A new combined prognostic model involving SLC44A4 improves the predictive ability for colorectal cancer patients

Introduction

The incidence rates of colorectal cancer (CRC) tend to increase in many countries with changes in lifestyle factors and diet (1,2). It was estimated that there were approximately more than 1.9 million new cases of CRC in 2020 worldwide, accounting for approximately one-tenth of cancer-related deaths; resulting in CRC ranking as second in the mortality rate (3). Since the decline in the incidence in the elderly population and the increase in the incidence in the younger population have occurred simultaneously, the CRC patient population has rapidly become younger (4). Early-stage CRC is difficult to detect because of the lack of clinical symptoms (5). Thus, many patients with CRC are diagnosed in the terminal stage. The clinical limitations of surgery, the toxic and side effects of radiotherapy and chemotherapy, and drug resistance have always been the main reasons restricting the diagnosis and treatment of this disease (6,7). Due to the increasing incidence and worse clinical prognosis of CRC, it is particularly significant to understand the development process of CRC and identify relevant prognostic factors.

The current CRC prognosis and treatment methods always follow the American Joint Committee on Cancer (AJCC) tumor lymph node metastasis (TNM) staging standard (8). Nevertheless, due to the high heterogeneity of CRC and the lack of accurate prediction of TNM staging, it is important to explore more new biomarkers for predicting prognosis and treatment. Meanwhile, publicly accessible large databases including The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO), provide an impetus to find new biomarkers. Although various biomarkers have been discovered in CRC, only a few have been translated into clinical practice (9). For example, KRAS mutations are associated with adverse reactions to anti-EGFR receptor therapies, including panitumumab and cetuximab (10,11). However, several potential biomarkers need to be further investigated.

Recently, the function of membrane transporters in different kinds of cancer has attracted more attention. The ATP-binding cassette family and the solute carrier (SLC) family are two main superfamily transporters (12). The SLC family participates in the transport of various solutes and metabolites (13). For example, previous studies have discovered that transporters of SLC1, SLC3 and some other SLC families play important roles in the absorption and reabsorption of amino acids required for protein synthesis, and the SLC family is expressed in the kidney and intestine (12). Meanwhile, the SLC2, SLC5, and SLC50 families participate in glucose intestinal absorption and glucose renal reabsorption (14,15). SLC30 and SLC39 families are zinc transporters, which regulate zinc levels in the body (12). Some transporters in the SLC family are upregulated in different cancers as tumor promoters, such as SLC6A14 (16), SLC7A5 (17) and SLC1A5 (18). On the contrary, SLC39A1 is downregulated in prostate cancer and could be a tumor suppressor gene. SLC39A1 inhibits the nuclear factor κB (NF-κB) signaling pathway to regulate the malignant potential of prostate cancer cells (19). Therefore, having a good command on the SLC family in different kinds of tumor tissues could provide a theoretical basis for formulating novel cancer treatment methods.

SLC44A4 is a member of the choline transporter-like family and is essential for the biosynthesis of acetylcholine (20). This neurotransmitter is vital for various cellular processes, including metabolism and energy production. The production of acetylcholine, facilitated by choline acetyltransferase (ChAT), plays a crucial role in mediating inflammatory responses, regulating blood pressure, and supporting neurological functions. For instance, circulating ChAT activity is responsive to inflammatory stimuli and can suppress the synthesis of cellular factors (21); moreover, mutations in ChAT have been shown to affect both acetylcholine production and temperature-dependent behavioral responses (22). These observations highlight the strong connection between acetylcholine production and cellular metabolism, indicating a possible relationship between the SLC44A4 transporter and both cellular metabolism and energy generation.

The translocator protein exhibits a crucial relationship with the metabolic processes of cancer cells, particularly highlighted by the significant overexpression of SLC35A2 in CRC tissues, especially within microsatellite stable tumors. This overexpression accelerates tumor growth by activating MYC target genes, including those involved in the V2 pathway, and facilitates the reprogramming of glucose metabolism through pathways such as the pentose phosphate pathway and the tricarboxylic acid cycle. The elevated levels of SLC35A2 are associated with a reduced recurrence-free survival duration in patients and correlate with resistance to irinotecan treatment. Additionally, the expression of SLC35A2 demonstrates an inverse relationship with the infiltration of CD8⁺ T cells and B cells within the tumor microenvironment, indicating its role in promoting tumor progression by inhibiting anti-tumor immunity (23). Moreover, ERRα overexpression results in the upregulation of SLC7A11, which disrupts the cellular redox equilibrium, characterized by diminished NADPH levels and elevated GSSG/GSH ratios. This metabolic imbalance allows tumors to thrive in the presence of sufficient glucose while rendering cancer cells susceptible to oxidative stress under conditions of microenvironmental stress, such as glucose deprivation (24). In the tumor budding area, CCL5 is secreted, which activates the CCR5 receptor on fibroblasts, leading to the latter’s overexpression of SLC25A24, the mitochondrial adenine nucleotide translocator. This protein enhances fibroblast activation via the pAkt-pmTOR signaling pathway, promoting angiogenesis through the upregulation of VEGFA and collagen synthesis, thereby fostering a pro-cancerous microenvironment (25).

