Abstract
Background/Aim: Pterygium is a fibrovascular ocular disease characterized by extracellular matrix (ECM) remodeling. Matrix metalloproteinase-3 (MMP-3), a key ECM-modulating enzyme, has been implicated to involved in pterygium progression, but its genetic marker has never been explored. This hospital-based case-control study investigated the association between MMP-3 polymorphisms and pterygium risk in a Taiwanese cohort.
Patients and Methods: Up to five MMP-3 genotypic patterns (rs3025058, rs522616, rs591058, rs650108, and rs679620) were identified among 160 pterygium cases and 320 age- and sex-matched controls.
Results: The MMP-3 rs3025058 5A allele was significantly associated with an elevated risk of pterygium (OR=1.94, 95%CI=1.39-2.71, p=0.0001). Compared to the wild-type 6A/6A genotype, individuals carrying MMP-3 5A/6A and 5A/5A exhibited a 1.66-fold and 4.36-fold elevated risk, respectively (p=0.0253 and 0.0014). Particularly, MMP-3 rs3025058 5A/5A genotype was significantly associated with elevated pterygium risk among elder (≥60 years old) individuals (p=0.0003). Furthermore, transcriptional and translational analyses revealed higher MMP-3 expression in carriers of the 5A allele, with the highest levels observed in 5A/5A homozygotes (all p<0.05). No significant associations were observed for the remaining four MMP-3 polymorphic variants.
Conclusion: MMP-3 rs3025058 may serve as a genetic biomarker for pterygium susceptibility prediction, potentially contributing to ECM dysregulation and disease progression.
Introduction
Pterygium, commonly referred to as surfer’s eye, is a non-malignant fibrovascular growth originating from the bulbar conjunctiva, frequently extending toward the corneal limbus (1). As one of the most prevalent ocular disorders worldwide, its underlying pathogenesis remains largely unresolved. A multinational study conducted in 2018, encompassing a cohort of 415,911 individuals across 24 countries, reported an average prevalence of approximately 12% (2). The condition was least common (3%) among individuals aged 10 to 20, whereas its occurrence peaked at 19.5% among those over 80 years old. Notably, the prevalence rates of pterygium were comparable between male and female sub-populations (2). The development of pterygium shares multiple hallmarks with neoplastic processes, including uncontrolled fibrovascular proliferation, progressive invasion of the cornea, extracellular matrix (ECM) remodeling, and a propensity for postoperative recurrence (3, 4). The encroachment of these wedge-shaped lesions onto the cornea exhibits striking parallels with the invasive behavior observed in solid tumors (5). Within the subepithelial regions of pterygium, substantial ECM accumulation comprising both fibrillar and amorphous components distinguishes it from the normal conjunctiva, where such deposits are absent. This dysregulated ECM turnover is hypothesized to play a pivotal role in pterygium pathogenesis (6). Despite significant advancements in high-throughput transcriptomic and proteomic analyses, pterygium research continues to face limitations due to consistently insufficient sample sizes (7-10). While investigations into genetic biomarkers offer promising avenues for enhancing both diagnostic precision and therapeutic strategies, the identification of universally applicable and clinically practical markers remains an ongoing challenge.
As noted above, dysregulated ECM remodeling is a critical feature of pterygium pathogenesis. However, the precise mechanisms underlying the development of pterygium remain insufficiently revealed. It is widely implicated matrix metalloproteinases (MMPs) in driving ECM remodeling, long-term inflammation, degradation of Bowman’s layer, and the invasive progression of pterygium into the cornea. Notably, studies conducted by Solomon’s and Li’s teams demonstrated that pro-inflammatory cytokines IL-1β and TNF-α significantly upregulated at both mRNA and protein levels of MMP-1 and MMP-3 in primary cultures of pterygium-derived fibroblasts (11, 12). Around the same period, Di Girolamo and his colleagues provided the first evidence of MMP-7 overexpression among eight cultured pterygium tissues (13). Subsequent studies further identified elevated expression of MMP-2 and MMP-9, the two MMPs strongly associated with cancer metastasis, among tissue samples from fifteen pterygium patients (14). A comprehensive analysis in 2010 confirmed the upregulation of MMP-3, alongside other MMPs including MMP-1, MMP-2, MMP-7, MMP-8, MMP-9, and MMP-14, in pterygium specimens (15). More recently, migrating fibroblasts within pterygium lesions were found to exhibit significantly elevated expression of MMP-3 and MMP-13 among 10 male and 10 female pterygium patients (16). Additionally, studies have reported heightened levels of MMP-1 and MMP-9 in 89 recurrent pterygium samples (17), as well as both upregulation and enhanced activation of MMP-14 in a cohort of 28 pterygium tissues (18). Despite over two decades of research, the precise contributions of various MMPs to pterygium etiology remain only partially elucidated, particularly from a genomic viewpoint. This knowledge gap underscores the urgent need for studies by ophthalmologists and translational medical researchers to further reveal the molecular mechanisms driving pterygium progression.
