Multi-omics analysis of BTF3L4 as a prognostic and immune biomarker in hepatocellular carcinoma

Introduction

Liver cancer, particularly hepatocellular carcinoma (HCC), is positioned as the sixth most prevalent cancer and represents the third major contributor to cancer-related mortality worldwide. The global incidence of HCC is increasing, becoming a leading cause of cancer-related morbidity and mortality (1,2). Although various proven therapeutic options exist, such as hepatectomy, radiotherapy and chemotherapy, the survival rates of patients with HCC remain dismally low owing to high recurrence and metastasis rates (3). The rapid evolution of therapeutic modalities for patients with advanced liver cancer raises numerous unresolved questions that require further exploration (4,5). Consequently, identifying novel therapeutic targets for liver cancer is of paramount importance (6).

The tumor microenvironment (TME) serves as a fundamental element in cancer advancement and plays a key role in cancer treatment (7,8). Tumor-infiltrating immune cells (TIICs) represent an essential component in tumor genesis (9). Mounting evidence suggests that immune cells in the TME are closely related to tumor growth and immunotherapy efficacy (10-12). However, their specific contribution to HCC progression remains unclear. Therefore, investigating immune-related therapeutic targets or prognostic markers within the HCC microenvironment is imperative.

The Cancer Genome Atlas (TCGA), a publicly funded initiative aimed at comprehensively cataloging and identifying major cancer-associated genomic alterations (13), has provided valuable insights into HCC. BTF3L4, located on human chromosome 1p32.3, was initially identified in the context of cartilage differentiation and is recognized as a pivotal regulator of cartilage formation (14). Previous studies have demonstrated that elevated BTF3L4 exacerbates liver injury and induces cellular apoptosis (15). However, the association between BTF3L4 and liver cancer, particularly HCC, has rarely been documented.

The multi-omics approach has gained significant traction in medicine and biological sciences (16). In the field of bladder cancer research, there have been a number of studies using multi-omics methods to study the correlation between target genes and the prognosis and immune cells of bladder cancer patients (17-19). In addition to assessing BTF3L4 mRNA expression through bioinformatics techniques, we conducted a more in-depth analysis of BTF3L4 protein abundance and its association with TIICs utilizing multiplex immunohistochemistry (mIHC) staining on tissue microarrays (TMAs). This comprehensive approach allowed us to investigate the association between BTF3L4 protein levels and various immune cell populations in the HCC microenvironment. We present this article in accordance with the REMARK reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2179/rc).

Methods

BTF3L4 mRNA expression levels analysis

We obtained liver HCC mRNA sequencing expression profiles through TCGA database (https://portal.gdc.cancer.gov), comprising 374 cancer cases and 50 benign samples. Subsequently, we performed a comparative analysis of BTF3L4 mRNA expression levels in HCC and adjacent non-tumor tissue, as well as its association with prognosis by Wilcoxon rank sum test and log-rank test.

RNA sequencing (RNA-seq) data generated via the Spliced Transcripts Alignment to a Reference (STAR) workflow from TCGA-Liver Hepatocellular Carcinoma (TCGA-LIHC) project were downloaded and curated from the TCGA database. Gene expression data in transcripts per kilobase of exon model per million mapped reads (TPM) format and clinically relevant data were extracted. The prognostic data among these were derived from a research article published in the journal Cell (20). Subsequently, normal tissue samples and those lacking complete clinical information were excluded. For data processing, log transformation was conducted by using the log₂ (value + 1) method, and the log-rank test was selected for statistical analysis. Finally, the proportional hazards assumption test was performed and a survival regression model was fitted using the survival R package. Meanwhile, we visualized the analytical results with the survminer and ggplot2 R packages.

Gene Expression Omnibus (GEO) database analysis

In this study, we utilized the GSE116174, GSE144269, and GSE109211 datasets. Specifically, the GSE116174 dataset contained 64 HCC tissue samples, with the detection platform being a microarray. The GSE144269 dataset included 70 pairs of HCC tissue samples and their matched adjacent non-tumor tissue samples, with RNA sequencing as the detection platform. The GSE109211 dataset consisted of 140 HCC tissue samples, using a microarray as the detection platform. For the data from the GEO database, relevant datasets were retrieved from the GEO database first. Subsequently, raw data preprocessing was performed. The specific procedures involved performing probe annotation and gene symbol conversion by means of the biomaRt R package combined with the annotation file corresponding to the GPL platform, followed by quality control filtering. Next, data normalization was performed using platform-specific strategies. Microarray data were normalized with the affy R package, while RNA-seq data were processed using the edgeR R package. Batch effects were then corrected using the ComBat algorithm. Finally, the proportions of immune infiltrating cells were estimated via the CIBERSORT tool, and the Spearman correlation analysis was adopted to investigate the association between target genes and immune cell infiltration. The analytical results were visualized as scatter plots using the ggplot2 R package.