Furthermore, SLC transport proteins are integral to the pathology of colon cancer. SLC5A8 serves as a critical factor in the dietary fiber-dependent suppression of colon tumors through the pathway of “dietary fiber intake → bacterial metabolites (short-chain fatty acids) → intestinal immune homeostasis”. This factor not only directly reduces inflammation and the occurrence of tumors by modulating immune tolerance but is also closely linked to dietary fiber intake due to its functional importance, thereby collaboratively preserving colonic health (26). SLC6A14 is identified as a pivotal tumor-promoting factor in colon cancer; its expression is directly modulated by the classical Wnt signaling pathway (TCF4/β-catenin complex), showing significant upregulation in colon cancer. It sustains the proliferation and invasive capabilities of cancer cells through amino acid transport and the activation of the mTOR signaling pathway. Targeting SLC6A14 inhibition can effectively obstruct these processes, rendering it a promising drug target for the treatment of colon cancer (27).

SLC44A4 (or CTL4) is a member of the family of solute carrier proteins known as SLC44A1-5 or choline transporter-like proteins (CTL1-5) (28,29). SLC44A4 is involved in synthesis and transport of acetylcholine (30) and uptake of thiamine pyrophosphate (TPP), the phosphorylated form of vitamin B1 (31). Colonic TPP transporter (cTPPT) is the product of SLC44A4 gene (32,33). It has been reported that SLC44A4 is significantly upregulated in various epithelial tumors, the most notable being pancreatic cancer and prostate cancer (34). Kang et al. reported that the messenger RNA (mRNA) expression level of SLC44A4 was significantly lower in clear cell renal cell carcinoma (ccRCC) tissues than normal in tissues. In the aforementioned research, the expression level of SLC44A4 was associated with the clinical stage of ccRCC patients (35) and the low expression level of SLC44A4 was related to the late clinical stage of ccRCC patients. Furthermore, Kang et al. found that high expression of SLC44A4 in ccRCC patients indicates better OS and disease-free survival (DFS) (35). Nevertheless, the prognostic value of SLC44A4 in CRC was yet to be evaluated.

In our study, we evaluated SLC44A4 using TCGA and GEO database data and assessed the expression level of the SLC44A4 protein in CRC tissues by tissue-microarray-based immunohistochemistry (IHC). Moreover, we proposed a combined prognostic model incorporating SLC44A4 expression, which was tested for its ability to improve the predictive ability for prognosis and clinical outcome of CRC patients. We present this article in accordance with the REMARK reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-973/rc).

Methods

Bioinformatics analysis by GEO database and TCGA database

The selection of candidate gene was as follows: first, we analyzed nine CRC datasets in the GEO database by limma package to find differentially expressed genes (DEGs). Then, the common DEGs of the nine datasets were identified by the Robust Rank Aggregation (RRA) method. Then the intersection was taken with DEGs related to prognosis in the TCGA database. Finally, we obtained 26 candidate genes, 18 of which were downregulated, incorporating SLC44A4. Next, we studied the differences in expression of SLC44A4 between CRC and normal tissue samples using the Oncomine, UALCAN, and GEO databases. GEO2R analyzed GEO data (GSE20842). We further evaluated the prognostic value of SLC44A4 in TCGA datasets using Kaplan-Meier survival analyses.

Analysis of immune infiltration

We performed an analysis on immune cell infiltration with the application of the single sample gene set enrichment analysis (ssGSEA) algorithm. We made a comparison between distinct gene sets of immune cells and turned sample gene expression values into enrichment fractions. we determined the relative abundance of immune cells with the help of the GSVA R package (Version 1.46).

Enrichment analyses of Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene set enrichment analysis (GSEA)

To uncover the functional significance of the dataset, comprehensive GO and KEGG pathway enrichment analyses were systematically performed using the “clusterProfiler” package (version 4.4.4) (36). In addition, The GSEA software (version 4.1.0) was utilized to delve deeper into gene functions associated with different levels of SLC44A4 expression, with a particular emphasis on comparing the profiles of high and low expression. The analyses were conducted in strict accordance with rigorous standards, which ensured that only results with an adjusted P value less than 0.05 and a false discovery rate (FDR) lower than 0.25 were deemed to be of statistical significance.

CRC patients and CRC tissue samples

CRC paraffin samples were collected from September 2011 to February 2013 in the Sir Run Run Shaw Hospital, an affiliate of Zhejiang University School of Medicine. The diagnostic criterion of all tissue samples was based on the 2019 World Health Organization tumor classification criteria and the clinical stage of tumors was based on the AJCC. The clinicopathological characteristics of CRC patients was shown in Table 1. The 140 patients with CRC consisted of 87 (62.1%) men and 53 (37.9%) women, and 119 (85.0%) of them were more than 60 years old. Seventy-six patients (54.3%) were identified as late stages (III and IV), while the other 64 patients (45.7%) were diagnosed as early stages (I and II). Twenty-five patients (17.9%) were transferred to other distant organs at the time of diagnosis and identified as pM1 stage. The electronic medical record was used to obtain the clinical information of these CRC patients. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Sir Run Run Shaw Hospital, Zhejiang University School of Medicine Institutional Review Board (No. 20220305-33) and individual consent for this analysis was waived due to the retrospective nature.