MMP-3 also referred to as stromelysin-1, is encoded by the MMP-3 gene located on chromosome 11q22.3 (19). This enzyme exhibits broad substrate specificity, capable of degrading various extracellular matrix components, including collagen types II, III, IV, IX, and X, as well as proteoglycans, fibronectin, laminin, elastin, and osteopontin (20, 21). As highlighted above, MMP-3 is upregulated in pterygium tissues and has been implicated in angiogenic processes, partly due to its role in generating angiostatin, a well-known inhibitor of neovascularization (22, 23). Conversely, pharmacological inhibition of MMP-3 has been shown to attenuate angiogenic responses (24). Beyond its involvement in angiogenesis, MMP-3 is also a key activator of several pro-MMPs, including MMP-1, MMP-7, MMP-9, and MMP-13, thereby contributing to ECM remodeling, cell migration, and tissue remodeling (25, 26). Despite these insights into its functional significance, no studies have explored the relationship between MMP-3 genetic variants and susceptibility to pterygium. To bridge this gap, we conducted the first hospital-based case-control study to systematically evaluate the association between MMP-3 polymorphisms and pterygium risk in a Taiwanese cohort. Specifically, we examined MMP-3 promoter (rs3025058, rs522616), intronic (rs591058, rs650108), and exonic (rs679620) polymorphisms among 160 pterygium cases and 320 age- and sex-matched controls without pterygium.
Materials and Methods
Recruitment of pterygium and non-pterygium participants. Enrollment of individuals with and without pterygium was conducted in accordance with an approved study protocol (IRB number: CMUH111-REC1-176) with the principles of the Helsinki’s Declaration. Prior to participation, written informed consent was secured from all the recruited subjects. A total of 160 patients diagnosed with pterygium were included, alongside a control cohort consisting of twice as many individuals without the condition. The overall flow chart of the study is illustrated in Figure 1. All participants were Taiwanese nationals who voluntarily completed a standardized questionnaire and donated peripheral blood samples. Strict criteria were applied to the selection of controls to exclude individuals with prior diagnoses of pterygium, endometriosis, leiomyoma, or any type of malignancy. A summary of patient characteristics is shown in Table I.
The overall flow chart of the study.
Demographics of the pterygium cases and the non-pterygium subjects.
MMP-3 genotyping processes. Each subject contributed a 3-5 ml blood sample, from which genomic DNA was isolated from peripheral leukocytes within 12 h. Extracted DNA was subsequently diluted and partitioned into aliquots for short-term storage at −20°C, adhering to our established protocols (27, 28). For extended preservation, samples were maintained at −80°C or cryopreserved in liquid nitrogen. The genotyping strategy for MMP-3 was developed at the Terry Fox Cancer Research Laboratory, encompassing primer design and enzyme selection. Detailed information regarding primer sequences, restriction enzymes employed and expected DNA fragment sizes pre- and post-digestion is presented in Table II. Amplifications was conducted using a T100 Thermal Cycler (Bio-Rad, Hercules, CA, USA) under the following conditions: initial denaturation at 94°C for 5 min, followed by 35 cycles of 94°C for 30 s (denaturation), 59°C for 30 s (annealing), and 72°C for 30 s (extension), with a final elongation step at 72°C for 10 min. Post-PCR, amplicons corresponding to MMP-3 rs522616 and rs650108 were subjected to direct sequencing, whereas those derived from MMP-3 rs3025058, rs591058, and rs679620 underwent enzymatic digestion before separation via 3% agarose gel electrophoresis. The physical locations of MMP-3 genotypes are illustrated in Figure 2. All genotyping assays were performed in duplicate as technical replicates. Approximately 10% of the samples were randomly selected and re-genotyped for quality control, and the concordance rate was 100%.
The primer sequences, methodologies for identifying MMP-3 rs3025058, rs522616, rs591058, rs650108 and rs679620 polymorphic genotypes.
