Colorectal Polyp Occurrence Associated With the ZMIZ1‐AS1 rs704017 Variant and Hypertension Among Taiwan Biobank Participants Authors: Bei‐Hao Shiu, Hsu‐Chih Huang, Oswald Ndi Nfor, Shu‐Yi Hsu, Yung‐Po Liaw Published: 2026 ABSTRACT ABSTRACTColorectal polyps (CPs) are recognized precursors of colorectal cancer (CRC), which remains a major global public health challenge, ranking third in incidence and second in cancer‐related mortality worldwide. This study investigated the associations of hypertension and the zinc finger MIZ‐type containing 1 antisense RNA 1 (ZMIZ1‐AS1) variant (rs704017) with the odds of CP among participants in the Taiwan Biobank (TWB). Data from the TWB (2008–2015) were linked to the National Health Insurance Research Database (NHIRD, 1998–2015), yielding 8970 participants aged 30–70 years. Multiple logistic regression models were used to estimate adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for CPs according to hypertension status and rs704017 genotype. The distribution of ZMIZ1‐AS1 rs704017 genotypes differed significantly between CP cases and controls (p = 0.001). Individuals carrying the rs704017 AG + GG genotype had higher odds of CPs than those with the AA genotype (aOR = 1.39, 95% CI: 1.16–1.66). Hypertension alone was not significantly associated with CP odds; however, a significant interaction was observed between hypertension and rs704017 (p = 0.029). Compared with non‐hypertensive individuals with the AA genotype, the aOR was 1.62 (95% CI, 1.29–2.03) for non‐hypertensive individuals with AG + GG, 1.45 (95% CI, 1.09–1.95) for hypertensive individuals with AA, and 1.56 (95% CI, 1.16–2.08) for hypertensive individuals with AG + GG. In conclusion, the ZMIZ1‐AS1 rs704017 AG + GG genotype was independently associated with increased odds of CP, whereas hypertension alone was not. However, the association between the rs704017 variant and CP appeared stronger among individuals with hypertension, suggesting a potential interaction between genetic susceptibility and hypertension. Abbreviations body mass index confidence interval colorectal polyp colorectal cancer International Classification of Diseases, Ninth Revision, Clinical Modification National Health Insurance Research Database odds ratio reference Taiwan Biobank zinc finger MIZ‐type containing 1 antisense RNA1 Introduction Colorectal cancer (CRC) represents a major global public health challenge and ranks among the leading causes of cancer incidence and mortality worldwide [1]. In 2020, approximately 1.9 million new cases and 935,000 deaths were reported worldwide [1]. Douaiher et al. [2] highlighted an increasing global prevalence of CRC, projecting that this trend will persist beyond 2035 [2], with Asia bearing the highest burden in terms of incidence and mortality [2]. Colorectal polyps are abnormal growths arising from the inner lining of the colon or rectum that protrude into the lumen. They are recognized as precursors to CRC [3, 4, 5, 6], with colorectal adenomas giving rise to the majority of CRC cases [3]. Consequently, the prompt identification and management of CPs are crucial for preventing CRC [7]. The progression from normal colonic mucosa to adenoma and eventually carcinoma typically occurs over approximately 10 years [8]. This transitional period allows for the implementation of preventive and timely interventions against cancer development [9]. Unfortunately, symptoms of CPs are often subtle and can be overlooked until cancer has developed. Therefore, effective screening strategies targeting asymptomatic individuals at elevated risk, coupled with timely intervention, could significantly reduce the incidence of CRC and associated mortality [8, 10, 11]. However, identifying asymptomatic individuals, particularly in the early stages of the disease, remains a challenge [12]. Current research is focused on identifying noninvasive biomarkers that could predict CPs at earlier stages [13]. Such biomarkers could aid in risk stratification and inform management strategies [14]. One effective approach to identifying and targeting high‐risk individuals is to explore the