IHC

CRC samples were fixed and then paraffin-embedded. Tissue sections (4 µm) were routinely processed and then immersed in xylene and graded ethanol series for deparaffinization and rehydration. Subsequently, the sections were treated with 1% hydrogen peroxide for 5 minutes and repaired with boiled EDTA repair solution for 10 min. Hereafter, they were blocked with 1.5% cattle serum, incubated with anti-SLC44A4 antibody (1:50 dilution; Sigma-Aldrich, Poole, Dorset, UK) at 4 ℃ for 18 hours, and secondary antibody conjugated with horseradish peroxidase (HRP) (DAKO) at 37 ℃ for 30 minutes. In the research performed by Ma et al. (37), the expression patterns related to pancreatic cancer were evaluated using a commercially sourced antibody in the framework of IHC, which effectively validated the dependability of the antibodies. Furthermore, the expression of the SLC44A4 protein was investigated in colon epithelial cell lines utilizing Western blotting, which further reinforced the antibodies’ reliability (38). Finally, the stained tissues were incubated with the DAB chromogen reagent and counterstained with hematoxylin.

IHC evaluation

The IHC results were collected on a microscope (Nikon, Japan) after sealing. Two experienced pathologists assessed the IHC scores according to the formula: SLC44A4 expression score = percentage score × intensity score (39). According to the percentages of positive tumor cells, the percentage score was classified into five levels, including 0 (while positive cells <5%), 1 (positive cells were at the range of 5–25%), 2 (positive cells were at the range of 25–50%), 3 (positive cells were at the range of 50–75%), and 4 (positive cells were >75%). The intensity score of SLC44A4 was classified into four levels: 0 (tumor cells were negative), 1 (tumor cells were weakly stained), 2 (tumor cells were moderately stained), and 3 (tumor cells were strongly stained). The IHC score ranging from 0 to 4 was defined as low SLC44A4 group, while others were defined as the high SLC44A4 group.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 6, SPSS 19.0, and R software (version 3.5.2, http://www.r-project.org/). χ² test was used to analyze the association between the expression of SLC44A4 and clinicopathological variables in CRC patients. The correlation between clinicopathological variables, DFS and OS of patients with CRC was evaluated by the Kaplan-Meier method in univariate analysis. The Cox regression model was used to perform multivariate analysis to identify independent prognostic factors in our study. The Harrell concordance index (C-index) was used to assess the new combined model’s prognostic accuracy. The difference with P<0.05 was considered statistically significant.

Results

Identification of SLC44A4 in CRC based on GEO database and TCGA database

Publicly available online databases were used to analyze gene expression in CRC. In our research, we selected SLC44A4 as a candidate gene that may be related to intestinal carcinogenesis. Oncomine analysis indicated that SLC44A4 was significantly downregulated in CRC and kidney cancer (Figure 1A). The UALCAN analysis results demonstrated that the expression of SLC44A4 was lower in CRC tissues than normal colon tissues (Figure 1B). Compared with normal colon tissues, SLC44A4 expression was downregulated in CRC in different clinical stages (Figure 1C). This was closely related to poor overall survival (OS) (Figure 1D). As shown in Figure 1E, a heat map of SLC44A4 mRNA expression was performed by GEO database. The expression of SLC44A4 was downregulated in CRC than that in normal tissues (Figure 1F).

Figure 1 SLC44A4 expression analysis in CRC from Online databases. (A) Analysis of SLC44A4 expression in human tumors compared with normal tissues from Oncomine database. 10% gene bank; P<0.0001; fold change >2. (B) Compared to 41 normal colonic tissue samples, SLC44A4 expression level was reduced in 286 CRC tissues in TCGA profile (P<0.001). (C) In TCGA profile, SLC44A4 expression was significantly downregulated compared with normal tissues in different clinical stages, including stage 1 to 4 (P<0.001). (D) CRC patients with high expression of SLC44A4 had a longer survival time than that with low SLC44A4 expression. n=350, P=0.007. (E) A heatmap of SLC44A4 mRNA expression from GEO database. (F) In GEO database, SLC44A4 expression was significantly downregulated in CRC (n=17) than that in normal tissues (n=17) (P<0.001). CRC, colorectal cancer; GEO, Gene Expression Omnibus; OS, overall survival; TCGA, The Cancer Genome Atlas.

SLC44A4 is correlated with immune infiltration in CRC

This research delves deeper into the intricate relationship between SLC44A4 expression and immune infiltration within CRC patients. Using ssGSEA algorithms, we analyzed immune-related infiltration to find the Pearson correlation between the immune microenvironment and SLC44A4 expression. Our analysis showed that SLC44A4 expression negatively correlated with the infiltration levels of T helper cells, Tem, Th1 cells, and macrophages (Figure 2A). In contrast, it positively correlated with Th17 cells, B cells, eosinophils, and NK CD56bright cells (Figure 2A-2E). Further analysis revealed significant differences in SLC44A4 gene expression among various immune cell types (Figure 2F-2I). By investigating different functional subsets of T cells, this research suggests that SLC44A4 may play an important role in the immune-inflamed microenvironment of CRC.

Figure 2 The relationship between SLC44A4 gene expression and immune cell infiltration. (A) There is an association between SLC44A4 gene expression and the status of immune cell infiltration. (B-E) Most immune-infiltrating cells show a positive correlation with SLC44A4 gene expression. (F-I) There are differences in the enrichment of specific immune cell subsets between the low- and high-expression groups of the SLC44A4 gene. ns, not significant; *, P<0.05; **, P<0.01; ***, P<0.001. TPM, transcripts per million.

