Application of fecal DNA multi-gene methylation testing in colorectal cancer screening in Haikou City

Abstract

BACKGROUND

Colorectal cancer (CRC) remains a leading cause of cancer-related mortality worldwide, with incidence and mortality rates continuing to rise. Early diagnosis and prevention are essential for reducing the disease burden. Fecal DNA methylation testing is a non-invasive method for CRC screening.

AIM

To evaluate the clinical utility of a fecal multi-gene methylation assay targeting three genes, syndecan 2 (SDC2), alcohol dehydrogenase, iron containing 1 (ADHFE1), and protein phosphatase 2 regulatory subunit B’ gamma (PPP2R5C), for early CRC screening in Haikou City and to identify risk factors associated with positive screening results.

METHODS

This study was conducted across five administrative districts of Haikou City, Hainan Province, from July 1, 2025 to May 1, 2026. Enrolled participants completed a questionnaire on CRC-related risk factors and provided fecal samples. The promoter methylation status of SDC2, ADHFE1, and PPP2R5C in exfoliated cells from stool samples was assessed using quantitative real-time polymerase chain reaction. With colonoscopy and histopathology as the gold standard, the positive predictive value (PPV) of the multi-gene methylation test was calculated for CRC, advanced adenoma, non-advanced adenoma, and polyps. Univariate χ² tests and binary logistic regression models were used to identify factors associated with positive screening results and independent risk factors.

RESULTS

A total of 4050 participants were included, with an initial screening positivity rate of 5.16% (209/4050). Among those with positive results, 128 (61.2%) underwent colonoscopy, which identified 4 cases of CRC (3.1%), 19 cases of advanced adenoma (14.8%), 86 cases of non-advanced adenoma or polyps (67.2%), and 19 cases with normal colonoscopy findings (14.8%). The PPV of the test for any colorectal lesion was 85.2% (109/128). Univariate and multivariate logistic regression analyses identified increasing age [odds ratio (OR) = 1.672, 95% confidence interval (CI): 1.245-2.245], Helicobacter pylori infection (OR = 1.777, 95%CI: 1.329-2.376), a long-term high-salt diet (OR = 2.596, 95%CI: 1.829–3.685), and consumption of pickled foods (OR = 1.827, 95%CI: 1.299-2.570) as independent risk factors for positive screening results (all P < 0.05).

CONCLUSION

The fecal multi-gene methylation test based on SDC2, ADHFE1, and PPP2R5C showed a high PPV (85.2%) for CRC screening in Hainan Province and effectively enriched for colorectal lesions. This non-invasive and convenient testing method may be a feasible strategy for large-scale CRC screening in Hainan Province. Future efforts should focus on improving compliance with follow-up colonoscopy.

Key Words: Colorectal cancer; Fecal DNA methylation; Syndecan 2; Alcohol dehydrogenase, iron containing 1; Protein phosphatase 2 regulatory subunit B’ gamma; Positive predictive value; Early screening

Core Tip: This cross-sectional study evaluated fecal DNA multi-gene methylation testing (syndecan 2/alcohol dehydrogenase, iron containing 1/protein phosphatase 2 regulatory subunit B’ gamma) in 4050 participants in Haikou City, Hainan Province, China. The test demonstrated a high positive predictive value (85.2%) for detecting colorectal lesions and identified independent risk factors associated with positive screening results, such as older age, Helicobacter pylori infection, high-salt diet, and pickled food consumption. These findings support its potential as a non-invasive colorectal cancer screening tool in Hainan Province.

INTRODUCTION

Colorectal cancer (CRC) is the third most common type of cancer globally and the second leading cause of death due to cancer[1]. In 2022, there were roughly 1.7 million cases and 900000 deaths related to CRC across the globe. Patients of early stages of CRC may present with non-specific signs and symptoms and are mostly diagnosed at advanced stages of CRC, making it difficult to manage. Development of precancerous conditions like polyps and adenomas to invasive CRC takes about 10 years or even more. Early diagnosis and timely management of CRC increases survival rates up to 90%[2,3]. Numerous studies have shown that there is clinical relevance in CRC screening. The findings from a screening project in the UK indicated a 24% decrease in CRC incidence and a 25% decrease in mortality in CRC patients undergoing screening[4]. In another prospective cohort study conducted in Sweden, the findings suggested that CRC screening was associated with a 14% decrease in mortality[5]. Overall, all these findings support the relevance of CRC screening on a population basis.