influence of genetic and environmental factors and their interactions [15]. Previously reported risk factors for CPs include the rs704017 variant [16, 17], hypertension [18], sex [19, 20], alcohol consumption [14, 18, 21], smoking [14, 18, 21, 22, 23, 24, 25], family history of CPs and CRC [8, 18, 20, 26, 27], obesity [14, 18, 28], diabetes [23, 29], and age [14, 18, 24]. Ethnic disparities exist concerning CP prevalence and risk [20]. CRC and CPs share several similarities regarding their genetic and nongenetic factors [29]. For instance, hypertension is associated with CPs and could cause CRC, especially if untreated [30]. The effects of genetic variants on disease pathogenesis can vary significantly across populations [31]. The rs704017 variant, located within the intronic region of ZMIZ1‐AS1 at chromosome 10q22.3, has been associated with CPs and CRC in several populations, including East Asians [16, 17, 32, 33]. While some studies have explored the relationship between rs704017 and CPs in East Asian groups, such as Japanese, Han Chinese, and Koreans, research involving Taiwanese individuals remains limited [17, 32, 33]. The relationship between hypertension and CPs is inconsistent, with various studies yielding conflicting results [18, 19, 22, 23, 24, 34, 35, 36]. Furthermore, the interaction between hypertension and rs704017 in relation to CP risk has remained largely unexplored. Therefore, this study aimed to evaluate the associations of hypertension and rs704017 with colorectal polyp risk among Taiwan Biobank participants. Materials and Methods Data Sources and Study Participants A total of 17,985 participants were initially enrolled in the TWB during the study period. Because dietary information was included as a study variable, only 9001 participants with complete dietary questionnaire data were eligible for analysis. Among these participants, 14 individuals with incomplete questionnaire data and 17 individuals with missing genotype data were excluded, leaving 8970 participants (536 CP cases and 8434 controls) for the final analysis. This study used data from the TWB (2008–2015) linked to the NHIRD (1998–2015). The TWB provided information on age, sex, educational level, physical activity, cigarette smoking, diet, genotype, alcohol consumption, weight, height, and tea and coffee consumption. However, the TWB does not contain diagnostic information on colorectal polyps. The NHIRD (1998–2015) included information on colorectal polyps. Therefore, the TWB and NHIRD datasets were linked using encrypted personal identification numbers. All participants provided written informed consent before enrollment. All methods were conducted in accordance with relevant guidelines and regulations, and ethical approval was granted by the Institutional Review Board of Chung Shan Medical University (IRB: CS1‐20009). Assessment of ZMIZ1‐AS1 rs704017, CPs, Hypertension, and Covariates The rs704017 variant was genotyped using custom‐designed Taiwan Biobank genotyping arrays on the Axiom Genome‐Wide Array Plate System (Affymetrix, Santa Clara, CA, USA). The rs704017 call rate was 99.83%, and the minor allele frequency (MAF) was 0.277. The genotype distribution was consistent with Hardy–Weinberg equilibrium (HWE). CPs were identified using lower gastrointestinal endoscopy and corresponding NHI reimbursement procedure codes, including 28017C, 49014C, 49025C, 49026C, 49023C, and 73008B. In the NHIRD, we identified colorectal polyps based on at least one admission or a single outpatient visit between 1998 and 2015, using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) codes, including 211.3, 211.4, and 569.0. Hypertension was defined by ICD‐9‐CM codes 401–405, diabetes mellitus with code 250, and hyperlipidemia with code 272, based on at least one hospitalization or two outpatient claims between 1998 and 2015. Cigarette smokers were defined as individuals who had smoked for more than 6 months. Alcohol drinkers were those who habitually consumed more than 150 mL of alcohol per week for over 6 months. Regular physical activity was defined as exercise lasting at least 30 min at least