GO and KEGG enrichment analysis

To explore the biological functions of SLC44A4, we performed pathway enrichment analyses based on GO and the KEGG. The GO analysis unveiled significant enrichment in several biological processes (BP), such as nucleosome assembly, protein-DNA complex assembly, chromatin remodeling, detection of abiotic stimulus, detection of external stimulus, and regulation of membrane potential. Prominent cellular components (CC) identified were nucleosome, protein-DNA complex, synaptic membrane, postsynaptic membrane, nuclear chromosome, and glutamatergic synapse. The molecular functions (MF) showed substantial enrichment in olfactory receptor activity, channel activity, gated channel activity, protein heterodimerization activity, cation channel activity, and peptide binding. The KEGG pathway enrichment analysis highlighted significant enrichment in pathways associated with systemic lupus erythematosus, neuroactive ligand-receptor interaction, alcoholism, neutrophil extracellular trap formation, olfactory transduction, and calcium signaling pathway (Figure 3A-3D). Collectively, these findings indicated the correlation between SLC44A4 with CRC.

Figure 3 GO and KEGG enrichment analysis of SLC44A4. (A) BP enrichment analysis of SLC44A4. (B) CC enrichment analysis of SLC44A4. (C) MF enrichment analysis of SLC44A4. (D) KEGG pathway enrichment analysis of SLC44A4. BP, biological process; CC, cellular component; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular function.

GSEA of the SLC44A4 gene expression

Using TCGA gene expression data, GSEA was performed to determine the biological and functional pathways between high- and low-SLC44A4 gene expression groups. Based on the normalized enrichment scores, the enrichment signaling pathway that was determined to be the most relevant for SLC44A4 gene expression was chosen (Figure 4). The GSEA analysis revealed that the low SLC44A4 gene expression phenotype was significantly and predominantly concentrated in the Olfactory signaling pathway, signaling by nuclear receptors, Tcf dependent signaling in response to Wnt, Esr mediated signaling, signaling by Nortch, signaling by Wnt, Eryth pathway and interleukin 7 signaling.

Figure 4 Results obtained from the gene set enrichment analysis. GSEA results indicated that colorectal cancer with low SLC44A4 mRNA level was significantly enriched with Olfactory signaling pathway, signaling by nuclear receptors, Tcf dependent signaling in response to Wnt, Esr mediated signaling, signaling by Nortch, signaling by Wnt, Eryth pathway and interleukin 7 signaling. FDR, false discovery rate; GSEA, gene set enrichment analysis; NES, normalized enrichment score.

The association between SLC44A4 expression and clinicopathological variables in CRC patients

To detect SLC44A4 expression in CRC tissues, CRC cells were stained for immunohistochemical analysis. The expression level of SLC44A4 protein in CRC was shown in Figure 5. SLC44A4 protein was strongly expressed in normal colorectal tissues (Figure 5A-5D). However, SLC44A4 expression varies in CRC tissues in different individuals. High scores of SLC44A4 in CRC were found in 63 cases (45.0%). As shown in Table 1, among all the available parameters, pN stage, clinical stage, and survival outcome significantly correlated to SLC44A4 expression (P<0.05). It also proved that there was no significant association between SLC44A4 expression and other clinicopathological variables, including tumor differentiation, pT stage, pM stage, and tumor relapse (P>0.05).

Figure 5 Expression of SLC44A4 protein in CRC tissues and normal colorectal tissues (immunohistochemical staining). (A-D) Strong expression of SLC44A4 in normal colorectal tissues (A,C: ×40; B,D: ×100). (E,I) Negative expression of SLC44A4 in CRC tissues (E: ×40, I: ×100); (F,J) low expression of SLC44A4 in CRC tissues (F: ×40, J: ×100); (G,K) moderate expression of SLC44A4 in CRC tissues (G: ×40, K: ×100); (H,L) strong expression of SLC44A4 in CRC tissues (H: ×40, L: ×100). CRC, colorectal cancer.

The association between SLC44A4 expression and the survival of CRC patients

As shown in Table 2, univariate analysis demonstrated that conventional clinicopathological prognostic factors significantly impacted the survival rate of CRC patients, such as pM stage, pN stage, tumor relapse and clinical stage (P<0.001). Meanwhile, low expression of SLC44A4 was significantly related to worse OS (P<0.001, Figure 6A). Moreover, a significant correlation was found between SLC44A4 expression and OS in CRC patients with N0 stage (without lymph node metastasis) or with M0 stage (without distant metastasis), suggesting that SLC44A4 is a prognostic, predictive factor in this subset of patients as well (Figure 6B,6C). Next, we identified that there was no significant association between SLC44A4 expression and DFS (P=0.10, Figure 6D). Furthermore, multivariate analysis by Cox regression model also proved that SLC44A4 expression is an independent prognostic factor for OS (hazard ratio: 0.489, 95% CI: 0.268–0.892, P=0.02; Table 3).

Figure 6 Kaplan-Meier survival analysis of SLC44A4 expression in patients with CRC for OS and DFS. (A) Kaplan-Meier curve of SLC44A4 expression level in total CRC patients for OS. (B) Kaplan-Meier curve of SLC44A4 expression level in CRC patients at M0 stage for OS. (C) Kaplan-Meier curve of SLC44A4 expression level in CRC patients at N0 stage for OS. (D) Kaplan-Meier curve of SLC44A4 expression level in total CRC patients for DFS. CRC, colorectal cancer; DFS, disease-free survival; OS, overall survival.