Physical map of MMP-3 rs3025058, rs522616, rs591058, rs650108 and rs679620 polymorphic sites. The rs3025058 and rs522616 locating in the promoter region are marked in green. The rs679620 locating in exon region is marked in blue. The rs591058 and rs650108 locating in intron region are marked in purple.
MMP-3 mRNA expression analysis via quantitative real-time PCR. A total of 49 blood samples were obtained, comprising 29 from individuals with pterygium and 20 from healthy controls (biological replicates). RNA extraction was conducted using the RNeasy Mini Kit (Qiagen Inc., Hilden, Germany), followed by RNA quantification utilizing the NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). To synthesize complementary DNA (cDNA), 1 μg of RNA was reverse-transcribed applying the RT2 First Strand Kit (Qiagen). Quantitative real-time PCR (qPCR) was then performed to evaluate MMP-3 mRNA levels on the FTC-3000 real-time PCR machine (Funglyn Biotech Inc., Toronto, ON, Canada). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) served as the internal control gene for normalization purposes. The primer sequences designed by Terry Fox Cancer Research Lab were as follows: GAPDH, forward 5′-GAAATCCCATCACCATC-TTCCAGG-3′ and reverse 5′-GAGCCCCAGCCTTCTCCATG-3′; MMP-3, forward 5′-GGAG TTCCTGATGTTGGTCA-3′ and reverse 5′-CTGGTGTATAA TTCACAATCCT-3′. PCR reactions were carried out in a total volume of 20 μl, containing IQ™ SYBR Green Supermix (Bio-Rad, Hercules, CA, USA), 0.5 μmol/l of each primer, and 9 ng of cDNA. Each sample was measured in triplicate as technical replicates to ensure the reliability of the results. To minimize bias, all PCR experiments were performed by double-blinded researchers. Relative changes in MMP-3 expression were quantified using the 2−ΔΔCt method, with normalization to the respective GAPDH CT values. The methodology followed in this study aligns with our previously published protocols (29).
MMP-3 protein expression level measured via western blotting assay. Blood samples from 29 pterygium patients and 20 healthy controls were processed using lysis buffer from Upstate Biotechnology Inc. (Lake Placid, NY, USA). After centrifugation, the supernatants were collected and subjected to Western blot analysis following our previously described protocols (30). In brief, 20 μg of protein from each sample were denatured at 95°C for 10 minutes and then separated by a fresh-prepared 12% sodium dodecyl sulfate polyacrylamide gel. The proteins were transferred to a nitrocellulose membrane (BioRad Laboratories, Hercules, CA, USA). To prevent non-specific binding, the membrane was treated by fresh-prepared 5% non-fat milk and incubated overnight at 4°C with a monoclonal anti-human MMP-3 antibody (1:1,000; Cat. sc-21732, Santa Cruz Biotechnology, Dallas, TX, USA). Following this, the membrane was exposed to a horseradish peroxidase-conjugated goat anti-mouse IgG secondary antibody (Chemicon, Temecula, CA, USA) at room temperature for 1 h. Afterward, protein bands were detected by enhanced chemiluminescence solution (Amersham, Arlington Heights, IL, USA) and visualized using a chemiluminescence imaging system (Syngene, Cambridge, UK). To re-probe the membrane, it was stripped at 56°C for 18 minutes using a stripping buffer (0.0626 M Tris-HCl, pH 6.7, 2% SDS, 0.1 M mercaptoethanol) and subsequently incubated with a monoclonal anti-GAPDH antibody (1:5,000, Cat. sc-137179, Santa Cruz Biotechnology) as an internal control. The intensity of each protein band was double-blinded quantified with the ImageJ analysis software by two well-trained researchers. Each Western blot experiment was independently repeated at least three times using the same protein samples.
Statistical analysis of MMP-3 variants. To assess the impact of MMP-3 polymorphisms, alleles, age- and sex-stratified genotypes on the likelihood of developing pterygium, Pearson’s chi-square test was employed for categorical data with an expected cell count of ≥5, while Fisher’s exact test was used for categories with fewer than 5 expected counts. The expression levels of MMP-3 at mRNA and protein were compared by unpaired Student’s t-test. The relationships between variables were further examined by computing odds ratios (ORs) along with their respective 95% confidence intervals (CIs). Any p-value less than 0.05 was considered statistically significant.
