Dietary Risk Factors and Odds of Colorectal Adenoma in Malaysia: A Case Control Study

Abstract Current evidence suggests that dietary and lifestyle factors may play an important role in colorectal cancer risk but there are only a few studies that investigated their relationship with colorectal adenomas (CRA), the precursors for colorectal cancer. A case-control study was conducted to determine the relationship between dietary and lifestyle factors associated with CRA risk among 125 subjects with CRA and 150 subjects without CRA at Hospital Canselor Tuanku Muhriz UKM (HCTM), Malaysia. We used dietary history questionnaire (DHQ) and International Physical Activity Questionnaire-Short Form (IPAQ) to estimate the diet and physical activity. The findings of this study showed that male gender [OR = 2.71 (95% CI= 1.01–7.27)], smoking [OR = 6.39 (95% CI= 1.04–39.30)], family history of cancer [OR = 6.39 (95% CI= 1.04–39.30)], high body fat percentage [OR = 1.25 (95% CI= 1.04–1.51)], high calorie and fat intake [OR = 1.03 (95% CI= 1.01–1.06)], [OR = 1.01 (95% CI= 0.95–1.09)] and red meat intake more than 100 g per day [OR = 1.02 (95% CI= 1.01–1.04)] increased CRA risk. High fiber [OR = 0.78 (95% CI= 0.64–0.95)] and calcium intake [OR = 0.78 (95% CI= 0.98–1.00)] was found to decrease CRA risk. Some of these modifiable risk factors could be advocated as lifestyle interventions to reduce risk of CRA.


Introduction
Colorectal cancer is one of the most common forms of gastrointestinal cancer around the world and its incidence has been increasing in Asian countries over the past few years (1). In Malaysia, colorectal cancer is the second most diagnosed cancer in males and the third most diagnosed cancer in females (2). Most of the patients who are aged 50 years and above were diagnosed with colorectal cancer at a very late stage because it is asymptomatic during early stages (3,4). Most colorectal cancer evolves from the progression of adenoma from small to large polyps and then to dysplasia and carcinoma. The duration for the progression of colorectal adenomas into carcinoma takes at least 10 years (5).
Colorectal adenoma can be categorized into two groups: conventional adenomas and sessile serrated polyps. In general, both are recognized as precursors of colorectal cancer (6). Ninety five percent of colorectal cancer arises from neoplastic adenomatous polyp. However, 40% of the cases are diagnosed in patients aged 60 and above (7,8). Early detection and colonoscopic removal of these precancerous polyps remain as the most effective procedures in reducing incidence and mortality rate of colorectal cancer (1). However, it is estimated that the number of cancer cases and cancer-related mortality rate will increase in the following decades due to population growth and aging, as well as an increase in the prevalence of existing risk factors such as diet, smoking, obesity, physical inactivity, and changing reproductive trends associated with urbanization and economic development (9). Dietary pattern also has increasingly evident to be one of the important risk factor for early onset of colorectal cancer (10)(11)(12)(13). Hence, it is important to take early preventive measures to reduce the number of colorectal cancer case.
Colorectal cancer is a multifactorial disease involving environmental and genetic factors. The development of malignant cells from colorectal epithelium is caused by long-term exposure to the interactive effects of these factors causing oncogenic mutations and abnormal cellular proliferation which finally develops into adenoma and carcinoma (14). Unhealthy dietary pattern, physically inactivity, obesity, high intake of alcohol and smoking habit are, among all other environmental and life-style factors, the most dominant in the development of colorectal cancer (15)(16)(17)(18)(19). Most of these dominant factors are recognized to influence systemic inflammation. Immune-related cells may release pro-inflammatory cytokines which causes chronic systemic inflammation to occur and high level of cytokines is associated with incidence of CRC (20). Diet is also known to have a direct association with inflammation as suggested from previous research (21)(22)(23). Dietary Inflammatory Index (DII) was developed to measure the relation between the inflammatory potential of individual diet (24). According to Shivappa et al., the highest quartile score of DII is linked to increased CRC risk by 40% as compared to the group with the lowest or most anti-inflammatory DII scores (25).
Meanwhile, recent systematic reviews and meta-analyses paper by Godos et al. revealed that dietary pattern labeled as 'unhealthy' and characterized by a high consumption of red meat, processed meat, high sugar, salty food and refined carbohydrates were associated with increased risk of colorectal adenomas, whereas dietary patterns considered to be healthier are characterized by high intake fruits, vegetables, whole grains and fishes were associated with a decreased risk of colorectal cancer (26). Keum et al. reported that nutrient-based chemoprevention could also ameliorate CRC risk (27). Supplements also are related to chemoprevention by modulating inflammatory status or another related pathway as described previously (28)(29)(30).
Numerous studies have reported on the association between dietary factors and colorectal cancer risk (31-33), and evident is now increasing on how dietary factors can relate to risk of colorectal adenoma (CRA), which is considered as a precursor of colorectal cancer (34)(35)(36)(37)(38)(39). Because colorectal cancer is now occurring in younger age groups and its prevention and treatment are much more efficacious when detected early, it is essential to know which risk factors affect CRA incidence (26). This study aimed to identify dietary and lifestyle factors associated with CRA risk in Malaysia using a case-control study method. Although no specific hypotheses were developed, it was expected that risk factors for CRAs seen in other populations such as dietary pattern and body composition would also be associated with CRAs in this Malaysian population.

