Search

The integration of artificial intelligence (AI) in medical diagnostics represents a significant advancement in managing upper gastrointestinal (GI) cancer, which is a major cause of global cancer mortality. Specifically for gastric cancer (GC), chronic inflammation causes changes in the mucosa such as atrophy, intestinal metaplasia (IM), dysplasia, and ultimately cancer. Early detection through endoscopic regular surveillance is essential for better outcomes. Foundation models (FMs), which are machine or deep learning models trained on diverse data and applicable to broad use cases, offer a promising solution to enhance the accuracy of endoscopy and its subsequent pathology image analysis. This review explores the recent advancements, applications, and challenges associated with FMs in endoscopy and pathology imaging. We started by elucidating the core principles and architectures underlying these models, including their training methodologies and the pivotal role of large-scale data in developing their predictive capabilities. Moreover, this work discusses emerging trends and future research directions, emphasizing the integration of multimodal data, the development of more robust and equitable models, and the potential for real-time diagnostic support. This review aims to provide a roadmap for researchers and practitioners in navigating the complexities of incorporating FMs into clinical practice for the prevention/management of GC cases, thereby improving patient outcomes.

Purpose This study investigated the safety and clinical activity of lumretuzumab, a humanised antihuman epidermal growth factor receptor 3 (HER3) monoclonal antibody, in combination with carboplatin and paclitaxel in first-line treatment of patients with squamous non-small cell lung cancer (sqNSCLC). HER3 ligand heregulin and HER3 protein expression were evaluated as potential biomarkers of clinical activity.Patients and methods This open-label, phase Ib/II study enrolled patients receiving lumretuzumab at 800 mg (flat) in combination with carboplatin (area under the curve (AUC) 6 mg/mL×min) and paclitaxel (200 mg/m2) administered intravenously on a every 3-week schedule. Adverse event (AE) rates and tumour responses were determined. Heregulin messenger RNA (mRNA) and HER3 protein expression were investigated in archival tumour biopsies.Results Altogether, 12 patients received lumretuzumab in combination with carboplatin and paclitaxel. The most frequent AEs were gastrointestinal, haematological and nervous system toxicities, which were generally mild and manageable. Partial responses were observed in 3 of 12 patients lasting 81, 177 and 207 days. All responses were achieved in tumours expressing higher heregulin mRNA levels.Conclusion Lumretuzumab in combination with carboplatin and paclitaxel was well tolerated. Objective responses were enriched in tumours expressing higher heregulin mRNA levels.

Abstract Somatic mutation analysis and evaluation of microsatellite instability (MSI) have become mandatory for selecting personalized therapy strategies for advanced colorectal cancer and are not available as routine methods in Paraguay. The aims of this study were to analyze the molecular profile as well as the microsatellite status in a series of advanced colorectal patients from two public hospitals from Paraguay, to introduce these methodologies in the routine practice to guide the therapeutic decisions. Thirty‐six patients diagnosed with advanced colorectal cancer from two referent public hospitals from Paraguay were recruited from May 2017 to February 2018. Sequenom Mass spectrometry, Oncocarta Panel V.1 was applied to analyze the mutational profile from formalin‐fixed paraffin‐embedded samples. The microsatellite status was tested by immunohistochemistry (IHC). The mean age of the patients was 52 years with a range from 20 to 74 years. Eighty‐three percent of the patients included in the study have advanced‐stage tumors at the moment of the diagnosis. Sixteen patients (44.4%) were wild‐type for all the oncogene regions analyzed with the Oncocarta panel. Thirty‐two hot‐spot pathogenic variants on seven oncogenes, among 20 patients (55.6%), were identified, including KRAS, NRAS, BRAF, PI3KCA, FGFR, epidermal growth factor receptor, and PDGFRA. Moreover, 14 (38.8%) of these patients presented pathogenic variants in KRAS/NRAS or BRAF genes that have implications in the clinical practice decisions. Five patients (14%) presented MSI. The IHC study for microsatellite status and the molecular profile analysis through Sequenom mass spectrometry are feasible and useful methods, due to identify those patient candidates for targeted therapies and for the budgetary calculations of the National Health Plans.

Background The molecular classification of gastric cancer recognises two subtypes prone to immune checkpoint blockade: the microsatellite unstable and the Epstein-Barr virus (EBV)-related tumours. We aim to assess the concordance between immunohistochemistry and PCR for microsatellite status evaluation, and explore the value of microsatellite instability (MSI) and EBV as predictive survival factors.Material and methods We collected 246 consecutively diagnosed gastric cancer cases in all stages and evaluated the microsatellite status using immunohistochemistry for mismatched repair (MMR) proteins and PCR. EBV expression was studied through in situ hybridisation.Results Forty-five (18%) cases presented MSI and 13 (6%) were positive for EBV. MSI was associated with female sex, older age, distal location and distal non-diffuse type of the modified Lauren classification. EBV expression was most frequent in proximal location and proximal non-diffuse type. The sensitivity, specificity, positive predictive value and negative predictive value of immunohistochemistry for the microsatellite study were 91%, 98%, 91% and 98%, respectively. In the multivariate analysis, MSI was an independent predictor of favourable tumour-specific survival (TSS) in stages I–III (MSI: HR: 0.37, 95% CI 0.12 to 0.95, p=0.04).Conclusions The MSI status and the EBV expression should be incorporated in routine pathological report for two reasons. First, MSI defines a different pathological entity with a better outcome. Second, MSI and EBV may be useful biomarkers to identify patients who will respond to immune checkpoint blockade inhibitors. For this purpose, immunohistochemical study for MMR proteins and in situ hybridisation study for EBV evaluation are feasible and cost-effective methods.