Purpose: The heavy chain of the CD98 protein (CD98hc) is encoded by the SLC3A2 gene. Together with the light subunit LAT1, CD98hc constitutes a heterodimeric transmembrane amino acid transporter. High SLC3A2 mRNA expression levels are associated with poor prognosis in patients with head and neck squamous cell carcinoma (HNSCC) treated with radiochemotherapy. Little is known regarding the CD98hc protein–mediated molecular mechanisms of tumor radioresistance. Experimental Design: CD98hc protein expression levels were correlated with corresponding tumor control dose 50 (TCD50) in HNSCC xenograft models. Expression levels of CD98hc and LAT1 in HNSCC cells were modulated by siRNA or CRISPR/Cas9 gene editing. HNSCC cell phenotypes were characterized by transcription profiling, plasma membrane proteomics, metabolic analysis, and signaling pathway activation. Expression levels of CD98hc and LAT1 proteins were examined by IHC analysis of tumor tissues from patients with locally advanced HNSCC treated with primary radiochemotherapy (RCTx). Primary endpoint was locoregional tumor control (LRC). Results: High expression levels of CD98hc resulted in an increase in mTOR pathway activation, amino acid metabolism, and DNA repair as well as downregulation of oxidative stress and autophagy. High expression levels of CD98hc and LAT1 proteins were significantly correlated and associated with an increase in radioresistance in HNSCC in vitro and in vivo models. High expression of both proteins identified a poor prognosis subgroup in patients with locally advanced HNSCC after RCTx. Conclusions: We found that CD98hc-associated signaling mechanisms play a central role in the regulation of HNSCC radioresistance and may be a promising target for tumor radiosensitization.

SLC3A2/CD98hc (solute carrier family 3 member 2) and its light chain subunits constitute the heterodimeric transmembrane complexes that mediate amino acid transport and regulate MTOR and macroautop...

T cells are the predominant immune cell population in the pancreatic tumor microenvironment. High CD8⁺ and Th1-polarized CD4⁺ T cell infiltration is associated with prolonged survival in human pancreatic ductal adenocarcinoma (PDAC). However, the expression pattern of co-stimulatory and inhibitory receptors by PDAC-infiltrating T cells and their prognostic significance are not well defined. In this study, we employed multiplex immunofluorescence to investigate the intratumoral expression of the co-stimulatory receptor inducible T-cell co-stimulator (ICOS), the inhibitory receptors lymphocyte-activation gene 3 (LAG-3), programmed death 1 (PD-1), and V-domain immunoglobulin suppressor of T cell activation (VISTA) by tumor-infiltrating T cells (CD3) in a cohort of 69 patients with resected PDAC. T cells were enriched particularly within the stromal area and were highly heterogeneous across tumors. Further, T cells were associated with prolonged disease-free survival (DFS). However, LAG-3 expression by PDAC-infiltrating T cells was correlated with reduced DFS. Our study highlights the biological importance of LAG-3 expression by tumor-infiltrating T cells. LAG-3⁺ T cells may represent a novel prognostic marker and a particularly attractive target for immunotherapeutic strategies in PDAC.

Purpose: Immunotherapy has led to a fundamental shift in the treatment of several cancers. However, its efficacy in pancreatic ductal adenocarcinoma (PDAC) is limited. Understanding the expression of inhibitory immune checkpoint receptors (ICR) by intratumoral T cells may help to unravel their involvement in insufficient T-cell–mediated antitumor immunity. Experimental Design: Using multicolor flow cytometry, we analyzed circulating and intratumoral T cells from blood (n = 144) and matched tumor samples (n = 107) of patients with PDAC. We determined the expression of programmed cell death protein 1 (PD-1) and T-cell immunoreceptor with Ig and immunoreceptor tyrosine-based inhibition motif (ITIM) domains (TIGIT) by CD8+ T-cells, conventional CD4+ T-cells (Tconv) and regulatory T cells (Treg) and their association with T-cell differentiation, tumor reactivity, and cytokine expression. A comprehensive follow-up was used to determine their prognostic value. Results: Intratumoral T cells were characterized by increased PD-1 and TIGIT expression. Both markers delineated distinct T-cell subpopulations. PD-1+TIGIT− T cells highly expressed proinflammatory cytokines and markers of tumor reactivity (CD39, CD103), whereas TIGIT expression was linked to antiinflammatory and exhausted phenotypes. In addition, the enhanced presence of intratumoral PD-1+TIGIT− Tconv was associated with improved clinical outcomes, while high ICR expression on blood T cells was a significant hazard for overall survival (OS). Conclusions: Our results uncover the association between ICR expression and T-cell functionality. PD-1 and TIGIT characterized intratumoral T cells with highly divergent phenotypes linked to clinical outcomes, further underscoring the relevance of TIGIT for immunotherapeutic approaches in PDAC. The prognostic value of ICR expression in patient blood may be a valuable tool for patient stratification.

