Diagnosis of adenocarcinoma in the liver is a frequent scenario in routine pathology and has a critical impact on clinical decision-making. However, rendering a correct diagnosis can be challenging and often requires the integration of clinical, radiological, and immunohistochemical information. We present a deep learning model (HEPNET) to distinguish intrahepatic cholangiocarcinoma (iCCA) from colorectal liver metastasis (CRM) as the most frequent primary and secondary forms of liver adenocarcinoma with clinical-grade accuracy using hematoxylin and eosin-stained whole-slide images. HEPNET was trained on 714 589 image tiles from 456 patients who were randomly selected in a stratified manner from a pool of 571 patients who underwent surgical resection or biopsy at Heidelberg University Hospital. Model performance was evaluated on a hold-out internal test set comprising 115 patients and externally validated on 159 patients recruited at Mainz University Hospital. On the hold-out internal test set, HEPNET achieved an area under the receiver operating characteristic curve (AUROC) of 0.994 (95% CI 0.989-1.000) and an accuracy of 96.522% (95% CI 94.521-98.694%) at the patient level. Validation on the external test set yielded an AUROC of 0.997 (95% CI 0.995-1.000), corresponding to an accuracy of 98.113% (95% CI 96.907-100.000%). HEPNET surpassed the performance of six pathology experts with different levels of experience in a reader study of 50 patients (P=.0005), boosted the performance of resident pathologists to the level of senior pathologists, and reduced potential downstream analyses. Here, we provide a ready-to-use tool with a clinical-grade performance that may facilitate routine pathology by rendering a definitive diagnosis and guiding ancillary testing. The incorporation of HEPNET into pathology laboratories may optimize the diagnostic workflow, complemented by test-related labor and cost savings.

Inhibition of the programmed cell death protein-1/ligand-1 (PD-1/PD-L1) axis has opened a new era in the treatment of solid cancers. However, there is no data on the expression and relevance of PD-L1 in Western gallbladder cancer (GBC). We assessed PD-L1 immunohistochemically in 131 GBC patients as Tumor Proportion Score (TPS), Immune Cell Score (IC) and Combined Positivity Score (CPS). Tumor cells expressed PD-L1 in a subset of 14.7% GBC patients at a TPS cut-off of 1%. Higher PD-L1 levels above 10% and 25% TPS were reached in 4.7% and 3.1% of GBC cases, respectively. At a 10% cut-off, TPS was associated with distinct histomorphological subtypes and correlated with poor tumor differentiation. Survival analysis revealed a TPS above 10% to be a highly significant and independent negative prognosticator in GBC. PD-L1 expression was associated with increased CD4⁺, CD8⁺ and PD-1⁺ immune cell densities. In 14.8% of the cases, scattered immune cells expressed T-cell immunoreceptor with Ig and ITIM domains (TIGIT), which was correlated to tumoral expression of its ligand CD155. We here show that a high PD-L1 expression confers a negative prognostic value in Western-world GBC and highlight the TIGIT/CD155 immune checkpoint as a potential new target for GBC immunotherapy.

Aims Intrahepatic cholangiocarcinoma (iCCA) is a diagnosis of exclusion that can pose a challenge to the pathologist despite thorough clinical workup. Although several immunohistochemical markers have been proposed for iCCA, none of them reached clinical practice. We here assessed the combined usage of two promising diagnostic approaches, albumin in situ hybridisation (Alb-ISH) and C reactive protein (CRP) immunohistochemistry, for distinguishing iCCA from other adenocarcinoma primaries. Methods We conducted Alb-ISH and CRP immunohistochemistry in a large European iCCA cohort (n=153) and compared the results with a spectrum of other glandular adenocarcinomas of different origin (n=885). In addition, we correlated expression patterns with clinicopathological information and mutation data. Results Alb-ISH was highly specific for iCCA (specificity 98.8%) with almost complete negativity in perihilar CCA and only rare positives among other adenocarcinomas (sensitivity 69.5%). CRP identified the vast majority of iCCA cases (sensitivity 84.1%) at a lower specificity of 86.4%. Strikingly, the combination of CRP and Alb-ISH boosted the diagnostic sensitivity to 88.0% while retaining a considerable specificity of 86.1%. Alb-ISH significantly correlated with CRP expression, specific tumour morphologies and small or large duct iCCA subtypes. Neither Alb-ISH nor CRP was associated with iCCA patient survival. 16 of 17 recurrent mutations in either IDH1, IDH2 and FGFR2 affected Alb-ISH positive cases, while the only KRAS mutation corresponded to an Alb-ISH negative case. Conclusions In conclusion, we propose a sequential diagnostic approach for iCCA, integrating CRP immunohistochemistry and Alb-ISH. This may improve the accuracy of CCA classification and pave the way towards a molecular-guided CCA classification.

Cholangiocarcinoma (CCA) is a heterogeneous malignancy with a dismal prognosis. Therapeutic options are largely limited to surgery and conventional chemotherapy offers limited benefit. As immunotherapy has proven highly effective in various cancer types, we have undertaken a quantitative immunohistopathological assessment of immune cells expressing the immunoinhibitory T cell immune response cDNA 7 receptor (TIRC7), an emerging immunoinhibitory receptor, in a cohort of 135 CCA patients. TIRC7⁺ immune cells were present in both the tumor epithelia and stroma in the majority of CCA cases with the highest levels found in intrahepatic CCA. While intraepithelial density of TIRC7⁺ immune cells was decreased compared to matched non-neoplastic bile ducts, stromal quantity was higher in the tumor samples. Tumors exhibiting signet ring cell or adenosquamous morphology were exclusively associated with an intraepithelial TIRC7⁺ phenotype. Survival analysis showed intraepithelial TIRC7⁺ immune cell density to be a highly significant favorable prognosticator in intrahepatic but not proximal or distal CCA. Furthermore, intraepithelial TIRC7⁺ immune cell density correlated with the number of intraepithelial CD8⁺ immune cells and with the total number of CD4⁺ immune cells. Our results suggest the presence and prognostic relevance of TIRC7⁺ immune cells in CCA and warrant further functional studies on its pharmacological modulation.