three times a week. Body mass index (BMI) was calculated as weight (kg) divided by height (m2). BMI categories included underweight (BMI < 18.5 kg/m2), normal weight (18.5 ≤ BMI < 24 kg/m2), overweight (24 ≤ BMI < 27 kg/m2), and obesity (BMI ≥ 27 kg/m2). A family history of CRC was defined as having a biological mother, father, sister, or brother with a doctor‐confirmed diagnosis of CRC. Tea drinkers consumed tea daily, while coffee drinkers consumed coffee three or more times per week. Vegetarians maintained a vegetarian diet for 6 months or more. Fat intake scores were assessed based on cooking methods (with or without oil), consumption of fried food, meat, poultry, and meat skin/fat over the past month. Scores ranging from 12 to 60 were grouped into quartiles: ≤ 35 (< Q1), 36–40 (Q1–Q2), 41–44 (Q2–Q3), and ≥ 45 (> Q3), with higher scores indicating diets lower in fat [37]. Statistical Analyses Statistical analyses were conducted using the Statistical Analysis System (SAS) software (version 9.4) and PLINK 1.09. Baseline characteristics were compared using the chi‐square test. ORs and 95% CIs for the associations of hypertension and the rs704017 variant with colorectal polyps were estimated using logistic regression models. Covariates included in the multivariable logistic regression models were selected a priori based on previously reported risk factors for CPs and potential confounders identified in the literature, rather than by automated variable selection procedures. Accordingly, all predefined covariates were entered simultaneously into the multivariable models. To assess potential multicollinearity among the independent variables, variance inflation factors (VIFs) and tolerance values were calculated. All VIFs ranged from 1.001 to 1.511, and tolerance values ranged from 0.662 to 0.999, indicating no evidence of substantial multicollinearity. Therefore, all predefined covariates were retained in the final regression models. A two‐sided p value < 0.05 was considered statistically significant. Results Table 1 summarizes the baseline characteristics of participants with and without CP. The distribution of the ZMIZ1‐AS1 rs704017 genotype between CP cases (n = 536) and controls (n = 8434) was significantly different (p = 0.001). In the control group, there were 4405 (52.23%) individuals with genotype AA, 3429 (40.66%) with AG, and 600 (7.11%) with GG. Among CP cases, 44.03%, 46.83%, and 9.14% carried the AA, AG, and GG genotypes, respectively. TABLE 1Baseline characteristics of 536 participants with colorectal polyps and 8434 polyp‐free participants.VariablesControlsCases p n % n %ZMIZ1‐AS1 rs704017 genotypes0.001AA440552.2323644.03AG342940.6625146.83GG6007.11499.14Hypertension< 0.001No648376.8733362.13Yes195123.1320337.87Sex0.166Female452453.6427150.56Male391046.3626549.44Age (years)< 0.00130–39204924.29438.0240–49234827.849818.2850–59242628.7620438.0660–70161119.1019135.63Educational level0.002Elementary school and below6047.166011.19Junior and senior high school353341.8922441.79University and above429750.9525247.01Cigarette smoking0.334No652077.3142479.1Yes191422.6911220.9Alcohol consumption0.798No753989.3948189.74Yes89510.615510.26Physical activity< 0.001No492958.4424345.34Yes350541.5629354.66Body mass index (kg/m2)0.024BMI < 18.5 (underweight)2182.5881.4918.5 ≤ BMI < 24 (normal weight)402647.7424044.7824 ≤ BMI < 27 (overweight)251029.7619035.45BMI ≥ 27 (obesity)168019.929818.28Diabetes< 0.001No727486.2540775.93Yes116013.7512924.07Hyperlipidemia< 0.001No592670.2627551.31Yes250829.7426148.69Family history of colorectal cancer< s0.001No797594.5646586.75Yes4595.447113.25Tea consumption0.006No525462.3036668.28Yes318037.7017031.72Coffee intake0.807No573267.9636768.47Yes270232.0416931.53Vegetarian0.045No754389.4449492.16Yes89110.56427.84Fat intake score0.0010–34212625.2111020.5235–39204524.2511822.0140–43182221.6010920.3444–60244128.9419937.13Abbreviations: BMI, body mass index; ZMIZ1‐AS1, zinc finger MIZ‐type containing 1 antisense RNA1. Baseline characteristics of 536 participants with colorectal polyps and 8434 polyp‐free participants. Abbreviations: BMI, body