Article Text


GI highlights from the literature
  1. Mairi H McLean, Education editor
  1. School of Medicine, Medical Sciences and Nutrition, Aberdeen University, Aberdeen, UK
  1. Correspondence to Dr Mairi H McLean, School of Medicine, Medical Sciences and Nutrition, Aberdeen University, Aberdeen AB24 3FX, UK; m.h.mclean{at}

Statistics from

Basic science

Stromal microenvironment is a key determinant of tumour biology

Lenos KJ, Miedema DM, Lodestijn SC et al. Stem cell functionality is microenvironmentally defined during tumour expansion and therapy response in colon cancer. Nat Cell Biol 2018;20:1193–1202.

Cancer stem cells (CSCs) drive solid tumour expansion and confer resistance to chemotherapy. The stem cell hierarchy is well defined in homoeostasis, but in cancer the plasticity of differentiated cells is less clear. Human colon cancer cell lines were injected subcutaneously into mice and lineage tracing introduced. Tumour growth was modelled stochastically and loss of clonogenicity used as a measure of intrinsic or environmental pressure; simultaneous loss of clonogenicity is driven by the environment and random loss driven by an intrinsic process. Lack of an intrinsic hierarchy was shown within individual clones. RNA sequencing of cells from xenograft edges showed enrichment of proliferative genes but no increase in CSC markers. Retransplantation of the tumour centre led to increased growth, revealing a difference between those CSC marker positive cells throughout the tumour and functional CSCs at the tumour edge. The edge cells were also enriched for stromal markers. Coculture of cancer cells with primary fibroblasts encouraged growth, suggesting that cancer-associated fibroblasts (CAFs) secrete key oncogenic factors. Osteopontin was the most abundantly expressed factor. When this was overexpressed in cancer cell lines, there was accelerated growth and less variation in clone sizes. When xenografts were treated with chemotherapy, there was an increase in CSC markers but no difference in clonal dynamics. Colorectal cancer lacks a stem cell hierarchy and stem cell function is defined by the environment. CAFs shape tumour biology and therapies aimed at CSCs are unlikely to be of benefit given cell plasticity but targeting CAF factors may be a new therapeutic strategy.

Differential regulation of non-alcoholic fatty liver disease-induced liver inflammation and tumourigenesis by STAT-1 and STAT-3 signalling

Grohmann M, Wiede F, Dodd GT, et al. Obesity Drives STAT-1-Dependent NASH and STAT-3-Dependent HCC. Cell 2018;175:1289–1306.

Hepatic outcomes in patients with non-alcoholic fatty liver disease (NAFLD) are heterogeneous; some patients develop simple steatosis, some develop liver inflammation, fibrosis and cirrhosis and some develop hepatocellular carcinoma (HCC) even in the absence of liver fibrosis. The mechanisms driving these disparate outcomes remain unclear. In this study, the authors investigated the role of STAT-1 and STAT-3 signalling in regulating liver inflammation and HCC development in NAFLD. Oxidation and inactivation of protein tyrosine phosphatases (PTPs) are known to regulate the activity of signalling cascades. Oxidation of T cell PTP (TCPTP) was increased in both murine models and liver biopsies of patients with NASH, and associated with increased STAT-1 and STAT-3 activation. A high-fat diet in transgenic mice with hepatocyte-specific knockout of TCPTP, resulted in enhanced activation of STAT-1 and STAT-3 signalling, accentuated hepatocyte damage, accumulation of intrahepatic TCs and B cells, liver fibrosis and HCC development. To assess the relative contributions of STAT-1 and STAT-3 signalling to this phenotype, these mice were then crossed to either STAT-1fl/- or STAT-3fl/- mice to facilitate selective reduction in each signalling pathway. Strikingly, reduced STAT-1 signalling in hepatocytes inhibited lymphocyte accumulation and liver fibrosis but did not affect HCC formation. In contrast, reduced hepatocyte STAT-3 signalling did not affect the inflammatory features of NASH but did completely repress HCC formation. Hence, this work has identified that discrete signalling pathways regulate different clinical sequelae of NAFLD. Future studies should aim to identify the drivers of STAT-1 and STAT-3 activation in NASH.