New prognostic model including pM classification, the absence or presence of relapse, and SLC44A4 expression level

Based on the multivariate analysis results, we proposed a combined clinicopathological prediction model, which contains three prognostic factors, including pM classification, the absence or presence of relapse, and SLC44A4 expression level. We assigned a high-risk group with 2 or 3 prognostic factors among the presence of the pM1 classification, with relapse and low expression level of SLC44A4, an intermediate-risk group with the presence of one factor (the presence of the advanced pM1 classification or with relapse or low expression level of SLC44A4), and a low-risk group with the presence of none (the pM0 classification, without relapse, and high expression level of SLC44A4). We demonstrated that the model we proposed can significantly stratify the risk of OS (high, medium, and low) (Figure 7, P<0.001).

Figure 7 The new proposed model (including pM classification, with or without relapse, and SLC44A4 expression level) can significantly stratify CRC patients’ risk for OS. The new proposed model can significantly stratify the risk of overall survival (high, medium, and low) (P<0.001). CRC, colorectal cancer; OS, overall survival; pM, pathological metastasis.

In our research, C-index was used to evaluate the proposed new prognostic model. We found that our prognostic model had a better predictive ability compared to the prognostic model without SLC44A4 expression level (C-index of 0.819 vs. 0.773).

Discussion

CRC is one of the most prevalent and lethal carcinomas, leading to about one in 10 cancer-related deaths (3). Therefore, it is necessary to explore ideal and individual treatment methods of patients with CRC. At the same time, understanding the prognosis of CRC patients could help improve the treatment methods. Traditional prognostic factors include the pM stage, pN stage, and TNM stage. However, to some extent, it is difficult to use only these traditional factors to predict the prognosis of CRC patients owing to the heterogeneity of the tumor.

This study comprehensively investigated the role of SLC44A4 in CRC. Through bioinformatics analysis of multiple databases and IHC on clinical samples, we found that SLC44A4 was downregulated in CRC tissues, which was consistent with its potential role as a tumor suppressor. The significant association between SLC44A4 expression and OS, especially in patients with N0 or M0 stage, indicated its value as a prognostic biomarker. In univariate and multivariate analyses, SLC44A4 expression emerged as an independent prognostic factor, suggesting that it could provide additional prognostic information beyond traditional clinicopathological factors. This research is valuable for clinicians to better classify the prognosis of patients and make more personalized treatment decisions.

The correlation between SLC44A4 expression and immune cell infiltration in CRC is an interesting aspect. SLC44A4 shows a negative correlation with Tcm, T helper cells, Tem, Th1 cells, and macrophages, and a positive correlation with Th17 cells, B cells, eosinophils, and NK CD56bright cells. This implies that SLC44A4 could play a crucial role in shaping the immune microenvironment of CRC. Tcm and Th1 cells are important for anti-tumor immunity (40), and the negative correlation with SLC44A4 suggests that low SLC44A4 expression might be linked to an unfavorable anti-tumor immune state. Conversely, the positive correlation with Th17 cells may be related to the complex role of Th17 cells in cancer, which can either promote or inhibit tumor growth depending on the circumstances (41,42) .These findings indicate that SLC44A4 is likely involved in the communication between cancer cells and the immune system, and further research is required to clarify the mechanisms involved.

The findings from GO and KEGG enrichment analyses, along with GSEA, provided insights into the biological functions of SLC44A4 in CRC. The enrichment in processes related to nucleosome assembly, chromatin remodeling, and regulation of membrane potential suggests that SLC44A4 may influence CRC development at the epigenetic and cellular membrane function levels. The significant enrichment in pathways such as systemic lupus erythematosus, neuroactive ligand-receptor interaction, and olfactory transduction is intriguing. Although the direct connection to CRC may not be immediately obvious, these pathways may be involved in the complex BP underlying CRC development and progression.

In the research conducted by Sabui et al. (43), it was demonstrated that SLC44A4 may function as a potential colon tumor suppressor through various molecular pathways. This transporter plays a crucial role in the uptake of TPP within the colon, facilitating the acquisition of active vitamin B1 by colon cells, which is vital for sustaining the equilibrium of energy metabolism and preventing metabolic disruptions that can lead to carcinogenesis. Furthermore, by ensuring adequate TPP absorption, SLC44A4 helps to suppress the aberrant activation of pro-inflammatory genes, including Tnf-α, IFN-γ, and Nos2, thereby mitigating chronic inflammation within the colon. A deficiency in this MF can directly result in metabolic dysregulation and heightened inflammation, significantly elevating the risk of colon cancer, which aligns with the fundamental characteristics associated with tumor suppressors. Additionally, our investigation corroborated that the expression of SLC44A4 is markedly diminished in cases of colon cancer. SLC44A4 is a member of the choline transporter-like (CTL) family, which is crucial for the production of acetylcholine. The biosynthesis of acetylcholine is vital for various cellular processes, including metabolism and energy production (44-46). ChAT-mediated acetylcholine synthesis is particularly significant in modulating inflammatory responses, regulating blood pressure, and supporting neurophysiological functions. For instance, ChAT activity in the bloodstream is responsive to inflammatory stimuli, subsequently inhibiting the secretion of cellular factors. Furthermore, mutations in ChAT have been shown to impact acetylcholine synthesis and influence behavior in temperature-regulated contexts. These observations underline the intricate relationship between acetylcholine production and cellular metabolism, indicating a probable connection between the SLC44A4 transporter and metabolic pathways involved in energy generation.