Clinical samples and data

This study was indeed based on prospectively collected surgical tissue specimens that were archived in our biobank. However, given that the follow-up data were fully gathered when we conducted the analysis, the study was ultimately designed as a retrospective cohort study. HCC tissues (n=208) were collected from the Affiliated Tumor Hospital of Nantong University and the Affiliated Hospital of Nantong University from 2012–2017. The date of last follow-up was October 1, 2021, with a maximum follow-up duration of 5 years and the median follow-up time was 3.82 years. None of the patients had received radiotherapy, chemotherapy, or immunotherapy prior to surgery, and all cases were pathologically confirmed as HCC postoperatively. We excluded patients who were pathologically diagnosed with other tumors after surgery and those who failed to complete the follow-up. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Human Research Ethics Committees of the Affiliated Tumor Hospital of Nantong University (No. 2021008) and the Affiliated Hospital of Nantong University (No. 2018-K020). Written informed consent was obtained from patients or their immediate family members. Clinical characteristics and monitoring data were partially derived from medical documentation. The survival duration was measured from the date of the surgical procedure until either mortality or the final documented check-up.

Multiplex immunohistochemical staining

We constructed TMAs using paraffin-embedded liver tissue samples and performed mIHC staining as previously described (21). Serial TMAs were employed in this study. All the TMAs were prepared from samples with consistent embedding time and identical storage conditions. Meanwhile, it was verified that the corresponding positions on all TMAs are anatomically identical. TMA slides were dewaxed, dehydrated, and exposed to antigen retrieval utilizing 10 mM sodium citrate buffer (pH 6.0) with microwave heating. The staining procedure involved sequential treatments with primary and secondary antibodies, followed by signal amplification using an Opal fluorophore-binding tyramide signal amplification system (Akoya Biosciences, Marlborough, MA, USA). This process was conducted iteratively until the biological specimens were tagged with a predetermined array of distinct identifiers.

The following primary antibodies were used: anti-cytokeratin (CK) antibody (1:1,000, Orb69073, Biorbyt, Cambridge, UK), anti-BTF3L4 antibody (1:250, Orb511711, Biorbyt), anti-CD3e antibody (1:3,000, 85061S, CST, Danvers, MA, USA), anti-CD4 antibody (1:100, ab133616, Abcam, Cambridge, UK), anti-CD8 antibody (1:200, 85336S, CST), anti-CD20 antibody (1:200, ab78237, Abcam), anti-CD66b antibody (1:10,000, arg66287, Arigo, Zhubei, China), anti-CD68 antibody (1:10,000, 76437S, CST), anti-CD86 antibody (1:100, MAB1774, R&D Systems, Minneapolis, MN, USA), anti-CD163 antibody (1:100, 93498S, CST), anti-Lamp3 antibody (1:500, ab111090, Abcam), anti-PD-1 antibody (1:500, 86163, CST), anti-PD-L1 antibody (1:100, 13684S, CST), anti-cytotoxic T-lymphocyte-associated antigen 4 (anti-CTLA-4) (1:2,000, Orb527271, Biorbyt). Following antibody staining, the cell nuclei underwent staining with a DAPI staining solution (Sigma, St. Louis, MO, USA), and specimens were prepared for examination. The prepared slides underwent scanning through the Vectra Automated Quantitative Pathology Imaging System (version 3.0; PerkinElmer, Shelton, CT, USA). All experimental reagents used in the experiment were from the same batch.

InForm image analysis software (version 2.6.0, PerkinElmer) was utilized to evaluate the distribution of the staining intensity across all tissue sections. For the scoring results, pathological experts were invited to conduct verification and correction. Meanwhile, to avoid the interference of subjective bias on experimental results, the researchers remained blinded to the survival status and various clinical characteristics. This approach was maintained throughout the entire experimental process of image analysis and scoring. The distribution of cells within each designated area was quantified and scored on a scale of 0 to 100.

Statistical analysis

We conducted a comparative analysis of BTF3L4 mRNA expression in HCC and non-tumor tissues utilizing the Wilcoxon rank-sum test. The association between BTF3L4 protein levels and clinicopathological characteristics was assessed utilizing Pearson’s χ² test. Subsequently, X-tile software (Yale University, USA) was utilized to establish the cut-off point for BTF3L4 protein expression in relation to the patients’ 5-year overall survival (OS) rates. We first designated the columns corresponding to survival time, survival status, and the variables to be analyzed. Then we calculated the P values corresponding to each candidate cutoff value via the log-rank test, and ultimately selected the cutoff value with the smallest P value as the optimal grouping node. Survival prognosis was assessed utilizing Cox regression and Kaplan-Meier survival curves. We pre-specified patients’ BTF3L4 expression level, age, gender, tumor-node-metastasis (TNM) stage, alpha-fetoprotein (AFP) level, presence of liver cirrhosis, presence of hepatitis B virus (HBV) infection, and tumor differentiation grade for inclusion in univariate and multivariate analyses. Regarding the variable selection method for the final multivariate model, we adopted a purposeful selection approach based on the results of univariate analysis, with a significance level of P<0.05 as the inclusion criterion. Data analysis was performed utilizing SPSS version 26.0 (IBM Corporation, Armonk, NY, USA). A significance threshold of P<0.05 was established.