Colonic mesenchymal cells: targeting the origin of inflammation in inflammatory bowel disease

Kinchen J, Chen HH, Parikh K, et al. Structural remodelling of the human colonic mesenchyme in inflammatory bowel disease. Cell 2018;175:372–386

Colonic mesenchymal cells (CMCs) play a role in maintaining cell turnover, innate immunity and mucosal barrier integrity. Their function is altered in inflammatory bowel disease (IBD), but their heterogeneity or disease-associated plasticity is not understood. Single-cell RNA sequencing was used to characterise subclasses of CMC from colonic biopsies of healthy volunteers (HVs) and newly diagnosed patients with ulcerative colitis (UC). This approach identified four clusters of stromal like cells, termed S1–S4. On further analysis, S2, located in the epithelial monolayer, expresses tumour necrosis factor-β (TNF) and Wnt ligand implying a role in epithelial reconstruction. The S2 subset was depleted in UC biopsies. S4 (barely detectable in HV but expanded in UC) shows enriched gene ontology for cytokine signalling pathways, particularly increased expression of CCL19 and IL-33. The authors then compared CMC from a UC murine model to human tissue: S2 and S3 in mice are comparable while S1 shows little correlation. In the murine dextran sodium sulfate (DSS) UC model an activated population with human-like S4 emerges. To replicate the S2 depletion and S4 expansion in human UC, the authors exposed colonic organoids to IL-6 and LIGHT (S4 secreted factors) with WNT withdrawal and demonstrated a decrease in LGR5, NOTCH1 and upregulation of damage-responsive stem cell markers. This suggests disease plasticity of CMC subclasses. In DSS colitic mice, Lox/Loxl1 (oxidative enzymes) are induced and when blocked by a specific inhibitor disease severity is reduced. This suggests S4 cells elicit redox imbalance perpetuating inflammation. These findings have revealed a potential new drug target for IBD.

Clinical practice

Prediction of response to biological therapy in IBD

Telesco SE, Brodmerkel C, Zhang H, et al. Gene expression signature for prediction of golimumab response in a Phase 2a open-label trial of patients with ulcerative colitis. Gastroenterology 2018;155:1008–1011.e8.

Prediction of response to targeted biological therapies, which would finally allow personalised treatment approaches, is currently unattainable in IBD. The challenge to establish and validate a predictive biomarker is again reflected in an exemplary prospective biomarker study by Telesco et al. The authors used a colonic 13-gene transcript panel that had previously shown an association with efficacy of anti-TNF therapy, to predict therapeutic response to the anti-TNF antibody golimumab in 103 patients with UC. The baseline gene expression signature identified patients with mucosal healing at week 6 and 30, respectively, with an area under the receiver operating curve of 0.688 (p=0.002) and 0.671 (p=0.006). The biomarker panel predicted mucosal healing with a sensitivity of 87%, but with a specificity of only 34%, indicative of a high false positive rate. The gene expression signature was not able to identify patients who would achieve clinical response or clinical remission. Altogether, the gene transcript panel cannot be regarded as a clinically applicable response prediction tool. The low specificity of the biomarker may reflect a lower golimumab drug level in predicted responders. Future trials that aim to validate a predictive biomarker of response must therefore also take into account other factors that have been shown to influence the efficacy of biological therapies, such as clinical phenotype, genetic factors, the microbiome, pharmacological factors, drug target expression and the immune cell infiltrate, reflecting the complexity of such an approach.

Prophylactic PPI in the intensive care unit setting

Krag M, Marker S, Perner A et al. Pantoprazole in Patients at Risk for Gastrointestinal Bleeding in the ICU. N Engl J Med 2018 doi: 10.1056/NEJMoa1714919. [Epub ahead of print 24 Oct 2018].

