Abstract
Background/Aim: Bladder cancer (BLCA) is a prevalent malignancy globally, particularly in Taiwan, where its incidence continues to rise. While environmental exposures such as smoking play critical roles in BLCA etiology, genetic predispositions also contribute significantly. Matrix metalloproteinase-2 (MMP-2), a key enzyme involved in extracellular matrix remodeling, has been implicated in BLCA progression. This study investigated the associations of two promoter polymorphisms in the MMP-2 gene, rs243865 and rs2285053, with BLCA susceptibility in a Taiwanese population.
Patients and Methods: A case-control study was conducted involving 375 BLCA patients and 375 cancer-free controls matched by age and sex. Genotyping was performed using PCR-RFLP, and associations with BLCA risk were evaluated using logistic regression, adjusting for confounders. Stratified analyses were performed based on demographic and clinical parameters.
Results: No significant association was observed between rs243865 and overall BLCA risk (OR=1.34, 95%CI=0.96-1.86, p=0.0953). In contrast, rs2285053 showed a significant allelic association with BLCA (OR=1.33, 95%CI=1.04-1.69, p=0.0263). Stratified analysis revealed that younger individuals (≤55 years) carrying the CT genotype of rs243865 had a significantly elevated BLCA risk (OR=2.48, 95%CI=1.39-4.42, p=0.0028). Among smokers, both CT and TT genotypes of rs2285053 were associated with increased risk (OR=1.68 and 4.01, p=0.0214 and 0.0045). Additionally, rs243865 TT genotype was associated with muscle-invasive BLCA (OR=3.68, 95%CI=1.96-6.91, p=0.0001).
Conclusion: MMP-2 rs2285053 and, to a lesser extent, rs243865 polymorphisms may influence BLCA susceptibility and aggressiveness, particularly in the context of age and smoking. These variants may serve as potential biomarkers for personalized risk assessment, pending validation in larger, multiethnic cohorts.
Introduction
Bladder cancer (BLCA) is among the most common malignancies of the urinary tract, and its global disease burden has been steadily rising in recent decades (1, 2). Data from the Global Cancer Observatory (GLOBOCAN) indicate that in 2022, approximately 613,000 new diagnoses and 220,000 deaths were attributed to BLCA worldwide, placing it ninth in incidence and thirteenth in cancer-related mortality (3). In Taiwan, BLCA ranks as the eleventh most frequent cancer in men and the sixteenth in women, with incidence rates continuing to rise (4). A striking male predominance is observed, with an estimated male-to-female ratio of 5:2, largely explained by differences in smoking prevalence (5). The etiology of BLCA is believed to involve complex interactions between inherited susceptibility and environmental exposures. Of these, tobacco smoking is consistently recognized as the predominant risk factor (6, 7). Other lifestyle and environmental contributors include alcohol consumption, occupational exposure to carcinogenic chemicals, fine particulate matter (PM2.5) pollution, prior radiotherapy, inappropriate use of certain Chinese herbal preparations, and recurrent urinary tract infections (8-12). Moreover, dietary habits have also been implicated, with meat consumption representing a potential dietary risk. The effect appears to depend on variables such as the type of meat, cooking methods, and the extent of thermal processing (13). In addition to environmental and lifestyle factors, hereditary influences also contribute to individual BLCA susceptibility. Although several genetic associations have been reported (14-16), the underlying molecular mechanisms remain incompletely elucidated, and the precise determinants of genetic risk are still under active investigation.
Matrix metalloproteinases (MMPs), through their collagenase activity, are key mediators of connective tissue remodeling and basement membrane degradation, processes that critically influence angiogenesis, tumor invasion, and metastatic progression (17, 18). Elevated expression of several MMP family members, particularly MMP-2 and MMP-9, has been linked to enhanced tumor growth, invasion, and dissemination (19, 20). MMP-2, located on chromosome 16 (21), encodes gelatinase A, an enzyme that hydrolyzes substrates such as gelatin, collagen type V, and collagen type VI (22). Functionally relevant polymorphisms within the promoter region of MMP-2 have been shown to alter its transcriptional activity. The rs243865 variant, positioned at −1306, involves a T allele that reduces MMP-2 expression (23). Similarly, the rs2285053 polymorphism, located at −735, has been reported to diminish transcription by disrupting a putative Sp1 binding site, thereby further reducing MMP-2 expression (23).
