Stimulated Raman Histology (SRH) employs the stimulated Raman scattering (SRS) of photons at biomolecules in tissue samples to generate histological images. Subsequent pathological analysis allows for an intraoperative evaluation without the need for sectioning and staining. The objective of this study was to investigate a deep learning-based classification of oral squamous cell carcinoma (OSCC) and the sub-classification of non-malignant tissue types, as well as to compare the performances of the classifier between SRS and SRH images. Raman shifts were measured at wavenumbers k₁ = 2845 cm⁻¹ and k₂ = 2930 cm⁻¹. SRS images were transformed into SRH images resembling traditional H&E-stained frozen sections. The annotation of 6 tissue types was performed on images obtained from 80 tissue samples from eight OSCC patients. A VGG19-based convolutional neural network was then trained on 64 SRS images (and corresponding SRH images) and tested on 16. A balanced accuracy of 0.90 (0.87 for SRH images) and F1-scores of 0.91 (0.91 for SRH) for stroma, 0.98 (0.96 for SRH) for adipose tissue, 0.90 (0.87 for SRH) for squamous epithelium, 0.92 (0.76 for SRH) for muscle, 0.87 (0.90 for SRH) for glandular tissue, and 0.88 (0.87 for SRH) for tumor were achieved. The results of this study demonstrate the suitability of deep learning for the intraoperative identification of tissue types directly on SRS and SRH images.

Hepatocellular carcinoma (HCC) is the fastest growing cause of cancer-related mortality and there are limited therapies ¹ . Although endoplasmic reticulum (ER) stress and the unfolded protein response (UPR) are implicated in HCC, the involvement of the UPR transducer ATF6α remains unclear ² . Here we demonstrate the function of ATF6α as an ER-stress-inducing tumour driver and metabolic master regulator restricting cancer immunosurveillance for HCC, in contrast to its well-characterized role as an adaptive response to ER stress ³ . ATF6α activation in human HCC is significantly correlated with an aggressive tumour phenotype, characterized by reduced patient survival, enhanced tumour progression and local immunosuppression. Hepatocyte-specific ATF6α activation in mice induced progressive hepatitis with ER stress, immunosuppression and hepatocyte proliferation. Concomitantly, activated ATF6α increased glycolysis and directly repressed the gluconeogenic enzyme FBP1 by binding to gene regulatory elements. Restoring FBP1 expression limited ATF6α-activation-related pathologies. Prolonged ATF6α activation in hepatocytes triggered hepatocarcinogenesis, intratumoural T cell infiltration and nutrient-deprived immune exhaustion. Immune checkpoint blockade (ICB) ⁴ restored immunosurveillance and reduced HCC. Consistently, patients with HCC who achieved a complete response to immunotherapy displayed significantly increased ATF6α activation compared with those with a weaker response. Targeting Atf6 through germline ablation, hepatocyte-specific ablation or therapeutic hepatocyte delivery of antisense oligonucleotides dampened HCC in preclinical liver cancer models. Thus, prolonged ATF6α activation drives ER stress, leading to glycolysis-dependent immunosuppression in liver cancer and sensitizing to ICB. Our findings suggest that persistently activated ATF6α is a tumour driver, a potential stratification marker for ICB response and a therapeutic target for HCC.

We describe a sequential multistaining protocol for immunohistochemistry, immunofluorescence and CyTOF imaging for formalin-fixed, paraffin-embedded specimens (FFPE) in the formalin gas-phase (FOLGAS), enabling sequential multistaining, independent from the primary and secondary antibodies and retrieval. Histomorphologic details are preserved, and crossreactivity and loss of signal intensity are not detectable. Combined with a DAB-based hydrophobic masking of metal-labeled primary antibodies, FOLGAS allows the extended use of CyTOF imaging in FFPE sections.