Study Design and Setting
This case-control study was conducted at Colonoscopy Department, Hospital Canselor Tuanku Muhriz UKM (HCTM). Ethics approval was obtained from the Medical Research Ethics of the Universiti Kebangsaan Malaysia (UKM. PPI. 800-1/1/5/JEP-2019-243). Consent from subjects or their caregivers was obtained prior to data collection.

Sampling and Subjects
We conducted a hospital-based case-control study in Hospital Canselor Tunku Mukhriz in Cheras, Malaysia. CRA cases were patients who had colorectal polyps as revealed in the colonoscopy and with histologically confirmed adenomatous polyp were assigned as they have been referred to the Endoscopy Center. Controls were selected randomly from patients admitted to the same hospitals as cases at the same time and same setting, with non-neoplastic conditions, not afflicted with diet-related chronic diseases assessed by doctors and aged 30-79. Both groups were matched in terms of age (±5 years).

Data Collection
Sociodemographic data, family history with cancer, smoking habit and alcohol consumption were some of the data obtained from interviews via questionnaires. For sociodemographic data we also included their monthly income, we classify them as B40, M40 and T20, respectively. According to the 2019 statistics in Malaysia, our government classified our population into three main groups based on their household income: Bottom (low income-less than RM 4850/ USD1192), Medium (average income -RM 4850/ USD1192-RM 10959/USD2695), and Top (high income-more than RM 10959/USD2695). These three groups had a percentage of 40%, 40% and 20%, respectively, thus creating the terms B40, M40, T20. Anthropometry measurements included body weight, height, body mass index (BMI), waist circumference, hip circumference, waist-hip ratio and fat percentage were performed. According to WHO (2006) guidelines, BMI was calculated using the formula [weight (kg)/height (m 2 )] (40). Dietary intake was also being assessed based on a validated dietary history questionnaire (DHQ) (41).

Dietary Data
A validated version of DHQ by Shahar et al. (2000) was used in this study to determine dietary patterns. Detailed interviews to gather information about the food taken by subjects including the type, amount, method of cooking and frequency were conducted by the dietitian. This includes all foods and beverages taken from waking up in the morning until before going to bed. Subjects were asked in detail so that any snacks or beverages taken in between are also included. In addition, the subjects were also asked about the frequency of having each mealtime, ranging from every day, almost every day (five to six times per week), some time (three to four times per week), seldom (once to twice per week), very seldom (less than once a week) and never. The time, venue and with whom the meal were taken and recorded.
Information gathered from DHQ was analyzed by using Nutritionist Pro (Axxya Systems Stafford, USA) software. In addition, to detect under-reporting, energy intake was divided by the basal metabolic rate (energy intake/basal metabolic rate). Basal metabolic rate was calculated using the following formulas (42): Basal metabolic rate for men (aged 30 to 60 years) = 0.0432 (weight in kg) + 3.112 ÷ 4.2 × 1000.
The classification of dietary intake based on the formula above was as shown in the below table.
Please refer to the age group in the reference below to get the exact equation. Binary logistic regression was used to determine the association between dietary factors and other related factors with colorectal adenoma risk. All models were adjusted for confounding variables. The odds ratio (OR) and 95% confidence interval (CI) for colorectal adenoma were adjusted for age, family history of cancer, smoking and alcohol intake.