Results
Comparison of age and sex between the pterygium and control cohorts. The preliminary assessment indicated no significant variation in age distribution between individuals with pterygium and those in the control group (p=0.9006). This finding remained unchanged even when participants were categorized using 60 years as a cutoff (p=0.2509). Additionally, to uphold methodological precision, sex matching was meticulously implemented between cases and controls, effectively minimizing potential bias (p=1.0000).
Association between MMP-3 genetic variants and pterygium susceptibility. Genotypic distributions of MMP-3 polymorphisms rs3025058, rs522616, rs591058, rs650108, and rs679620 in the control cohort conformed to Hardy-Weinberg equilibrium (p=0.6246, 0.4755, 0.1635, 0.4563, and 0.9594, respectively; Table III). Notably, in individuals with pterygium, the prevalence of rs3025058 genotypes (6A/6A, 5A/6A, and 5A/5A) was 58.8%, 32.5%, and 8.7%, respectively, differing from the control group, where these genotypic frequencies were 73.1%, 24.4%, and 2.5% (p for trend=0.0006). Compared to the 6A/6A genotype, individuals harboring 5A/6A and 5A/5A exhibited a 1.66-fold and 4.36-fold increased likelihood of developing pterygium, respectively (95%CI=1.09-2.54 and 1.77-10.73, p=0.0253 and 0.0014). Under a dominant inheritance model, carriers of either the 5A/6A or 5A/5A genotype demonstrated a 1.91-fold greater risk of pterygium (95%CI=1.28-2.85, p=0.0020) (Table III, upper section). Conversely, no statistically significant differences were detected in the distribution of rs522616, rs591058, rs650108, or rs679620 genotypes between cases and controls under any genetic model (all p>0.05) (Table III, lower sections).
Genotypic frequency distributions of matrix metalloproteinase-3 genotypes among the pterygium cases and the non-pterygium subjects.
Association between MMP-3 allelic variants and pterygium susceptibility. Analysis of allelic distributions demonstrated a significant correlation between the MMP-3 rs3025058 5A allele and an increased risk of pterygium compared to the 6A allele (25.0% vs. 14.7%, OR=1.94, 95%CI=1.39-2.71, p=0.0001) (Table IV, upper section). In contrast, for rs522616, rs591058, rs650108, and rs679620, the presence of variant alleles showed no statistically significant association with pterygium susceptibility (OR=1.00, 1.16, 0.94, and 1.10, respectively; 95%CI=0.76-1.32, 0.87-1.54, 0.71-1.23, and 0.83-1.46; p=1.0000, 0.3601, 0.6902, and 0.5448, respectively; Table IV, lower sections).
Allelic frequencies for matrix metalloproteinase-3 polymorphisms among the pterygium cases and healthy subjects.
Association between MMP-3 genotypes and pterygium risk across different age and sex groups. To explore the potential interaction between MMP-3 genotypes and age in influencing pterygium risk, further stratified analyses were conducted for MMP-3 rs3025058 across age groups (Table V). No significant association with pterygium risk was observed among individuals aged less than 60 years, regardless of whether they carried the heterozygous or homozygous variant genotypes (OR=1.17 and 1.84, 95%CI=0.60-2.28 and 0.36-9.46, p=0.7830 and 0.6670, respectively (Table V, left section). However, among individuals older than 60 years, the homozygous variant 5A/5A genotype of MMP-3 rs3025058 was significantly associated with an increased risk of pterygium (OR=7.10, 95%CI=2.34-21.54, p=0.0003; Table V, right section). The analyzing results remain significant after adjusting with the factor of sex (OR=6.29, 95%CI=2.12-18.37; Table V, right section). Similarly, the homozygous variant 5A/5A genotype of MMP-3 rs3025058 was significantly associated with an increased risk of pterygium among both males and females (OR=3.91 and 5.10, 95%CI=1.22-12.49 and 1.22-21.38, p=0.0326 and 0.0235; Table VI).
Distribution of matrix metalloproteinase-3 rs3025058 genotypes among 160 pterygium cases and 320 non-pterygium controls after stratification by age.
Distribution of matrix metalloproteinase-3 rs3025058 genotypes among 160 pterygium cases and 320 non-pterygium controls after stratification by sex.