mass index; ZMIZ1‐AS1, zinc finger MIZ‐type containing 1 antisense RNA1. Regarding hypertension, the control group included 1951 (23.13%) hypertensive and 6483 (76.87%) non‐hypertensive individuals, whereas the case group included 203 (37.87%) hypertensive and 333 (62.13%) non‐hypertensive individuals. Table 2 illustrates the relationship between the rs704017 genotype, hypertension, and other variables associated with CPs. There was an increased odds of CPs among individuals with the AG + GG genotype compared to those with the AA genotype (reference group): aOR = 1.39 (95% CI: 1.16–1.66). Hypertension alone was not independently associated with CP aOR = 1.16 (95% CI: 0.94–1.44). However, a significant interaction between hypertension and the rs704017 genotype was observed (p = 0.029). Age was also significantly associated with the odds of CP. Compared with participants aged 30–39 years, the aOR (95% CI) for individuals aged 40–49 was 1.84 (1.27–2.66), 3.11 (2.17–4.45) for those aged 50–59, and 3.72 (2.53–5.47) for those aged 60–69. Additionally, hyperlipidemia and a family history of colorectal cancer (CRC) were significantly associated with an increased odds of CPs, with aOR (95% CI) of 1.41 (1.14–1.73) for hyperlipidemia and 2.31 (1.76–3.03) for a family history of CRC. Notably, tea consumption was significantly associated with a lower odds of CPs (aOR = 0.81, 95% CI: 0.66–0.98). TABLE 2Association between the ZMIZ1‐AS1 rs704017 variant and the odds of colorectal polyps.VariablesOR95% CI p ZMIZ1‐AS1 rs704017 (ref.: AA)AG + GG1.391.16–1.66< 0.001Hypertension (ref.: No)Yes1.160.94–1.440.170Sex (ref.: Female)Male1.230.99–1.530.066Age (ref.: 30–39 years)40–491.841.27–2.660.00150–593.112.17–4.45< 0.00160–703.722.53–5.47< 0.001Educational level (ref.: Elementary school and below)Junior and senior high school0.930.68–1.280.667University and above1.070.77–1.480.692Cigarette smoking (ref.: No)Yes0.850.66–1.100.218Alcohol consumption (ref.: No)Yes0.950.69–1.300.740Physical activity (ref.: No)Yes1.110.92–1.350.272BMI (ref.: 18.5 ≤ BMI < 24)BMI < 18.50.750.36–1.560.44624 ≤ BMI < 271.080.88–1.330.445BMI ≥ 270.900.69–1.160.402Diabetes (ref.: No)Yes1.170.92–1.470.201Hyperlipidemia (ref.: No)Yes1.411.14–1.730.001Family history of colorectal cancer (ref.: No)Yes2.311.76–3.03< 0.001Tea consumption (ref.: No)Yes0.810.66–0.980.032Coffee consumption (ref.: No)Yes1.090.90–1.330.379Vegetarian (ref.: No)Yes0.830.60–1.150.266Fat intake score (ref.: 0–34)35–391.020.78–1.340.88040–430.960.73–1.280.79744–601.150.89–1.500.282 Note: The p‐value for the interaction (*) between rs704017 and hypertension = 0.029.Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio; ref, reference; ZMIZ1‐AS1, zinc finger MIZ‐type containing 1 antisense RNA1. Association between the ZMIZ1‐AS1 rs704017 variant and the odds of colorectal polyps. Note: The p‐value for the interaction (*) between rs704017 and hypertension = 0.029. Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio; ref, reference; ZMIZ1‐AS1, zinc finger MIZ‐type containing 1 antisense RNA1. Table 3 details the ORs for colorectal polyps according to the combined ZMIZ1‐AS1 rs704017 genotype and hypertension status. Compared to the reference group (non‐hypertensive individuals with the AA genotype), the odds of CPs was significantly higher in non‐hypertensive individuals with AG + GG (aOR = 95% CI: 1.62; 1.29–2.03), hypertensive patients with AA (aOR; 95% CI: 1.45; 1.09–1.95), and hypertensive patients with AG + GG (aOR = 1.56, 95% CI: 1.16–2.08). TABLE 3Odds ratios for colorectal polyps according to the combined ZMIZ1‐AS1 rs704017 genotype and hypertension status.VariablesOR95% CI p‐valueHypertension and rs704017 genotypes ref.: No hypertension and AANo hypertension and AG + GG1.621.29–2.03< 0.001Hypertension and AA1.451.09–1.950.012Hypertension and AG + GG1.561.16–2.080.003Sex (ref.: Female)Male1.230.99–1.530.061Age (ref.: 30–39 years)40–491.841.27–2.660.00150–593.112.17–4.46< 0.00160–703.722.53–5.46< 0.001 Note: Adjusted for educational level, smoking, alcohol consumption, physical activity, BMI, diabetes, hyperlipidemia, family history of colorectal cancer, tea and coffee consumption, vegetarian diet, and fat intake.Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio, ref, reference; ZMIZ1‐AS1, zinc finger MIZ‐type containing 1 antisense RNA1. Odds ratios for colorectal polyps according to the combined ZMIZ1‐AS1 rs704017 genotype and hypertension status. Note: Adjusted for educational level, smoking, alcohol consumption, physical activity, BMI, diabetes, hyperlipidemia, family history of colorectal cancer, tea and coffee consumption, vegetarian diet, and fat intake. Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio, ref, reference; ZMIZ1‐AS1, zinc finger MIZ‐type containing 1 antisense RNA1. Discussion In the current study, we combined data from the TWB and the NHIRD and found a significant association between the rs704017 AG + GG genotype and CPs compared with the AA genotype. Initially, hypertension showed no significant association with CPs. However, subgroup analyses that incorporated both rs704017 genotypes and hypertension (using non‐hypertensive individuals with the AA genotype as the reference) revealed higher odds of CPs in the following categories: (1) non‐hypertensive individuals with AG + GG, (2) hypertensive patients with AA, and (3) hypertensive patients with AG + GG. Ethnic disparities in CP prevalence and risk have been documented [20]. Genetic mutations play a critical role in the colorectal adenoma‐carcinoma cascade [12], and the predisposition to adenomas contributes significantly to the genetic basis of CRC [38]. Therefore, tailored interventions for genetically predisposed individuals represent an essential public health approach aimed at individualized management of CRC [7]. Several studies have established a significant association between the ZMIZ1‐AS1 rs704017 variant and CRC. For instance, a large‐scale genetic study involving East Asians identified rs704017 as a risk variant for CRC [32]. Additionally, a genome‐wide association study (GWAS) among Japanese populations identified rs704017 as an efficient lifetime predictor of CRC risk [33]. In Korea, a study linked this SNP to a higher risk of CRC among individuals aged 50 and above [33]. Notably, some studies suggest that CRC and CPs share a common genetic architecture regarding the rs704017 variant. Keiko Hikino and colleagues [17] conducted a multi‐trait analysis of GWAS and identified rs704017 as a colorectal polyp SNP [17]. They also found that colorectal polyps and cancer exhibit overlapping polygenic traits [17]. A case–control study in Spain found significant associations between serrated polyposis syndrome and seven CRC SNPs, including rs704017, with a similar direction of association for both the SNPs and CRC [16]. The rs704017 variant is thought to contribute to colorectal carcinogenesis through its effects on ZMIZ1 gene expression and regulation [32, 39]. The relationship between CPs and hypertension has yielded conflicting results. For instance, a cross‐sectional study in Japan found no significant association between blood pressure and CPs; however, it did report a significant positive association between antihypertensive drug use and CPs [24]. Conversely, a study in Korea found no significant link between hypertension and CPs [36]. In a Taiwanese study [19], hypertension was significantly associated with an increased risk of both non‐advanced and advanced adenomatous polyps, although prehypertension was not linked to advanced adenomatous polyps. In the USA, significant associations were found between hypertension and CPs in both men and women [18, 35]. Similarly, a cross‐sectional study involving Taiwanese participants found a significant positive association between hypertension and advanced adenomatous polyps [23]. Moreover, a retrospective cohort study indicated that hypertension is a key predictor of colorectal adenoma [22], while a case–control study in Israel also found a significant positive relationship between hypertension and CPs [34]. The proposed mechanism is that hypertension and colorectal polyps share the underlying pathophysiology of insulin resistance. In our study, hypertension and CPs