Background: Collagens are the most abundant proteins in extra cellular matrix and important components of tumor microenvironment. Recent studies have showed that aberrant expression of collagens can influence tumor cell behaviors. However, their roles in hepatocellular carcinoma (HCC) are poorly understood. Methods: In this study, we screened all 44 collagen members in HCC using whole transcriptome sequencing data from the public datasets, and collagen type IV alpha1 chain (COL4A1) was identified as most significantly differential expressed gene. Expression of COL4A1 was detected in HCC samples by quantitative real-time polymerase chain reaction (qRT-PCR), western blot and immunohistochemistry (IHC). Finally, functions and potential mechanisms of COL4A1 were explored in HCC progression. Results: COL4A1 is the most significantly overexpressed collagen gene in HCC. Upregulation of COL4A1 facilitates the proliferation, migration and invasion of HCC cells through FAK-Src signaling. Expression of COL4A1 is upregulated by RUNX1 in HCC. HCC cells with high COL4A1 expression are sensitive to the treatment with FAK or Src inhibitor. Conclusion: COL4A1 facilitates growth and metastasis in HCC via activation of FAK-Src signaling. High level of COL4A1 may be a potential biomarker for diagnosis and treatment with FAK or Src inhibitor for HCC.

Background and aims: Diagnosis of leptomeningeal metastasis (LM) is challenging. In our previous study, CEACAM6 mRNA was found to be highly expressed in the circulating tumor cells of cerebrospinal fluid (CSF) from patients with lung adenocarcinoma with LM (LUAD‐LM). The aim of this study was to identify whether CEACAM6 could be used as a biomarker for LUAD‐LM.Materials and methods: The level of CEACAM6 was determined by enzyme‐linked immunosorbent assay (ELISA) in CSF from 40 LUAD‐LM and 44 normal controls, and additional serum samples from 138 LUAD patients, including 12 LUAD‐LM patients, and 30 healthy controls. Carcinoembryonic antigen (CEA), cytokeratin 19 fragment (CYFRA 21‐1) and neuron‐specific enolase (NSE) levels in the CSF and sera were detected by chemiluminescent immunoassay. Receiver operating characteristic curve was plotted to evaluate the diagnostic performance for LUAD‐LM.Results: CSF CEACAM6 level was higher in LUAD‐LM than that in normal controls. In serum, LUAD patients had a higher level of CAECAM6 than healthy controls, and LM patients had the highest level among them. Serum CEACAM6 had a higher AUC than CEA in differentiating LM from non‐LM in LUAD patients (0.95 vs. 0.64, p < 0.001).Conclusion: CEACAM6 may serve as a potential biomarker in diagnosing LUAD‐LM.

Background: The diagnosis of lung adenocarcinoma (LUAD) leptomeningeal metastasis (LM) remains a clinical challenge. Human epididymis protein 4 (HE4) functions as a novel tumor biomarker for cancers. This study aimed to assess the diagnostic value of cerebrospinal fluid (CSF) HE4, and combined with CEACAM6, for LUAD LM.Methods: The CSF HE4 protein level was measured in two independent cohorts by electrochemiluminescence. Test cohort included 58 LUAD LM patients, 22 LUAD patients without LM (Wiot-LM), and 68 normal controls. Validation cohort enrolled 50 LUAD LM patients and 40 normal controls, in parallel with Wiot-LM patients without brain metastases (19 Wiot-LM/BrM patients) or with BrM (26 BrM patients). The CSF level of CEA, CA125, CA153, CA199, CA724, NSE and ProGRP of these samples was measured by electrochemiluminescence, whereas the CSF CEACAM6 level was detected by enzyme-linked immunosorbent assay (ELISA). In addition, the serum level of these biomarkers was detected by same method as CSF.Results: The level of HE4 or CEACAM6 in CSF samples from LUAD LM patients was significantly higher than those from normal controls and Wiot-LM patients. The HE4 or CEACAM6 level in CSF was higher than that in serum of LM patient. The CSF HE4 or CEACAM6 level for distinguished LM from Wiot-LM showed good performance by receiver-operating characteristic analysis. The better discriminative power for LM was achieved when HE4 was combined with CEACAM6. In addition, the CSF HE4 and CEACAM6 level showed little or no difference between Wiot-LM/BrM and BrM patients, the BrM would not significantly influence the HE4 or CEACAM6 level in CSF. The diagnostic power of CSF CA125, CA153, CA199, CA724, NSE and ProGRP for LUAD LM were not ideal.Conclusion: The combination with HE4 and CEACAM6 has the promising application for the diagnosis of LUAD LM.