Results

BTF3L4 mRNA expression in HCC tissues

We assessed the BTF3L4 mRNA levels in HCC and non-tumor tissues using the TCGA database. Our research results show that BTF3L4 is differentially expressed in various tumors, including HCC (Figure 1A). Compared with non-tumor tissues, the expression of BTF3L4 mRNA in HCC tissues was significantly increased (Figure 1B, P<0.001). Furthermore, based on the downloaded original TCGA mRNA data (expression range, 2.122–5.440), we selected the median value of 3.456 as the cut-off. Scores ≤3.456 were classified as low expression, and scores >3.456 were classified as high expression. Then we identified a significant association between elevated BTF3L4 mRNA levels and adverse clinical outcomes in HCC patients (Figure 1C, P=0.001). These observations highlight the possible prognostic value of BTF3L4 mRNA expression in individuals with HCC.

Figure 1 BTF3L4 mRNA levels in tissues. (A) Expression of BTF3L4 mRNA in different types of tumors by Wilcoxon rank sum test from TCGA database. (B) BTF3L4 mRNA levels in HCC tissues (n=374) and adjacent non-tumor tissues (n=50) by Wilcoxon rank sum test from the TCGA database. (C) Kaplan-Meier plot of overall survival between BTF3L4 mRNA high expression group (n=212) and BTF3L4 mRNA low expression group (n=212) by log-rank test from the TCGA database. **, P<0.01; ***, P<0.001. HCC, hepatocellular carcinoma; mRNA, messenger RNA; TCGA, The Cancer Genome Atlas; TPM, transcripts per million.

BTF3L4 protein levels in HCC tissues

We performed mIHC staining analysis on the TMA sections to assess BTF3L4 protein levels in HCC. The BTF3L4-positive signals were primarily detected within the cytoplasm of the HCC cells and hepatocytes. Moreover, the positive signal of BTF3L4 in HCC tissues was significantly stronger than that in non-tumor tissues (Figure 2A,2B). Statistical analysis showed that the fluorescence staining score of BTF3L4 protein in HCC tissues was significantly higher compared to that in adjacent non-tumor liver tissues (Figure 2C). These findings suggest that the expression level of BTF3L4 protein is markedly elevated in HCC tissues relative to non-tumor tissues. Additionally, Kaplan-Meier curve analysis demonstrated an association between elevated BTF3L4 protein levels and poor prognosis in individuals with HCC (Figure 2D).

Figure 2 BTF3L4 protein levels in tissues. Representative mIHC images suggest that the fluorescence intensity of BTF3L4 staining is stronger in (A) HCC tissues than in (B) adjacent non-tumor tissue. (C) BTF3L4 protein levels in HCC tissues and adjacent non-tumor tissues by Wilcoxon rank sum test. (D) Kaplan-Meier survival curves showing that patients with high BTF3L4 protein levels (n=74) had worse overall survival than those with low BTF3L4 protein levels (n=134). ***, P<0.001. CI, confidence interval; CK, cytokeratin; HCC, hepatocellular carcinoma; HR, hazard ratio; mIHC, multiplex immunohistochemistry.

Association between BTF3L4 protein levels and clinical characteristics of HCC

According to mIHC staining results, the quantitative score for BTF3L4 protein expression in HCC tissues, based on the positive percentage, was determined using the InForm image analysis software (version 2.6.0, PerkinElmer), with scores ranging from 0 to 100. Using X-tile software, we determined a critical value of 54.33 (cutoff) for BTF3L4 protein expression. Scores ≤54.33 were classified as low expression, and scores >54.33 were classified as high expression. We stratified 208 patients with HCC into high-expression (n=74) and low-expression (n=134) groups to assess the association between BTF3L4 protein levels and clinical features. Our analysis revealed a significant association between BTF3L4 protein levels and the TNM stage (Pearson’s χ²=9.630, P=0.008). Nevertheless, no statistically significant correlations were identified between BTF3L4 protein levels and other clinical parameters, encompassing gender, age, liver cirrhosis, differentiation status, AFP levels, and HBV infection (Table 1). The specific patient enrollment process is illustrated in the flow chart (Figure 3A). For cases with missing data, we adopted the complete case analysis method to address the missing values. However, we must also objectively acknowledge that this will exert a certain impact on the generalizability of the research conclusions.