Sociodemographic Profile
Based on Table 1, a total of 275 subjects (125 cases and 150 controls) were included in the study. All the subjects were aged 60 years and above. The majority of the CRA cases were male (53.3%) whereas most of the controls were female (75.6%). Most of the subjects were Chinese and married. The majority of the male subjects had secondary education, 35.7% in case and 56.3% in control. Most of the females in the case group had secondary education (45.5%) but most of the control group had tertiary education (47.1%). Table 2 shows the smoking history, alcohol intake and family history of cancer among subjects in case and control groups. There was a significant difference in smoking status among subjects in case and control groups (p < 0.05). The majority of the subjects in the case group consumed more than 20 cigarettes per day (57.1%). No significant difference was found in the history of alcohol intake. Data on smoking and alcohol intake are not reported in females as they did not report any intake of alcohol and cigarette smoking, respectively. As shown in Table 2, the majority of subjects in the case group had a family history of cancer; 92.9% in males and 81.8% in females. Majority of the subjects in the control group had no family history of cancer, which were 56.3% in males and 61.8% in females. There was a significant difference in family history factor among those with and without colorectal adenoma (p < 0.05). Table 3 shows the anthropometry measurement status among subjects in case and control groups. The mean weight of the male category was significantly higher among subjects in control group (78.4 ± 9.8 kg) compared to subjects in the case group (71.1 ± 8.1 kg) (p < 0.05). Mean height of the subjects showed a similar trend as it was significantly higher in control groups (173.2 ± 3.7 cm) as compared to case group (167.8 ± 6.3 cm) (p < 0.01). In contrast, subjects in the case group of male categories had a significantly higher mean of fat percentage (27.2 ± 2.2%) as compared to the control group (25.4 ± 2.0%) (p < 0.05). Table 4 shows the energy, macronutrients, micronutrients, red meat, vegetables and fruits intake of the subjects in case and control groups. The median of energy, carbohydrates, protein and fat intake among both genders in the case groups was significantly higher as compared to the control group.

Dietary Factors
Moreover, the median intake of saturated fatty acids, monounsaturated fatty acids and polyunsaturated fatty acids among male subjects were significantly higher in the case group as compared to subjects in control group. On the other hand, subjects in the control group of both genders showed a significantly higher intake of dietary fiber and calcium as compared to the subjects in the case group. As shown in Table 4, the fruits intake of male subjects in the case group was significantly lower than the recommended servings from Malaysian Dietary Guideline (2010), in which 92.9% of case and 56.2% control group ate less than two servings per day (p < 0.05). However, the number of subjects who ate more than 100 g red meat per day were significantly higher in case group; 64.3% in male and 72.7% in female as compared to those subjects in Significant at *(p < 0.05), **(p < 0.01) and ***(p < 0.001) by using Pearson Chi Square test.   Significant at *(p < 0.05), **(p < 0.01) and ***(p < 0.001) by using Mann-Whitney u test. Significant at *(p < 0.05), **(p < 0.01) and ***(p < 0.001) by using Pearson Chi Square test. the control group which is 0% for male and 11.8% for female (p < 0.001).

