PDF

J Cancer 2026; 17(10):1709-1719. doi:10.7150/jca.139096 This issue Cite

Research Paper

Associations of HMGA2 single-nucleotide polymorphisms with clinicopathological features of tongue squamous cell carcinoma in Taiwanese men with distinct lifestyle-related risk factors

Yi-Chieh Yang¹, Chia-Hwa Lee², Yi-Fang Ding³, Chiao-Wen Lin^4,5, Lun-Ching Chang⁶, Yu-Ching Wen^7, , Shun-Fa Yang^8,9, , Ming-Hsien Chien^{2,10,11,12,13,}

1. School of Oral Hygiene, College of Oral Medicine, Taipei Medical University, Taipei, Taiwan.
2. School of Medical Laboratory Science and Biotechnology, College of Medical Science and Technology, Taipei Medical University, Taipei, Taiwan.
3. Department of Otolaryngology, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan.
4. Department of Otolaryngology, Chung Shan Medical University Hospital, Taichung, Taiwan.
5. Institute of Oral Sciences, Chung Shan Medical University, Taichung, Taiwan.
6. Department of Mathematics and Statistics, Florida Atlantic University, Boca Raton, FL, USA.
7. Department of Urology, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan.
8. Institute of Medicine, Chung Shan Medical University, Taichung, Taiwan.
9. Department of Medical Research, Chung Shan Medical University Hospital, Taichung, Taiwan.
10. Graduate Institute of Clinical Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan.
11. Pulmonary Research Center, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan.
12. Traditional Herbal Medicine Research Center, Taipei Medical University Hospital, Taipei, Taiwan.
13. TMU Research Center of Cancer Translational Medicine, Taipei Medical University, Taipei, Taiwan.

✉ Corresponding authors: Ming-Hsien Chien, PhD, Email: d002089002edu.tw. Shun-Fa Yang, PhD, Email: ysfedu.tw. Yu-Ching Wen, MD, PhD, Email: 95207tmu.edu.tw.

Received 2026-6-9; Accepted 2026-8-21; Published 2026-9-11

Abstract

Background: Oral squamous cell carcinoma (OSCC) is a predominant malignancy of the oral cavity, with tongue squamous cell carcinoma (TSCC) accounting for a substantial proportion of cases worldwide. High-mobility group A protein 2 (HMGA2) is a nonhistone chromatin-binding factor implicated in aggressive malignant phenotypes of various human cancers. However, the effects of HMGA2 variants on OSCC susceptibility and tumor clinicopathological aggressiveness remain unclear.

Methods: We examined the associations of functional HMGA2 single-nucleotide polymorphisms (SNPs) with TSCC susceptibility and aggressive tumor characteristics in Taiwanese men. Four tagging SNPs (rs6581658 A>G, rs10573247 T>del, rs968697 T>C, and rs8756 A>C) were genotyped using a TaqMan allelic discrimination assay. In addition, HMGA2 expression patterns, as well as their associations with disease aggressiveness and patient survival outcomes, were analyzed using multiple independent datasets, including TCGA, CPTAC, and GEO.

Results: Our results indicated no significant association between the four SNPs and TSCC susceptibility. However, carriers of the rs6581658 G allele had significantly increased risks of large tumors (>T2) and advanced clinical stage (stage III or IV) in the dominant model. Moreover, the rs10573247 variant was significantly associated with an increased risk of lymph node metastasis in patients exposed to environmental carcinogens, including betel quid and cigarette smoke. By contrast, the rs968697 variant exerted a protective effect against advanced disease and lymph node metastasis in patients without such environmental exposure. Analyses of clinical datasets pertaining to an American cohort with head and neck squamous cell carcinoma and a Taiwanese cohort with OSCC revealed that HMGA2 expression was upregulated in tumor tissues and was associated with poor prognosis.

Conclusion: These findings suggest potential exposure-stratified association in TSCC and highlight HMGA2 polymorphisms as promising biomarkers for risk assessment and disease monitoring in male patients with TSCC.

Keywords: oral squamous cell carcinoma, high-mobility group A protein 2, single-nucleotide polymorphism, lifestyle-related risk, aggressive tumor characteristics, male population

Introduction

Oral squamous cell carcinoma (OSCC) is the most common malignancy of the oral cavity, accounting for approximately 90% of all types of oral cancer. OSCC predominantly originates in the tongue, where it is referred to as tongue squamous cell carcinoma (TSCC). Poor prognosis in OSCC is largely attributable to its aggressive invasion and metastasis patterns, adverse clinicopathological features, and limited treatment response. OSCC is a complex and heterogeneous disease whose development is influenced by genetic, epigenetic, environmental, and lifestyle factors [1-3]. Therefore, identifying reliable genetic biomarkers for TSCC is crucial, and further research is required to determine the roles of candidate genes in TSCC progression.