Figure 3 Patients enrollment flow and sensitivity analysis in the study. (A) Flow chart illustrating the inclusion and exclusion criteria for HCC cases during the study period. (B) Association between BTF3L4 protein levels and overall survival by Spearman correlation analysis. (C) Kaplan-Meier survival curves showing that patients with high BTF3L4 protein levels (n=104) had worse overall survival than those with low BTF3L4 protein levels (n=104). CI, confidence interval; HCC, hepatocellular carcinoma; HR, hazard ratio.

Prognostic significance of BTF3L4 protein levels in HCC

Univariate and multivariate Cox regression analyses were performed to explore the prognostic markers in individuals with HCC. Univariate Cox regression analysis revealed marked correlations between the 5-year survival rate of individuals with HCC and BTF3L4 protein levels [hazard ratio (HR): 1.011, P=0.03], liver cirrhosis (HR: 0.568, P=0.007), TNM stage (HR: 1.556, P=0.02), and AFP level (HR: 2.204, P=0.004) (Table 2). Furthermore, multivariate Cox regression analysis identified BTF3L4 protein levels (HR: 1.025, P=0.043), TNM stage (HR: 1.706, P=0.02), and AFP level (HR: 2.388, P=0.01) as independent prognostic factors markedly associated with diminished OS in individuals with HCC (Table 2).

In the sensitivity analysis, we first treated the BTF3L4 protein expression level as a continuous variable to re-analyze its correlation with clinical outcomes. The results showed that high expression of BTF3L4 protein was significantly associated with shortened survival time in patients (r=−0.229, P=0.001) (Figure 3B). Subsequently, we used the median value of BTF3L4 protein expression as an alternative cutoff value to stratify patients into a high-expression group and a low-expression group. Survival analysis results indicated that there was a significant difference in survival outcomes between the high-expression group and the low-expression group (HR: 1.64, P=0.02) (Figure 3C).

BTF3L4 mRNA levels: relationships with TIICs and immunoregulatory genes in HCC

The clinical outcomes of diverse cancer patients are substantially affected by the amount and distribution of TIICs (22). We analyzed the relationship between BTF3L4 mRNA level expression and TIICs and immunoregulatory genes through GEO database. We obtained the datasets GSE116174, GSE144269, and GSE109211 from the GEO database. Spearman correlation analysis revealed that the expression level of BTF3L4 was positively associated with the expression of CTLA-4, while it showed a negative correlation with the levels of CD66b⁺ neutrophils and CD4⁺ T cells (Figure 4A-4J). Notably, the signs of the correlation coefficients in Figure 4B,4D,4E are inconsistent with the directions of the trend lines, which can be attributed to the influence of a small number of extreme outliers. Accordingly, the directions of the linear trend lines are adopted as the primary basis for characterizing the correlations between variables.

Figure 4 Correlation of BTF3L4 mRNA expression with tumor-infiltrating immune cells and immunoregulatory genes in HCC. (A-J) Correlation between BTF3L4 mRNA expression and (A) CTLA-4, (B) CD274 (PD-L1), (C) PDCD1 (PD-1), (D) CD8⁺ T cells, (E) M1-like macrophages, (F) CD20⁺ B cells, (G) CD66b⁺ neutrophils, (H) LAMP3⁺ dendritic cells, (I) CD4⁺ T cells, and (J) M2-like macrophages by Spearman correlation analysis based on the GEO database. GEO, Gene Expression Omnibus; HCC, hepatocellular carcinoma; mRNA, messenger RNA; PD-1, programmed cell death-1; PD-L1, programmed death-ligand 1.

BTF3L4 protein levels: relationships with TIICs and immunoregulatory genes in HCC

Subsequently, we performed mIHC staining for BTF3L4, CD3, CD4, CD8, CD68, CD86, CD163, CD20, CD66b, PD-1, PD-L1, CTLA-4 and LAMP3.The results of mIHC staining indicated distinct differences in the positive expression levels of BTF3L4, CD66b, CTLA-4, and CD4⁺ T cells (CD3⁺CD4⁺) between HCC tissues and non-tumor tissues (Figure 5A,5B). Compared with non-tumor tissues, there was no significant difference in the staining fluorescence intensity of CD8⁺ T cells (CD3⁺CD8⁺), CD20⁺ B cells, M1-like macrophages (CD68⁺CD86⁺), M2-like macrophages (CD68⁺CD163⁺), LAMP3⁺ dendritic cells, PD-1, and PD-L1 (Figure 6A,6B).

Figure 5 Representative mIHC images of tumor-infiltrating immune cells and immunoregulatory genes. Representative mIHC images show the fluorescence intensities of the composite of BTF3L4, CTLA-4, CD66b, CD3, CD4, CK and DAPI through seven colors in (A) HCC tissues and (B) adjacent non-tumor tissues. CK, cytokeratin; HCC, hepatocellular carcinoma; mIHC, multiplex immunohistochemistry.