At present, CRC screening still faces many practical challenges. The main screening methods include colonoscopy, fecal occult blood testing, fecal immunochemical testing (FIT), and fecal genomic methylation testing[6]. Colonoscopy is recognized as the gold standard for CRC diagnosis because of its high diagnostic accuracy and its ability to directly visualize intestinal lesions and allow biopsy or lesion removal under endoscopic guidance[7]. However, colonoscopy requires strict bowel preparation and is an invasive procedure associated with risks such as bleeding and perforation, which limits its use in large-scale population screening to some extent[8]. FIT is widely used in CRC screening. This method is based on the antigen-antibody reaction of hemoglobin and is not affected by factors such as diet or medications. Although FIT has advantages in large-scale screening, its sensitivity for advanced adenomas and CRC remains suboptimal[9]. Therefore, a non-invasive and convenient screening method is needed to improve participation in early CRC screening and increase lesion detection rates.

In recent years, fecal multi-gene methylation testing has become an active research area in CRC screening because it is non-invasive, convenient, and highly sensitive[10]. DNA methylation is an important molecular event in the occurrence and development of CRC. Studies have shown that methylation levels of relevant genes in exfoliated cells from patients with CRC are significantly higher than those in healthy individuals. Therefore, CRC and precancerous lesions can be identified at an early stage by detecting the methylation status of specific genes in feces. Fecal samples from patients with CRC often contain exfoliated tumor cells derived from the natural shedding of intestinal tumor tissues[11]. Compared with normal colorectal mucosa, CRC tissue releases more exfoliated cells into the intestinal lumen, and these cells often carry the same genetic and epigenetic alterations as the primary tumor. Therefore, tumor cell DNA analysis in fecal specimens using certain techniques can provide valuable information for the early diagnosis of CRC. The basic premise behind fecal DNA methylation testing is to detect abnormal methylation of certain genes. Studies show that fecal DNA testing has greater sensitivity in detecting advanced adenoma and CRC than conventional non-invasive screening tests. Compared to traditional screening techniques, fecal multi-gene methylation test is non-invasive, easy to perform, and offers high sensitivity and specificity without any invasiveness[12]. Furthermore, it does not cause any pain to the patient. In 2014, Cologuard was the first Food and Drug Administration-approved fecal multi-target CRC screening test in the United States. This test includes markers such as hemoglobin, mutation in Kirsten rat sarcoma viral oncogene homolog gene, and methylation of bone morphogenetic protein 3 and N-myc downstream regulated gene 4. Important studies show that the sensitivity of Cologuard test in detecting CRC is above 90%, while sensitivity in detecting precancerous lesions is significantly higher than that of conventional fecal occult blood tests[13].

Currently, methylation analysis of multiple genes in the stool is considered an emerging non-invasive test for CRC screening. In terms of these tests, the joint identification of syndecan 2 (SDC2), alcohol dehydrogenase, iron containing 1 (ADHFE1), and protein phosphatase 2 regulatory subunit B’ gamma (PPP2R5C) received much attention owing to their high sensitivity and specificity observed in previous studies. However, it should be noted that the effectiveness of using the combination of these genes in the real-life screening process can depend on the particular epidemiologic features of a region, genetic background of the population, and other relevant factors. Therefore, this research was carried out in Haikou city in Hainan Province in order to determine the PPV of this methylation test.

MATERIALS AND METHODS

Research design and procedure

This research study was performed on five districts of Haikou City in Hainan province from July 1, 2025 to May 1, 2026. This protocol was approved by the Ethics Committee of the Second Affiliated Hospital of Hainan Medical University (Approval No. 2026-K49-01). Written informed consent was obtained from all the participants according to the committee-adopted consent form.