Our results of the SLC44A4 expression level in tumor and normal tissues and the association between SLC44A4 expression and clinical outcome of CRC patients are similar to previous research. In this study, we demonstrated that low expression of SLC44A4 was associated with worse OS by univariate analysis. One study showed that the mRNA expression level of SLC44A4 in ccRCC was significantly reduced, and the high expression level of SLC44A4 was related to the improved OS and DFS (35). Nevertheless, there was no significant difference in the protein expression level of SLC44A4 (35). Nabokina et al. studied the molecular mechanism of SLC44A4 expression, focusing on the basic transcriptional activity of SLC44A4 in colonic epithelial cells (47). They demonstrated that cAMP-response element (CRE)-binding protein-1 (CREB-1) transcription factors and ELF3 play an important role in regulating the basic activity of the SLC44A4 promoter in colonic cells (47). Recently, it was reported that SLC44A4 may be a potential “ulcerative colitis susceptibility” by gene genome-wide association studies (GWAS) (48,49). Meanwhile, Anthonymuthu et al. found the expression level of cTPPT (product of SLC44A4 gene) in colonic tissues of patients with active ulcerative colitis was lower than that in healthy controls (33). Moreover, they demonstrated that cTPPT expression levels are inhibited via JNK and ERK1/2 signaling pathways when colonocytes are exposed to TNFα, thereby colonic uptake of TPP and free thiamin was inhibited (33). TNFα is an important pro-inflammatory cytokine whose level is significantly induced in conditions related to intestinal inflammation, such as inflammatory bowel diseases (IBD) (50).

Our results also supported the view that pM classification, in the absence or presence of relapse and SLC44A4 expression level, can improve the predictive ability to determine the prognosis of CRC patients. The main prognostic factor of CRC is postoperative tumor lymph node metastasis (pTNM) staging, which currently drives the therapeutic strategies after surgical operation (51). However, based on pTNM staging, the prognosis of CRC patients is not always predictable. In addition, depending on the specific clinicopathological characteristics and the extent of the tumor, the pTNM stage may have limitations in providing key information that affects prognosis and treatment methods. Patients with tumors of the same clinical stage may have different clinical outcomes and long-term prognoses (52). Hence, it is necessary to find more novel prognostic factors and methods that may effectively distinguish patients with different clinical outcomes to receive the most appropriate therapeutic strategies. We demonstrated that the SLC44A4 expression level evaluated by IHC could predict the prognosis of CRC patients, and SLC44A4 could be regarded as an independent favorable prognostic factor. The predictive ability of established clinicopathologic models can be improved when combined with novel biomarkers. Therefore, the combined prognosis model may become more meaningful and useful for predicting the clinical outcomes of CRC patients.

However, there are some limitations in this study. First, the sample size of the IHC analysis was relatively small, which may limit the generalizability of the results. Second, the study was mainly based on retrospective data, and potential confounding factors may not have been fully controlled. Future studies should aim to enroll larger, prospectively collected cohorts to validate the prognostic value of SLC44A4 and the new combined model. Mechanistic studies are also needed to clarify how SLC44A4 exerts its effects on CRC development, progression, and the immune response. Moreover, exploring the potential of SLC44A4 as a therapeutic target or a biomarker for guiding targeted therapy in CRC could be an exciting direction for future research.

Conclusions

In conclusion, our results demonstrated that SLC44A4 was an independent favorable prognostic factor of patients with CRC. Furthermore, the new prognostic model we proposed had a better predictive value for patients with CRC than previous models. Hence, the SLC44A4 expression level can be regarded as an additional effective tool for predicting patients with CRC at high risk of tumor progression. Moreover, SLC44A4 can be a potential target for the clinical diagnosis, prognosis, and treatment of CRC patients in the future.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the REMARK reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-973/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-973/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-973/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-973/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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Sir Run Run Shaw Hospital, Zhejiang University School of Medicine Institutional Review Board (No. 20220305-33) and individual consent for this analysis was waived due to the retrospective nature.

