Introduction

Patients with chronic liver disease (CLD), particularly those with cirrhosis, exhibit increased risk of developing severe illness when infected with SARS-CoV-2, primarily due to their underlying immune dysfunction and coexisting conditions. As such, vaccination plays a crucial role in safeguarding this population. Here, to fully assess the efficacy COVID-19 vaccination in patients with CLD, we thoroughly characterized the overall and SARS-CoV-2-specific cellular immune response of these patients.

Aims & Methods

Blood samples were systematically collected from patients with CLD (n=22) and healthy volunteers (n=12) at various intervals spanning one year after the initial administration of the BNT162b2 vaccine. Peripheral blood mononuclear cells (PBMCs) were isolated and stained with a panel of 22 surface markers and a viability dye, allowing for the detection of 28 immune cell subsets. In parallel, we analysed the presence of cells specific against SARS-CoV-2 peptide tetramers. Samples were analysed by flow cytometry with acquisition on a four-laser Cytek Aurora® spectral analyzer.

Results

Prior to vaccination, patients with CLD patients were already displaying fewer T cells and an increased number of B cells, comparing to healthy subjects. Within the T cell subsets, differences were significant for CD4+ T cells and the Naïve and Central Memory CD4+ subsets. At both 6-months and 1- year after vaccination, variances between groups were no longer observed. Rather, at 1-year postvaccination, T, CD8+ and CD4+ cells, as well as monocytes were found increased in both groups, suggesting that the vaccination is critical for inducing a strong cellular immune response. Analysis of SARS-CoV-2 epitope-specific cells demonstrated significantly lower frequencies of monocytes, CD4+ and CD8+ T cells, B cells, NK T cells, and NK cells including the regulatory and cytotoxic subsets recognizing the peptide tetramers in CLD patients, in comparison with healthy controls.

Conclusion

Together, these results indicate that CLD patients are capable of mounting a satisfactory immune cellular response following COVID-19 vaccination. However, the SARS-CoV-2-specific response is significantly weaker in CLD patients comparing with healthy individuals. These results reinforce the need for prioritization and/or booster vaccination for CLD patients. (Supported by 2022.08837.PTDC, FCT).

Introduction

Esophageal variceal bleeding is a serious and fatal complication of cirrhosis. Currently, efficient noninvasive monitoring strategies are lacking. By applying artificial intelligence (AI) to computed tomography (CT) imaging, we developed an innovative tool to meet clinical needs.

Aims & Methods

This is an ambispective cohort study conducted at West China hospital between April 2021 and December 2021, patients underwent endoscopy and enhanced CT were registered. Furthermore, 97 patients with radiomic and hepatic venous pressure gradient data were enrolled from a portal hypertension database. High-risk esophageal varices (HREV) were defined as either moderate-severe varicose (HREV-1) or EV requiring intervention (HREV-2). Deep learning was used to develop segmentation and EV quantification models. The EV bleeding prediction model was established using least absolute shrinkage and selection operator (LASSO) regression and logistic regression analysis. The area under the curve (AUC) was calculated for efficiency evaluation. This trial is registered with the Chinese Clinical Trial Registry (ChiCTR2300069865) and is completed.

Results

A total of 2291 patients in the retrospective (n=780) and prospective (n=1511) cohorts were included. The internal and external validation cohort AUCs were 0.976 and 0.951 for EV vs. non-EV, 0.986 and 0.951 for HREV-1, and 0.978 and 0.898 for HREV-2, respectively. We found that volume (p=0.0348) and AI-10 (p=0.0196) were closely related to portal pressure. We further analyzed the AI features and bleeding events for patients with cirrhosis in the retrospective (n=254) and prospective (n=569) cohorts and applied the EV bleeding prediction model (validation cohort AUCs of 0.669, 0.701 and 0.705 at 1, 3, and 6 months, respectively), which was not inferior to that of endoscopic signs (p > 0.05).

Conclusion

Enhanced CT assisted by AI is a promising noninvasive strategy for quantifying EV, recognizing HREV and predicting short-term bleeding risk and has the potential to replace simple gastroscopy for diagnosis or monitoring. This approach could decrease unnecessary gastroscopy to some extent.

Disclosure

The authors have declared no competing interests.