Inclusion criteria: (1) Subjects from Haikou City who were between 40 years old and 74 years old, voluntary participants who provided written informed consent irrespective of gender; and (2) Absence of any history of CRC, ulcerative colitis, or any other forms of malignancies.

Exclusion criteria: (1) Age < 40 years or > 74 years; (2) Cognitive impairment; (3) History of CRC or other malignant tumors, or previous or current chemotherapy or immunotherapy; (4) Inability to provide qualified stool samples; and (5) Inability or unwillingness to provide written informed consent.

Questionnaire

Data were collected using structured questionnaires. Researchers who had received standardized training interviewed the participants and completed the questionnaires item by item. The questionnaire included the following information: (1) Demographic characteristics, including age and sex; (2) Lifestyle and dietary habits, including smoking history, drinking history, high-salt diet, intake of pickled foods, excessive red meat intake, and insufficient dietary fiber intake; (3) Personal medical history, including Helicobacter pylori infection, hypertension, and diabetes; and (4) Family history of CRC. All questionnaire data were independently entered by two researchers and cross-checked to ensure data accuracy and reliability.

Smoking was defined as smoking at least one cigarette per day or any smoking behavior within the past year. Drinking was defined as alcohol consumption of ≥ 100 g per week during the past year. A high-salt diet was defined as daily salt intake of ≥ 5 g. Excessive red meat intake was defined as red meat consumption at least three times per week, which was classified as high-frequency intake[14]. Low dietary fiber intake was defined as daily dietary fiber intake of < 18 g for women and < 22 g for men[15]. Frequent consumption of pickled foods was defined as consumption at least four times per week. Hypertension was defined as self-reported physician-diagnosed hypertension or current use of antihypertensive medication. Diabetes was defined as self-reported physician-diagnosed diabetes or current use of hypoglycemic medication. Helicobacter pylori infection was defined as a diagnosis based on a ¹³C or ¹⁴C urea breath test or gastroscopic biopsy. Family history of CRC was defined as at least one first-degree relative with CRC.

Collection of stool samples and subsequent procedures

A total of 4050 participants were included in this study, and stool samples were collected from all participants. Samples were placed in collection tubes containing 5 mL of preservation solution and transported to the designated testing laboratory in Haikou City within 48 hours. Before sample collection, participants completed electronic informed consent through a WeChat mini-program. Stool gene methylation test results were provided to participants through the mini-program 10-14 days after sampling. For participants with positive test results, the researchers informed them individually by phone or text message and recommended colonoscopy at Hainan Cancer Hospital or Hainan 928 Hospital. If participants or their family members refused colonoscopy, the researchers conducted up to five follow-up communications to encourage completion of the examination at a designated hospital.

Stool DNA methylation detection and analysis

All collected stool samples were sent to the Haikou Testing Laboratory of Shenzhen BGI Corporation within 48 hours for genomic DNA extraction and methylation analysis. The BGI Huangchang^® Stool DNA Methylation Assay Kit was used to detect the methylation status of three genes: SDC2, ADHFE1, and PPP2R5C.

The participants took samples of stool in specific collection tubes following standard operating procedures and sent the samples within 48 hours after defecation for laboratory tests. The samples received in the lab were taken 5 g of stool and transferred to 50 mL collection tubes, each tube filled with 15 mL of preservation solution and kept at -80 °C. For the analysis, samples were first defrosted, homogenized for 1 minute with oscillator and centrifuged for 15 minutes at 12000 × g. Genomic DNA extraction was performed by the TIANamp Stool DNA Extraction Kit and then the DNA underwent bisulfite conversion by the DNA bisulfite conversion kit in order to perform the methylation-specific test. Purification and conversion of DNA was measured by the NanoDrop 2000 spectrophotometer and the samples stored at -20 °C until further use. Quantitative analysis was performed by the SLAN-96S real-time fluorescence quantitative polymerase chain reaction instrument and ADHFE1, SDC2, and PPP2R5C methylation assay kits provided by BGI with methylation-specific polymerase chain reaction. The delta crossing point (Cp) value (Cp target gene - Cp glyceraldehyde-3-phosphate dehydrogenase) was used to evaluate the methylation level of each gene, with a lower delta Cp value indicating a higher degree of methylation. Based on the preset clinical threshold, 26 methylation sites across the three genes were comprehensively interpreted. Samples with more than five methylation sites were considered positive; otherwise, they were considered negative.