Correlation between genotype and phenotype of MMP-3. The transcriptional level of MMP-3 in both the non-pterygium and pterygium groups showed that the heterozygous 5A/6A carriers had higher levels than the wild-type 6A/6A carriers (p=0.0377 and 0.0004, Figure 3A and B). Additionally, the homozygous 5A/5A group exhibited even higher levels than the heterozygous 5A/6A group (Figure 3A, B). Furthermore, the pterygium group had higher levels than the non-pterygium group. When combining the heterozygous 5A/6A and homozygous 5A/5A groups and comparing them to the wild-type 6A/6A, there was a borderline significant difference in the non-pterygium group (p=0.0627, Figure 3A), while a significant difference was observed in the pterygium group (p=0.0065, Figure 3B). The findings demonstrate that the translational level of MMP-3 in both the non-pterygium and pterygium groups was higher in heterozygous 5A/6A carriers compared to wild-type 6A/6A carriers, and that homozygous 5A/5A carriers exhibited even higher levels than heterozygous 5A/6A (Figures 3C, D). Additionally, pterygium patients had higher levels than non-pterygium patients. When combining the heterozygous 5A/6A and homozygous 5A/5A groups and comparing them to the wild-type 6A/6A, significant differences were observed in both the non-pterygium group and the pterygium group (p=0.0420 and 0.0048, Figure 3C, D).
Serum mRNA levels among (A) 20 healthy control subjects and (B) 29 pterygium patients according to their MMP-3 rs3025058 genotypes. The serum MMP-3 mRNA levels were measured by quantitative PCR. The quantitative results of Western blotting for the 20 healthy control subjects (C) and 29 pterygium patients (D) according to their MMP-3 rs3025058 genotypes. The relative levels of serum MMP-3 protein levels were based on folds of those carrying MMP-3 rs3025058 6A/6A genotype for the controls. Statistically significant p-values between compared groups are shown in red. Data are presented as mean±SD from independent biological samples, and each measurement was performed in triplicate for qRT-PCR. Western blot analyses were independently repeated at least three times.
Discussion
There remains considerable debate among ophthalmologists regarding the precise molecular mechanisms driving pterygium development. A widely accepted hypothesis suggests that disruptions in the balance of ECM components contribute to pterygium pathogenesis, with MMPs playing a pivotal role in ECM remodeling. MMPs have been implicated in both pterygium and various malignancies (31-34). However, their precise functions in pterygium remain incompletely understood. Previous studies have documented the overexpression of multiple MMPs in pterygium, including MMP-1 (11, 12, 17, 35, 36), MMP-2 (14, 36), MMP-9 (14, 37), MMP-10 (37), MMP-14 (18), and particularly MMP-3 (11, 12, 36), Despite these findings, studies on other MMP family members remain scarce, and previous investigations have often been constrained by limited sample sizes. From a genomic perspective, research on MMPs in pterygium has been relatively limited. Recently, our team has examined the genotypic distributions of MMP-1 (38), MMP-2 (39), MMP-7 (40), MMP-8 (41), MMP-9 (38) and MMP-11 (42) in a representative Taiwanese pterygium cohort. The highlight is that we found MMP-1 rs1799705 2G and MMP-11 rs738792 C alleles could serve as predictive biomarkers for pterygium (38, 42).
To date, the involvement of MMP-3 at the genetic or proteomic level in pterygium pathogenesis remains unexplored. This study firstly investigated the potential association between pterygium susceptibility and MMP-3 genetic variants, including promoter (rs3025058, rs522616), intronic (rs591058, rs650108), and exonic (rs679620) polymorphisms, in a well-defined Taiwanese cohort comprising 160 pterygium cases and 320 matched controls (Table I). Notably, analysis revealed a significant correlation between the presence of the heterozygous 5A/6A and homozygous 5A/5A genotypes at rs3025058 and an elevated risk of developing pterygium (Table III). In contrast, no significant associations were observed for the variant genotypes of rs522616, rs591058, rs650108, or rs679620 with disease susceptibility (Table III). Under a dominant inheritance model, individuals carrying either the 5A/6A or 5A/5A genotype exhibited an odds ratio of 1.91 for pterygium, positioned between the separate risks conferred by the heterozygous (1.66-fold) and homozygous (4.36-fold) forms of rs3025058. Furthermore, allele frequency analysis demonstrated a strong association between the 5A allele and pterygium occurrence (p=0.0001) (Table IV). These findings indicate that carrying at least one copy of the 5A allele at rs3025058 increases the likelihood of developing pterygium.
