SAMD4B: a novel prognostic biomarker of endometrial carcinoma

Abstract

Background

Endometrial cancer (EC) is a prevalent gynecologic cancer; however, the EC risk factors based on alterations in individual gene expression require further investigation. Sterile alpha motif domain-containing protein 4 (SAMD4) is an evolutionary conserved gene that has been linked to various cancers. Here, we aimed to investigate the relationship between SAMD4B expression and EC.

Methods

Microarray data of SAMD4B were obtained from the Gene Expression Omnibus datasets GSE17025 and GSE29981; RNA sequencing data and clinical and survival data were downloaded from the Cancer Genome Atlas cohort. The datasets were analyzed using Affymetrix GeneChip^TM Human Genome U133 Plus 2.0 Array, and the patients were divided into four sub-groups based on DNA polymerase epsilon mutation, microsatellite instability/mismatch repair deficiency, TP53 gene mutation, and microsatellite stability status.

Results

SAMD4B expression was higher in EC tissues than that in normal endometrial tissues. SAMD4B overexpression was associated with poor overall survival (OS) of patients with EC [P<0.001, hazard ratio (HR) =2.254, 95% confidence interval (CI): 1.447–3.510]. Cox regression analysis showed that high SAMD4B expression, clinical stage, histologic grade, and positive cytology were independent prognostic factors for poor OS in patients with EC. In the subgroup analysis, SAMD4B mean expression was the highest in the TP53 mutation group, which showed the worst OS.

Conclusions

SAMD4B may serve as a novel prognostic biomarker for patients with EC.

Highlight box.

Key findings

• This study found that SAMD4B is significantly overexpressed in endometrial cancer (EC) tissues compared to normal tissues, and its high expression is independently associated with poor overall survival, particularly in patients with TP53 mutations.

What is known and what is new?

• SAMD4B has been previously linked to the development and progression of various types of cancers.

• This study provides new evidence that SAMD4B is overexpressed in EC and may serve as a novel prognostic biomarker, particularly in genetically defined subgroups such as those with TP53 mutations.

What is the implication, and what should change now?

• The findings suggest that SAMD4B expression could guide prognostication and risk stratification in EC. Further research and clinical validation could support its inclusion in molecular diagnostic panels and inform personalized treatment strategies.

Introduction

Background

Endometrial cancer (EC) is one of the most common gynecologic malignancies and ranks as the fourth most prevalent cancer among women in the United States (1). Globally, its rapidly increasing incidence has become a major health concern (2). Notably, the incidence of EC varies geographically (3). For example, it is high in developed countries like North America (4) but low in North African countries (3). This difference can be partly explained by the fact that EC is largely driven by risk factors such as obesity, which is a major health concern in developed countries (5,6). Other well-known risk factors for EC include exposure to unopposed estrogen, age, diabetes, and endometrial hyperplasia (5,7). In addition, recent molecular analyses have identified four distinct subtypes of EC based on molecular features and risk stratification (8). These subtypes are DNA polymerase epsilon (POLE) mutations, microsatellite instability (MSI) or mismatch repair deficiency (MMRd), microsatellite stability (MSS) or low copy number variation (CNV), and TP53 gene mutations or high CNV (9,10). However, risk factors for EC based on alterations in individual gene expression require further investigation to enable molecular risk assessment.

The sterile alpha motif domain containing protein 4 (SAMD4) is an evolutionarily conserved gene that plays crucial roles in various biological functions (11). In mammalian cells, the SAMD4 protein family comprises SAMD4A and SAMD4B (12); the genes that encode these proteins (SAMD4A and SAMD4B) are located on human chromosomes 14 and 19, respectively (12). SAMD4A and SAMD4B proteins contain 629 and 694 amino acids, respectively (13,14). SAMD4 directly binds to target mRNA and mediates their transcription and translation (15). Dysregulation or mutations in mRNA-binding proteins can lead to various diseases (16) and several studies have reported a relationship between aberrant SAMD4 expression and diseases such as myopathy and cancers (17-19). However, few studies have been conducted on SAMD4B and its role in cancer development. One such study revealed that SAMD4B overexpression inhibits p53 and p21 (14), which leads to cell cycle dysregulation and malignant proliferation (20). This finding highlights the importance of investigating the relationship between SAMD4B expression and cancer development.

Rationale and knowledge gap

EC is a rising global health concern and its incidence is influenced by factors like obesity and hormonal imbalance. While molecular subtypes of EC have been defined, gene-specific risk factors remain underexplored. SAMD4B, an mRNA-binding protein linked to cancer-related pathways, may influence tumor progression, but its role in EC is unknown. This study addresses the lack of research on SAMD4B in the context of EC.