Colonoscopy and histopathological results

Participants with positive test results were advised to undergo colonoscopy at Hainan Cancer Hospital or Hainan 928 Hospital, performed by experienced endoscopists. Abnormal lesions, such as polyps, identified during colonoscopy were removed and submitted for pathological examination. Pathological results were independently interpreted by two senior pathologists. For participants with positive pathological findings, clinicians formulated individualized treatment plans according to the patient’s condition. For participants with negative colonoscopy findings, repeat colonoscopy every 1-2 years was recommended. Histopathological diagnoses were classified as CRC, advanced adenoma, non-advanced adenoma, or polyps. Advanced adenoma was defined as the presence of any of the following features: (1) Diameter ≥ 10 mm; (2) High-grade intraepithelial neoplasia or high-grade dysplasia; or (3) Villous components, including villous adenoma or tubulovillous adenoma. Non-advanced adenoma was defined as a lesion with a diameter < 10 mm, low-grade intraepithelial neoplasia or low-grade dysplasia, and a pathological type of tubular adenoma without villous components. Polyps included hyperplastic polyps, inflammatory polyps, and other non-neoplastic lesions. The study flowchart is shown in Figure 1.

Figure 1  Study flowchart.

Statistical analysis

Data were analyzed using SPSS 27.0 statistical software (IBM Corp., Armonk, NY, United States). Categorical variables were described as n (%), and between-group comparisons were performed using the χ² test or Fisher’s exact test. Continuous variables were expressed as the mean ± SD. Binary logistic regression analysis was used to identify independent risk factors associated with positive screening results. The two-sided significance level was set at α = 0.05, and P < 0.05 was considered statistically significant.

RESULTS

Screening results of fecal gene methylation

In this study, the kit for assessing the methylation levels of genes SDC2, ADHFE1, and PPP2R5C using the fecal multi-gene methylation test kit was manufactured by Hua Chang Kang Company. 4180 participants from Haikou City were included in this study, but 130 were eliminated owing to having an unqualified fecal sample. Therefore, the final number of participants was 4050. The screening outcome showed that there were 209 positive cases, which accounted for 5.16% positivity rate.

Among those tested positive for the disease, 122 participants were male and accounted for 7.04% (122/1,733) and 87 participants were female and accounted for 3.75% (87/2317). There was a higher positivity rate among the male participants compared to the female ones, and the difference was significant (χ² = 21.86, P < 0.01). The average age of the participants was 54.22 ± 9.26. Age-stratified analysis showed that the positivity rate increased with age. The positivity rates were 3.92% (57/1453) in the 40 years group to 49 years group, 4.84% (66/1363) in the 50 years group to 59 years group, 5.63% (55/977) in the 60 years group to 69 years group, and 12.1% (31/257) in the 70-74-year group, with the highest rate observed in the 70 years group to 74 years group. The differences among age groups were statistically significant (χ² = 30.282, P < 0.001). The trend χ² test further confirmed that the positivity rate increased significantly with age (χ² trend = 19.901, P < 0.01; Table 1).

Table 1 Trend χ² test for participant age groups.

Age group	Total cases	Positive cases	Positive rate
40-49 years	1453	57	3.92%
50-59 years	1363	66	4.84%
60-69 years	977	55	5.63%
70-74 years	257	31	12.1%
Total	4050	209	5.16%

Trend χ² test: χ² trend = 19.901, P < 0.01.

Analysis of positive fecal methylation results and related factors

To explore factors associated with positive fecal gene methylation results, univariate analysis was performed using the χ² test for demographic characteristics, including age and sex; lifestyle factors, including smoking, drinking, high-salt diet, and pickled food intake; and clinical factors, including Helicobacter pylori infection, family history of CRC, history of diabetes, and history of hypertension. The results showed that sex, age, Helicobacter pylori infection, smoking, drinking, pickled food intake, high-salt diet, and family history of CRC were significantly associated with the test positivity rate (P < 0.05; Table 2).

