Introduction

Primary sclerosing cholangitis (PSC) is an inflammatory disorder of the bile ducts, in which Inflammatory bowel disease (IBD) concomitantly occurs (PSC-IBD). Substantial differences in clinical presentation are observed between PSC-IBD and ulcerative colitis (UC).

Aims & Methods

In this study we aim to find distinct pathomechanisms for PSC-IBD using single-cell RNA sequencing. Forty-seven colonic samples of PSC-IBD (n=24), UC (n=18) and non-IBD subjects (n=5) were collected. Whole biopsies were dissociated into single cells using one-step collagenase digestion. Library preparation was done with 10x Genomics and sequencing was performed on an MGI2000 sequencer. Then, differential abundance, differential expression, cell-cell interaction analyses, and heritability analyses were performed. Stainings for DUOX2 and HLA-DR were performed on paraffin embedded colon mucosal slides

Results

In total, 71,798 high quality cells identified 54 distinct cell types, including a new type of epithelial cells: DUOX2+ enterocytes. DUOX2+ enterocytes are almost exclusively present in inflamed colon and can act as antigen presenting cells through the expression of MHC2 on its surface. A general shift in cell type composition upon inflammation was observed, comparable between PSC and UC samples. However, an increased abundance was seen for stem cells in non-inflamed (NI) PSC, and not in UC-NI nor non-IBD controls. Additionally, we found distinct gene expression patterns between PSC and UC inflammation: while activation of inflammatory monocytes was observed in PSC inflammation, several types of fibroblasts were activated upon UC inflammation. Through heritability analysis, we also highlighted how mucosal follicular B and CD4+ activated Fos-Hi cells carry the heaviest burden of PSC risk genes.

Conclusion

We describe DUOX2+ enterocytes, which may play a role in both PSC and UC colitis inflammation through antigen presentation. Furthermore, we show that intestinal inflammation in PSC-IBD, as compared to UC, is characterized by distinct mucosal cell composition and cell-specific gene expression patterns. PSC colitis, and not UC, is driven by activation of inflammatory monocytes, with antigen presentation through HLA-DRB1, and activation of specific CD4+ T cell subsets. Additionally, an increased abundance of stem cells in PSC-NI could be related with the increased colorectal cancer risk in PSC-IBD.

Disclosure

S.S. received a grant from China, J.B. received a grant from Searave Foundation, R.K.W., E.A.M.F. and K.N.F. received unrestricted research grants from Takeda, R.K.W. received an unrestricted grant from Tramedico and Ferring and is supported by a Diagnostics Grant from the Dutch Digestive Foundation (D16-14). W.T.C.U.V. and E.A.M.F. received a Gastrostart Grant from the Dutch Society of Gastroenterology. E.A.M.F. received a personalized medicine grant form the University of Groningen a personalized grant a Clinical Fellowship from the ZonMw, and is supported by a MLDS Career Development grant (CDG 14-04).

A.B., P.P., S.d.J., E.B., G.M.P.J.V., F.G.M.K., G.K., M.A.H., A.V.V. and M.G.P.v.d.W. disclose no conflicts of interest.

Introduction

The incidence of oesophago-gastric cancer has an age-dependent male predominance mirroring physiological changes in sex hormonal levels. Some previous studies suggest that menopausal hormone therapy (MHT) counteracts the occurrence of these tumours, but larger studies with longer follow-up and adjustment for confounders are needed, which prompted the present study.

Aims & Methods

To investigate the hypothesis that MHT decreases the risk of developing oesophago-gastric cancer, especially oesophageal adenocarcinoma.
This population-based case-control study included women aged ≥45 years in all five Nordic countries (Denmark, Finland, Iceland, Norway, Sweden) between 1995-2020. Prospectively collected data came from national complete registries for medications, cancer, diagnoses, total populations, and death. The study included 19,518 case patients with oesophago-gastric cancer and 195,094 randomly identified control participants matched for age, country and calendar year. The participants were categorised into three equal-sized groups (tertiles) of defined daily doses (DDD) of MHT, with non-users as the reference group. MHT was also categorized as systemic or local, and oestrogen only or oestrogen combined with progesterone MHT. Oesophago-gastric cancer was divided into oesophageal or cardia adenocarcinoma, oesophageal squamous cell carcinoma, and gastric non-cardia adenocarcinoma as separate outcomes. Conditional logistic regression provided odds ratios (OR) with 95% confidence intervals (CI), adjusted for age, calendar year, country, obesity, gastroesophageal reflux disease, smoking, alcohol, Helicobacter pylori eradication, use of statins and non-steroidal anti-inflammatory drugs.

Results

​Compared to non-users of MHT, the adjusted ORs of oesophageal or cardia adenocarcinoma were 0.74 (95% CI 0.67-0.81) for low MHT-use (<158 DDDs), 0.68 (95% CI 0.61-0.75) for intermediate MHT-use (158-848 DDDs), and 0.68 (95% CI 0.61-0.75) for high MHT-use (>848 DDDs). The corresponding ORs were 0.69 (95% CI 0.62-0.77), 0.70 (95% CI 0.62-0.77), and 0.71 (95% CI 0.64-0.79) for oesophageal squamous cell carcinoma, and 0.90 (95% CI 0.84-0.96), 0.85 (95% CI 0.79-0.91), and 0.80 (95% CI 0.74-0.86) for gastric non-cardia adenocarcinoma.

Table. Odds ratios (OR) and 95% confidence intervals (CI) of oesophago-gastric cancer by tertiles of defined daily doses (DDD) of menopausal hormone therapy (MHT), MHT hormonal constitution and administration of MHT.

Conclusion

This study provides support for the hypothesis of an inverse association between MHT and risk of developing oesophago-gastric cancer, and particularly for oesophageal adenocarcinoma where the association was seemingly dose-dependent and stronger for combined oestrogen and progesterone MHT and for systemic MHT. A limitation to the study was the inability to adjust for socioeconomic factors as confounders.