High-mobility group A protein 2 (HMGA2) is a 160-kb gene located on chromosome 12. As an architectural transcription factor, HMGA2 contains three AT-hook domains that bind to AT-rich sequences in the minor groove of DNA, which alters the chromatin architecture and modulates the assembly and maintenance of enhancer complexes, ultimately regulating the transcription of numerous genes [4, 5]. In addition, HMGA2 functions as a transcriptional coregulator by recruiting other transcription-associated proteins [6]. Overexpression of HMGA2 variants promotes the growth of benign tumors, indicating that HMGA2 promotes cell proliferation and contributes to tumorigenesis [7]. Indeed, HMGA2 overexpression has been implicated in tumor aggressiveness, and chemotherapy resistance across human malignancies, including OSCC [8-11]. The molecular mechanisms underlying HMGA2-mediated carcinogenesis have been extensively studied [8]. For example, HMGA2 interacts with protein phosphatase 4 regulatory subunit 1 to facilitate cell migration and metastasis by activating epithelial-mesenchymal transition (EMT) in non-small cell lung cancer [12]. In OSCC, tumorigenesis is further promoted by the histone lysine methyltransferase SMYD3, which enhances HMGA2 transcription through H3K4me3-mediated epigenetic regulation [13].

Recently, researchers have increasingly investigated the role of genetic risk factors, particularly single-nucleotide polymorphisms (SNPs), in the early prediction and prognosis of OSCC [4]. SNPs refer to single-base substitutions, insertions, or deletions that occur at specific genomic loci, with an average frequency of approximately 1 in every 800 base pairs [5]. By definition, these variants must be present in at least 1% of the general population. Genome-wide association studies have identified significant associations between height and HMGA2 SNPs [14]. However, molecular genetic evidence regarding the role of HMGA2-related SNPs in cancer remains limited. Very few studies have examined the association between HMGA2 polymorphisms and cancer risk [15-17], and data specific to OSCC remain lacking. In the present case-control study, we investigated the associations of HMGA2 polymorphisms with TSCC susceptibility and clinicopathological characteristics in Taiwanese men.

Materials and Methods

Study cohort

This study included 383 men with TSCC who received treatment at Chung Shan Medical University Hospital, Taichung, Taiwan. The study protocol was approved by the Ethics Committee of Chung Shan Medical University Hospital (approval number: CS1-21151). All participants provided written informed consent before enrollment. They also provided detailed information on lifestyle-related risk factors for cancer—for example, alcohol consumption, cigarette smoking, and betel quid chewing. This study also included 911 anonymized healthy controls randomly selected from the Taiwan Biobank Project. None of these controls had a history of cancer at any site. Patients with oral precancerous conditions were excluded from the control group. Clinical data pertaining to patients with TSCC were retrieved from medical records and included tumor-node-metastasis clinical stage, primary tumor size, lymph node (LN) involvement, distant metastasis status, and tumor histologic grade. All participants included in the control group were male, and the case and control groups were comparable in terms of ethnicity and geographic background, as all participants were Taiwanese individuals recruited from the same population. Therefore, the potential influence of population stratification is expected to be limited.

Genomic DNA extraction and HMGA2 SNP selection

Whole blood samples were collected in EDTA-coated tubes and centrifuged at 3,000 rpm to isolate the buffy coat. Genomic DNA was extracted from peripheral blood leukocytes by using a QIAamp DNA Mini Kit (Qiagen, Valencia, CA, USA), following the manufacturer's instructions. DNA purity and concentration were measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), and the samples were stored at -20°C until genotyping. On the basis of data obtained from the SNPinfo (https://snpinfo.niehs.nih.gov/snpinfo/snpfunc.html) and dbSNP (https://www.ncbi.nlm.nih.gov/snp/) databases, we selected four potentially functional polymorphisms of HMGA2: rs6581658 A>G, rs10573247 T>del, rs968697 T>C, and rs8756 A>C. These SNPs were also selected based on functional predictions, minor allele frequency (MAF), linkage disequilibrium (LD) criteria, and prior cancer association studies, as prior studies have demonstrated their association with cancer susceptibility or aggressive tumor characteristics across malignancies [15-20]. Based on the dbSNP database, a MAF threshold of > 0.10 was set for the Asian reference population to ensure sufficient statistical power. The MAFs of rs6581658 (A>G), rs10573247 (T>del), rs968697 (T>C), and rs8756 (A>C) in the Asian population were 0.219 (G), 0.391 (T), 0.138 (C), and 0.135 (C), respectively. In cancer cohorts, the rs8756 A>C and rs6581658 A>G polymorphisms have been reported to decrease and increase the risk of brain tumors in Chinese children, respectively [15, 18]. The rs968697 T>C polymorphism has been linked to Wilms tumor susceptibility [16], and its CC genotype is associated with a lower frequency of advanced-stage colorectal cancer[19]. Likewise, individuals carrying the TC/CC genotypes of rs968697 exhibit a significantly reduced risk of hepatoblastoma [20]. Additionally, the HMGA2 rs10573247 polymorphism has been implicated in breast cancer susceptibility [17]. To ensure independent genetic representation across the HMGA2 locus and eliminate redundant tagSNPs, pairwise LD analysis was performed. All pairwise LD values among these four variants were below 0.30 (the highest LD was observed between rs10573247 and rs8756, with an LD value of 0.288), indicating that these selected variants represent distinct, non-redundant genetic variations in HMGA2. Two of these SNPs—rs8756 A>C and rs10573247 T>del—are located in the 3′ untranslated region (3′ UTR) of HMGA2, where they affect microRNA (miRNA) binding and thus modulate gene expression and mRNA stability. By contrast, the other two SNPs—rs6581658 A>G and rs968697 T>C—are located in the 5′ near-gene region, where they modify transcription factor binding and thus regulate HMGA2 transcription. In the present study, no significant linkage disequilibrium was observed among the selected SNPs (R² < 0.8).