Objective

This study aimed to investigate the relationship between SAMD4B expression and EC, and assess its potential as a prognostic biomarker. We present this article in accordance with the REMARK reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1648/rc).

Methods

Acquisition of gene expression data

Gene expression microarray datasets from patients with EC were obtained from the Gene Expression Omnibus database (National Institutes of Health, Bethesda, MD, USA; http://www.ncbi.nlm.nih.gov/geo). The datasets GSE29981 and GSE17025 included data from 20 healthy endometrial tissues and 79 EC tissues, respectively (21,22). Both datasets were analyzed using the Affymetrix GeneChip^TM Human Genome U133 Plus 2.0 Array (Affymetrix, Santa Clara, CA, USA). These data were merged using the Robust Multiarray Average algorithm and quantile normalized using R (version 3.4.1; http://cran.r-project.org). To calculate the fold change (FC) in SAMD4B expression levels between normal and cancerous tissues, DESeq2 and other Bioconductor packages were used (https://bioconductor.org/packages/release/bioc/html/DESeq2.html). The results were expressed as log₂FC, and an adjusted P value <0.0001 was considered statistically significant.

Acquisition of clinical and RNA expression data

Clinical and RNA expression data were downloaded from the UCSC Xena Browser (http://xenabrowser.net). Clinical outcomes followed the standardized definitions and curation of the TCGA Pan-Cancer Clinical Data Resource (23). These included: clinical data (dataset ID: TCGA-UCEC.GDC_phenotype), survival data (dataset ID: TCGA-UCEC.survival), and RNA-sequencing data for gene expression (dataset ID: TCGA-UCEC.Htseq_fpkm) from patients with EC. Among the RNA sequencing data samples, only 532 primary tumor samples were included in the analyses after excluding data with insufficient clinical information. The clinical data included age, clinical stage and histologic grade of cancer, presence or absence of diabetes, hypertension, hormone replacement therapy (HRT), menopause, cytology results, and histologic subtype. On the basis of each clinical variable, patients were grouped as follows: age <60 years and ≥60 years; clinical stage I or II, and stage III or IV; with or without hypertension; with or without diabetes; history of using HRT or not; menopausal state; histologic grade 1 or 2, and grade 3; positive or negative cytology; and adenocarcinoma subtype—endometrioid or serous. Fragments per kilobase million (FPKM) were used to quantify the gene expression levels. This study was performed in accordance with the publication guidelines provided by The Cancer Genome Atlas (http://www.cancer.gov/about-nci/organization/ccg/research/structural-genomics/tcga/using-tcga/citing-tcga) and was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Subgroup analysis

The patients were categorized into four groups: POLE mutation, MSI-h/MMRd, TP53 gene mutation, and MSS (24). By analyzing somatic mutation data downloaded from the USCS Xena browser (dataset ID: TCGA-UCEC.muse_snv), patients who had any of the following were grouped into the POLE mutation group: intron variants, missense variants, synonymous variants, splice acceptor variants, or splice region variants of the POLE gene. Methylation data downloaded from the USCS Xena browser (dataset ID: TCGA-UCEC.methylation450) were analyzed to identify patients with MLH1 promoter hypermethylation; using the Illumina Infinium Human Methylation450 BeadChip, the obtained methylation beta values >0.9512 were categorized into the MSI-h/MMRd subgroup. Patients with TP53 gene mutations—including intron variants, missense variants, synonymous variants, splice acceptor variants, or splice region variants, were grouped together. The remaining patients who did not match any of the above criteria were grouped into the MSS or low copy number group.

Survival analysis

Survival data of patients with EC were obtained from the UCSC Xena browser (dataset ID: TCGA-UCEC.survival). The survival data of 528 patients were analyzed after excluding the available data. Using the cut-off value of the median SAMD4B expression level in patients with EC, patients were grouped into high and low expression groups. The data were analyzed using Kaplan-Meier survival and Cox regression analyses with the survival package in R (version 3.5-5; http://CRAN.Rproject.org/package=survival). A log-rank test was conducted, and P<0.05 was considered statistically significant.

Statistical analysis

For the analysis of clinical parameters, one-way analysis of variance (ANOVA) and t-tests were used. Levene’s test was conducted before the t-test, and P<0.05 indicated a non-parametric distribution of variances. Where significant results were observed in the ANOVA, a Bonferroni post-hoc test was conducted. Statistical significance was set as P<0.05 for both the t-test and ANOVA, and at P<0.025 for the post-hoc test. Univariate and multivariate analyses were performed using Cox proportional hazard models. Variables with P values <0.05 in the univariate analysis were subsequently analyzed in the multivariate analysis.

