Introduction

Hepatocellular carcinoma (HCC) constitutes a major global health burden, with its high metastatic potential and heterogeneous metastatic sites serving as critical determinants of postoperative survival outcomes. While recent research has elucidated the involvement of ncRNAs in the metastatic progression of HCC, the precise mechanisms underlying their regulation within the tumor immune microenvironment remain incompletely understood. This study employs cutting-edge spatial transcriptomics to investigate the spatial dynamics of these interactions in HCC.

Aims & Methods

To define the role of ncRNAs in facilitating metastasis within the HCC, we selected key ncRNAs for further investigation through BLAST homology analysis and differential expression profiling between primary and metastatic liver cancer specimens. Subsequently, utilizing spatial single-cell transcriptomic data, we conducted cell-cell communication and neighborhood analysis to explore the complex interactions among tumor cells, immune cells, and stromal cells.

Results

Our findings reveal that cancer-associated fibroblasts (CAFs) promote tumor metastasis through exosome-mediated communication with tumor cells and tumor-associated macrophages (TAMs), driving epithelial-mesenchymal transition (EMT) and enhancing the migratory and invasive capacities of cancer cells. Specifically, through the spatial interaction map, we identified hsa-miR-548d-5p and POT1-AS1 as critical molecular nodes involved in the metastatic process. Tumor cells exposed to both CAFs and TAMs exhibited significantly altered gene expression profiles that favored metastasis, with CAF-derived signals priming TAMs to release pro-metastatic factors that ultimately reprogrammed neighboring cancer cells. Furthermore, we validated our findings through transwell experiments and analysis of ncRNA sequencing.

Conclusion

This study utilized spatial transcriptomics to elucidate how CAFs influence tumor cells and TAMs through exosome-mediated ncRNA signaling, promoting EMT and enhancing the migratory and invasive potential of tumor cells. These findings suggest novel therapeutic strategies targeting CAFs and their interactions with TAMs, providing new avenues for optimizing HCC immunotherapy.

Introduction

The inflammatory bowel disease (IBD) subgroup classification of Crohn’s disease (CD) and ulcerative colitis (UC) is based on clinical, endoscopic, histological and radiological findings. A non-invasive serological tool to aid the diagnosis of IBD is currently lacking. Extracellular matrix (ECM) proteins have been proposed as promising biomarkers of inflammatory activity, mucosal damage, and fibrosis development in IBD and may have diagnostic utility.

Aims & Methods

We aimed to investigate the diagnostic accuracy of 13 serum ECM protein epitopes representing inflammation and collagen formation and degradation. Serum from 518 newly diagnosed and treatment-naïve adult IBD patients (180 CD, 338 UC) and 205 symptomatic controls (SC) from the IBSEN III cohort (discovery cohort) were collected at diagnosis. The serum was analysed for a panel of ECM protein epitopes (C4M, C6M, C7M, PROC3, PROC4, PROC5, PROC6, PROC11, PROC16, PROC22, VICM, CTXIII, and CPa9HNE) using enzyme-linked immunosorbent assay (ELISA). We used log2-transformed protein concentrations to fit logistic models for IBD vs SC, CD vs SC, UC vs SC, and CD vs UC. We fitted models for each individual protein and for all 13 proteins combined. Backward regression with the Bayesian information criterion, starting from the full model with 13 proteins, identified models with three to four proteins.
The models fitted to the discovery cohort were validated in the independent NORDTREAT cohort of 130 newly diagnosed IBD patients (56 CD, 74 UC) and 103 SC using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve.

Results

The coefficients of models fitted to the discovery cohort identified C6M, C7M, C4M, PRO-C3, and PRO-C6 as the most important proteins for separating disease from controls. In the validation cohort, the combination of C6M, C7M and PRO-C3 reached the highest AUC of 0.65 for distinguishing IBD from SC. When fitting the model to differentiate CD from SC, the collagen VI epitope C6M was superior, reaching an AUC of 0.65 and increasing to 0.70 when combined with C7M, PRO-C3, and PRO-C6. A model with all 13 markers gave an AUC of 0.71 for CD vs SC. Only C4M showed capacity of differentiating between UC and CD with
an AUC 0.66. Lastly, differentiation of UC from SC using only C6M had an AUC of 0.62, increasing to 0.64 when adding C7M and PRO-C3 to the model.

Conclusion

This study revealed a moderate diagnostic accuracy of predicting IBD versus SC. However, the ability to separate patients with CD from SC was promising, motivating further exploration of biomarker utility for CD subgroup distinctions.

Disclosure

JH has received consulting and/or advisory board fees from: AbbVie, Alfasigma, Aqilion, Bristol Myers Squibb, Celgene, Celltrion, Eli Lilly, Ferring, Galapagos, Gilead, Hospira, Index Pharma, Janssen, Johnson & Johnson, MEDA, Medivir, Medtronic, Merck, Merck Sharp & Dohme, Novartis, Pfizer, Prometheus Laboratories Inc., Sandoz, Shire, STADA, Takeda, Thermo Fisher Scientific, Tillotts Pharma, Vifor Pharma, UCB; and speaker’s fees from: AbbVie, Alfasigma, Bristol Myers Squibb, Celgene, Eli Lilly, Ferring, Galapagos, Gilead, Hospira, Janssen, Johnson & Johnson, Merck Sharp & Dohme, Novartis, Pfizer, Shire, Takeda, Thermo Fisher Scientific, Tillotts Pharma; and research grant support from Janssen, Merck Sharp & Dohme and Takeda.
AWM has received and researcher-initiated grant from Takeda
VS has received research funding from Takeda
VA is a member of the Merck advisory board and has consulted for Janssen
JDF reports travel grants from Tillotts Pharma
VAK has received consulting and/or advisory board fees from Takeda, Tillotts Pharma, Janssen-Cilag, and speaker fees from Thermo Fisher Scientific.
MLH has received investigator-initiated research grants from Takeda, Pfizer, Tillotts, Ferring and Janssen, speaker fees from Takeda, Tillotts, Ferring, AbbVie, Pfizer, Galapagos, MSD and Meda, and advisory board fees from Takeda, Galapagos, MSD, Lilly, Janssen, Pfizer and AbbVie