Given this background, dysregulated MMP-2 expression may be strongly influenced by underlying genetic variation. In particular, promoter region polymorphisms appear to play a pivotal role in shaping individual susceptibility to BLCA. On this basis, the present study investigated whether the MMP-2 rs243865 and rs2285053 genotypes contribute to BLCA risk in a Taiwanese population, consisting of 375 patients with BLCA and an equal number of cancer-free controls. Furthermore, we examined whether these genetic variants interact with demographic and clinical characteristics to modulate BLCA susceptibility.
Patients and Methods
Recruitment of BLCA patients and cancer-free controls. This hospital-based case-control study, involving both BLCA patients and cancer-free controls, was approved by the Institutional Review Board of China Medical University Hospital (CMUH111-REC1-176), and written informed consent was obtained from all participants. The study was conducted in accordance with the ethical standards of the Declaration of Helsinki. A total of 375 patients with pathologically confirmed BLCA were recruited, and tumor classification was performed by pathologists. Tumors restricted to the urothelium or lamina propria were defined as non-muscle-invasive, whereas those invading the muscularis propria or beyond were categorized as muscle invasive. Each patient completed a structured questionnaire and provided 3-5 ml of peripheral blood. The control group consisted of 375 cancer-free individuals randomly selected from the hospital’s Health Examination Cohort (n=15,000). Controls were frequency-matched to cases by age, sex, smoking, and alcohol consumption. Individuals with a previous malignancy, metastatic disease, tumors of unknown origin, or hereditary/genetic disorders were excluded. Information on demographic characteristics, lifestyle behaviors, and environmental exposures was collected, with particular emphasis on smoking and alcohol use. “Ever smokers” were defined as those who smoked daily or nearly daily, accumulating at least five pack-years over one year. “Ever alcohol drinkers” were defined as individuals who either experienced intoxication at least twice or consumed more than three drinks per week for at least one year; intoxication was characterized as an impaired ability to walk in a straight line. Baseline characteristics of cases and controls are summarized in Table I. The mean ages were 61.4 years for patients and 62.9 years for controls, with no significant age difference (p=0.7315). The case group displayed a male-to-female ratio of approximately 3:1, while sex distribution did not significantly differ between cases and controls (p=0.5525). Similarly, smoking and alcohol consumption were comparable across groups (p=0.3063 and 0.3807, respectively). Among BLCA patients, 62.7% presented with non-muscle-invasive tumors and 37.3% with muscle-invasive disease. Regarding histopathological grade, 40.3% of tumors were low-grade and 59.7% were high-grade (Table I).
Basic characteristics of the 375 bladder cancer patients and 375 non-cancer controls.
MMP-2 genotyping methods and experimental conditions. Peripheral blood samples were obtained from all participants, and genomic DNA was extracted within 24 h using standardized protocols established by our group (24-26). Genotypes for MMP-2 rs243865 and rs2285053 were performed using polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis. Specific primers for MMP-2 genotyping were designed and optimized by Terry Fox Cancer Research Lab, and PCR amplification was carried out according to published protocols (27). Amplified products were digested overnight with Xsp I (for rs243865) or Hinf I (for rs2285053), and restriction fragments were resolved on 3% agarose gels to determine genotype profiles. To ensure accuracy, each sample was independently and blindly genotyped by at least two investigators, yielding 100% concordant results. The physical locations and nearby sequences of the rs243865 and rs2285053 polymorphic sites are depicted in Figure 1.
Physical maps of the sequences nearby matrix metalloproteinase-2 (MMP-2) rs243865 and rs2285053 polymorphic sites.