External in-silico validation

To externally validate our findings, we analyzed data using UALCAN portal (http://ualcan.path.uab.edu). Patients were stratified into high versus low/medium SAMD4B expression groups according to the portal’s default algorithm, and Kaplan-Meier survival curves were generated. In addition, transcript level of SAMD4B expression was compared between TP53 mutation and non-mutation groups as provided by UALCAN.

Results

SAMD4B overexpression is associated with poor survival of patients with EC

Expression level analysis of SAMD4B in normal endometrial and EC tissues showed that the SAMD4B expression was higher in EC tissues than that in normal endometrial tissues (Log₂FC =1.149783, adjusted P=5.18e−07). Kaplan-Meier survival analysis was performed using data from the 528 patients with EC. These patients were grouped based on the median cut-off value of SMAD4B expression, resulting in 262 and 266 patients with EC in the high and low expression groups, respectively. The results showed that SAMD4B overexpression was associated with poor overall survival (OS) of patients with EC [P<0.001, hazard ratio (HR) =2.254, 95% confidence interval (CI): 1.447–3.510] (Figure 1). To further validate our results, we utilized UALCAN portal. Patients with high SAMD4B expression (n=135) had significantly worse OS compared to those with low/medium expression (n=408; log-rank P<0.001) (Figure S1).

Figure 1.

Survival analysis of patients with endometrial cancer showing SAMD4B expression. Kaplan-Meier analysis was conducted on patients with endometrial cancer showing relative high or low expression of SAMD4B. Data are presented as HR with 95% confidence interval. CI, confidence interval; HR, hazard ratio.

SAMD4B is independently correlated with poor OS in patients with EC

The results of both univariate and multivariate Cox regression analyses for each clinical variable in patients with EC are presented in Table 1. In the univariate analysis, a higher expression level of SAMD4B, clinical stage, histologic grade, and positive cytology were associated with poor OS in patients with EC (SAMD4B expression level: HR =1.909, 95% CI: 1.427–2.2554, P<0.001; clinical stage: HR =4.094, 95% CI: 2.672–6.275, P<0.001; histologic grade: HR =3.595, 95% CI:2.060–6.274, P<0.001; positive cytology: HR =4.809, 95% CI: 2.918–7.925, P<0.001). Similarly, in the multivariate analysis, high SAMD4B expression levels, clinical stage, histologic grade, and positive cytology were identified as independent prognostic factors for poor OS (SAMD4B expression level: HR =1.589, 95% CI:1.050–2.404, P=0.03; clinical stage: HR =2.256, 95% CI: 1.272–4.004, P=0.005; histologic grade: HR =2.340, 95% CI: 1.204–4.548, P=0.01; positive cytology, HR =2.647, 95% CI: 1.479–4.737, P=0.001).

Table 1. Cox regression analysis of variables.

Variables	Univariate				Multivariate
	HR	95% CI	P value		HR	95% CI	P value
SAMD4B level	1.909	1.427–2.554	<0.001		1.589	1.050–2.404	0.03
Age (years)
<60	1	–	–		–	–	–
≥60	2.126	1.265–3.573	0.09
Stage
Stage I and II	1	–	–		–	–	–
Stage III and IV	4.094	2.672–6.275	<0.001		2.256	1.272–4.004	<0.001
Diabetes
No	1	–	–		–	–	–
Yes	1.196	0.695–2.060	0.51		–	–	–
Hypertension
No	1	–	–		–	–	–
Yes	1.143	0.685–1.910	0.60		–	–	–
HRT
No	1	–	–		–	–	–
Yes	0.923	0.410–2.080	0.85		–	–	–
Menopause
No	1	–	–		–	–	–
Yes	1.129	0.545–2.341	0.74		–	–	–
Grade
1, 2	1	–	–		–	–	–
3	3.595	2.060–6.274	<0.001		2.340	1.204–4.548	0.01
Cytology
Negative	1	–	–		–	–	–
Positive	4.809	2.918–7.925	<0.001		2.647	1.479–4.737	<0.001
Histology
Endometrioid	1	–	–		–	–	–
Serous	1.181	0.746–1.870	0.48

CI, confidence interval; HR, hazard ratio; HRT, hormone replacement therapy.