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. Siegel RL, Miller KD, Goding Sauer A, et al. Colorectal cancer statistics, 2020. CA Cancer J Clin 2020;70:145-64. [Crossref] [PubMed]
2. Arnold M, Abnet CC, Neale RE, et al. Global Burden of 5 Major Types of Gastrointestinal Cancer. Gastroenterology 2020;159:335-349.e15. [Crossref] [PubMed]
3. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
4. Xu Y, Liu J, Jiang T, et al. PDRG1 predicts a poor prognosis and facilitates the proliferation and metastasis of colorectal cancer. Exp Cell Res 2021;409:112924. [Crossref] [PubMed]
5. Zhu Y, Chen S, Ling Z, et al. Comprehensive Analysis of a tRNA-Derived Small RNA in Colorectal Cancer. Front Oncol 2021;11:701440. [Crossref] [PubMed]
6. Ogunwobi OO, Mahmood F, Akingboye A. Biomarkers in Colorectal Cancer: Current Research and Future Prospects. Int J Mol Sci 2020;21:5311. [Crossref] [PubMed]
7. Maertens V, Stefan S, Rawlinson E, et al. Emergency robotic colorectal surgery during the COVID-19 pandemic: A retrospective case series study. Laparoscopic, Endoscopic and Robotic Surgery 2022;5:57-60. [Crossref] [PubMed]
8. Weiser MR. AJCC 8th Edition: Colorectal Cancer. Ann Surg Oncol 2018;25:1454-5.
9. Beklen H, Yildirim E, Kori M, et al. Systems-level biomarkers identification and drug repositioning in colorectal cancer. World J Gastrointest Oncol 2021;13:638-61. [Crossref] [PubMed]
10. Amado RG, Wolf M, Peeters M, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol 2008;26:1626-34. [Crossref] [PubMed]
11. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 2008;359:1757-65. [Crossref] [PubMed]
12. Lin L, Yee SW, Kim RB, et al. SLC transporters as therapeutic targets: emerging opportunities. Nat Rev Drug Discov 2015;14:543-60. [Crossref] [PubMed]
13. Diallinas G. Understanding transporter specificity and the discrete appearance of channel-like gating domains in transporters. Front Pharmacol 2014;5:207. [Crossref] [PubMed]
14. Mueckler M, Thorens B. The SLC2 (GLUT) family of membrane transporters. Mol Aspects Med 2013;34:121-38. [Crossref] [PubMed]
15. Wright EM. Glucose transport families SLC5 and SLC50. Mol Aspects Med 2013;34:183-96. [Crossref] [PubMed]
16. Babu E, Bhutia YD, Ramachandran S, et al. Deletion of the amino acid transporter Slc6a14 suppresses tumour growth in spontaneous mouse models of breast cancer. Biochem J 2015;469:17-23. [Crossref] [PubMed]
17. Wang Q, Holst J. L-type amino acid transport and cancer: targeting the mTORC1 pathway to inhibit neoplasia. Am J Cancer Res 2015;5:1281-94.
18. Nicklin P, Bergman P, Zhang B, et al. Bidirectional transport of amino acids regulates mTOR and autophagy. Cell 2009;136:521-34. [Crossref] [PubMed]
19. Golovine K, Makhov P, Uzzo RG, et al. Overexpression of the zinc uptake transporter hZIP1 inhibits nuclear factor-kappaB and reduces the malignant potential of prostate cancer cells in vitro and in vivo. Clin Cancer Res 2008;14:5376-84. [Crossref] [PubMed]
20. Zheng PF, Yin RX, Guan YZ, et al. Association between SLC44A4-NOTCH4 SNPs and serum lipid levels in the Chinese Han and Maonan ethnic groups. Nutr Metab (Lond) 2020;17:105. [Crossref] [PubMed]
21. Revathikumar P, Estelius J, Karmakar U, et al. Microsomal prostaglandin E synthase-1 gene deletion impairs neuro-immune circuitry of the cholinergic anti-inflammatory pathway in endotoxaemic mouse spleen. PLoS One 2018;13:e0193210. [Crossref] [PubMed]
22. Arredondo J, Lara M, Gospe SM Jr, et al. Choline Acetyltransferase Mutations Causing Congenital Myasthenic Syndrome: Molecular Findings and Genotype-Phenotype Correlations. Hum Mutat 2015;36:881-93. [Crossref] [PubMed]
23. Tsai KY, Wei PL, Lee CC, et al. Solute Carrier Family 35 A2 (SLC35A2) Promotes Tumor Progression through MYC-Mediated Pathways in Colorectal Cancer. Int J Med Sci 2025;22:1992-2009. [Crossref] [PubMed]
24. Dong Z, Gu W, Xu L. Exploring the mechanism of miR320a in regulating PDL1 upon lung cancer pathogenesis. Acta Biochim Pol 2022;69:573-8. [Crossref] [PubMed]
25. Gao LF, Zhong Y, Long T, et al. Tumor bud-derived CCL5 recruits fibroblasts and promotes colorectal cancer progression via CCR5-SLC25A24 signaling. J Exp Clin Cancer Res 2022;41:81. [Crossref] [PubMed]
26. Gurav A, Sivaprakasam S, Bhutia YD, et al. Slc5a8, a Na+-coupled high-affinity transporter for short-chain fatty acids, is a conditional tumour suppressor in colon that protects against colitis and colon cancer under low-fibre dietary conditions. Biochem J 2015;469:267-78. [Crossref] [PubMed]
27. Sikder MOF, Sivaprakasam S, Brown TP, et al. SLC6A14, a Na+/Cl--coupled amino acid transporter, functions as a tumor promoter in colon and is a target for Wnt signaling. Biochem J 2020;477:1409-25. [Crossref] [PubMed]
28. Ma Z, Xia W, Liu F, et al. SLC44A4 mutation causes autosomal dominant hereditary postlingual non-syndromic mid-frequency hearing loss. Hum Mol Genet 2017;26:383-94. [Crossref] [PubMed]