Table 2 Univariate analysis of factors associated with positive fecal gene methylation test results in Haikou City (n = 4050).

Variable	Participants, n	Positive cases, n	χ2 value	P value
Gender			21.860	< 0.01
Male	1733	122
Female	2317	87
Age (years)			30.282	< 0.01
40-49	1453	57
50-59	1363	66
60-69	977	55
70-74	257	31
H. pylori infection			13.672	< 0.01
Yes	1382	96
No	2668	113
Preserved food intake			13.31	< 0.01
Frequent	628	51
Infrequent	3422	158
High-salt diet			37.05	< 0.01
Yes	448	50
No	3602	159
Long-term smoking history			8.658	0.03
Yes	589	45
No	3461	164
Excessive alcohol consumption			6.395	0.011
Yes	707	50
No	3343	159
Hypertension			27.538	< 0.01
Yes	597	57
No	3453	152
Diabetes			38.776	< 0.01
Yes	250	34
No	3800	175
Family history			11.826	< 0.01
Yes	384	34
No	3666	175
Excessive consumption of red meat			5.352	0.021
Yes	2146	127
No	1904	82
Consuming too little dietary fiber			12.173	< 0.01
Yes	2380	147
No	1670	62

H. pylori: Helicobacter pylori.

Variables with statistical significance in the univariate analysis were included in the multivariate binary logistic regression model. The results showed that increasing age, Helicobacter pylori infection, long-term high-salt diet, consumption of pickled foods, and family history of CRC were independent risk factors for positive screening results (P < 0.05; Table 3).

Table 3 Binary logistic regression analysis of factors associated with positive fecal gene methylation test results.

Factor	β coefficient	SE	Wald χ2	P value	OR	95%CI
						Lower limit	Upper limit
Smoking	0.518	0.182	8.114	0.004	1.679	1.175	2.398
Drinking alcohol	0.399	0.175	5.221	0.022	1.491	1.058	2.100
Pickled foods	0.603	0.174	11.969	0.001	1.827	1.299	2.570
High-salt diet	0.954	0.179	28.483	0.000	2.596	1.829	3.685
Excessive consumption of red meat	0.312	0.150	4.310	0.038	1.366	1.018	1.834
Insufficient dietary fiber	0.576	0.160	13.008	0.000	1.778	1.301	2.431
Hypertension	0.873	0.169	26.656	0.000	2.393	1.718	3.333
Diabetes	1.254	0.209	35.918	0.000	3.505	2.326	5.282
Family history of colorectal cancer	0.644	0.205	9.862	0.002	1.903	1.274	2.844
HP infection	0.575	0.148	15.063	0.000	1.777	1.329	2.376
Gender	-0.662	0.148	19.865	0.000	0.516	0.386	0.690
Year (≥ 60 years vs < 60 years)	0.514	0.150	11.685	0.001	1.672	1.245	2.245

OR: Odds ratio; CI: Confidence interval.

Analysis of colonoscopy compliance

Among the 209 participants with positive initial screening results, 128 completed colonoscopies, with an overall compliance rate of 61.2%. Stratified by sex, the compliance rate was 70.1% in females and 54.9% in males, and the difference was statistically significant (χ² = 4.941, P = 0.026). Stratified by age, the compliance rate decreased with age, and the difference was statistically significant (χ² = 15.218, P = 0.02; Table 4).

Table 4 Comparison of Colonoscopy Follow-up Rates Among Participants with Positive Fecal Methylation Test Results.

Factor	Positive cases, n	Follow-up cases, n	χ2 value	P value
Gender			4.941	0.026
Female	87	61
Male	122	67
Age group (years)			15.218	0.02
40-49	57	44
50-59	66	44
60-69	55	27
70-74	31	13
Total	209	128

Colonoscopy results

Among the 128 screening-positive participants who underwent colonoscopy, 109 intestinal lesions were detected, including 4 cases of CRC, 19 cases of advanced adenoma, and 86 cases of non-advanced adenoma or polyps. No obvious abnormalities were found on colonoscopy in 19 cases. The PPV of fecal multi-gene methylation testing was 3.1% (4/128) for CRC, 14.8% (19/128) for advanced adenoma, and 67.2% (86/128) for non-advanced adenoma and polyps. The overall PPV of this test for all intestinal lesions was 85.2% (109/128).