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) has a dismal prognosis. At diagnosis, only 20% of patients with PDAC are eligible for primary resection. Neoadjuvant chemotherapy can enable surgical resection in 30%–40% of patients with locally advanced and borderline resectable PDAC. The effects of neoadjuvant chemotherapy on the cytokine production of tumor-infiltrating T cells are unknown in PDAC. METHODS: We performed multiplex immunofluorescence to investigate T cell infiltration in 91 patients with PDAC. Using flow cytometry, we analyzed tumor and matched blood samples from 71 patients with PDAC and determined the frequencies of T cell subsets and their cytokine profiles. Both cohorts included patients who underwent primary resection and patients who received neoadjuvant chemotherapy followed by surgical resection. RESULTS: In human PDAC, T cells were particularly enriched within the tumor stroma. Neoadjuvant chemotherapy markedly enhanced T cell density within the ductal area of the tumor. Whereas infiltration of cytotoxic CD8 ⁺ T cells was unaffected by neoadjuvant chemotherapy, the frequency of conventional CD4 ⁺ T cells was increased, and the proportion of Tregs was reduced in the pancreatic tumor microenvironment after neoadjuvant treatment. Moreover, neoadjuvant chemotherapy increased the production of proinflammatory cytokines by tumor-infiltrating T cells, with enhanced TNF-α and IL-2 and reduced IL-4 and IL-10 expression. CONCLUSION: Neoadjuvant chemotherapy drives intratumoral T cells toward a proinflammatory profile. Combinational treatment strategies incorporating immunotherapy in neoadjuvant regimens may unleash more effective antitumor responses and improve prognosis of pancreatic cancer. FUNDING: This work was supported by the Jung Foundation for Science and Research, the Monika Kutzner Foundation, the German Research Foundation (SE2980/5-1), the German Cancer Consortium, and the Faculty of Medicine Carl Gustav Carus, Technische Universität Dresden.

Background: The ongoing shortage of medical personnel highlights the urgent need to automate clinical documentation and reduce administrative burden. Large vision-language models (VLMs) offer promising potential for supporting surgical documentation and intraoperative analysis. Methods: We conducted an observational, comparative performance study of two general-purpose VLMs – GPT-4o (OpenAI) and Gemini-1.5-pro (Google) – from June to September 2024, using 15 cholecystectomy and 15 appendectomy videos (1 fps) from the CholecT45 and LapApp datasets. Tasks included object detection (vessel clips, gauze, retrieval bags, bleeding), surgery type classification, appendicitis grading, and surgical report generation. In-context learning (ICL) was evaluated as an enhancement method. Performance was assessed using descriptive accuracy metrics. Results: Both models identified vessel clips with 100% accuracy. GPT-4o outperformed Gemini-1.5-pro in retrieval bag (100% vs. 93.3%) and gauze detection (93.3% vs. 60%), while Gemini-1.5-pro showed better results in bleeding detection (93.3% vs. 86.7%). In surgery classification, Gemini-1.5-pro was more accurate for cholecystectomies (93% vs. 80%), with both models achieving 60% accuracy for appendectomies. Appendicitis grading showed limited performance (GPT-4o: 40%, Gemini-1.5-pro: 26.7%). For surgical reports, GPT-4o produced for CCE more complete outputs (CCE: 90.4%, APE: 80.1%), while Gemini-1.5-pro achieved higher correctness overall (CCE: 71.1%, APE: 69.6%). ICL notably improved tool recognition (e.g., in APE step 4, GPT-4o improved from 69.2% to 80%), though its effect on organ removal step recognition was inconsistent. Conclusion: GPT-4o and Gemini-1.5-pro performed reliably in object detection and procedure classification but showed limitations in grading pathology and accurately describing procedural steps, which could be enhanced through in-context learning. This shows that domain-agnostic VLMs can be applied to surgical video analysis. In the future, VLMs with domain knowledge can be envisioned to enhance the operating room in the form of companions.