Figure 6 Representative mIHC images of tumor-infiltrating immune cells and immunoregulatory genes. Representative mIHC images show the fluorescence intensities of the composite of CD3, CD8, CD20, CD68, CD86, CD163, LAMP3, PD-1 and PD-L1 in (A) HCC tissues and (B) adjacent non-tumor tissues. CK, cytokeratin; HCC, hepatocellular carcinoma; mIHC, multiplex immunohistochemistry; PD-1, programmed cell death-1; PD-L1, programmed death-ligand 1.

Next, we conducted a quantitative analysis of the fluorescence intensity in mIHC staining. Notably, statistical analysis revealed that BTF3L4 protein expression in HCC positively associated with that of CTLA-4 (r=0.219, P=0.003). On the contrary, a notable inverse association was noted between CD4⁺ T cells (CD3⁺CD4⁺) (r=−0.321, P<0.001) and CD66b⁺ neutrophils (r=−0.269, P<0.001). Additionally, no significant correlation between BTF3L4 protein levels and the TIICs of the remaining staining in the TME was found. There was no notable association noted between BTF3L4 protein expression and PD-1 or PD-L1 expression either (Figure 7A-7K). The discrepancy between the correlation coefficient sign and the trend line direction in Figure 7G is attributable to the influence of extreme outliers on the linear fit. Although we obtained promising experimental results, these analyses are exploratory in nature. Their findings offer valuable insights for research on the prognostic assessment and immunotherapeutic implications of BTF3L4 in HCC. However, this conclusion warrants further validation through additional experiments.

Figure 7 Correlation of BTF3L4 protein expression with tumor-infiltrating immune cells (TIICs) and immunoregulatory genes in HCC. (A) Heatmap of the association between BTF3L4 protein levels and TIICs as well as immunoregulatory genes. (B-K) Correlation between BTF3L4 protein levels and (B) CD66b⁺ neutrophils, (C) CD4⁺ T cells (CD3⁺CD4⁺), (D) CTLA-4, (E) CD20⁺ B cells, (F) M1-like macrophages (CD86⁺CD68⁺), (G) M2-like macrophages (CD163⁺CD68⁺), (H) CD8⁺ T cells (CD3⁺CD8⁺), (I) LAMP3⁺ dendritic cells, (J) PD-1, and (K) PD-L1 by Spearman correlation analysis. HCC, hepatocellular carcinoma; PD-1, programmed cell death-1; PD-L1, programmed death-ligand 1; TIICs, tumor-infiltrating immune cells.

Discussion

Given the poor prognosis associated with liver cancer, discovering new therapeutic targets remains essential for enhancing clinical outcomes in individuals with HCC (22,23). Simultaneously, several studies utilizing multi-omics approaches to identify prognostic and immune cell-related targets in prostate cancer have also provided valuable insights (24,25). Our study revealed that the BTF3L4 mRNA level was markedly elevated in HCC tissues in contrast to non-tumor tissues, as determined through bioinformatics analyses. Kaplan-Meier analysis demonstrated that individuals with elevated BTF3L4 mRNA levels experienced notably poorer survival outcomes, consistent with previous research correlating BTF3L4 mRNA expression with OS in individuals with glioma and gastric cancer (26,27). Given that mRNA levels may not consistently reflect protein abundance (28,29), we employed mIHC staining to investigate BTF3L4 protein levels in HCC and non-tumor tissues. As anticipated, BTF3L4 protein levels were markedly elevated in HCC tissues in contrast to the non-tumor tissues, corroborating our mRNA findings.

Immunotherapy has revolutionized cancer treatment and reinvigorated tumor immunology research (30,31). The complex interplay between immune cells and tumors is highly intricate, with diverse subtypes exerting distinct functions (32). Subsequently, substantial evidence indicates the pivotal role of immune checkpoints in numerous tumors and their impact on treatment outcomes (33). We analyzed the association between BTF3L4 protein levels, immune cells, and immune checkpoints using mIHC technology. The results showed that in HCC tissues, BTF3L4 protein expression was positively correlated with CTLA-4, but significantly negatively correlated with both CD4⁺ T cells and CD66b⁺ neutrophils. Furthermore, the function of lymphocytes in tertiary lymphoid structures is crucial for HCC pathogenesis (34). CD4⁺ T cells primarily regulate tumor progression through cytokine secretion and associated helper cell functions (35). Research indicates that a low density of CD66b⁺ neutrophils in peritumoral regions of colorectal and gastric cancers correlates with an unfavorable prognosis and an elevated risk of disease recurrence (36). We hypothesized that high levels of BTF3L4 protein expression may interact with CD4⁺ T cells and CD66b⁺ neutrophils through specific mechanisms, thereby accelerating tumor growth and enhancing metastasis to facilitate a malignant phenotype.

