Diet is Linked to Colorectal Cancer Risk among Asian Adults: A Scoping Review

This review aimed to map current evidence on the association between dietary factors and colorectal cancer (CRC) risk in Asia. This review was conducted based on Arksey and O’Malley methodological framework. Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) flow diagram was used to record the review process. For the purpose of searching for articles, three electronic databases namely PubMed, EBSCOHost and ScienceDirect were employed. The inclusion criteria for articles selection were articles with association analysis between diet and CRC risk among Asians, had adults as participants, articles were written in English, open-accessed and published between years 2009 and 2021. Thus, 35 out of 369 screened articles were eventually included in this review which covered 28 case-control studies, six prospective cohort studies and one randomised clinical trial. Foods such as meats, alcohol and westernised diet have been shown to be associated with increase of CRC risk while fruits, vegetables and traditional meals decreased the risk of CRC. Only a few interventional and dietary patterns studies were identified. Specific single foods and nutrients and dietary patterns have been found to increase the risk but also protected the Asian population against CRC. The findings of this review will guide health professionals, researchers and policy makers to conduct a suitable study design and topic for future research.


Introduction
Colorectal cancer (CRC) is currently a major public health concern worldwide due to the increasing trend of new cases from 1.36 million in 2015 (GLOBOCAN 2015) to 1.8 million in 2018 (GLOBOCAN 2018) (1,2). CRC is the second leading cause of cancer-related mortality and the third most common cancer globally (1,3). Despite having a lower overall frequency of CRC than Western countries, Asia has the highest number of prevalent instances, according to the International Agency for Research on Cancer's analysis of the WHO database (4). The 5-year prevalence rate was reportedly higher in China, Japan, Korea, Malaysia, Singapore and Turkey than that of other Asian countries (≥ 46.5/100,000) while India, Indonesia, Vietnam and Iran had relatively low prevalence rate (5). The reason for these differences between Review Article | Diet is linked to colorectal cancer risk Outcomes, Study type (PICOS) framework or model was applied to develop a review question and search strategy (18,19). Three PICOS elements developed were: Asia (populations), diet (intervention or exposure) and CRC (outcomes) while another two elements (comparison and study type) were not applicable. The proposed review question was 'What is the association between diet and CRC risk among Asians?' Three key terms identified through the review question were diet, CRC and Asia. This is because the dietary habits of each country and region were different (20), Asia had the highest number of prevalent cases as compared to Western countries (4) and investigating the Asians habits may help to identify foods, nutrients and dietary patterns that causes CRC among the population.

Identifying Relevant Studies
This review used three electronic databases namely PubMed, EBSCOHost and ScienceDirect for article searching. A simple search using the key terms: diet and CRC was performed in the electronic databases to determine the keywords (synonyms). Keywords were also obtained from Medical Subject Headings (MeSH) database.
The key terms and keywords (Supplementary 1) obtained were then used in PubMed, EBSCOHost and ScienceDirect databases to search for the articles. The Boolean operator of AND was used to combine key terms and keywords (1 AND 2 AND 3) during the articles search. The number of studies available and selected based on the key terms and keywords search is shown in Supplementary 1.

Study Selection
The article selections were conducted in two stages. For the first stage, two researchers were responsible to screen the titles and abstracts of the articles independently based on the inclusion and exclusion criteria and search terms. The inclusion criteria for the article's search included articles that reported the association between diets and CRC risk among Asians, adults (aged 18 years old and above), all types of study designs except review articles, systematic reviews, narrative reviews, opinion papers, letters and conference proceedings with abstract only, the articles have to be written in English, open-accessed and published in between 1 January 2009 and 7 October 2021. Any nonhuman, animal studies and studies involving benign colorectal diseases were excluded.
Asian countries is not entirely clear. However, the large population size in China might be a contributing factor to its highest number of prevalent cases, new cases and deaths (4).
The root causes of CRC can be due to nonmodifiable and environmental risk factors (4). Non-modifiable factors include genetic factors, age, gender, ethnicity, body height and family history of CRC while environmental risk factors consisted of smoking, alcohol drinking, overweight and obesity, westernised diet, physical inactivity, chronic diseases and microbiota (4). A genetic predisposition was estimated to cause 25% of CRC while 5% had reported to be inherited factors that contributed to its growth. Approximately 70% of CRC cases were sporadic cases and they were affected by environmental factors (6,7). A systematic review and meta-analysis by Magalhães et al. (8) reported that a westernised diet characterised by a high intake of red and processed meat had increased the colon cancer risk. However, some foods might be protective factors against CRC risk such as whole grains, dairy products and foods containing dietary fibre based on a recent report from the continuous update project (CUP) (9).
Studies from various diet-related to CRC were conducted widely around the world (10)(11)(12)(13)(14)(15). However, the literature mapping and research gaps especially in Asia are limited. Therefore, this review proposed to map the current evidence on the linkage between diet and CRC risk in Asia.

