Background Tumour microenvironment, including tumour–stroma ratio (TSR), might help identifying high-risk cancer patients. However, the significance of TSR in gastric cancer is unclear, especially in the intestinal and diffuse subtypes. The aim of this study was to investigate the tumour–stroma ratio in gastric adenocarcinoma, and its intestinal and diffuse histological subtypes, in relation to prognosis. Methods Five hundred and eighty-three gastric adenocarcinoma patients who underwent surgery in Oulu University hospital during years 1983–2016 were included in this retrospective cohort study. TSR was analysed from the slides that were originally used for diagnostic purposes. Patients were divided into stroma-poor (≤50% stroma) and stroma-rich (>50% stroma) groups and TSR was analysed in relation to 5-year mortality and overall mortality. Results Patients with stroma-rich tumours had worse 5-year prognosis (HR 1.80, 95% CI 1.41–2.28) compared to stroma-poor tumours. Stratified analysis showed that stroma-rich tumours had worse 5-year prognosis in both intestinal (HR 1.68, 95% CI 1.24–2.27) and diffuse histological types (HR 2.09, 95% CI 1.35–3.23) compared to stroma-poor tumours, respectively. Conclusions High proportion of stroma is an independent prognostic factor in both intestinal and diffuse histological subtypes of gastric adenocarcinoma.

Tumour-stroma ratio (TSR) is a novel potential prognostic factor in cancers and based on the proportions of stroma and tumour area. The prognostic value of TSR in gastric cancer is incompletely known. The aim of this study was to estimate prognostic significance of TSR in gastric adenocarcinoma. A search of PubMed (MEDLINE), Web of Science, EMBASE, Cochrane and Scopus databases was performed. A meta-analysis was conducted on five-year survival in gastric cancer patients using inverse variance random-effects methods. The literature search yielded 5329 potential titles, of which a total of seven studies were eligible. Results of six studies including a total of 1779 patients were pooled in the meta-analysis. Only 23 (1.3%) of the patients received neoadjuvant therapy. All six studies had a cut-off of 50% for the proportion of stroma when dividing the patients into low- and high stroma groups. Low TSR (high amount of stroma) was strongly associated with increased five-year mortality (hazard ratio 2.19, 95% CI 1.69–2.85). In conclusion, TSR is a strong prognostic factor in gastric cancer. It could be used to estimate prognosis of gastric cancer patients not receiving neoadjuvant chemotherapy. Further studies including patients receiving neoadjuvant therapy are recommended.

Tertiary lymphoid structures (TLSs) are part of immune response against cancer. Their high density and high diameter have been shown to be associated with prognosis in different cancer types. The aim of this study was to examine the prognostic significance of TLS density and diameter in gastric cancer and reproducibility of their assessments. TLS densities and maximal TLS diameter were assessed from hematoxylin–eosin (HE) stained slides of 721 surgically treated gastric cancer patients from two hospitals in Finland. Mortality hazard ratios (HRs) for TLS densities and maximal TLS diameter were analyzed. TLS densities and maximal TLS diameter were assessed with moderate interobserver agreement (Cohen's kappa 0.50–0.62). Maximal TLS density was not associated with survival (adjusted HR 0.85, 95% CI 0.70–1.02) and neither was hotspot TLS density (adjusted HR 0.85, 95% CI 0.70–1.02). High maximal TLS diameter was associated with longer survival in overall study population (adjusted HR 0.74, 95% CI 0.61–0.89) and in diffuse type subgroup (adjusted HR 0.65, 95% CI 0.50–0.85). In conclusion, high maximal TLS diameter is associated with improved survival in gastric cancer and can be assessed from HE‐stained slides. Its prognostic value might be limited to diffuse histological type.

Histological analysis with microscopy is the gold standard to diagnose and stage cancer, where slides or whole slide images are analyzed for cell morphological and spatial features by pathologists. The nuclei of cancerous cells are characterized by nonuniform chromatin distribution, irregular shapes, and varying size. As nucleus area and shape alone carry prognostic value, detection and segmentation of nuclei are among the most important steps in disease grading. However, evaluation of nuclei is a laborious, time-consuming, and subjective process with large variation among pathologists. Recent advances in digital pathology have allowed significant applications in nuclei detection, segmentation, and classification, but automated image analysis is greatly affected by staining factors, scanner variability, and imaging artifacts, requiring robust image preprocessing, normalization, and segmentation methods for clinically satisfactory results. In this paper, we aimed to evaluate and compare the digital image analysis techniques used in clinical pathology and research in the setting of gastric cancer. A literature review was conducted to evaluate potential methods of improving nuclei detection. Digitized images of 35 patients from a retrospective cohort of gastric adenocarcinoma at Oulu University Hospital in 1987-2016 were annotated for nuclei (n = 9085) by expert pathologists and 14 images of different cancer types from public TCGA dataset with annotated nuclei (n = 7000) were used as a comparison to evaluate applicability in other cancer types. The detection and segmentation accuracy with the selected color normalization and stain separation techniques were compared between the methods. The extracted information can be supplemented by patient's medical data and fed to the existing statistical clinical tools or subjected to subsequent AI-assisted classification and prediction models. The performance of each method is evaluated by several metrics against the annotations done by expert pathologists. The F1-measure of 0.854 ± 0.068 is achieved with color normalization for the gastric cancer dataset, and 0.907 ± 0.044 with color deconvolution for the public dataset, showing comparable results to the earlier state-of-the-art works. The developed techniques serve as a basis for further research on application and interpretability of AI-assisted tools for gastric cancer diagnosis.

Purpose: To examine and compare the prognostic value of immune cell score (ICS) and Klintrup–Mäkinen (KM) grade in gastric cancer. Methods: Gastric adenocarcinoma tissues from samples of 741 patients surgically treated in two hospitals in Finland were assessed for ICS and KM grade. Cox regression with adjustment for confounders provided hazard ratios (HRs) and 95% CIs. Subgroup analyses were performed in intestinal and diffuse type subgroups. The primary outcome was 5-year overall survival. Results: High ICS was associated to longer 5-year survival (adjusted HR 0.70, 95% CI 0.52–0.94), compared to low ICS. The difference was significant in intestinal type subgroup (adjusted HR 0.54, 95% CI 0.36–0.81) but not in diffuse type subgroup (adjusted HR 0.92, 95% CI 0.58–1.46). High KM grade was an independent prognostic factor for longer 5-year overall survival (adjusted HR 0.59, 95% CI 0.45–0.77) in both intestinal (adjusted HR 0.61, 95% CI 0.44–0.85) and diffuse subgroups (adjusted HR 0.52, 95% CI 0.31–0.86). ICS and KM grade were moderately correlated (ρ = 0.425). When both immune cell score and KM grade were included in the regression analysis, only KM grade remained prognostic. Conclusions: Both ICS and KM grade are prognostic factors in gastric adenocarcinoma, but immunohistochemistry-based ICS might not have additional prognostic value over hematoxylin–eosin-based KM grade.