Genotyping of HMGA2 SNPs

HMGA2 SNPs were genotyped using an ABI StepOnePlus RT-PCR system (Applied Biosystems, Foster City, CA, USA) with the following individual TaqMan SNP probes: rs6581658 (assay ID: C__44861231_10), rs10573247 (assay ID: C__63497439_20), rs968697 (assay ID: C___8702860_20), and rs8756 (assay ID: C___1155510_20). The resulting genotypes were analyzed using SDS version 3.0 (Applied Biosystems). Detailed protocols for these procedures have been described elsewhere [21].

Bioinformatic analyses

Data for this study were obtained from the UALCAN online platform (http://ualcan.path.uab.edu; accessed February 23, 2026) [22], developed by the University of Alabama at Birmingham. This platform was used both to analyze HMGA2 mRNA and protein expression levels in normal and primary tumor tissues and to identify their associations with clinicopathological parameters, including tumor grade, in patients with head and neck squamous cell carcinoma (HNSCC). Analyses were conducted using data from The Cancer Genome Atlas (TCGA) and the Clinical Proteomic Tumor Analysis Consortium (CPTAC). Transcriptomic data from 40 Taiwanese paired OSCC and adjacent nontumor tissue specimens were analyzed using the microarray dataset GSE37991—obtained from the Gene Expression Omnibus (GEO)—to determine whether HMGA2 expression (probe ID: ILMN _1666236) is associated with OSCC tumorigenesis. HMGA2 expression and survival data for patients with HNSCC were retrieved from the KM Plotter database (accessed on February 23, 2026), and the prognostic significance of HMGA2 expression was evaluated through the Kaplan-Meier survival analysis. Finally, potential miRNAs targeting HMGA2 were identified using data from the miRDB database (http://mirdb.org/).

Statistical analysis

Demographic variables were compared between cancer-free individuals (healthy controls) and patients with TSCC by using the Mann-Whitney U test or Fisher's exact test. Logistic regression models were used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) to identify the associations of genotype frequency with TSCC susceptibility and clinicopathological parameters. Betel quid chewing, cigarette smoking, and alcohol consumption were included as covariates in the adjusted analyses. All statistical analyses were conducted using SAS (version 9.1; SAS Institute, San Francisco, CA, USA).

Results

General characteristics of the study cohort

In this study, we examined the association between HMGA2 polymorphisms and TSCC clinicopathological features by analyzing the data of 383 Taiwanese men with TSCC (TSCC group) and 911 cancer-free individuals (control group). Their demographic and lifestyle characteristics are summarized in Table 1. Age distribution was similar between the TSCC and control groups. Analysis of lifestyle data revealed that the prevalence of betel quid chewing, alcohol consumption, and cigarette smoking was significantly higher in the TSCC group than in the control group. These findings are consistent with those from other Asian OSCC cohorts [23, 24], highlighting the aforementioned lifestyle behaviors as major environmental risk factors for TSCC carcinogenesis. Clinically, most patients with TSCC exhibited no LN involvement (60.8%) or distant metastasis (99.5%), and most tumors exhibited moderate to poor differentiation (88.8%).

Effects of HMGA2 variants on TSCC susceptibility

To determine whether the selected HMGA2 SNPs (rs6581658 A>G, rs10573247 T>del, rs968697 T>C, and rs8756 A>C) were associated with TSCC risk, adjusted ORs (AORs) with 95% CIs were calculated using multiple logistic regression models adjusted for betel quid chewing, cigarette smoking, and alcohol consumption. The distributions of HMGA2 genotypes revealed that the most frequent alleles were homozygous A/A for the rs6581658 and rs8756 loci and homozygous T/T for the rs10573247 and rs968697 loci (Table 2). Unlike corresponding wild-type genotypes, HMGA2 SNPs exhibited no significant difference in frequency between the TSCC and control groups.

Effects of HMGA2 variants on TSCC clinicopathological features

To assess the clinical relevance of HMGA2 polymorphisms in TSCC, we examined the associations between HMGA2 SNPs and clinicopathological features, including primary tumor size, clinical stage, LN metastasis, and histopathological grade (Tables 3 and 4). Patients carrying the rs6581658 minor allele (AG or GG) were at significantly higher risks of presenting with an advanced clinical stage (stage III or IV; AOR = 1.661, 95% CI = 1.086-2.541, p = 0.019) and large primary tumors (>T2; AOR = 1.556, 95% CI = 1.025-2.362, p = 0.038) than were those carrying the wild-type AA genotype (Table 3). These findings suggest that rs6581658 facilitates tumor aggressiveness by promoting tumor growth or increasing aggressiveness in TSCC. Patients carrying the rs10573247 deletion polymorphism (TT/Del or Del/Del) exhibited a trend toward an increased risk of LN metastasis (p = 0.077; Table 3). By contrast, patients carrying the rs968697 minor allele (TC or CC) exhibited a protective trend against advanced clinical stage (p = 0.074) and LN metastasis (p = 0.079; Table 4).