Clinical Case Summary

A 13-year-old boy presented with abdominal distention, recurrent hypoglycemia, and self-resolving jaundice. Clinical examination revealed growth retardation and hepatosplenomegaly. Blood work revealed anemia, thrombocytopenia, elevated aspartate and alanine transaminase levels, and raised plasma cholesterol and triglycerides. Ultrasound showed an enlarged liver with surface irregularity, splenomegaly, and a dilated portal vein. Etiological work-up for cirrhosis was positive for hepatitis B (HBV) surface antigen (HbsAg) and antibodies for HBV core and envelope antigens (anti Hbc, antiHbe) but an undetectable HBV DNA level. HBV was considered to be an unlikely cause of cirrhosis due to the young age of the patient, hence further metabolic and genetic work-up was done. Tests for Wilson’s and autoimmune liver disease were negative. Liver biopsy revealed a distorted lobular architecture with nodules, thin septae, and portal fibrosis. Hepatocytes showed marked swelling and pyknotic nuclei, and negative HbsAg immunostaining. Molecular analysis revealed a novel mutation in the exon 3 of the PHKG2 gene (c.98G>T, p.Gly33Val). The patient was diagnosed with cirrhosis due to GSD IXc with inactive chronic hepatitis B. His symptoms resolved with an optimized diet of corn starch, high-protein and vitamin supplementation. The importance of liver transplantation was also discussed with the family.

Introduction

Gastric cancer is the 5^th most common cancer globally(1). Europe's Beating Cancer Plan recognises the urgent need for a gastric cancer screening strategy(2). One promising approach, outlined in the Maastricht VI/Florence consensus report, suggests integrating screening modalities for gastric cancer, such as gastroscopy with colorectal cancer screening (3). This method holds the potential for cost-saving benefits through combined screening efforts (4).

Aims & Methods

We aimed to determine the prevalence of preneoplastic and neoplastic lesions in a high-risk age group, in those who underwent gastroscopy at the time of colonoscopy. Our secondary aims were to determine a)the quality of endoscopy performed and b) the endoscopic burden for pathology that required follow-up.
This was a single-centre review and audit of key performance indicators in a colorectal cancer screening centre. The age group targeted was 50-74, i.e. the target group for colon cancer screening in Europe. We included those undergoing colonoscopy for polyp surveillance or as part of the national colon cancer screening programme. Participants with red flag symptoms (iron deficiency anaemia, weight loss, dysphagia), family history of gastric cancer, established diagnosis of Barrett’s, intestinal metaplasia, dysplasia or who had a gastroscopy in the last 3 years were excluded. The sample size required for a 95% confidence interval with a presumed positivity rate of 5-7% was 101. Descriptive statistics, correlations and associations were examined using Stata software.

Results

101 patients were included; the median age was 59;38% were male. Indications for gastroscopy included abdominal pain (45%), chronic reflux (37%), chronic dyspepsia (7%) and other (9%). All patients had a technically complete gastroscopy (D2 intubation rate 100%, retroflexion 100%), the median procedure length was 7 minutes(range 4-21). 93% of patients had a macroscopic finding at the time of gastroscopy. Findings included gastritis(chronic superficial, chemical)(60%), hiatus hernia (50%), oesophagitis (28%), duodenitis (24%), gastric body polyps (10%), Barrett’s mucosa (8%), benign gastric or duodenal ulceration(9%), Schatzki ring (4%), erosions (4%), intestinal metaplasia (IM)(1%), oesophageal candidiasis (1%) and varices (1%). 59% had biopsies taken from two topographical sites as per MAPS guidelines. On biopsy, Helicobacter pylori infection was found in 15% (15/101), 7% had intestinal metaplasia(IM); no atrophy, dysplasia or cancer was detected. In those with IM who had MAPS biopsies (5/7), all were focal IM. In those with suspected Barrett’s (n=8), 88% had biopsies taken in accordance with the Seattle protocol and the Prague classification was used in 50% of cases. Additional pathology encountered included histological evidence of antral gastritis (56%), body gastritis (53%), reflux oesophagitis (21%), Barrett’s without dysplasia (4%), duodenitis (3%), and eosinophilic oesophagitis (1%). There was no statistical difference in age (p<0.82) or gender (p<0.32)in those with preneoplastic changes on gastroscopy.

Conclusion

11% of individuals within the target age bracket exhibited a pre-neoplastic condition on gastroscopy and/or histology (i.e. gastric intestinal metaplasia, gastric atrophy or Barrett's oesophagus), a further 15% were diagnosed and treated for Helicobacter pylori infection, the leading cause of gastric cancer. Among these cases, 4-7% necessitated subsequent surveillance gastroscopy. Our findings support the role of dual endoscopic screening with quality gastroscopy as a potential strategy for gastric cancer screening.