Methods
This review was conducted following the methodological framework developed by Arksey and O'Malley (16) which included identifying the research question, identifying relevant studies, study selection, charting data and summarising the results. This review will be following the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) (17). A protocol for the review does not exist and ethics approval was not required as the review relied solely on publicly available information.

Identifying the Research Question
The review or research question was determined prior to identifying relevant studies. Population, Intervention, Comparison,

Results
A total of 367 articles were identified from three search engines namely PubMed (n = 93), EBSCOHost (n = 27) and ScienceDirect (n = 247) while another two articles were obtained from other sources (Figure 1). Five duplicate articles were removed and the remaining 364 articles were screened based on the title and abstracts. Out of 364 articles, 290 were excluded because the titles and abstract did not match the research question of the study. The remaining 74 articles were assessed for eligibility based on their full text. A total of 39 articles were not eligible and excluded from the study as the articles did not analyse the association between CRC and diet. The final total articles included in this review were 35 and the outcome was presented in Table 1.
The results showed that various single foods and nutrients and dietary patterns (multiple food groups and multiple foods groups and nutrients) were related to the risk of CRC. A total of 24 primary studies indicated the association of CRC risk with single foods and nutrients while 11 studies, with dietary patterns. Single foods and nutrients and dietary patterns that increased risks were red, white and total meat, dry-fish, white bread, spices, sugar, tapioca, saturated fat, dietary glycemic index (GI), total energy, protein, carbohydrate, percentage of energy from fat, dietary cholesterol, alcohol, sodium, westernised diet and inflammatory factors. On the other hand, single foods and nutrients, and dietary patterns that decreased risks were fruits, vegetables, antioxidants, isoflavone, fish, fresh fish, whole bread, total fibre, fatty acids, index of nutritional quality (INQ) of fibre, folate, riboflavin, calcium, and vitamin C, folate, vitamin E, caffeine, coffee, selenium, traditional, prudent, healthy, Mediterranean, Chinese Food Pagoda (CHFP) and Healthy Eating Index-2010 (HEI-2010) ( Figure 2). The PRISMA-ScR diagram was used as a guide to record the review process (17). After the selected titles and abstracts were screened and checked to ensure their relevance to the review question, researchers then retrieved the full text articles of selected abstracts and excluded unrelated abstracts.
In the second stage, the full articles were checked and reviewed by two researchers independently to ensure it meets the review objective. Furthermore, articles that do not reflect the association between dietary intake and CRC risk in Asia and articles that do not meet the review objective were excluded. Then, relevant articles were assessed to answer the review question. Discrepancies between researchers were discussed and a consensus was sought at the end. Throughout the process, EndNote® software was used to manage the results from the search.

Charting the Data
A standardised data charting form (Table  1) was developed and reviewed by the study team for relevance and appropriateness. The form's extraction fields captured relevant information on study characteristics, including author(s), year of publication, aims of the study, study design, type of scale or instrument, sample size, participant's characteristics and results that were relevant to the objectives of the review. The charting form was pilot tested by researchers with a random sample of five sources of evidence to ensure all relevant data were captured. Eligible sources of evidence were charted using Microsoft Excel. The goal of this scoping review was to provide an overview of the existing literature regardless of quality and a formal appraisal of the methodological quality of sources of evidence that were included in the review was not performed.