Influence of environmental factors on the association between HMGA2 variants and TSCC clinicopathological features

Cigarette smoking and betel quid chewing are well-established risk factors for OSCC. Exposure to these carcinogens can induce distinct molecular changes and alter gene expression profiles in the oral epithelium, thereby dysregulating the expression of various proteins and biomolecules [25]. In the present study, we examined whether betel quid chewing and cigarette smoking modify the association between HMGA2 SNPs and TSCC clinicopathological features. Stratified analysis revealed that among patients who neither chewed betel quid nor smoked, those carrying ≥1 minor allele of rs968697 were at a significantly reduced risk of LN metastasis or advanced-stage disease. However, this protective effect was not observed in patients exposed to either environmental risk factor (Tables 5 and 6). Conversely, among patients with a history of betel quid chewing or cigarette smoking, those carrying ≥1 rs10573247 deletion allele were at a significantly increased risk of LN metastasis (Tables S1 and S2). These findings underscore the complex interplay between genetic susceptibility and environmental exposure, suggesting that HMGA2 variants contribute to TSCC aggressiveness in the presence and absence of traditional risk factors. Further research is required to elucidate the biological mechanisms underlying these exposure-stratified associations. Insights from such research may inform personalized prevention strategies and targeted therapeutic approaches for genetically susceptible individuals.

Correlations of upregulated HMGA2 expression in HNSCC tissues with tumor aggressiveness and poor prognosis

HMGA2 polymorphisms influence gene expression in tumor cells [19]. In this study, we analyzed the TCGA-HNSCC and CPTAC-HNSCC datasets to respectively evaluate HMGA2 mRNA and protein expression in normal and HNSCC tissues and investigate its potential association with tumor aggressive tumor characteristics and prognosis. Expression levels of HMGA2 mRNA (Fig. 1A) and protein (Fig. 1B) were significantly higher in HNSCC tissues than in noncancerous tissues. Consistent with these findings, analysis of the GSE37991 dataset from the GEO database revealed that HMGA2 expression was markedly higher in OSCC specimens than in corresponding matched normal tissues from a Taiwanese OSCC N/T paired cohort with betel quid chewing and cigarette smoking habits (Fig. 1C). Furthermore, the levels of both HMGA2 transcript (Fig. 1D) and protein (Fig. 1E) were higher in patients with high-grade tumors (grades 2 or 3) than in those with low-grade tumors (grade 1). Kaplan-Meier survival analysis indicated that higher HMGA2 expression was associated with shorter overall survival in patients with HNSCC (Fig. 1F). Together, these findings indicate that HMGA2 SNPs are associated with TSCC aggressiveness, and that elevated HMGA2 expression is also correlated with tumor aggressiveness and poor prognosis in TSCC.

 Figure 1 

Clinical relevance of HMGA2 expression in patients with HNSCC. Analyses were performed using data from TCGA, CPTAC, and the GEO databases. Expression levels of HMGA2 (A) gene and (B) protein in normal and HNSCC tissues, analyzed using data from TCGA and the CPTAC. Statistical significance was determined using an unpaired t test. (C) HMGA2 gene expression levels in an OSCC normal/tumorous (N/T) paired cohort with the habits of betel quid chewing and smoking (GSE37991). Statistical significance was determined using a paired t-test. HMGA2 gene (D) and protein (E) expression in tumor tissues from patients with HNSCC stratified by tumor grade. Statistical significance was determined using an unpaired t test. Significance: ^***p < 0.001 versus the control group and ^##p < 0.01 and ^###p < 0.001 versus patients with grade 1 or 2 tumors. (F) The correlation between HMGA2 expression and overall survival in patients with HNSCC was evaluated using the Kaplan-Meier Plotter database. Gene expression levels were dichotomized into high and low groups by using the median value as the cutoff. HR, hazard ratio; HNSCC, head and neck squamous cell carcinoma; TCGA, The Cancer Genome Atlas; GEO, Gene Expression Omnibus; CPTAC, Clinical Proteomic Tumor Analysis Consortium.

Discussion

HMGA2 plays key roles in multiple aspects of cancer development, including cell proliferation [26], protection against apoptosis [27], cancer stemness [28], angiogenesis [9], metastasis mediated by EMT [29], and drug resistance [11]. Upregulated HMGA2 expression has been observed in various malignancies, including bladder [30], brain [31], colorectal [32], and breast [33] cancers, and is often associated with poor prognosis. The present study revealed that HMGA2 expression was significantly upregulated in HNSCC and OSCC cohorts and was positively associated with advanced tumor grades and poor prognosis in patients with HNSCC, supporting the oncogenic role of HMGA2 in HNSCC clinicopathological aggressiveness. OSCC is a complex disease driven by multiple genetic alterations [34]. Genetic variations such as SNPs may alter gene function and influence disease susceptibility, thereby contributing to increased cancer risk or distinct pathological characteristics. Several genetic polymorphisms have been associated with OSCC susceptibility and aggressive tumor characteristics, often in conjunction with environmental risk factors [35-37]. Although elevated HMGA2 expression contributes to OSCC aggressiveness [9], the potential interaction effects of environmental risk factors and HMGA2 polymorphisms on OSCC aggressiveness remain largely unclear. To the best of our knowledge, the present study is the first to demonstrate that HMGA2 polymorphisms are significantly associated with TSCC clinicopathological characteristics in Taiwanese patients with different lifestyle-related risk factors.