References

1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209-49. doi: 10.3322/caac.21660. Epub 2021 Feb 4.
2. Commission E. Europe's Beating Cancer Plan - Communication from the commission to the European Parliament and the Council. 2021.
3. Malfertheiner P, Megraud F, Rokkas T, Gisbert JP, Liou JM, Schulz C, et al. Management of Helicobacter pylori infection: the Maastricht VI/Florence consensus report. Gut. 2022.
4. Libanio D, Antonelli G, Marijnissen F, Spaander MC, Hassan C, Dinis-Ribeiro M, Areia M. Combined gastric and colorectal cancer endoscopic screening may be cost-effective in Europe with the implementation of artificial intelligence: an economic evaluation. Eur J Gastroenterol Hepatol. 2024;36(2):155-61. doi: 10.1097/MEG.0000000000002680. Epub 2023 Nov 15.

Introduction

Arsenic, a potent environmental toxin, has been widely reported to cause significant health problems, including liver damage and disturbances in the metabolic processes. One of the lesser-understood facets of arsenic toxicity is its impact on gut microbiota, the community of microorganisms living in the intestinal tract, which plays a crucial role in host metabolism and overall health. Previous studies have indicated that arsenic exposure can alter the composition of gut microbiota, potentially affecting the host’s metabolism of arsenic and exacerbating its toxic effects. Despite ongoing research, the mechanisms underlying the interaction between arsenic exposure, gut microbiota changes, and arsenic metabolism remain inadequately defined.

Aims & Methods

The primary objective of this study was to elucidate the dynamics of gut microbiota changes under varying levels of arsenic exposure and to understand how these changes affect arsenic metabolism and subsequent liver toxicity. We used C57BL/6 mice as a model system, exposing them to 0, 5, or 10 ppm sodium arsenite (NaAsO2) in their drinking water over a period of six months. We conducted a comprehensive analysis of the gut microbiota, focusing on folic acid (FA)/vitamin B12 (VB12)- and butyrate-synthesizing communities. Additionally, molecular interactions between arsenic and methionine synthase (MTR), an enzyme crucial for FA/VB12 utilization, were assessed using 3D molecular simulations and biochemical assays. The intervention model involved supplementation of FA and VB12 to evaluate their protective effects against arsenic-induced disturbances.

Results

Exposure to NaAsO2 significantly altered gut microbiota, with a notable increase in the abundance of FA/VB12- and butyrate-synthesizing microbiota. This shift was hypothesized to compensate for the disrupted utilization of these nutrients due to arsenic exposure. In higher arsenic exposure levels, however, the microbiota changes were inconsistent, suggesting a potential threshold beyond which the compensatory mechanism may be overwhelmed. Biochemical assays confirmed that arsenic binds to MTR, disrupting its function and further impacting FA/VB12 metabolism. The interaction was substantiated by SEC-UV-DAD and fluorescence-labeled arsenic studies, which demonstrated that arsenic directly interacts with MTR in both non-cellular and cellular systems. Our intervention model using FA and VB12 supplementation showed that it was possible to mitigate the disturbances in gut microbiota, restore MTR levels, enhance arsenic metabolism, and reduce liver injury.

Conclusion

This study provides significant insights into the interaction between arsenic exposure and gut microbiota. We demonstrated that changes in gut microbiota are not merely a consequence of arsenic exposure but are intricately linked to the metabolic processes affected by arsenic, particularly through the disruption of MTR function. The supplementation of FA and VB12 proved effective in blocking these disturbances, suggesting potential therapeutic strategies to combat arsenic toxicity. The findings emphasize the importance of considering gut microbiota in the management and treatment of arseniasis and underscore the complex interplay between environmental toxins, gut microbiota, and host metabolism. Future studies should focus on further delineating the specific pathways and mechanisms by which microbiota influence arsenic metabolism and exploring the potential for microbiota-targeted therapies in arsenic-related diseases.

Introduction

The development of gastric cancer is strongly linked to Helicobacter pylori infection. Therefore, early diagnosis and eradication of H. pylori infection can significantly lower the risk, as studies consistently show reduced cancer rate with these measures. The success in achieving such a goal depends on widespread early detection and treatment, plus effective prevention of new infections. To date, there is no effective certified vaccine against H. pylori available. This study describes the development of a novel multi-epitope H. pylori vaccine consisting of B- and T-cell epitopes from six conserved virulence factors of the bacteria.