Background/Objectives: Purinergic signaling, which involves extracellular ATP (eATP), its metabolites, purinergic receptors and ectonucleotidases, plays a pivotal role in the tumor microenvironment (TME), impacting tumor progression and the antineoplastic immune response. In this study, the CD39, CD73, P2X4, and P2X7 expression in NSCLC tumor cells and the surrounding stroma of 139 resected patients was examined. Methods: The study included tissue samples from 139 NSCLC patients. Tissue microarrays (TMAs) were constructed using 1.0 mm cores from annotated tumor regions. Immunohistochemical staining for CD39, CD73, P2X4, and P2X4 was performed on 2 µm sections. TMA slides were digitized and analyzed with QuPath, where staining intensity was evaluated using a semi-quantitative H-score. Statistical analysis, including survival analysis, was performed using R, to assess the impact of biomarker expression on patient outcomes. Results: High CD39 expression in both tumor and stromal cells was significantly associated with prolonged PFS (respectively: p = 0.0058 and p = 0.0067), particularly in adenocarcinoma (ADC) patients (respectively: p = 0.01 and p = 0.023). In the multivariable Cox model, low CD73 expression in tumor cells correlated with longer PFS (HR: 0.47; 95% CI: [0.28, 0.8], p = 0.005), while low CD73 expression in stromal cells was linked to increased progression risk (HR: 4.81; 95% CI: [1.61, 14.4], p = 0.001). Neither P2X7 nor P2X4 demonstrated a consistent effect on PFS in univariable analyses; however, multivariable analyses suggested that P2X4 might play a prognostic role in NSCLCs (HR: 0.37; 95% CI: [0.19, 0.73], p = 0.003). Conclusions: These findings underscore the importance of purinergic signaling in NSCLC prognosis and highlight the role of the ectonucleotidases CD39 and CD73 as potential therapeutic targets to enhance antineoplastic immune responses.

Immunohistochemical (IHC) studies of formalin-fixed paraffin-embedded (FFPE) samples are a gold standard in oncology for tumor characterization, and the identification of prognostic and predictive markers. However, despite the abundance of archived FFPE samples, their research use is limited due to the labor-intensive nature of IHC on large cohorts. This study aimed to create a high-throughput workflow using modern technologies to facilitate IHC biomarker studies on large patient groups. Semiautomatic constructed tissue microarrays (TMAs) were created for two tumor patient cohorts and IHC stained for seven antibodies (ABs). AB expression in the tumor and surrounding stroma was quantified using the AI-supported image analysis software QuPath. The data were correlated with clinicopathological information using an R-script, all results were automatically compiled into formatted reports. By minimizing labor time to 7.7%—compared to whole-slide studies—the established workflow significantly reduced human and material resource consumption. It successfully correlated AB expression with overall patient survival and additional clinicopathological data, providing publication-ready figures and tables. The AI-assisted high-throughput TMA workflow, validated on two patient cohorts, streamlines modern histopathological research by offering cost and time efficiency compared to traditional whole-slide studies. It maintains research quality and preserves patient tissue while significantly reducing material and human resources, making it ideal for high-throughput research centers and collaborations.

Intrahepatic cholangiocarcinoma (ICC) features poor survival due to frequent recurrences and limited prognostic markers. Using mass spectrometry-based proteomics, we analyze two independent cohorts comprising 80 and 62 treatment-naive ICC tumors, along with 9 independent patient-derived xenografts (PDX). In the first cohort, we identify two subclusters with distinct times-to-recurrence (TTR): An extracellular matrix (ECM)-enriched cluster (mean TTR 859 days) and a proliferation cluster (mean TTR 229 days). A 4-protein classifier trained on our cohort accurately stratifies these clusters in the Dong et al. dataset (2022) and in our second cohort, revealing similar proteomic motifs and clinical outcomes. The translation regulator EIF4A1, enriched in ICCs of both clusters, emerges as a therapeutic target, as its inhibition with eFT226 significantly reduces tumor growth in an ICC PDX model. Proteomic analyses of various PDX models also emphasize the critical role of tumor-stroma interactions in ICC. Overall, this study establishes two prognostic proteomic clusters, validates their relevance across datasets, and highlights EIF4A1 inhibition as a potential therapeutic strategy.