Acutely unwell patients treated in the intensive care unit (ICU) setting are at risk of gastrointestinal bleeding. Based on historical data, many guidelines recommend prophylactic acid suppression in this patient population. Krag et al performed a multicentre, double-blind, randomised controlled trial of once daily intravenous pantoprazole or placebo in 3298 patients admitted to ICU. Primary outcome was 90-day mortality with a composite secondary outcome of gastrointestinal bleeding, new-onset pneumonia, Clostridium difficile infection or acute myocardial ischaemia occurring in ICU. Adjusted 90-day mortality was 510/1642 (31.1%) in the pantoprazole group and 499/1640 (30.4%) in controls (relative risk (RR) 1.02, 95% CI 0.91 to 1.13, p=0.76). The 90-day morbidity was 360/1644 (21.9%) in the pantoprazole group and 372/1647 (22.6%) in controls (RR 0.96, 95% CI 0.83 to 1.11). There was a small difference in unadjusted episodes of gastrointestinal bleeding (41/1644 (2.5%) in the pantoprazole group and 69/1647 (4.2%) in controls (RR 0.58, 95% CI 0.40 to 0.86) but no difference in infectious events (276/1644 (16.8%) in the pantoprazole group and 279/1647 (16.9%) in controls (RR 0.99, 95% CI 0.84 to 1.16). Overall, this study shows that the prevalence of acute gastrointestinal bleeding is low in contemporary ICU care and a blanket approach to proton pump inhibition (PPI) prophylaxis does not confer any significant risk reduction. PPI does appear safe in this setting. PPI prophylaxis could be best focused on those at higher risk of bleeding such as need for renal replacement therapy, coagulopathy, liver disease and other factors. There now needs to be identification of those patient groups at highest risk to aid this approach.

Avatrombopag before procedures reduces need for platelet transfusion in patients with chronic liver disease and thrombocytopaenia

Terrault N, Chen YC, Izumi N, et al. Avatrombopag before invasive procedures in patients with liver cirrhosis: is it time to ADAPT? Gastroenterology 2018;155:705–718.

Thrombocytopaenia is common in patients with chronic liver disease (CLD), with severity increasing as liver disease progresses. There is an increased bleeding risk in patients with severe thrombocytopaenia (defined as platelets <50×109/L) undergoing invasive procedures with some guidelines advising prophylactic platelet transfusions to minimise this risk. Avatrombopag is a thrombopoietin receptor agonist to increase platelet counts. Across two international multicentre randomised, placebo controlled phase 3 trials (ADAPT-1 and ADAPT-2), Terrault et al assessed the efficacy of avatrombopag in patients with CLD undergoing invasive procedures. The primary endpoint was proportion of patients who did not require platelet transfusion or rescue procedure for bleeding up to 7 days postprocedure. Patients in ADAPT 1 (n=231, 75 sites in 20 countries) and ADAPT 2 (n=204, 74 sites in 16 countries) were divided into two cohorts according to baseline platelet count and randomised (2:1) to receive five daily doses of avatrombopag (60 mg if platelets <40×109/L vs 40 mg if platelets 40 to <50×109/L) or placebo. The primary endpoint was achieved in 65.6% (ADAPT 1) and 68.6% (ADAPT 2) patients in the 60 mg avatrombopag group and 88.1% (ADAPT 1) and 87.9% (ADAPT 2) in the 40 mg avatrombopag group, versus 22.9% and 38.2% (ADAPT 1), and 34.9% and 33.3% (ADAPT 2) patients in the placebo groups (p<0.0001 for all). Avatrombopag was well tolerated with a safety profile comparable to placebo. The authors conclude that avatrombopag is a safe and effective alternative to platelet transfusion in patients with CLD undergoing invasive procedures. Further large studies are required to validate these findings.


Sujata Biswas, Prakash Ramachandran, Francesca Moroni, Raja Atreya, John Leeds, Norma McAvoy.

Journals reviewed

Nature Cell Biology, Cell, Gastroenterology, New England Journal of Medicine.

View Abstract


  • Patient consent for publication Not required.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.