The Gleason score is the most important prognostic marker for prostate cancer patients, but it suffers from significant observer variability. Artificial intelligence (AI) systems based on deep learning can achieve pathologist-level performance at Gleason grading. However, the performance of such systems can degrade in the presence of artifacts, foreign tissue, or other anomalies. Pathologists integrating their expertise with feedback from an AI system could result in a synergy that outperforms both the individual pathologist and the system. Despite the hype around AI assistance, existing literature on this topic within the pathology domain is limited. We investigated the value of AI assistance for grading prostate biopsies. A panel of 14 observers graded 160 biopsies with and without AI assistance. Using AI, the agreement of the panel with an expert reference standard increased significantly (quadratically weighted Cohen's kappa, 0.799 vs. 0.872; p = 0.019). On an external validation set of 87 cases, the panel showed a significant increase in agreement with a panel of international experts in prostate pathology (quadratically weighted Cohen's kappa, 0.733 vs. 0.786; p = 0.003). In both experiments, on a grouplevel, AI-assisted pathologists outperformed the unassisted pathologists and the standalone AI system. Our results show the potential of AI systems for Gleason grading, but more importantly, show the benefits of pathologist-AI synergy. Members of the ISUP Pathology Imagebase Expert Panel are listed below Acknowledgements.

Background The literature on the prognostic relevance of signet-ring cell (SRC) histology in gastric cancer (GC) is controversial which is most likely related to inconsistent SRC classification based on haematoxylin-eosin staining. We hypothesised that mucin stains can consistently identify SRC-GC and predict GC patient outcome. Methods We performed a comprehensive literature review on mucin stains in SRC-GC and characterised the mucin expression in 851 Caucasian GC and 410 Asian GC using Alcian Blue (AB)-Periodic Acid-Schiff (PAS), MUC2 (intestinal-type mucin), and MUC5AC (gastric-type mucin). The relationship between mucin expression and histological phenotype [poorly cohesive (PC) including proportion of SRCs, non-poorly cohesive (non-PC), or mucinous (MC)], clinicopathological variables, and patient outcome was analysed. Results Depending on mucin expression and cutoffs , the positivity rates of SRC-GC reported in the literature varied from 6 to 100%. Patients with MUC2 positive SRC-GC or SRC-GC with (gastro)intestinal phenotype had poorest outcome. In our cohort study, PC with ≥ 10% SRCs expressed more frequently MUC2, MUC5AC, and ABPAS (p < 0.001, p = 0.004 and p < 0.001, respectively). Caucasians with AB positive GC or combined ABPAS-MUC2 positive and MUC5AC negative had poorest outcome (all p = 0.002). This association was not seen in Asian patients. Conclusions This is the first study to suggest that mucin stains do not help to differentiate between SRC-GC and non-SRC-GC. However, mucin stains appear to be able to identify GC patients with different outcome. To our surprise, the relationship between outcome and mucin expression seems to differ between Caucasian and Asian GC patients which warrants further investigations.

Introduction Targeted therapy against tumor angiogenesis is widely used in clinical practice for patients with colorectal liver metastases (CRLM). Possible predictive biomarkers for tumor angiogenesis, such as, microvessel density (MVD), hypoxia and cell proliferation, can be determined using immunohistochemical staining. However, patients ineligible for surgical treatment need to undergo invasive diagnostic interventions in order to determine these biomarkers. CT perfusion (CTP) is an emerging functional imaging technique, which can non-invasively determine vascular properties of solid tumors. The purpose of this study was to evaluate CTP with histological biomarkers in CRLM. Material and methods Patients with CRLM underwent CTP one day before liver surgery. CTP analysis was performed on the entire volume of the largest metastases in each patient. Dual-input maximum slope analysis was used and data concerning arterial flow (AF), portal flow (PF) and perfusion index (PI) were recorded. Immunohistochemical staining with CD34, M75/CA-IX and MIB-1 was performed on the rim in the midsection of the tumor to determine respectively MVD, hypoxia and cell proliferation. Results Twenty CRLM in 20 patients were studied. Mean size of the largest CRLM was 37 mm (95% CI 21–54 mm). Mean AF and PF were respectively 64 ml/min/100ml (95% CI 48–79) and 30 ml/min/100ml (95% CI 22–38). Mean PI was 68% (95% CI 62–73). No significant correlation was found between tumor growth patterns and CTP (p = 0.95). MVD did not significantly correlate to AF (r = 0.05; p = 0.84), PF (r = 0.17; p = 0.47) and PI (r = -0.12; p = 0.63). Cell proliferation also did not significantly correlate to AF (r = 0.07; p = 0.78), PF (r = -0.01; p = 0.95) and PI (r = 0.15; p = 0.52). Hypoxia did not significantly correlate to AF (r = -0.05; p = 0.83), however, significantly to PF (r = 0.51; p = 0.02) and a trend to negative correlation with PF (r = -0.43; p = 0.06). However, after controlling the false discovery rate, no significant correlation between CTP and used immunohistochemical biomarkers was found. Conclusion In conclusion, this feasibility study found a trend to negative correlation between PI and hypoxia, CTP might therefore possibly evaluate this prognostic marker in CRLM non-invasively. However, CTP is not an appropriate technique for the assessment of microvessels or cell proliferation in CRLM.

Oesophageal adenocarcinomas may show different histopathological patterns, including excessive acellular mucin pools, signet‐ring cells (SRCs), and poorly cohesive cells (PCCs). These components have been suggested to correlate with poor outcomes after neoadjuvant chemoradiotherapy (nCRT), which might influence patient management. However, these factors have not been studied independently of each other with adjustment for tumour differentiation grade (i.e. the presence of well‐formed glands), which is a possible confounder. We studied the pre‐ and post‐treatment presence of extracellular mucin, SRCs, and/or PCCs in relation to pathological response and prognosis after nCRT in patients with oesophageal or oesophagogastric junction adenocarcinoma. A total of 325 patients were retrospectively identified from institutional databases of two university hospitals. All patients were scheduled for ChemoRadiotherapy for Oesophageal cancer followed by Surgery Study (CROSS) nCRT and oesophagectomy between 2001 and 2019. Percentages of well‐formed glands, extracellular mucin, SRCs, and PCCs were scored in pre‐treatment biopsies and post‐treatment resection specimens. The association between histopathological factors (≥1 and >10%) and tumour regression grade 3–4 (i.e. >10% residual tumour), overall survival, and disease‐free survival (DFS) was evaluated, adjusted for tumour differentiation grade amongst other clinicopathological variables. In pre‐treatment biopsies, ≥1% extracellular mucin was present in 66 of 325 patients (20%); ≥1% SRCs in 43 of 325 (13%), and ≥1% PCCs in 126 of 325 (39%). We show that pre‐treatment histopathological factors were unrelated to tumour regression grade. Pre‐treatment presence of >10% PCCs was associated with lower DFS (hazard ratio [HR] 1.73, 95% CI 1.19–2.53). Patients with post‐treatment presence of ≥1% SRCs had higher risk of death (HR 1.81, 95% CI 1.10–2.99). In conclusion, pre‐treatment presence of extracellular mucin, SRCs, and/or PCCs is unrelated to pathological response. The presence of these factors should not be an argument to refrain from CROSS. At least 10% PCCs pre‐treatment and any SRCs post‐treatment, irrespective of the tumour differentiation grade, seem indicative of inferior prognosis, but require further validation in larger cohorts.