SAMD4B overexpression is associated with higher histologic grade and advanced stage of EC

The mean expression level of SAMD4B, according to each clinical variable, was analyzed using a t-test. SAMD4B expression levels did not differ significantly based on age, presence/absence of hypertension, diabetes, HRT, menopausal status, cytology, or histology. The mean expression level of SAMD4B was positively correlated with histologic grade and clinical stage (Table 2, Figure 2).

Table 2. Mean SAMD4B expression level in each clinical parameter.

Parameters	Number	Mean gene expression (FPKM)	P value
Age (years)
<60	179	4.050	0.15
≥60	353	4.114
Clinical stage
Stage I and II	332	4.026	<0.001
Stage III and IV	200	4.208
Histological grade
Grade 1	98	3.855	<0.001
Grade 2	118	3.965
Grade 3	316	4.215
Hypertension
No	157	4.029	0.06
Yes	227	4.144
Diabetes
No	259	4.060	0.15
Yes	99	4.148
HRT
No	240	4.105	0.86
Yes	38	4.120
Menopausal status
Pre/Peri	54	4.024	0.38
Post	452	4.093
Cytology
Negative	345	4.112	0.17
Positive	57	4.221
Histology
Endometrioid	392	4.070	0.13
Serous	140	4.151

FPKM, fragment per kilobase million; HRT, hormone replacement therapy.

Figure 2.

Box plots demonstrating mean expression level of SAMD4B in patients with endometrial cancer, based on (A) histologic grade and (B) clinical stage. *, significant difference based on Bonferroni’s post-hoc test.

SAMD4B is overexpressed in the TP53 mutation group

Patients with EC were sub-grouped into four categories: POLE mutation, MSI-h/MMRd, MSS, and TP53 mutation, including 80, 98, 244, and 110 patients. Survival analysis of each group showed that the POLE mutation group had the most favorable OS, whereas the TP53 mutation group showed the worst OS (Figure 3A). A comparison of the mean expression level between all groups showed that SAMD4B expression was the highest in the TP53 mutation group (Figure 3B). This finding was further supported by an independent analysis using UALCAN, which showed significantly higher SAMD4B transcript levels in TP53 mutant cases (n=196) compared with nonmutant cases (n=345; P<0.001) (Figure S2).

Figure 3.

Expression level of SAMD4B observed in molecular subgroup analysis. (A) Kaplan-Meier analysis of subgroups of patients with endometrial cancer: POLE mutation, MSI-h/MMRd, MSS, TP53 mutation. (B) Mean expression level of SAMD4B in subgroups patients having POLE mutation, MSI-h/MMRd, MSS, TP53 mutation. MMRd, mismatch repair deficiency; MSI-h, micro-satellite instability high; MSS, micro-satellite stable; POLE, polymerase epsilon.

Discussion

The incidence of EC is increasing rapidly worldwide (2). The potential risk factors for this significant increase in EC incidence include obesity, diabetes, and HRT (25). Other well-known risk factors of EC include age, obesity, race, family history, hypertension, smoking, and parity (26). Recent advances in bioinformatics tools have enabled investigators to identify potential therapeutic targets or prognostic markers of EC (27,28). Moreover, large-scale genomic analyses, such as The Cancer Genome Atlas, have enabled the identification of novel classes of EC based on its molecular features, providing new insights into EC management (29,30). However, the association between numerous individual genes and EC risk remains unclear. In this study, we focused on the potential prognostic biomarker SAMD4B in association with EC.

Key findings

In this study, we compared the mean SAMD4B expression levels between normal endometrial and EC tissues using the microarray datasets GSE17025 and GSE29981, respectively. The mean SAMD4B expression level was higher in EC tissues than in normal endometrial tissues. Survival analysis showed that SAMD4B overexpression was associated with poor OS in patients with EC. Mean expression level of SAMD4B was higher in histologic grade 3 than in histologic grades 1 and 2. Moreover, the mean expression level of SAMD4B was higher in EC stages 3 and 4 than in stages 1 and 2. In subgroup analysis, we observed that the mean expression level of SAMD4B was significantly higher in the TP53 mutation group, which showed the worst survival outcome in survival analysis.

Strengths and limitations

The novelty of our study lies in the comprehensive analysis of multiple microarray and RNA sequencing datasets. This study remains a retrospective single-gene investigation. Future research should extend this work by performing genome-wide analyses, co-expression network studies, and pathway enrichment analyses to validate and expand our findings. In addition, larger prospective studies are warranted to confirm the prognostic role of SAMD4B in EC.