were not found to be significantly related. However, incorporating the rs704017 genotypes revealed a significant risk of CPs, implying that genetic profiles might help explain the differences in hypertension risk among individuals. Although independent associations of hypertension, rs704017, and colorectal polyps have been reported, the specific interaction between these factors has been investigated less extensively. Our findings suggest that hypertension and the ZMIZ1‐AS1 rs704017 variant may have additive or synergistic effects on the odds of CPs, potentially through shared mechanisms involving insulin resistance and dysregulation of tumor suppressor pathways [19, 32]. Nevertheless, further research is needed to fully clarify the nature of this interaction and the underlying mechanisms. In our study, age, hyperlipidemia, and family history of CRC were significantly associated with increased odds of CPs, whereas tea consumption was associated with lower odds. A previous study in Taiwan reported similar relationships between CPs and age [19] and hyperlipidemia [40]. Other studies have also corroborated these findings regarding age, family history of CRC, and CPs [18, 20, 26]. Despite the value of these results, several limitations should be acknowledged. First, the study sample was drawn from a specific population, which may limit the generalizability of the findings to other ethnic groups. Second, because this was an observational analysis of linked registry and biobank data, causal inference remains limited by residual confounding. Third, several lifestyle covariates were self‐reported at enrollment, creating potential recall and reporting bias. Finally, the Taiwan‐based sample may limit generalizability to other populations. Despite these limitations, this study meaningfully contributes to understanding the genetic and environmental factors influencing the risk of colorectal polyps (CPs). By examining the interaction between the ZMIZ1‐AS1 rs704017 variant and hypertension among Taiwanese individuals, we provide evidence that certain genetic profiles can increase the odds of CPs, particularly among individuals with hypertension. This highlights the importance of personalized medicine, in which genetic information may guide screening and prevention strategies tailored to individual risk profiles, potentially reducing the incidence of colorectal cancer. Conclusion In summary, the ZMIZ1‐AS1 rs704017 variant was independently associated with higher odds of CPs, whereas hypertension alone was not. However, the interaction between hypertension and the rs704017 variant was significant. Our findings suggest that individuals with the AG + GG genotype are at increased odds of CPs, independent of hypertension. Nonetheless, the presence of the rs704017 variant may further increase the odds of CPs among hypertensive individuals. Therefore, genetic profiles may play an important role in CP management and risk assessment. Author Contributions Oswald Ndi Nfor: conceptualization, methodology, writing – review and editing, writing – original draft. Bei‐Hao Shiu: conceptualization, writing – original draft, methodology. Yung‐Po Liaw: conceptualization, formal analysis, methodology, supervision, funding acquisition, writing – review and editing, software. Hsu‐Chih Huang: conceptualization, data curation, methodology, writing – original draft. Shu‐Yi Hsu: conceptualization, methodology, formal analysis, writing – review and editing. Funding The National Science and Technology Council (NSTC) (NSTC115‐2121‐M‐039‐002, NSTC114‐2121‐M‐039‐002, NSTC113‐2121‐M‐040‐001, NSTC112‐2121‐M‐040‐002, NSTC115‐2811‐M‐039‐002, NSTC114‐2811‐M‐040‐001, NSTC113‐2811‐M‐040‐001 and NSTC112‐2811‐M‐040‐001). Ethics Statement The Institutional Review Board of the Chung Shan Medical University Hospital granted ethical approval for this study (IRB: CS1‐20009). Consent Participants signed an informed consent letter before enrolling in the Taiwan Biobank project. All methods were carried out in accordance with relevant guidelines and regulations. Conflicts of Interest The authors declare no conflicts of interest.