Statistical analysis. Hardy-Weinberg equilibrium for genotype distribution in the controls was evaluated using a chi-square goodness-of-fit test. Differences in age between BLCA patients and controls were expressed as mean±standard deviation (SD) and compared using an unpaired Student’s t-test. The distribution of MMP-2 genotypes was analyzed by Pearson’s chi-square test with Yates’ correction when all expected cell counts exceeded five, or by Fisher’s exact test when any expected count was ≤5. Associations between MMP-2 genotypes and BLCA risk were estimated using odds ratios (ORs) with corresponding 95% confidence intervals (CIs) under various stratified analysis models. Statistical significance was defined as p≤0.05. All analyses were conducted using SPSS software, version 16.0 (SPSS Inc., Chicago, IL, USA).
Results
Association between MMP-2 genotypes and BLCA risk. The distribution of MMP-2 rs243865 and rs2285053 genotypes, along with their associations with BLCA risk, is summarized in Table II. Genotype frequencies in the control group conformed to Hardy-Weinberg equilibrium (pHWE=0.2902 and 0.3687). Neither polymorphism showed a statistically significant association with overall BLCA risk (Ptrend=0.2354 and 0.0846, respectively). For rs243865, neither the heterozygous nor the homozygous variant genotype was significantly associated with BLCA risk (p=0.2229 and 0.3729). Similarly, no significant associations were observed for rs2285053 heterozygotes or homozygotes (p=0.1234 and 0.1023). In the dominant genetic model, rs243865 remained non-significant (p =0.2354). Notably, rs2285053 demonstrated a borderline association under the dominant model, with the p-value approaching statistical significance (p =0.0517).
Genotypic frequency distributions of matrix metalloproteinase-2 rs243865 and rs2285053 among the bladder cancer patients and the non-cancer subjects.
Association between MMP-2 alleles and BLCA risk. To complement the genotypic results shown in Table II, allelic frequency distributions were further examined (Table III). For MMP-2 rs243865, no significant association was detected, as the variant allele showed only a marginally elevated but non-significant risk of BLCA (OR=1.34, 95%CI=0.96-1.86, p=0.0953). In contrast, rs2285053 demonstrated a significant association, with carriers of the variant allele exhibiting an increased risk of BLCA (OR=1.33, 95%CI=1.04-1.69, p=0.0263; Table III).
Allelic frequencies for matrix metalloproteinase-2 rs243865 and rs2285053 polymorphisms among the bladder cancer patients and the non-cancer subjects.
Stratified analysis of MMP-2 rs243865 genotypes based on demographic and clinical characteristics. To further clarify the potential role of MMP-2 rs243865, stratified analyses were conducted according to demographic (age, sex, smoking, and alcohol consumption; Table IV) and clinical factors (stage and grade; Table V). Age-stratified analysis revealed a significant effect in the younger subgroup (≤55 years). Individuals carrying the heterozygous CT genotype exhibited a markedly elevated risk of BLCA (OR=2.48, 95% CI=1.39-4.42, p=0.0028). Although the homozygous TT genotype showed a similar trend (OR=2.96, 95% CI=0.56-15.55), the association did not reach statistical significance. In contrast, no significant associations were observed among participants >55 years. When stratified by sex, rs243865 variant genotypes were not significantly associated with BLCA risk in either males or females. In the smoking-stratified analysis, however, clear differences emerged. Among smokers, the CT genotype was significantly associated with increased risk (OR=1.76, 95% CI=1.05-2.93, p=0.0414), while the TT genotype conferred an even higher, though borderline, risk estimate (OR=6.37, 95% CI=0.76-53.53, p=0.0643). No associations were found among non-smokers. Finally, stratification by alcohol consumption revealed no significant differences in genotype distributions between drinkers and non-drinkers (Table IV).
Distribution of matrix metalloproteinase-2 rs243865 genotypes with bladder cancer risk stratified by demographic and clinical indexes.