LncRNA is a key epigenetic regulator in biological processes. In the human cancer transcriptome library MiTranscriptome, we identified GAU1 as the top upregulated lncRNA in colorectal cancer (CRC) by sample set enrichment analysis (overexpression ranking %, ), which is coexpressed with the potential oncogene GALNT8 (Spearman , , TCGA dataset ). Experimental data revealed that GAU1 regulates the expression of GALNT8. The overexpression of either GAU1 or GALNT8 significantly promotes the cell cycle and proliferation of CRC cell lines and correlates with poor prognosis in patients with CRC ( ), while silencing of GAU1 or GALNT8 suppressed the cancer cell proliferation and induced the CRC cell line resistance to oxaliplatin in vitro treatment. Our results suggested that the previously less studied GAU1 and GALNT8 may play as CRC prognosis markers and potential targets for chemotherapy treatment.

This study aimed to assess the diagnostic performance of carcinoembryonic antigen-related adhesion molecule 6 (CEACAM6) and human epididymis protein 4 (HE4) in pleural fluid for the detection of malignant pleural effusion (MPE). In this prospective study, pleural levels of CEACAM6 and HE4 were measured in two independent cohorts. The test cohort included 182 patients with exudative pleural effusions (123 malignant and 59 benign), and the validation cohort comprised 117 patients with exudative pleural effusions (65 malignant and 52 benign). Receiver operating characteristic (ROC) curves were used to assess the diagnostic performance of CEACAM6 and HE4 for MPE. Both CEACAM6 and HE4 levels were significantly elevated in MPE compared to benign pleural effusion (BPE) in both cohorts (p < .001). In the test cohort, CEACAM6 and HE4 demonstrated areas under the curve (AUC) values of 0.862 and 0.826, respectively. The combination of CEACAM6 and HE4 yielded a higher AUC of 0.938 compared to either marker alone. In the validation cohort, both CEACAM6 (AUC = 0.811) and HE4 (AUC = 0.721), along with their combination (AUC = 0.834), exhibited strong diagnostic performance for MPE. Notably, in cytology-negative cases, the combination of CEACAM6 and HE4 also demonstrated a favourable diagnostic efficacy, with an AUC of 0.800. The addition of CEA to the CEACAM6/HE4 combination further improved the AUC to 0.819 (p = .04). Pleural CEACAM6 and HE4 are promising biomarkers for distinguishing MPE from BPE. Their combination improves diagnostic accuracy, offering a valuable tool for MPE diagnosis, especially in challenging cases with cytology-negative pleural effusion.

Lung adenocarcinoma (LUAD)-derived leptomeningeal metastases (LM) represent a predominant subtype among all LM cases. Nevertheless, the cerebrospinal fluid (CSF) profile of LUAD-LM patients remains poorly characterized and reliable CSF diagnostic biomarkers for LUAD-LM have yet to be established. Using single-cell RNA sequencing data of CSF cells from six LUAD-LM patients, we drew a systematic transcriptomic atlas of the CSF cellular landscape. Our analysis revealed that LUAD-LM reprograms CSF into an immunosuppressive state, marked by the emergence of pro-tumoral LGMN-SELENOP^high macrophages and proliferating CSF circulating tumor cells (CSF-CTC). Cell-cell communication analysis showed that CSF-CTC reinforces immunosuppression by co-inhibitory checkpoint axis NECTIN2_TIGIT axis with the CD8⁺T/NK cells, and via CD47_SIRPA axis with antigen-presenting cells. Furthermore, we identified the single-cell transcriptomic difference between CSF-CTC and tumor cells of parenchymal brain metastases (PBM). Notably, Trophoblast cell surface antigen 2 (TROP2) levels in CSF were significantly elevated in LUAD-LM patients versus both normal controls (NC) and LUAD patients without LM (Non-LM). It showed strong diagnostic accuracy for distinguishing LUAD LM from Non-LM or NC, and PBM did not influence the CSF TROP2 level. Collectively, our findings advance the understanding of LUAD-LM pathogenesis and highlight the potential of CSF TROP2 as a diagnostic biomarker for LM.