Discussion
This study is one of the first few studies conducted in Malaysia to investigate dietary pattern and CRA. It is well established that almost all colorectal cancers arise from benign, neoplastic adenomatous polyps. The multivariable analysis performed demonstrated that gender, body fat percentage, smoking, red meat intake, high calories and high fat, low fiber and low calcium intake were significantly associated with risk of colorectal adenoma. This study showed that male subjects had a significantly higher risk of CRA compared to female subjects. Similar results had been reported by Jun, Tan and Jingyu (2019), where the detection rate of CRA cases was significantly higher in male population (22.5%) as compared to female (17.1%) (p < 0.05) (1). In Malaysia, colorectal cancer was the second most common cancer among male but the third most common cancer among female (2). Differences in the CRA incidence between males and females in our study may also reflect different effects of lifestyle and hormonal factors on anatomical location of CRAs given that colorectal cancer is generally more prevalent in the proximal colon for females, while for males it is more frequent in the distal colon (44). Furthermore, a recent meta-analysis found gender differences relating to the protective effect of fiber depending on the proximal or distal location of colorectal cancer (45).
Our study also showed that those who were working had a reduced risk of CRA relative to those who were retired or unemployed. Studies suggest that this is probably due to the increased physical activity level Significant at *(p < 0.05), **(p < 0.01) and ***(p < 0.001) by using Binary Logistic regression.
of the population during work. Individuals who are physically active were negatively associated with colorectal adenoma risk (26.0%). In contrast, those with sedentary working schedules were associated with higher risk of CRA (44.0%) (46). Based on our findings, smoking is one of the modifiable risk factor of CRA and the result was consistent with the result from Bailie, Loughrey & Coleman (2017) which reported 2.5-fold increased risk of serrated polyps (SP) in smokers (47). The potential mechanisms for the increased risk can be explained through molecular level via specific mutations induced by carcinogens in cigarette smoke. It has been shown that BRAF gene mutations correlate with lung cancer risk and colorectal cancer in smokers (48,49). SP is more likely to contain mutation in BRAF gene as compared to non-SP and raises the possibility that smoking status may be a contributing factor (47). In general, smoking may increase the risk of DNA mutations in colon cells which leads to malignant transformation via a serrated pathway. Based on this hypothesis we might expect that serrated polyps may be more prevalent in smokers relative to nonsmokers.
In addition, people who had a family history of cancer especially colorectal cancer or adenomatous polyps were considered more prominent to risk of colorectal cancer. Risk of developing colorectal cancer among those who had a family history of cancer were two folds higher than those who without (50). In a local case-control study, an association between variant allele and genotypes of IL-8-251 T > A and TNF-α-308 G > A polymorphisms and colorectal cancer susceptibility risk was observed suggesting that these two Single nucleotide polymorphisms in inflammatory response genes which undoubtedly contribute to individual risks to colorectal cancer susceptibility may be considered as potential genetic predisposition factors for colorectal cancer in Malaysian population (51). In addition, a previous study in 2012 reported that in colorectal cancer patients, frequency of KRAS mutation and PTEN loss, lower BRAF mutation rate, higher PIK3CA amplification frequency, and rare PTEN mutation were observed (52). Although both studies did not address gene polymorphism and mutation in colorectal adenoma, it had provided baseline data for gene mutation happened in colorectal cancer patients in Malaysia. We have limited studies addressing on gene mutations in Malaysia that may predispose to CRA, which hopefully will be addressed in future study design.
Our study reported that high body fat percentage was positively associated with risk of CRA. This result was supported by a systematic review and meta-analysis on adult weight gain and CRA which revealed that the summary OR was 1.39 for CRA occurrence and with each 5 kg weight gain the odds increased by 7% (53). High body fat percentage may stimulate inflammatory response in the body, which can promote the development of colorectal cancer (54). Our study also reported that high energy and high fat intake were associated with higher risk of CRA. According to a case-control study by Sun et al. (2012), high-calorie intake associated with increased risk of colorectal cancer [Adjusted OR = 1.56 (95% CI= 1.21-2.01)] (p < 0.05) (55). This might be due to our westernized diet that has been reflected in nutrition transition in Malaysia. Additionally, excessive energy intake and low energy expenditure may lead to excess body weight.
In a review conducted recently, dietary patterns were suggested to have an association with the risk of colorectal adenomas (35,36). In Ramadas and Kandiah (2009), a pre-tested quantitative food frequency questionnaire (FFQ) has been used to determine the dietary pattern of CRA subjects with good cognition and who were at least 30 years at the time of interview and have undergone colonoscopy in Hospital Kuala Lumpur (56). The results found that the protective factors against colorectal adenomas were soy, fruits and vegetables while on the other hand high intake of red meat and tubers increases the risk by two and a half fold. These findings is in agreement with our study which red meat intake more than 100 g per day (approximately half cup of chopped red meat) was significantly associated with increased risk of CRA. This was further supported by systematic and meta-analysis study by Zhao et al., (2018) which reported that there was positive association between red meat intake and incidence of CRA (RR = 1.23, 95% CI = 1.15-1.31) (57). Besides, a dose-response analysis had showed a consistent result with this study, whereby there was 14% increase in incidence of CRA with a daily intake of 100 g red meat, 1.14 (1.07-1.20) (p < 0.01).