We found that carriers of the mutant G allele of rs6581658 (AG+GG) were at significantly increased risks of large tumors (>T2) and advanced clinical stage (III or IV). Among patients who neither chewed betel quid nor smoked, those carrying the rs968697 mutant C allele (TC+CC) were at significantly reduced risks of advanced clinical stage or LN metastasis. Because rs6581658 and rs968697 are located in the 5ʹ near-gene region, the nucleotide changes introduced by these SNPs may alter transcription factor-binding sites, thereby regulating HMGA2 transcription. The rs968697 CC genotype is associated with low HMGA2 expression and low high-stage tumor frequency in patients with colorectal cancer [19]. Moreover, the rs968697 TC/CC genotypes are strongly associated with a reduced risk of hepatoblastoma in men [20]. Furthermore, the rs968697 C allele was significantly associated with reduced susceptibility to gastric cancer (GC). Patients with GC carrying the C allele were less likely to present with advanced-stage disease (stage III or IV) and have longer overall survival than did those not carrying this allele. Consistently, GC tissues carrying the rs968697 TC and CC genotypes exhibited low levels of HMGA2 mRNA expression [38]. Taken together, the results of our study are consistent with those of the aforementioned studies, suggesting that the C allele of rs968697 downregulates HMGA2 expression in TSCC and subsequently inhibits disease aggressiveness in patients with TSCC. In addition to rs968697, the rs6581658 AG/GG genotypes has been associated with an increased risk of glioma in Chinese children [18]. However, whether rs6581658 polymorphisms affect HMGA2 mRNA expression remains unclear.

The deletion polymorphism rs10573247, located in the 3′ UTR region of HMGA2, has been identified as a regulatory variant that may influence mRNA stability and miRNA binding and modulate HMGA2 expression. A study demonstrated that miR-3125 exhibits a lower binding affinity for the DelTT allele than for the TT allele, whereas miR-4476 exhibits a higher binding affinity for the DelTT allele than for the TT allele [17]. Individuals carrying the TT/Del or Del/Del genotypes are at a significantly higher risk of breast cancer than are those carrying the TT/TT genotype [17]. Consistent with these findings, our results indicated that patients with TSCC carrying the rs10573247 deletion polymorphism (TT/Del or Del/Del), particularly those who smoked cigarettes and chewed betel quid, had a significantly increased risk of LN metastasis. Exposure to tobacco smoke or betel quid can lead to the widespread dysregulation of miRNAs in HNSCC [25, 39]. For instance, treatment with areca nut extract or arecoline has been demonstrated to upregulate several miRNAs (miR-23a, miR-26a, miR-26b, miR-30a-5p, miR-143, and miR-1915) and downregulate others (miR-886-3p, miR-1914, and miR-1305) in human oral fibroblasts. Betel quid chewing has been associated with miR-23a overexpression in patients with OSCC [40]. Notably, several of the aforementioned miRNAs (miR-23a, miR-26a, miR-26b, miR-143, and miR-1305) are predicted to target HMGA2 according to the miRNA target prediction database miRDB [41] (Table 7). Together, these findings suggest that the rs10573247 deletion polymorphism influences the binding of betel-quid-regulated miRNAs, thereby modulating HMGA2 expression and promoting TSCC aggressiveness. Nonetheless, this hypothesis and the precise mechanisms underlying this interaction require further investigation.

Our study has several limitations that should be acknowledged. First, the analyzed cohorts differed in their ethnic composition. All TSCC cases included in the SNP analysis were derived from a Taiwanese population, whereas the clinicopathological associations of HMGA2 expression were evaluated using the TCGA-HNSCC and CPTAC-HNSCC cohorts, which predominantly consist of White/Caucasian individuals. Although the GSE37991 dataset was included as an independent validation cohort, it comprised only 40 Taiwanese OSCC specimens. Therefore, additional validation in larger independent Taiwanese cohorts is warranted to strengthen the clinical relevance and generalizability of our findings. Second, environmental risk factors—specifically betel quid chewing, smoking, and alcohol consumption—were evaluated as binary variables due to the lack of detailed quantitative data on exposure duration, intensity, and former versus current status. Consequently, potential dose-response interactions between lifestyle factors and HMGA2 polymorphisms could not be analyzed.

Third, although we identified potential associations between the candidate SNPs and various clinicopathological parameters, including clinical stage, tumor size (T stage), and lymph node metastasis (N status), these parameters are biologically and clinically interconnected. Because overall clinical stage is intrinsically determined by T and N categories, treating them as entirely independent endpoints may overstate the consistency of our findings. Thus, these clinicopathological evaluations might be regarded as secondary endpoints, and the observed secondary associations must be interpreted with caution until validated through comprehensive multivariable models in larger independent cohorts. In addition, although significant associations were identified, multiple testing was not formally adjusted in the present exploratory study. Therefore, some associations may represent false-positive findings arising from multiple comparisons. These results should be regarded as hypothesis-generating and require confirmation in independent cohorts.