Over the past decade, immunotherapy has revolutionized oncology and found indications in many cancers, including HCC (37). The application of immune checkpoint inhibitors represents a promising strategy for enhancing the therapeutic outcomes in cancer treatment (38). CTLA-4, a member of the immunoglobulin-associated receptor family, is widely acknowledged as a vital immune checkpoint and has emerged as a significant therapeutic target in autoimmunity and oncology (39). Predominantly expressed by T cells, CTLA-4 binds to CD80/CD86 and inhibits T-cell proliferation signal transduction provided by CD28 binding to CD80/CD86 during the initiation phase (40). Our research revealed that elevated BTF3L4 protein levels were associated with an increased presence of CTLA-4 in the TME. This finding suggests that BTF3L4 high tumors appear to display a CTLA-4 enriched microenvironment and that this hypothesis should be tested in cohorts treated with immune checkpoint inhibitors.

In recent years, BTF3L4 has garnered significant attention due to its potential role in various cancers. Its expression has been associated with multiple tumor types. Recent studies have identified BTF3L4 as a carcinogenic driver in gastric cancer, where it is significantly overexpressed (41). This overexpression correlates with poor differentiation, malignant progression, and adverse prognosis. In gliomas, BTF3L4 has been shown to enhance malignant phenotypes by modulating tumor cell function and the immune microenvironment (26). Additionally, BTF3L4 overexpression has been observed in pancreatic ductal adenocarcinoma and colorectal cancer, establishing it as a key prognostic marker (15).

Emerging evidence from recent studies indicates that BTF3L4 overexpression directly impairs the structure and function of mitochondrial membranes, leading to electron transport chain dysfunction and subsequent excessive production of reactive oxygen species (ROS), which activate c-Jun N-terminal kinases (JNK), promote c-Jun phosphorylation, and induce the formation of activator protein-1 (AP-1) transcription factor complexes. Upon binding to the promoter regions of target genes, these complexes upregulate the expression of pro-inflammatory cytokines, including interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α) (15,42). Furthermore, in gliomas, high BTF3L4 expression can directly enhance JNK phosphorylation and boost AP-1 transcriptional activity, thereby facilitating the process of tumor immune evasion (26). Collectively, these findings elucidate the potential mechanisms underlying the role of BTF3L4 in immune regulation.

Despite the significance of our findings, several limitations should be acknowledged. First, this study was retrospective rather than prospective, which may have introduced certain biases. Second, relevant cytological or zoological experiments to support our findings are lacking. Additionally, there is a need for mechanistic studies to elucidate the specific molecular pathways through which BTF3L4 influences immune cell populations and checkpoint expression in HCC. Among the 208 patients included in the study, there were 101 deaths. Finally, as this study lacked specific clinical validation, there is uncertainty regarding whether the study’s statistical power is sufficient to detect clinically meaningful HRs associated with BTF3L4. In the subsequent studies, we plan to employ a multicenter prospective design to enhance the reliability of our research data. Additionally, we will further investigate the correlation between BTF3L4 and TIICs through molecular cytological experiments and in vivo animal models.

Conclusions

In conclusion, this investigation revealed that BTF3L4 demonstrates markedly elevated expression in HCC tissues, suggesting that it is a promising prognostic indicator for HCC. This comprehensive analysis underscores the clinical relevance of BTF3L4 in HCC, positioning it as a potential therapeutic and prognostic target for HCC patients. Although BTF3L4 is a promising candidate prognostic and immune-associated biomarker in HCC, this biomarker still requires prospective validation and mechanistic studies. Given the limitations of this study, the applicability of BTF3L4 in other types of liver cancer, such as cholangiocarcinoma, remains to be validated. The observed correlation between BTF3L4 expression and immunological parameters in HCC tissues suggests a potential role for this protein in modulating the TME. These findings provide preliminary exploratory significance for predicting the prognosis of patients with HCC and developing new therapeutic strategies.

Acknowledgments

We extend our sincere gratitude to all members of the laboratory. We thank Bullet Edits Limited for the linguistic editing and proofreading of the manuscript.

Footnote

Reporting Checklist: The authors have completed the REMARK reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2179/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2179/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2179/prf