Distribution of matrix metalloproteinase-2 rs243865 genotypes with bladder cancer risk stratified by clinical indexes.
Analysis by clinical parameters revealed significant associations between rs243865 variants and tumor aggressiveness (Table V). In the muscle-invasive subgroup, the homozygous TT genotype was more prevalent and conferred a markedly elevated risk compared with the non-muscle-invasive group (OR=3.68, 95% CI=1.96-6.91, p=0.0001). Similarly, when stratified by tumor grade, TT homozygotes were significantly overrepresented in patients with high-grade disease, corresponding to an approximately twofold increase in risk compared with low-grade cases (OR=2.01, 95% CI=1.11-3.65, p=0.0310).
Stratified analysis of MMP-2 rs2285053 genotypes based on demographic and clinical characteristics. Subgroup analyses were conducted for MMP-2 rs2285053, stratified by the same demographic (Table VI) and clinical parameters (Table VII) as for rs243865. No significant differences in genotype distribution were observed with respect to age, sex, or alcohol consumption (Table VI). In contrast, smoking status revealed a pronounced effect. Among non-smokers, the variant CT and TT genotypes appeared to confer a protective trend against BLCA (OR=0.97 and 0.70; 95% CI=0.62-1.51 and 0.28-1.78), although these associations did not reach statistical significance (p=0.9715 and 0.6087). Conversely, in smokers, both CT and TT genotypes were significantly associated with elevated BLCA risk (OR=1.68 and 4.01, 95% CI=1.09-2.59 and 1.55-10.35, p=0.0214 and 0.0045, respectively).
Distribution of matrix metalloproteinase-2 rs2285053 genotypes with bladder cancer risk stratified by demographic and clinical indexes.
Distributions of matrix metalloproteinase-2 rs2285053 genotypes with bladder cancer risk stratified by clinical indexes.
Analysis by clinical features further indicated that rs2285053 variants are associated with more aggressive disease. The heterozygous CT genotype was significantly more prevalent in muscle-invasive BLCA, conferring increased risk (OR=1.62, 95% CI=1.04-2.55, p=0.0441). Additionally, the homozygous TT genotype was significantly overrepresented in high-grade tumors compared with low-grade cases, corresponding to a 2.5-fold increased risk (OR=2.47, 95% CI=1.07-5.71, p=0.0087; Table VII).
Discussion
BLCA remains the most recurrent malignancy worldwide, with incidence rates continuing to rise, particularly in Taiwan. Its development is influenced by both genetic predispositions and environmental exposures, with cigarette smoking being a major contributing factor (28, 29). The expression and activity profiles of MMPs and their tissue inhibitors (TIMPs) play critical roles in BLCA progression, particularly in angiogenesis and metastasis under pathological conditions. Consequently, numerous studies have explored the functions and expression patterns of MMPs in BLCA in both in vivo and in vitro models. Previous reports have documented dysregulation of several MMPs, including MMP-1 (30), MMP-9 (31-33), MMP-11 (34), as well as TIMP-2 (35, 36), which may serve as predictors of tumor stage, grade, and clinical outcomes. Among these, MMP-2 alterations have been reported most frequently (33, 37, 38). Despite these insights, relatively few studies have examined the genomic associations of MMP and TIMP genes with BLCA susceptibility (39-42). Notably, recent evidence from a Taiwanese population further supports the robustness of MMP-2 promoter polymorphism analyses. Hung et al. investigated the associations of MMP-2 rs243865 and rs2285053 with breast cancer risk in a large Taiwanese cohort and reported no significant associations with overall cancer susceptibility, while confirming Hardy-Weinberg equilibrium in the control group. These findings, derived from a comparable PCR-RFLP genotyping approach and population background, are consistent with our observation that MMP-2 promoter variants may exert context-dependent effects rather than serving as universal cancer susceptibility markers (43). In this context, the present study evaluated the associations between MMP-2 rs243865 and rs2285053 genotypes and BLCA risk, as well as their interactions with demographic and clinical factors. To our knowledge, only a limited number of studies have addressed the relationship between MMP-2 polymorphisms and BLCA susceptibility (44-48). In the following discussion, we summarize their findings, compare them with our results, and highlight the relevant data in Table VIII.