Fourth, future studies should perform genotype-expression analyses, such as expression quantitative trait locus (eQTL) analyses, using matched DNA and RNA samples from the same TSCC patients to more accurately determine the effects of HMGA2 polymorphisms on HMGA2 expression. Such samples would also facilitate the investigation of alterations in miRNA expression associated with different lifestyle-related risk factors. Moreover, future studies involving allele-specific binding assay and luciferase reporter assay are warranted to determine whether the rs10573247 deletion polymorphism alters the binding affinity of these miRNAs to the HMGA2 3'UTR and consequently contributes to dysregulated HMGA2 expression. Finally, the lack of long-term follow-up data precluded an assessment of the prognostic significance of HMGA2 polymorphisms. Future studies with extended clinical follow-up are therefore needed to clarify their associations with patient survival and clinical outcomes.

Conclusions

Although HMGA2 has been implicated in cancer development, the impacts of HMGA2 genetic variation in OSCC remain largely unexplored. The novelty of the present study lies not only in identifying distinct allelic effects of rs6581658 on the clinicopathological characteristics of TSCC in a Taiwanese population, but also, importantly, in revealing exposure-stratified associations between HMGA2 polymorphisms and lifestyle-related risk factors. Specifically, rs968697 and rs10573247 were associated with TSCC aggressiveness among individuals with different exposure profiles, including betel quid chewing and cigarette smoking. These findings extend the current understanding of HMGA2 genetic variation in TSCC by suggesting that lifestyle-related exposures may modify the associations between HMGA2 polymorphisms and TSCC progression.

Abbreviations

CIs: Confidence intervals

CPTAC: Clinical Proteomic Tumor Analysis Consortium

EMT: Epithelial-mesenchymal transition

GEO: Gene Expression Omnibus

HMGA2: High-mobility group A protein 2

HNSCC: Head and neck squamous cell carcinoma

miRNA: microRNA

OR: Odds ratio

OSCC: Oral squamous cell carcinoma

SNPs: Single-nucleotide polymorphisms

TCGA: The Cancer Genome Atlas

TSCC: Tongue squamous cell carcinoma

UTR: Untranslated region

Supplementary Material

Supplementary tables.

Acknowledgements

Funding

This research was supported by the Taipei Medical University Research Center of Cancer Translational Medicine (under the Featured Areas Research Center Program, supported by the Ministry of Education of Taiwan's Higher Education Sprout Project (awarded to M.-H. Chien) and Taipei Medical University Wan Fang Hospital (grant number: 115-wf-swf-01; awarded to M.-H. Chien, S.-F. Yang, and Y.-C. Wen).

Author contributions

The authors confirm their contributions to the paper as follows: study conception and design: Yu-Ching Wen, Shun-Fa Yang, and Ming-Hsien Chien; data collection: Shun-Fa Yang, and Ming-Hsien Chien; analysis and interpretation of results: Yi-Chieh Yang, Chia-Hwa Lee, and Lun-Ching Chang; draft manuscript preparation: Yi-Chieh Yang, Shun-Fa Yang, and Ming-Hsien Chien; formal analysis: Yi-Fang Ding and Chiao-Wen Lin; resources: Shun-Fa Yang. All authors reviewed the results and approved the final version of the manuscript.

Availability of data and materials

The data supporting the findings of this study are available within the article and its supplementary materials. Additional original data are available from the corresponding author upon reasonable request.

Ethics approval

Experiments involving clinical samples were approved by the Institutional Review Board of Chung Shan Medical University Hospital (approval number: CS1-21151).

Competing Interests

The authors have declared that no competing interest exists.

References

1. Michcik A, Wojciechowska B, Tarnawski J, Choma P, Polcyn A, Garbacewicz Ł, Sikora M, Iacoviello P, Wach T, Drogoszewska B. Diagnostic, Prognostic, and Predictive Molecular Biomarkers in Head and Neck Squamous Cell Carcinoma: A Comprehensive Review. J Clin Med. 2026;15(2):769

2. Irani S. New Insights into Oral Cancer-Risk Factors and Prevention: A Review of Literature. Int J Prev Med. 2020;11:202

3. Jumparway D, Su CW, Yen AM, Liu YT, Chen MK, Liu KJ, Sarakarn P, Chen SL. Quantifying Genetic and Environmental Factors Accounting for Multistage Progression of Precancerous Lesions and Oral Cancer: Applications to Risk-Guided Prevention. Cancers (Basel). 2025;17(13):2114

4. Fusco A, Fedele M. Roles of HMGA proteins in cancer. Nat Rev Cancer. 2007;7(12):899-910

5. Pfannkuche K, Summer H, Li O, Hescheler J, Dröge P. The high mobility group protein HMGA2: a co-regulator of chromatin structure and pluripotency in stem cells? Stem Cell Rev Rep. 2009;5(3):224-230

6. Fedele M, Visone R, De Martino I, Troncone G, Palmieri D, Battista S, Ciarmiello A, Pallante P, Arra C, Melillo RM. et al. HMGA2 induces pituitary tumorigenesis by enhancing E2F1 activity. Cancer Cell. 2006;9(6):459-471