To assess potential bias introduced by participants who did not undergo colonoscopy, sensitivity analyses were performed under three scenarios for the 81 participants with positive initial screening results who did not complete colonoscopy. In the most conservative scenario, assuming that none of these participants had intestinal lesions, the PPV of the test for all intestinal lesions was 52.2%. In the most optimistic scenario, assuming that all of these participants had intestinal lesions, the PPV was 90.9%. Even under the most conservative assumption, the PPV of this test remained relatively high.

Age-stratified colonoscopy results showed that, in the 40-49-year group, 2 cases of CRC, 8 cases of advanced adenoma, 29 cases of non-advanced adenoma or polyps, and 5 normal cases were detected. In the 50 years group to 59 years group, the corresponding numbers were 1 case, 2 cases, 32 cases, and 9 cases; in the 60 years group to 69 years group, 1 case, 6 cases, 17 cases, and 3 cases; and in the 70 years group to 74 years group, 0 case, 3 cases, 8 cases, and 2 cases, respectively. The overall lesion detection rates in these age groups were 88.6% (39/44), 79.5% (35/44), 88.9% (24/27), and 84.6% (11/13), respectively. Stratified by sex, the overall lesion detection rates were 83.6% (56/67) in males and 86.9% (53/61) in females (Table 5).

Table 5 Distribution of colonoscopy findings by age and sex.

Group	Number of cases	CRC	AA	NAA/polyps	Normal	Detection rate of lesions
Age group
40-49 years old	44	2	8	29	5	88.6%
50-59 years old	44	1	2	32	9	79.5%
60-69 years old	27	1	6	17	3	88.9%
70-74 years old	13	0	3	8	2	84.6%
Gender
Man	67	3	11	42	11	83.6%
Woman	61	1	8	44	8	86.9%

CRC: Colorectal cancer; AA: Advanced adenoma; NAA: Non-advanced adenoma.

DISCUSSION

This study provides a real-life application of the fecal multi-gene methylation test that involved the use of SDC2, ADHFE1, and PPP2R5C genes. The rate of positivity for the test was 5.16%. In terms of PPV, the detection of any intestinal lesion had a PPV of 85.2%, while the PPV for CRC was 3.1%. These findings have significant importance from a public health perspective with respect to CRC screening programs in Hainan City.

The key results of the study mentioned above agree with those found in similar studies performed around the world. The real-life study carried out in China, which involved 113209 individuals, confirmed that fecal SDC2 methylation had beneficial PPVs in case of all kinds of colorectal lesions, including CRC, advanced adenomas, and non-advanced lesions[16]. Though there were some differences in terms of specific values because of differences in the studied population, many studies proved that fecal DNA methylation-based detection methods have the advantage in selection of high-risk groups of patients. There was another study carried out in Dongguan City, Guangdong Province (China), involving 10578 asymptomatic residents[17]. In particular, compared to previous screening studies performed in Hainan Province, the multigene methylation test used in this research proved its significant role in lesion detection. As was established in earlier FIT research in Hainan City, the prevalence of positivity in the average risk group was 4.4%, while the PPV for CRC was 3.6%[18]. While it is not possible to make a direct statistical comparison because of different detection methods, in this study, the percentage of detected lesions among initially positive subjects reached up to 85.2%. This finding suggests that multi-gene methylation testing may help reduce unnecessary use of colonoscopy resources.