Literature reporting the genotypes of matrix metalloproteinase-2 rs243865 and rs2285053 among bladder cancer populations.
In 2006, Kader and his colleagues explored the potential relationship between MMP-2 rs243865 polymorphisms and BLCA risk within a diverse Caucasian cohort from the United States, comprising 542 BLCA patients and 545 matched controls (44). Their analysis revealed that individuals harboring either the CT or TT genotypes exhibited a reduced susceptibility to BLCA. Additionally, they assessed gene-environment interactions, particularly with smoking behavior, but did not observe any statistically significant associations. The same investigation extended to evaluating whether rs243865 variants influenced tumor invasiveness, yet no meaningful differences were detected between muscle-invasive and non-muscle-invasive cases. The study’s robustness was supported by its relatively large sample size and rigorously controlled questionnaire-based data collection. Building on this research, Kader’s team in 2007 further examined the role of another promoter polymorphism, MMP-2-1585, in BLCA progression. However, their findings once again indicated no association with disease invasiveness (49). In contrast, a subsequent study conducted in 2013 by Srivastava and his colleagues in an Indian population provided novel insights. Analyzing both MMP-2 rs243865 and rs2285053 genotypes in 200 BLCA patients and an equal number of controls, they reported that individuals homozygous for the TT allele at rs243865 faced an elevated risk of developing BLCA. This risk appeared to be further amplified among smokers. Moreover, patients with the CT genotype showed increased likelihood of tumor recurrence, and those carrying either CT or TT variants experienced shorter overall survival (45). In the same year, Wieczorek and his colleagues conducted a validation study among a Polish cohort, consisting of 241 patients and 199 controls. Their findings did not support the use of MMP-2 rs2285053 as a predictive marker for BLCA risk. Furthermore, no synergistic effects with smoking were detected, suggesting limited utility of this SNP in assessing genetic susceptibility in this population (46). More recently, in 2023, Alkanli and his colleagues investigated MMP-2 rs243865 polymorphisms in a Turkish cohort comprising 98 BLCA cases and 81 controls. Their results demonstrated a significant association between the TT genotype and increased BLCA risk. Allelic frequency analysis further supported the T allele as a risk factor. Interestingly, the CT genotype was found more frequently among controls, suggesting a potential protective effect (48). However, the limited sample size warrants cautious interpretation and emphasizes the need for larger, independent validation studies. In a parallel study conducted by the same group in the same year, the role of MMP-2 rs2285053 in BLCA susceptibility was revisited using a similar Turkish cohort. No significant association was found (47). Notably, the authors proposed that a combinatorial analysis involving rs2285053 and other polymorphisms, such as interleukin-8 rs2227306, may enhance predictive power for early detection, disease progression, and therapeutic outcomes in BLCA.
In our investigation, we analyzed the relationship between MMP-2 rs243865 and rs2285053 polymorphisms and the risk of BLCA, but no statistically significant associations were initially observed (Table II). However, for rs2285053, a borderline significance in genotypic distribution prompted further allelic frequency analysis, which revealed a clear and statistically significant association (Table III). Moreover, specific genotypic variants of both rs243865 and rs2285053 demonstrated potential utility as predictive markers when considered alongside demographic and clinical parameters, suggesting a role in personalized early detection strategies (Tables IV-VII).