7. Zaidi MR, Okada Y, Chada KK. Misexpression of full-length HMGA2 induces benign mesenchymal tumors in mice. Cancer Res. 2006;66(15):7453-7459

8. Mansoori B, Mohammadi A, Ditzel HJ, Duijf PHG, Khaze V, Gjerstorff MF, Baradaran B. HMGA2 as a Critical Regulator in Cancer Development. Genes (Basel). 2021;12(2):269

9. Sakata J, Hirosue A, Yoshida R, Kawahara K, Matsuoka Y, Yamamoto T, Nakamoto M, Hirayama M, Takahashi N, Nakamura T. et al. HMGA2 Contributes to Distant Metastasis and Poor Prognosis by Promoting Angiogenesis in Oral Squamous Cell Carcinoma. Int J Mol Sci. 2019;20(10):2473

10. Ouyang X, Li K, Wang J, Zhu W, Yi Q, Zhong J. HMGA2 promotes nasopharyngeal carcinoma progression and is associated with tumor resistance and poor prognosis. Front Oncol. 2023;13:1271080

11. Krafft U, Tschirdewahn S, Hess J, Harke NN, Hadaschik B, Olah C, Krege S, Nyirády P, Szendröi A, Szücs M. et al. Validation of survivin and HMGA2 as biomarkers for cisplatin resistance in bladder cancer. Urol Oncol. 2019;37(11):810.e817-810.e815

12. Wang B, Pan LY, Kang N, Shen XY. PP4R1 interacts with HMGA2 to promote non-small-cell lung cancer migration and metastasis via activating MAPK/ERK-induced epithelial-mesenchymal transition. Mol Carcinog. 2020;59(5):467-477

13. Yang Z, Liu F, Li Z, Liu N, Yao X, Zhou Y, Zhang L, Jiang P, Liu H, Kong L. et al. Histone lysine methyltransferase SMYD3 promotes oral squamous cell carcinoma tumorigenesis via H3K4me3-mediated HMGA2 transcription. Clin Epigenetics. 2023;15(1):92

14. Weedon MN, Lettre G, Freathy RM, Lindgren CM, Voight BF, Perry JR, Elliott KS, Hackett R, Guiducci C, Shields B. et al. A common variant of HMGA2 is associated with adult and childhood height in the general population. Nat Genet. 2007;39(10):1245-1250

15. Liu J, Hua RX, Cheng Y, Zhu J, Zhang J, Cheng J, Zhou H, Xia H, Bian J, He J. HMGA2 Gene rs8756 A>C Polymorphism Reduces Neuroblastoma Risk in Chinese Children: A Four-Center Case-Control Study. Onco Targets Ther. 2020;13:465-472

16. Cheng J, Zhuo Z, Yang L, Zhao P, Zhang J, Zhou H, He J, Li P. HMGA2 gene polymorphisms and Wilms tumor susceptibility in Chinese children: a four-center case-control study. Biotechnol Appl Biochem. 2020;67(6):939-945

17. Najibi K, Moghanibashi M, Naeimi S. Association of deletion polymorphism rs10573247 in the HMGA2 gene with the risk of breast cancer: bioinformatic and experimental analyses. World J Surg Oncol. 2024;22(1):142

18. Zhou J, Wang P, Zhang R, Huang X, Dai H, Yuan L, Ruan J. Association of HMGA2 Polymorphisms with Glioma Susceptibility in Chinese Children. Pharmgenomics Pers Med. 2021;14:601-607

19. Gao X, Wang X. HMGA2 rs968697 T > C polymorphism is associated with the risk of colorectal cancer. Nucleosides Nucleotides Nucleic Acids. 2021;40(8):821-828

20. Li L, Zhuo Z, Yang Z, Zhu J, He X, Yang Z, Zhang J, Xin Y, He J, Zhang T. HMGA2 Polymorphisms and Hepatoblastoma Susceptibility: A Five-Center Case-Control Study. Pharmgenomics Pers Med. 2020;13:51-57

21. Lin YW, Wang SS, Wen YC, Tung MC, Lee LM, Yang SF, Chien MH. Genetic Variations of Melatonin Receptor Type 1A are Associated with the Clinicopathologic Development of Urothelial Cell Carcinoma. Int J Med Sci. 2017;14(11):1130-1135

22. Chandrashekar DS, Bashel B, Balasubramanya SAH, Creighton CJ, Ponce-Rodriguez I, Chakravarthi B, Varambally S. UALCAN: A Portal for Facilitating Tumor Subgroup Gene Expression and Survival Analyses. Neoplasia. 2017;19(8):649-658

23. Goud E, Malleedi S, Ramanathan A, Wong GR, Hwei Ern BT, Yean GY, Ann HH, Syan TY, Zain RM. Association of Interleukin-10 Genotypes and Oral Cancer Susceptibility in Selected Malaysian Population: A Case- Control Study. Asian Pacific journal of cancer prevention: APJCP. 2019;20(3):935-941