The follow-up compliance rate among screening-positive participants in this study was 61.2%, and compliance decreased with age (P < 0.01). This rate is comparable to that reported in similar studies. A real-world study involving 113209 participants reported a colonoscopy compliance rate of 36.4%, which increased to 62.7% in organized screening programs[16]. The compliance rate in the Dongguan City study was 63.8%[17]. A large-scale study in the United States involving 362646 patients showed that colonoscopy compliance among participants with positive multi-target stool DNA test results was 77.1%, which was significantly higher than that among those with positive FIT/fecal occult blood test results (45.1%)[19]. To further assess whether incomplete follow-up introduced selection bias, baseline characteristics were compared between participants who completed colonoscopy (n = 128) and those who did not (n = 81). Among non-completers, the proportions of males (χ² = 4.941, P = 0.026) and older participants (χ² = 15.218, P = 0.02) were significantly higher. Because male sex and advanced age are known risk factors for colorectal tumors, the higher proportion of these high-risk groups among non-completers suggests that the observed PPV of 85.2% may underestimate, rather than overestimate, the true lesion detection rate. Sensitivity analysis further supported this inference. Even under the most conservative assumption, in which all non-completers were assumed to have no lesions, the PPV of this test remained 52.2%. The potential impact of incomplete follow-up on the main conclusion of this study was limited, but further verification in prospective studies is needed. In organized screening programs, greater attention should be given to follow-up mobilization among older adults and males to reduce loss-to-follow-up bias and improve the overall effectiveness of screening.

This study found that increasing age, Helicobacter pylori infection, long-term high-salt diet, consumption of pickled foods, excessive red meat intake, insufficient dietary fiber intake, and family history of CRC were independent risk factors for positive fecal multi-gene methylation results (all P < 0.05). The association between Helicobacter pylori infection and positive screening results is consistent with previous meta-analyses. Li et al[20] showed that Helicobacter pylori infection increased the risk of CRC by 59% and the risk of adenoma by 47%. Animal experiments have suggested that Helicobacter pylori infection may promote carcinogenesis by activating the signal transducer and activator of transcription 3 signaling pathway in intestinal epithelial cells[21]. Therefore, eradication of Helicobacter pylori infection may be valuable for CRC prevention.

High-salt diet and consumption of pickled foods were dietary risk factors identified in this study. The concentration of nitrites in pickled foods is higher, and they can interact with amines in the stomach or intestines to produce nitrosamines. The World Health Organization has classified nitrosamines as carcinogenic substances[22]. Higher levels of fecal nitrosamines have been found among people with CRC compared to people without the disease, and these results have been supported in animals as well[23,24]. Moreover, there is evidence that a salt-rich diet, along with being associated with an increased risk for gastric cancer, might affect CRC due to altering the intestinal microenvironment[25,26]. Long-term intake of a high-sodium diet can disturb the balance of gut microbiota, leading to a reduction in beneficial microbes like Lactobacillus and proliferation of pathogenic organisms. This disturbance may weaken the intestinal barrier function and reduce the production of metabolites that are important for the maintenance of the intestines, hence causing and maintaining inflammation in the gut and development of CRC[26].

The consumption of red meat in excess is a major dietary risk factor for CRC. In a study involving 1162 people who screened positive using FIT for CRC, it was found that an increased intake of red and processed meat was associated with advanced colorectal lesions, with statistical significance. This means that every additional 100 g consumed daily increased the risk of advanced lesions by 32% [odds ratio (OR) = 1.32, 95% confidence interval (CI): 1.09-1.60][27]. Moreover, in a study conducted on the European Prospective Investigation into Cancer and Nutrition-Norfolk population, it was found that each 19 g increase in daily intake of processed meat, which is equal to one standard deviation, increased the risk of CRC associated with adenomatous polyposis coli gene GC-to-AT mutations by 68% (OR = 1.68, 95%CI: 1.03-2.75). This suggests that N-nitroso compounds in processed meat may contribute to colorectal carcinogenesis by inducing adenomatous polyposis coli gene mutations[28]. In the present study population, the risk of screening positivity was increased among participants with excessive red meat intake (OR = 1.366, 95%CI: 1.018-1.834), consistent with the above findings.