Funding: This work was supported by the Jiangsu Province Capability Improvement Project through Science, Technology and Education (No. ZDXK202234), Jiangsu Provincial Research Hospital, China (No. YJXYY202204), the Science and Technology Project of Nantong, Jiangsu, China (No. MS2023049), and the Scientific Research Project of Nantong Municipal Health Commission (No. MSZ2022088).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2179/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Human Research Ethics Committees of the Affiliated Tumor Hospital of Nantong University (No. 2021008) and the Affiliated Hospital of Nantong University (No. 2018-K020). Written informed consent was obtained from patients or their immediate family members.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Llovet JM, Pinyol R, Yarchoan M, et al. Adjuvant and neoadjuvant immunotherapies in hepatocellular carcinoma. Nat Rev Clin Oncol 2024;21:294-311. [Crossref] [PubMed]
2. Llovet JM, Castet F, Heikenwalder M, et al. Immunotherapies for hepatocellular carcinoma. Nat Rev Clin Oncol 2022;19:151-72. [Crossref] [PubMed]
3. Komoll RM, Hu Q, Olarewaju O, et al. MicroRNA-342-3p is a potent tumour suppressor in hepatocellular carcinoma. J Hepatol 2021;74:122-34. [Crossref] [PubMed]
4. Rimassa L, Finn RS, Sangro B. Combination immunotherapy for hepatocellular carcinoma. J Hepatol 2023;79:506-15. [Crossref] [PubMed]
5. Andrisani O. Epigenetic mechanisms in hepatitis B virus-associated hepatocellular carcinoma. Hepatoma Res 2021;7:12. [Crossref] [PubMed]
6. Kurebayashi Y, Ojima H, Tsujikawa H, et al. Landscape of immune microenvironment in hepatocellular carcinoma and its additional impact on histological and molecular classification. Hepatology 2018;68:1025-41. [Crossref] [PubMed]
7. Yang P, Yang W, Wei Z, et al. Novel targets for gastric cancer: The tumor microenvironment (TME), N6-methyladenosine (m6A), pyroptosis, autophagy, ferroptosis and cuproptosis. Biomed Pharmacother 2023;163:114883. [Crossref] [PubMed]
8. Bejarano L, Jordāo MJC, Joyce JA. Therapeutic Targeting of the Tumor Microenvironment. Cancer Discov 2021;11:933-59. [Crossref] [PubMed]
9. Sun P, Xu H, Guo C, et al. TMEM115 as an Oncogenic and Immunological Biomarker in Hepatocellular Carcinoma. Liver Int 2025;45:e70048. [Crossref] [PubMed]
10. Hinshaw DC, Shevde LA. The Tumor Microenvironment Innately Modulates Cancer Progression. Cancer Res 2019;79:4557-66. [Crossref] [PubMed]
11. Chen J, Lin Z, Liu L, et al. GOLM1 exacerbates CD8(+) T cell suppression in hepatocellular carcinoma by promoting exosomal PD-L1 transport into tumor-associated macrophages. Signal Transduct Target Ther 2021;6:397. [Crossref] [PubMed]
12. Dong S, Li X, Chen Z, et al. MMP28 recruits M2-type tumor-associated macrophages through MAPK/JNK signaling pathway-dependent cytokine secretion to promote the malignant progression of pancreatic cancer. J Exp Clin Cancer Res 2025;44:60. [Crossref] [PubMed]
13. Tomczak K, Czerwińska P, Wiznerowicz M. The Cancer Genome Atlas (TCGA): an immeasurable source of knowledge. Contemp Oncol (Pozn) 2015;19:A68-77. [Crossref] [PubMed]
14. Ji YH, Ji JL, Sun FY, et al. Quantitative proteomics analysis of chondrogenic differentiation of C3H10T1/2 mesenchymal stem cells by iTRAQ labeling coupled with on-line two-dimensional LC/MS/MS. Mol Cell Proteomics 2010;9:550-64. [Crossref] [PubMed]
15. Lin J, Fan A, Yifu Z, et al. BTF3L4 Overexpression Mediates APAP-induced Liver Injury in Mouse and Cellular Models. J Clin Transl Hepatol 2024;12:245-56. [Crossref] [PubMed]
16. Wang Y, Zhu H, Zhang L, et al. Common immunological and prognostic features of lung and bladder cancer via smoking-related genes: PRR11 gene as potential immunotherapeutic target. J Cell Mol Med 2024;28:e18384. [Crossref] [PubMed]
17. Xu Y, Mao S, Fan H, et al. LINC MIR503HG Controls SC-β Cell Differentiation and Insulin Production by Targeting CDH1 and HES1. Adv Sci (Weinh) 2024;11:e2305631. [Crossref] [PubMed]
18. Pang ZQ, Wang JS, Wang JF, et al. JAM3: A prognostic biomarker for bladder cancer via epithelial-mesenchymal transition regulation. Biomol Biomed 2024;24:897-911. [Crossref] [PubMed]
19. Wang Y, Wang J, He J, et al. Comprehensive analysis of PRPF19 immune infiltrates, DNA methylation, senescence-associated secretory phenotype and ceRNA network in bladder cancer. Front Immunol 2023;14:1289198. [Crossref] [PubMed]