This finding is particularly significant and novel, highlighting the need for expanded investigations into MMP-3 genetic variations, especially rs3025058, in a larger cohort of individuals affected by pterygium. To the best of our knowledge, this study is the first to identify a potential link between MMP-3 rs3025058 genotypes and pterygium susceptibility at a global level. The observed frequency of the 5A allele at rs3025058 in our cohort closely aligns with previous reports from various populations, including 48.4% in a Slovakian cohort of 308 individuals (43), 40.0% among 140 Egyptian subjects (44), 31.7% in a Brazilian group of 169 participants (45), 19.3% in 150 Chinese individuals (46), 14.6% in a Turkish cohort (47), and 14.0% among 332 Taiwanese subjects (48). The 5A allele frequency of 14.7% in our study is consistent with prior findings in the Taiwanese population but exhibits considerable variation across ethnic groups. We have to emphasized that the cited reports above were investigating other diseases, not pterygium. These differences emphasize the necessity of investigating pterygium-associated genetic markers in diverse populations to further validate the role of the MMP-3 rs3025058 5A allele as a potential risk factor for this condition.
Study limitations. First, our collected samples were obtained from a single medical center rather than from institutions across Taiwan. This may introduce a slight sampling bias. Fortunately, our control data were sourced from a hospital-based health examination database, which contains over 15,000 control samples for selection. Second, due to the rarity and small size of pterygium specimens, it was not feasible to obtain a matched normal counterpart from exactly the same patient under the existing IRB regulations. Consequently, the normal samples we collected were not only scarcer than those from the pterygium group but also originated from different individuals. This limitation prevents us from definitively determining whether the elevated MMP-3 expression is a pathogenic factor in pterygium. Third, we were unable to elucidate in detail the mechanisms by which MMP-3 contributes to pterygium etiology. Based on previous literature and the findings of this study, we speculate that MMP-3 may be a highly relevant candidate in the pathogenesis and progression of pterygium. MMP-3 exhibits broad-spectrum enzymatic activity, enabling it to degrade collagen (types II, III, IV, IX, and X), proteoglycans, fibronectin, laminin, and elastin (49). Both the expression and activity of MMP-3 have been reported to be significantly increased in pterygium fibroblasts, which dominate the stromal compartment (12). The upregulation of MMP-3 in the pterygium stroma may contribute to the elastotic degeneration of various collagens frequently observed in diseased tissue (26), and consistent with our findings (Figure 3). The genotype-phenotype correlation of MMP-3, which showed that 5A allele had a higher expression of mRNA and/or protein, was validated by Du’s team (50) and Huang’s team (51). More interesting, MMP-3 can activate other MMPs, including MMP-1, MMP-7, and MMP-9, indicating that MMP-3 functions upstream in the activation cascade of multiple latent MMPs and plays a crucial role in extracellular matrix remodeling (52). A more in-depth investigation into the mechanistic involvement of MMP-3 in pterygium development is warranted.
In summary, the MMP-3 rs3025058 5A allele is significantly associated with an increased risk of pterygium, exhibiting a gene-dosage effect on susceptibility. Individuals carrying 5A/6A and 5A/5A genotypes have a 1.66-fold and 4.36-fold elevated risk, respectively. Functionally, these genotypes correlate with increased MMP-3 transcription and translation, with homozygous 5A/5A carriers exhibiting the highest expression levels. These findings suggest that MMP-3 rs3025058 may contribute to pterygium pathogenesis via extracellular matrix remodeling and could serve as a potential biomarker or therapeutic targets.
Acknowledgements
The Authors are grateful to the Tissue-bank of China Medical University Hospital and doctors/nurses for their blood sampling and questionnaire collection. The technical help from Yu-Cheng Luo and Yu-Hsin Yen was appreciated by the authors. This research was funded by Asia University & China Medical University, and Hospital (DMR-114-039 and CMU113-ASIA-02), and Taichung Veterans General Hospital (TCVGH-VHCY1138602). The funding sources did not influence the study’s design, data acquisition and interpretation, publication decision, or manuscript preparation.
Footnotes
Conflicts of Interest
All the Authors declare no conflicts of interest regarding this study.
Authors’ Contributions
Conceptualization: NYH, HCC and CLT; Data curation: NYH, CLT, JCL and TCH; Formal analysis: HCC, JCL and YCW; Funding acquisition: NYH, HCC and CLT; Investigation: HYS, CWT and WSC; Methodology: YCW, HYS, CWT, WSC and DTB; Project administration: CWT and DTB; Resources: NYH, HCC, PSH and TCH; Supervision: DTB; Validation: TCH, CWT and DTB; Writing - original draft: NYH, CWT and DTB; Writing - review & editing: CWT and DTB.
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 5, 2026.
- Revision received January 27, 2026.
- Accepted February 5, 2026.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