These findings highlight the importance of integrating both genetic predisposition and lifestyle-related risk factors when evaluating individual susceptibility to BLCA. This approach is particularly relevant for high-risk populations, such as those in Taiwan, and offers a potential path toward tailored preventive interventions. Although our cohort size was relatively large and provides a credible representation of the studied population, the novel implication that rs2285053 variant genotypes may serve as biomarkers for BLCA warrants validation in larger, multicenter cohorts across diverse ethnic groups to determine population specificity. Regarding MMP-2 rs243865, our findings align with those reported by Kader’s and Wieczorek’s teams (44, 46), but differ from the results presented by Alkanli’s and Srivastava’s teams (45, 48). These discrepancies may reflect underlying ethnic or population-specific genetic variations, further emphasizing the need for expanded research across different groups. Ultimately, advancing the clinical utility of MMP-2 genotyping in early detection and risk prediction of BLCA will require continued collaboration between clinicians and translational researchers. From a biological perspective, a comprehensive review by Dofara et al. summarized that promoter polymorphisms of MMP-2, including rs243865 and rs2285053, may influence cancer susceptibility through modulation of transcriptional activity and alterations in the balance between latent and active forms of MMP-2 (50). Importantly, the review highlighted that the functional consequences of these variants appear to be cancer-type specific and may vary across populations. Such mechanistic complexity provides a plausible explanation for the heterogeneous associations reported among different malignancies and supports the interpretation that MMP-2 polymorphisms may contribute to tumor behavior in a context-dependent manner rather than exerting uniform effects across cancers (50). Coordinated efforts through well-designed prospective trials are essential to evaluate the feasibility and accuracy of incorporating these genetic markers into precision oncology frameworks.
Several limitations of this study should be acknowledged. First, although the sample size of BLCA patients was relatively large, it may still have been underpowered to detect subtle effects of genetic variants. Consequently, we refrained from performing complex stratified analyses involving multiple layers of subgrouping. Second, the study cohort was limited to individuals from Taiwan, which may constrain the generalizability of the findings to other ethnic or geographic populations. Nevertheless, this study represents a rare contribution from East Asia in the context of genetic research on BLCA, where such data remain scarce. To validate and extend our findings, future research should include larger and more ethnically diverse populations. Third, we were unable to evaluate the prognostic significance of the examined SNPs due to incomplete or insufficient follow-up data regarding patient survival outcomes. Lastly, exploring gene-environment interactions and the clinical relevance of these polymorphisms, such as associations with tumor grade or stage, remains particularly challenging in BLCA research. This difficulty stems primarily from the reduced and potentially unrepresentative sample sizes available for analysis after stratification, especially for SNPs with very low minor allele frequencies.
Conclusion
In summary, our study adds to the accumulating evidence that genetic variants in MMP-2 may serve as potential predictive markers for the diagnosis and prognosis of BLCA. While our findings require further validation in larger, multiethnic cohorts, the results offer important insights into the potential utility of MMP-2 genotypes as genetic biomarkers in BLCA. These contributions may inform future efforts toward the development of personalized approaches for early detection and clinical management of the disease.
Acknowledgements
The Authors are grateful to the Tissue-bank of China Medical University Hospital and doctors/nurses for their blood sample and questionnaire collection. The assistances from Ai-Chia Tung, Huai-Mei Hsu, and Hong-Xue Ji are highly appreciated. This study was supported by research grants from China Medical University Hospital (DMR-115-125) and Taichung Armed Forces General Hospital (TCAFGH_D_115022).
Footnotes
Authors’ Contributions
Research design: Liao CH, Tsai CW, Chang WS, Bau DT; patient and questionnaire summaries: Liao CH, Chang SY, Chang CH, Chen WC; experimental work: Chang WS, Wang YC, Shih HY, Tsai CW; statistical analysis: Hsia TC, Tsai CW, Chang WS; manuscript writing: Tsai CW, Chang WS, Bau DT; manuscript checking and discussing: Liao CH, Tsai CW, Chang WS, Wang YC, Shih HY, Hsia TC, Chang SY, Chang CH, Chen WC, Bau DT.
Conflicts of Interest
All Authors declare no conflicts of interest regarding this study.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received January 6, 2026.
- Revision received January 30, 2026.
- Accepted February 13, 2026.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.