Adequate dietary fiber intake is considered protective against CRC. Dietary fiber may exert protective effects through several mechanisms. First, fermentable fibers are metabolized by intestinal bacteria to produce short-chain fatty acids, such as butyric acid, which is the preferred energy source for colonocytes and can inhibit histone deacetylase activity, regulate gene expression, and suppress tumor development[29]. Second, fiber can increase fecal volume, dilute carcinogen concentrations, and shorten intestinal transit time[30]. Studies have shown a clear dose-response relationship between dietary fiber intake and CRC risk[31]. In this study, participants with insufficient dietary fiber intake had an increased risk of screening positivity (OR = 1.778, 95%CI: 1.301-2.431), suggesting that increasing dietary fiber intake may be valuable for CRC prevention.

This study found that hypertension was an independent risk factor for positive fecal multi-gene methylation testing (OR = 2.393, 95%CI: 1.718-3.333), consistent with the findings of a meta-analysis by Xuan et al[32]. That meta-analysis included 25 observational studies involving 1.95 million participants and showed a significant positive association between hypertension and CRC risk (pooled relative risk = 1.15, 95%CI: 1.08-1.23).

This study also found that diabetes was an independent risk factor for positive fecal multi-gene methylation testing (OR = 3.505, 95%CI: 2.326-5.282). A classic meta-analysis by Sun and Yu[33], based on 39 studies, confirmed that diabetes increased the risk of CRC by 29% (relative risk = 1.29, 95%CI: 1.23-1.35). The main mechanism by which diabetes promotes colorectal carcinogenesis involves hyperinsulinemia associated with insulin resistance. Zhou et al[34] demonstrated through polygenic risk score analysis that the hyperinsulinemic phenotype linked to insulin resistance is a key driver of CRC risk, with a more pronounced effect in obese individuals (OR = 1.75). In addition, Mahboob et al[35] systematically described the mediating role of gut microbiota, showing that diabetes-related dysbiosis and metabolic disturbances may jointly create a pro-carcinogenic microenvironment.

This study found that the risk of positive fecal DNA methylation testing was lower in females than in males (OR = 0.516, 95%CI: 0.386-0.690), consistent with the report by Martinez et al[36] based on a French population (OR for males = 1.54). These sex differences may be driven by biological factors, such as sex hormones, as well as behavioral factors, including screening participation and access to treatment[37]. Increasing age and family history of CRC are recognized risk factors for CRC[38,39]. In this study, the positivity rate increased linearly with age (χ² trend = 19.901, P < 0.01) and reached 12.1% in the 70 years group to 74 years group. Having a first-degree relative with CRC significantly increases an individual’s risk of developing the disease. Regular screening using colonoscopy from the age of 40 years is advocated for those having a family history of CRC. All the results above demonstrate that public health initiatives should encourage people to consume a healthy diet by consuming fewer amounts of red meat, pickles, and salted foods and eating more dietary fibers, fruits, and vegetables. Additionally, it is important to improve screening management of people who are infected with Helicobacter pylori and those with a family history of CRC.

There are certain limitations with this study. One is that due to the small total number of subjects as a result of the limited number of people within Hainan Province and because of the higher number of females than males, there might be selection bias present. Second, colonoscopy follow-up compliance among participants with positive initial screening results was suboptimal, and some positive participants did not complete gold-standard confirmation, which may have affected the accurate assessment of detection performance. Third, this study had a cross-sectional design and could not establish causal relationships between risk factors and positive screening results. Fourth, colonoscopy follow-up was not performed in the double-negative group, which may have led to underestimation of the false-negative rate of the test. These limitations suggest that large-sample prospective cohort studies are needed to further verify the findings.

CONCLUSION

This study showed that fecal multi-gene methylation testing had a high PPV for overall intestinal lesions among participants in Haikou City, Hainan Province (China). This method showed promising lesion enrichment performance among screen-positive participants, but prospective studies with complete colonoscopy verification are needed to further confirm its diagnostic accuracy. Public health education on CRC screening and colonoscopy should be strengthened to improve the willingness of screen-positive individuals to undergo colonoscopy. Improved dietary patterns, eradication of Helicobacter pylori infection, and regular colonoscopy may contribute to better screening outcomes. Targeted intervention measures may help reduce the incidence of CRC.