In addition, Aune et al. (2013) concluded an increase in CRA risk with high consumption of red meat (58). In general, there were 19 case-control studies and seven prospective studies revealed that the overall relative risk, RR for each 100 g/d of red meat intake was 1.27 (95% CI = 1.16-1.40) followed by 1.20 (95% CI = 1.06-1.36) in prospective studies and 1.34 (95% CI = 1.12-1.59) in case-control studies (58). According to IARC (2015), a daily intake of 100 g red meat or more, enhances the risk of getting colorectal cancer by 17% (59). The possible mechanisms on the carcinogenic effects on red meat are the effects of heme iron. Red meat which is high in heme iron induces cytotoxic damage on colonial epithelial cells that causes hyperplasia, promotion of oxidative stress and lipid peroxidation as well as formation of n-nitroso compounds (NOC), alteration of gut microbiota and leads to colorectal carcinogenesis (60). Besides that, red meat is pro-inflammatory which may lead to CRAs. There was a significant association between a high intake of red and processed meat with the increase in inflammation biomarkers that is c-reactive protein (CRP) (61,62). Additionally, cooking meat under high temperature induce the production of carcinogenic compounds such as nitroso compounds (NOC), polycyclic aromatic hydrocarbons (PAH) and heterocyclic aromatic amines (HCA) which had been associated with colorectal carcinogenesis. Recent studies also showed the link between red meat intake with colorectal cancer adenoma (63)(64)(65)(66). High intakes of red and processed meats are strongly and associated with sessile serrated lesion risk and part of the association may be due to heterocyclic amine intake (64).
Numerous studies reported on the protective effects of fruits and vegetables against colorectal adenoma and cancer which was consistent with the result of this study (27,67,68). According to a case-control study conducted by Bahrami et al. (2019) on dietary patterns and the risk of colorectal cancer and adenoma had reported that healthy dietary pattern was characterized by high consumption of anti-inflammatory foods such as vegetables, fruits, fish, legumes and poultry was associated with a decreased colorectal adenoma risk [Adjusted OR = 0.43 (95% CI= 0.27-0.69)] (69). Apart from containing high fiber which of high interest due to its anti-inflammatory effects, fruits are also rich in Vitamin C and E as well as a variety of bioactive compounds which may have anti-tumorigenic potential. These include folate, flavonoids, polyphenols and limonene which exert anti-oxidative properties and anti-inflammatory that can inhibit cellular damage and exposure to reactive oxygen species (70). However, WCRF/AICR (2018) report had revealed that the evidence suggesting that low consumption of fruit increase the risk of colorectal cancer is limited (54).
There was significantly lower risk of colorectal adenoma in those who had high dietary fiber intake (68). Based on WCRF/AICR (2018) report, consumption of food containing high dietary fiber probably protects against colorectal cancer (54). Several biologically plausible mechanisms had been proposed on the protective effect of dietary fiber against colorectal cancer. Dietary fiber which is fermented in the bowel will form short-chain fatty acids, such as butyrate which exert anti-proliferation effects. Besides, dietary fiber may exert anti-carcinogenic effects by reducing transit time and contact of carcinogens with the colonic mucosa, increase the binding of carcinogens and production of short-chain fatty acids as well as decrease the concentration of secondary bile acids (71). All these effects contribute to reduce the colorectal adenoma and cancer risk. Diet high in dietary fiber may also reduce insulin resistance, which is also a risk factor for colorectal cancer (72,73).
Calcium also exerts protective effects against colorectal cancer. According to Bailie, Loughrey and Coleman (2017), individuals who had high intake of calcium, folate and fiber had lower risk of colorectal cancer (RR = 0.65, 95% CI = 0.49-0.85) (47). The protective effects of calcium may be explained by two mechanisms. Firstly, calcium will bind with secondary bile acids and fatty acids in the lumen and form insoluble calcium salts which will be then excreted from the body and thereby reduces the carcinogenic effects of all these bile acids on the colonial cells. Next, calcium can bind with calcium-sensing receptors of the apical membrane on the colorectal enterocytes and activates intracellular calcium signaling pathways which inhibits proliferations and enhance differentiation and apoptosis (74).
The current evidence is consistently proving that westernized diet affects risk for CRA as reflected by meat intake, high energy and high fat diet. This study is not without limitations. The study population was relatively small, partly due to COVID-19 pandemic that have impacted the whole study including data collection, and it is possible that some associations were not detected due to insufficient power. The fact that it focused on subjects in one location may limit the extrapolation of these findings to the entire Malaysian population hence warranted future studies for analyzing dietary pattern covering all geographical locations in Malaysia. Further subgroup analysis or expanding the data collection could not be possible due to COVID-19 pandemic that starts in December 2019 until now. The characteristics of the CRA were also not reported as access to the hospital are very limited. This could provide a clearer view on the risk factors associated with CRA. The possibility that the associations may be confounded or modified by other genetic or dietary factors could not be excluded. Also, this study had not explored dietary risk factors with colorectal subsites which could be worth exploring.
Despite these few limitations, the results of our study may be crucial for Malaysian CRA and CRC as guidance in dietary intake aiming to prevent or delay the occurrence of CRA and CRC. The comprehensive collection on individual dietary history and exclusion of subjects who reported a changed in their dietary lifestyle after diagnosis contributes to the strength for this study.

Conclusion
Risk factors which include gender, body fat percentage, smoking, family history of cancer, red meat intake, high fat, high energy, low fiber and inadequate calcium have a significant association with risk of colorectal adenoma in the Malaysian population. The results of this study will help to inform the design of healthy lifestyle promotions which are necessary to reduce the risk of colorectal adenoma.