Aims & Methods

Through comprehensive bioinformatic analysis, sixteen B- and T-cell epitopes were carefully selected, considering both MHC classes I and II for the T-cell epitopes. The recombinant Multi-Epitope-Unit antigen (MEU) was combined with flagellin as adjuvant. Additionally, recombinant Modified Vaccinia virus Ankara (MVA) encoding the MEU antigen was generated. The vaccine candidates were tested in mice. To this end, two sets of animal experiments were designed. One set to evaluate the immunogenicity of the new MEU construct, while the second set to assess the protective effect of the vaccine candidates. In the vaccination protocol, both protein- and MVA-based vaccine candidates were administered intramuscularly (IM) to C57/Bl6 mice on days 0, 7 and 14. Two weeks after the last immunization, mice in the prevention groups were challenged with three doses of the mouse adapted H. pylori SS1 strain. The groups, allocated for evaluating the immunogenicity of the new vaccine, were not exposed to infection with H. pylori or any further treatments prior to the analysis. Humoral (antibody) and cellular immune responses, along with the levels of protection and infection, were evaluated at two or four weeks post-immunization or -infection respectively. Bacterial load in prevention groups was determined through culturing of the gastric samples. The reactivity of anti-MEU antibodies, raised in immunized animals, to the H. pylori clinical isolates was also assessed via ELISA employing crude antigens of a set of 30 clinical isolates.

Results

The MEU antigen, whether adjuvanted with flagellin or in MVA construct, was observed to elicit significant antibody levels in mice, capable of reacting with crude antigens from H. pylori. The most substantial antibody response was detected towards the individual epitopes from Urease and FliD (hook-associated protein 2 homologue).
Both vaccine protocols, using the flagellin-adjuvanted MEU and the recombinant MVA, demonstrated complete protection against H. pylori infection compared to non-immunized animals. This protection was achieved through either a schedule of immunization with the protein-based vaccine candidate followed by MVA boost or two administered doses of MVA.

Conclusion

The novel MEU-based vaccine demonstrated significant immunoreactivity against H. pylori antigens. Our findings indicate a complete protection against H. pylori infection in the mice immunized with the multi-epitope unit antigen, regardless of the vaccination protocols. The reactivity of these antibodies with the bacteria native crude antigens suggests the antigen’s correct processing by the mammalian immune system. Moving forward, our next step involves assessing the therapeutics effects of the vaccine candidate.

Introduction

Oesophageal squamous cell carcinoma (OSCC) represents approximately 90% of the nearly half-million annual cases of oesophageal cancer globally. The invasiveness and high costs associated with traditional upper endoscopy limit its use in the routine surveillance of at-risk populations, including individuals with a history of head and neck cancer (HNC). Previous data suggest that copy number abnormalities (CNAs) are a feature of OSCC development.

Aims & Methods

This study explored the efficacy of shallow whole-genome sequencing (sWGS) to detect genome-wide CNAs in cells collected using a non-endoscopic sponge pan-oesophageal sampling device. Our objective was to develop and validate a predictive model for the early detection of oesophageal squamous dysplasia and neoplasia, aiming to improve disease prognosis through timely intervention.
We prospectively collected oesophageal specimens from 178 participants using the Cytosponge device as part of the EDEN and OSCAR projects: OSCC (n = 24), high-risk individuals (n = 61), and healthy controls with dyspeptic symptoms (n = 93). The samples underwent 1.5-2x sWGS as well as H&E staining for atypia assessment and p53 immunohistochemistry (IHC) which were performed by two gastrointestinal (GI) pathologists who were blinded to the reference endoscopic diagnoses. Through penalised regression analysis, we identified significant CNAs indicative of dysplasia or early neoplasia. These critical genomic insights were then used to develop and refine a logistic regression model aimed at predicting dysplasia or early neoplasia in cytosponge-derived specimens. The predictive accuracy of the model was subsequently confirmed through validation against sWGS in microdissected ESD samples which represented known pathological states.

Results

Key genomic indicators for oesophageal squamous dysplasia and early neoplasia were identified, including significant CNAs in chromosome arms chr2q, chr3q, chr9p, and chr11q along with the Genome-Wide Instability Score (GWIS), a percentage-based metric quantifying genomic instability across the genome. Our logistic regression model, which integrated these genomic biomarkers, achieved an Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.947, with sensitivity and specificity rates of 89% and 88.2%, respectively. This model outperformed another that utilised only pathological data (definite squamous atypia and p53 IHC), which achieved an AUROC of 0.880, with a sensitivity and specificity of 83.6% and 84.6%, respectively. However, when pathological data were combined with the genomic indicators, the diagnostic performance of our model improved significantly, resulting in an AUROC of 0.969 with enhanced sensitivity and specificity of 95.9% and 90.5%, respectively. Further validation of the genomic model against microdissected ESD samples with known pathological states reinforced its robustness, achieving an AUROC of 0.879, which improved to 0.905 when low grade dysplasia (LGD) samples were excluded, highlighting the diagnostic challenges and variability in LGD classification.

Conclusion

The integration of shallow whole-genome sequencing with non-endoscopic sponge sampling has the potential for the early detection of OSCC. This approach is particularly valuable in high-risk populations and settings where traditional endoscopy is impractical.