24. Senghore T, Chien HT, Wang WC, Chen YX, Young CK, Huang SF, Yeh CC. Polymorphisms in ERCC5 rs17655 and ERCC1 rs735482 Genes Associated with the Survival of Male Patients with Postoperative Oral Squamous Cell Carcinoma Treated with Adjuvant Concurrent Chemoradiotherapy. Journal of clinical medicine. 2019;8(1):33

25. Wang TH, Hsia SM, Shih YH, Shieh TM. Association of Smoking, Alcohol Use, and Betel Quid Chewing with Epigenetic Aberrations in Cancers. Int J Mol Sci. 2017;18(6):1210

26. Tan L, Wei X, Zheng L, Zeng J, Liu H, Yang S, Tan H. Amplified HMGA2 promotes cell growth by regulating Akt pathway in AML. J Cancer Res Clin Oncol. 2016;142(2):389-399

27. Natarajan S, Hombach-Klonisch S, Dröge P, Klonisch T. HMGA2 inhibits apoptosis through interaction with ATR-CHK1 signaling complex in human cancer cells. Neoplasia. 2013;15(3):263-280

28. Kaur H, Ali SZ, Huey L, Hütt-Cabezas M, Taylor I, Mao XG, Weingart M, Chu Q, Rodriguez FJ, Eberhart CG. et al. The transcriptional modulator HMGA2 promotes stemness and tumorigenicity in glioblastoma. Cancer Lett. 2016;377(1):55-64

29. Hawsawi O, Henderson V, Burton LJ, Dougan J, Nagappan P, Odero-Marah V. High mobility group A2 (HMGA2) promotes EMT via MAPK pathway in prostate cancer. Biochem Biophys Res Commun. 2018;504(1):196-202

30. Yang GL, Zhang LH, Bo JJ, Hou KL, Cai X, Chen YY, Li H, Liu DM, Huang YR. Overexpression of HMGA2 in bladder cancer and its association with clinicopathologic features and prognosis HMGA2 as a prognostic marker of bladder cancer. Eur J Surg Oncol. 2011;37(3):265-271

31. Zhang S, Zhang H, Yu L. HMGA2 promotes glioma invasion and poor prognosis via a long-range chromatin interaction. Cancer Med. 2018;7(7):3226-3239

32. Wang X, Liu X, Li AY, Chen L, Lai L, Lin HH, Hu S, Yao L, Peng J, Loera S. et al. Overexpression of HMGA2 promotes metastasis and impacts survival of colorectal cancers. Clin Cancer Res. 2011;17(8):2570-2580

33. Wu J, Zhang S, Shan J, Hu Z, Liu X, Chen L, Ren X, Yao L, Sheng H, Li L. et al. Elevated HMGA2 expression is associated with cancer aggressiveness and predicts poor outcome in breast cancer. Cancer Lett. 2016;376(2):284-292

34. Barroso L, Veiga P, Melo JB, Carreira IM, Ribeiro IP. Molecular and Genetic Pathogenesis of Oral Cancer: A Basis for Customized Diagnosis and Treatment. Biology (Basel). 2025;14(7):842

35. Ding YF, Wen YC, Chuang CY, Lin CW, Yang YC, Liu YF, Chang WM, Chang LC, Yang SF, Chien MH. Combined Impacts of Genetic Variants of Long Non-Coding RNA MALAT1 and the Environmental Carcinogen on the Susceptibility to and Progression of Oral Squamous Cell Carcinoma. Front Oncol. 2021;11:684941

36. Debata T, Swain A, Jena SR, Das SN, Mishra N, Samanta L. Vitamin D receptor gene polymorphism in oral cancer as a function of tobacco consumption: an evidence based systematic review and meta-analysis. Front Oral Health. 2025;6:1550683

37. Gupta N, Yumnam G, Sharma C, Patel A, Sharma R, Dev S, Ghadage M. Relationship among Tobacco Habits, Human Papilloma Virus (HPV) Infection, p53 Polymorphism/Mutation, and the Risk of Oral Squamous Cell Carcinoma. J Pharm Bioallied Sci. 2024;16(Suppl 4):S3424-s3426

38. Hu J, Wang X. The relationship between rs968697 genetic polymorphism and gastric cancer susceptibility, TNM stage and survival prognosis. Eur J Inflamm. 2023;21:1-7

39. Bhat MY, Nanjappa V, Brait M, Sidransky D, Chatterjee A, Advani J. Smoking and alcohol specific alterations in miRNA expression in head and neck squamous cell carcinoma. Discov Oncol. 2025;16(1):1458

40. Tsai YS, Lin CS, Chiang SL, Lee CH, Lee KW, Ko YC. Areca nut induces miR-23a and inhibits repair of DNA double-strand breaks by targeting FANCG. Toxicol Sci. 2011;123(2):480-490

41. Chen Y, Wang X. miRDB: an online database for prediction of functional microRNA targets. Nucleic Acids Res. 2020;48(D1):D127-d131

Author contact

Corresponding authors: Ming-Hsien Chien, PhD, Email: d002089002edu.tw. Shun-Fa Yang, PhD, Email: ysfedu.tw. Yu-Ching Wen, MD, PhD, Email: 95207tmu.edu.tw.

Citation styles

©2026 Ivyspring International Publisher. All rights reserved.
Open access content is subject to applicable licences. By accessing or using this platform, you agree to our Terms of Use.