristics of the patients' tumors were retained in the established PDXs, and the liquid diets were consumed as intended by the mice. Some of the older mice developed lymphomas that originated from human Ki67+, CD45+, and EBV+ lymphoid cells. We present a detailed description of the process and methodology, as well as possible issues that may arise, to refine the method and improve PDX engraftment rate for future studies. The established PDX model for CRC can be used for exploring different cancer treatment regimes, and liquid diets enriched with fish oil may be successfully delivered to the mice through the drinking flasks.