29. O’Regan S, Traiffort E, Ruat M, et al. An electric lobe suppressor for a yeast choline transport mutation belongs to a new family of transporter-like proteins. Proc Natl Acad Sci U S A 2000;97:1835-40. [Crossref] [PubMed]
30. Song P, Rekow SS, Singleton CA, et al. Choline transporter-like protein 4 (CTL4) links to non-neuronal acetylcholine synthesis. J Neurochem 2013;126:451-61. [Crossref] [PubMed]
31. Nabokina SM, Inoue K, Subramanian VS, et al. Molecular identification and functional characterization of the human colonic thiamine pyrophosphate transporter. J Biol Chem 2014;289:4405-16. [Crossref] [PubMed]
32. Sabui S, Romero JM, Said HM. Developmental maturation of the colonic uptake process of the microbiota-generated thiamin pyrophosphate. Am J Physiol Gastrointest Liver Physiol 2021;320:G829-35. [Crossref] [PubMed]
33. Anthonymuthu S, Sabui S, Sheikh A, et al. Tumor necrosis factor α impedes colonic thiamin pyrophosphate and free thiamin uptake: involvement of JNK/ERK(1/2)-mediated pathways. Am J Physiol Cell Physiol 2022;323:C1664-80. [Crossref] [PubMed]
34. Mattie M, Raitano A, Morrison K, et al. The Discovery and Preclinical Development of ASG-5ME, an Antibody-Drug Conjugate Targeting SLC44A4-Positive Epithelial Tumors Including Pancreatic and Prostate Cancer. Mol Cancer Ther 2016;15:2679-87. [Crossref] [PubMed]
35. Kang W, Zhang M, Wang Q, et al. The SLC Family Are Candidate Diagnostic and Prognostic Biomarkers in Clear Cell Renal Cell Carcinoma. Biomed Res Int 2020;2020:1932948. [Crossref] [PubMed]
36. Yu G, Wang LG, Han Y, et al. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 2012;16:284-7. [Crossref] [PubMed]
37. Ma W, Li T, Wu S, et al. LOX and ACSL5 as potential relapse markers for pancreatic cancer patients. Cancer Biol Ther 2019;20:787-98. [Crossref] [PubMed]
38. Anandam KY, Srinivasan P, Subramanian VS, et al. Molecular mechanisms involved in the adaptive regulation of the colonic thiamin pyrophosphate uptake process. Am J Physiol Cell Physiol 2017;313:C655-63. [Crossref] [PubMed]
39. Chu Y, Li J, Zeng Z, et al. A Novel Model Based on CXCL8-Derived Radiomics for Prognosis Prediction in Colorectal Cancer. Front Oncol 2020;10:575422. [Crossref] [PubMed]
40. Haabeth OA, Lorvik KB, Hammarström C, et al. Inflammation driven by tumour-specific Th1 cells protects against B-cell cancer. Nat Commun 2011;2:240. [Crossref] [PubMed]
41. Wilke CM, Kryczek I, Wei S, et al. Th17 cells in cancer: help or hindrance? Carcinogenesis 2011;32:643-9. [Crossref] [PubMed]
42. Ye J, Livergood RS, Peng G. The role and regulation of human Th17 cells in tumor immunity. Am J Pathol 2013;182:10-20. [Crossref] [PubMed]
43. Sabui S, Anthonymuthu S, Ramamoorthy K, et al. Effect of knocking out mouse Slc44a4 on colonic uptake of the microbiota-generated thiamine pyrophosphate and colon physiology. Am J Physiol Gastrointest Liver Physiol 2024;327:G36-46. [Crossref] [PubMed]
44. Gabalski AH, Tynan A, Tsaava T, et al. Circulating extracellular choline acetyltransferase regulates inflammation. J Intern Med 2024;295:346-56. [Crossref] [PubMed]
45. Duerr JS, McManus JR, Crowell JA, et al. Analysis of Caenorhabditis elegans acetylcholine synthesis mutants reveals a temperature-sensitive requirement for cholinergic neuromuscular function. Genetics 2021;218:iyab078. [Crossref] [PubMed]
46. Stiegler A, Li JH, Shah V, et al. Systemic administration of choline acetyltransferase decreases blood pressure in murine hypertension. Mol Med 2021;27:133. [Crossref] [PubMed]
47. Nabokina SM, Ramos MB, Valle JE, et al. Regulation of basal promoter activity of the human thiamine pyrophosphate transporter SLC44A4 in human intestinal epithelial cells. Am J Physiol Cell Physiol 2015;308:C750-7. [Crossref] [PubMed]
48. Gupta A, Juyal G, Sood A, et al. A cross-ethnic survey of CFB and SLC44A4, Indian ulcerative colitis GWAS hits, underscores their potential role in disease susceptibility. Eur J Hum Genet 2016;25:111-22. [Crossref] [PubMed]
49. Gupta A, Thelma BK. Identification of critical variants within SLC44A4, an ulcerative colitis susceptibility gene identified in a GWAS in north Indians. Genes Immun 2016;17:105-9. [Crossref] [PubMed]
50. Ruder B, Atreya R, Becker C. Tumour Necrosis Factor Alpha in Intestinal Homeostasis and Gut Related Diseases. Int J Mol Sci 2019;20:1887. [Crossref] [PubMed]
51. Barresi V, Reggiani Bonetti L, Ieni A, et al. Poorly Differentiated Clusters: Clinical Impact in Colorectal Cancer. Clin Colorectal Cancer 2017;16:9-15. [Crossref] [PubMed]
52. De Divitiis C, Nasti G, Montano M, et al. Prognostic and predictive response factors in colorectal cancer patients: between hope and reality. World J Gastroenterol 2014;20:15049-59. [Crossref] [PubMed]