Comparison with similar research

SAMD is a conserved protein module that plays a crucial role in diverse cellular processes such as binding to RNA or DNA (31). SAMD4 is a member of the SAMD family and consists of the SAMD4A and SAMD4B proteins. SAMD4 binds directly to the target mRNA and regulates its translation (15,32). Recent studies have identified several biological functions of SAMD4. For example, SAMD4 regulates mRNA stabilization and degradation (15,33) SAMD4 also represses mRNA translation, which is crucial for regulating gene expression (11). Luo et al. reported that SAMD4B is a potential transcriptional repressor that inhibits transcriptional activity in humans (14). Thus, mutations or aberrant gene expression that regulate mRNA function may cause many diseases, specifically human cancers (19). SAMD4 is overexpressed in topotecan-resistant ovarian cancer cells, suggesting that SAMD4 gene is related to the development of drug resistance in ovarian cancer (34,35). Additionally, a study has reported that SAMD4B overexpression promotes colorectal cancer progression, suggesting that SAMD4B is a potential therapeutic target in colorectal cancer (36). Li et al. reported that SAMD4A might be a biomarker of poor prognosis in gastric cancer (17). However, some studies have suggested that SAMD4A is a novel breast cancer angiogenesis inhibitor (13). These studies indicate that SAMD4 is strongly associated with various cancers.

In this study, we found that the mean SAMD4B expression level was higher in EC tissues than in normal endometrial tissues. These findings are consistent with previous reports in colorectal, ovarian, and gastric cancers, where SAMD4B was associated with tumor progression. This suggests that SAMD4B may similarly play a pro-oncogenic role in EC, although the precise molecular mechanisms in endometrial tissue remain to be elucidated. While our study focused on a single gene, such an approach can still yield clinically meaningful insights. For instance, L1CAM expression alone has been validated as a powerful prognostic marker in early-stage EC (37,38). This precedent supports the notion that SAMD4B, despite being evaluated as a single gene, may similarly hold independent prognostic value and contribute to risk stratification in EC.

Explanations of findings

The observed overexpression of SAMD4B in EC and its association with poor survival outcomes may reflect its functional role in tumor progression through post-transcriptional regulatory mechanisms. SAMD4B is known to bind to target mRNAs and repress their translation or influence their stability, thereby exerting widespread control over gene expression (11-13). This mRNA-level regulation may contribute to tumor cell survival, proliferation, and adaptation under cellular stress, such as hypoxia or genotoxic insult.

One possible explanation for the association between high SAMD4B expression and poor prognosis is its interaction with the p53 signaling pathway. Previous studies have shown that SAMD4B can suppress the transcriptional activity of p53 and p21, which are key regulators of cell cycle arrest and apoptosis (15,21). In our study, SAMD4B expression was significantly higher in the TP53 mutation subgroup, which also exhibited the worst survival outcomes among molecular subtypes. This suggests that SAMD4B may exacerbate oncogenic effects in tumors with impaired p53 function, either by further disrupting cell cycle control or by promoting resistance to cell death.

Furthermore, the elevated expression of SAMD4B in advanced-stage and high-grade EC implies its involvement in tumor aggressiveness. As tumors progress, they often acquire post-transcriptional modifications that favor proliferation, invasion, and immune evasion (17,34). SAMD4B may play a role in this process by stabilizing oncogenic transcripts or suppressing tumor suppressor mRNAs. Although not directly assessed in this study, this hypothesis is supported by previous findings in other malignancies, where SAMD4B has been implicated in chemoresistance and metastatic potential (32-34).

Taken together, these observations suggest that SAMD4B contributes to EC progression through multifaceted mechanisms, including disruption of apoptotic regulation, enhancement of malignant transcript stability, and potential interaction with key tumor suppressor pathways. These findings underscore the importance of further functional studies to elucidate the direct downstream targets of SAMD4B and to determine whether it can serve as a therapeutic target in EC.

Implications and actions needed

To the best of our knowledge, this is the first report to suggest that SAMD4B overexpression indicates poor prognosis in patients with EC. Implicating SAMD4B in the novel molecular classification of EC may offer new insights into the understanding of EC. Our findings suggest that SAMD4B expression could guide prognostication and risk stratification in EC. Further research and clinical validation could support its inclusion in molecular diagnostic panels and inform personalized treatment strategies.

Conclusions

In this study, we revealed SAMD4B as a novel prognostic biomarker for patients with EC. We demonstrated that SAMD4B overexpression is significantly associated with poor prognosis in patients with EC, particularly in advanced stages and high-grade tumors. Our findings suggest that SAMD4B may serve as a potential therapeutic target, warranting further large-scale validation studies.

Supplementary

The article’s supplementary files as

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.