20. Cancer Genome Atlas Research Network. Electronic address: wheeler@bcm; . Comprehensive and Integrative Genomic Characterization of Hepatocellular Carcinoma. Cell 2017;169:1327-1341.e23. [Crossref] [PubMed]
21. Jiang S, Zhang Y, Zhang X, et al. GARP Correlates With Tumor-Infiltrating T-Cells and Predicts the Outcome of Gastric Cancer. Front Immunol 2021;12:660397. [Crossref] [PubMed]
22. Donne R, Lujambio A. The liver cancer immune microenvironment: Therapeutic implications for hepatocellular carcinoma. Hepatology 2023;77:1773-96. [Crossref] [PubMed]
23. Lei P, Li W, Xie D, et al. Analysis of novel therapeutic targets and construction of a prognostic model for hepatocellular carcinoma. PeerJ 2025;13:e19899. [Crossref] [PubMed]
24. Wang Y, Ma L, He J, et al. Identification of cancer stem cell-related genes through single cells and machine learning for predicting prostate cancer prognosis and immunotherapy. Front Immunol 2024;15:1464698. [Crossref] [PubMed]
25. Wang Y, He J, Zhao Q, et al. Evaluating the predictive value of angiogenesis-related genes for prognosis and immunotherapy response in prostate adenocarcinoma using machine learning and experimental approaches. Front Immunol 2024;15:1416914. [Crossref] [PubMed]
26. Lu B, Lu T, Shi J, et al. Basic Transcription Factor 3 Like 4 Enhances Malignant Phenotypes through Modulating Tumor Cell Function and Immune Microenvironment in Glioma. Am J Pathol 2024;194:772-84. [Crossref] [PubMed]
27. Huang P, Xia L, Guo Q, et al. Genome-wide association studies identify miRNA-194 as a prognostic biomarker for gastrointestinal cancer by targeting ATP6V1F, PPP1R14B, BTF3L4 and SLC7A5. Front Oncol 2022;12:1025594. [Crossref] [PubMed]
28. Agle K, Vincent BG, Piper C, et al. Bim regulates the survival and suppressive capability of CD8(+) FOXP3(+) regulatory T cells during murine GVHD. Blood 2018;132:435-47. [Crossref] [PubMed]
29. Wegler C, Ölander M, Wiśniewski JR, et al. Global variability analysis of mRNA and protein concentrations across and within human tissues. NAR Genom Bioinform 2020;2:lqz010. [Crossref] [PubMed]
30. Zhang Y, Zhang Z. The history and advances in cancer immunotherapy: understanding the characteristics of tumor-infiltrating immune cells and their therapeutic implications. Cell Mol Immunol 2020;17:807-21. [Crossref] [PubMed]
31. Song H, Chen L, Pan X, et al. Targeting tumor monocyte-intrinsic PD-L1 by rewiring STING signaling and enhancing STING agonist therapy. Cancer Cell 2025;43:503-518.e10. [Crossref] [PubMed]
32. Yu J, Fu L, Wu R, et al. Immunocytes in the tumor microenvironment: recent updates and interconnections. Front Immunol 2025;16:1517959. [Crossref] [PubMed]
33. Liu X, Li S, Wang L, et al. Microecological regulation in HCC therapy: Gut microbiome enhances ICI treatment. Biochim Biophys Acta Mol Basis Dis 2024;1870:167230. [Crossref] [PubMed]
34. Li J, Xu H, Han J, et al. Lymphocyte Function in Tertiary Lymphoid Structures Predicts Hepatocellular Carcinoma Outcome. Lab Invest 2024;104:102144. [Crossref] [PubMed]
35. Speiser DE, Chijioke O, Schaeuble K, et al. CD4(+) T cells in cancer. Nat Cancer 2023;4:317-29. [Crossref] [PubMed]
36. Li J, Xia Y, Sun B, et al. Neutrophil extracellular traps induced by the hypoxic microenvironment in gastric cancer augment tumour growth. Cell Commun Signal 2023;21:86. [Crossref] [PubMed]
37. Sensi B, Angelico R, Toti L, et al. Mechanism, Potential, and Concerns of Immunotherapy for Hepatocellular Carcinoma and Liver Transplantation. Curr Mol Pharmacol 2024;17:e18761429310703. [Crossref] [PubMed]
38. Lin A, Jiang A, Huang L, et al. From chaos to order: optimizing fecal microbiota transplantation for enhanced immune checkpoint inhibitors efficacy. Gut Microbes 2025;17:2452277. [Crossref] [PubMed]
39. Ward FJ, Kennedy PT, Al-Fatyan F, et al. CTLA-4-two pathways to anti-tumour immunity? Immunother Adv 2025;5:ltaf008. [Crossref] [PubMed]
40. Zhang H, Dai Z, Wu W, et al. Regulatory mechanisms of immune checkpoints PD-L1 and CTLA-4 in cancer. J Exp Clin Cancer Res 2021;40:184. [Crossref] [PubMed]
41. Cao H, Wang Z, Guo Q, et al. MIR194-2HG, a miRNA host gene activated by HNF4A, inhibits gastric cancer by regulating microRNA biogenesis. Biol Direct 2024;19:95. [Crossref] [PubMed]
42. Motonari H, Hayashi A, Tanaka Y, et al. Mitochondrial dysfunction exacerbates inflammatory responses via the JNK/p38-AP-1 pathway in primary cultured chondrocytes. Int Immunopharmacol 2025;165:115433. [Crossref] [PubMed]