Summarising the Results
A numerical and descriptive summary of the charting results was used to present findings. Besides that, a summary diagram was used to better visualise the linkage between diet and CRC risk in Asia. The limitations of this study were also assembled to allow identification of research gap and produce useful recommendations for future research.
Review Article | Diet is linked to colorectal cancer risk   were not related to CRC risk in men and women, respectively In men, sugars intake was associated with CRC risk inversely among never-smokers and positively among male ever-smokers (P for interaction = 0.01). Sugars intake was associated with an increased risk among men with no alcohol consumption but was unrelated to the risk among male alcohol drinkers   Besides food items and food groups, macro-and micronutrients were also identified as contributors to the CRC risk in this current review. High fats diet (32), saturated fat (22,33), dietary GI (34), total energy (33), protein (33), carbohydrate intakes (33), percentage of energy from fat (33), dietary cholesterol (33), alcohol (35) and sodium (33) intake were associated with increased risk of CRC where by, high intake of these nutrients among certain populations could be the reason why it leads to disease development.
However, there are certain food items and groups have been found to be protective against CRC. The consumption of fruits (24) and vegetables (24) including cabbage (23) and sprout (23) may reduce the development of CRC among the Indian population. A study by Xu et al. (36) reported that, antioxidants from fruits and vegetables such as flavan-3ols, flavanones, flavones and flavonols were associated with decreased CRC risks among Chinese population. According to Khankari et al. (37), higher intake of isoflavone could reduce the CRC risk among premenopausal women with BMI < 23.0 kg/m 2 at baseline assessment. Antioxidants trap the reactive oxygen molecules and free radicals at cellular level, thus intervening in a protective mechanism against oxidative damage (33). CRC risk could be reduced by approximately 25% among individuals who consume fish daily like the community in Kerala, India (24). They commonly consumed sea fish that were rich in vitamin D and selenium. According to Du and Fang (38), these nutrients may protect the pathogenesis and development of CRC. Another study conducted also in India showed that, freshfish intake had a 40% to 70% reduction in CRC development as compared to individuals who do not consume fresh-fish (23). Moreover, whole bread was classified as having a protective effect against CRC in a Jordanian (28) study and this is because whole grain products could be rich in antioxidants including trace minerals and phenolic compounds (39).
Macronutrients such as total fibre (34) were associated with a 53% reduction of CRC risk while the increased intake of polyunsaturated fatty acids (PUFA) (40) and linoleic acid (LA) (40) might decrease the risk in men and linolenic acid (ALA) (40) in women. For INQ of fibre, it was negatively associated with CRC risk (41). Micronutrients that could prevent the formation of CRC included folate (35,63), vitamin E (33),

Discussion
The present review aims to systematically identify the existing literature on the relationship between diet and CRC risk in Asia. Based on the included literature (Table 1), total meat (21) and red meat (22,23) including beef (24) and pork (25) are considered as single foods were associated with increase CRC risk in Asian countries. According to Hur et al. (26) the harmful substance known as heterocyclic amines (HCAs), polycyclic aromatic hydrocarbons (PAHs) and N-nitroso compounds (NOCs) that were typically produced during cooking and processing of red meat might induce cancer. Furthermore, Dao et al. (27) reported that heterocyclic amines 2-amino-1-methyl-6-phenylimidazo pyridine (PhIP) which was produced from cooked beef was associated with an increase in CRC risk. Besides red meat, white meat (23) including chicken (28) was found to contribute to CRC development. In addition, Clostridium perfringens/histolyticum spp. was found abundant in chicken meat than beef or salmon based on in vitro faecal batch culture investigation (29). It has been proclaimed to produce other enzymes involved in genotoxic or carcinogenic metabolites production in the colon such as β-glucosidase, nitroreductase and azoreductase (29).
Ganesh et al. (23) have observed that the consumption of dry-fish increased the risk of CRC among non-vegetarian Indian population. Dry-fish was considered as a risk factor probably due to the methods of preservation which consisted of salting and drying (23). Besides that, the high GI of white bread (grains) might explain why this diet was also associated with CRC risk among Jordanians (28). Moreover, Franceschi et al. (30) stated that refined carbohydrate was responsible for CRC based on their positive relationship with GI and glycaemic load (GL). Although the exact mechanism for cancer genesis with high GI type of dietary pattern is not clear, but insulin resistance or insulin growth factor might be involved (31). Spices such as chillies and pepper, sugar and tapioca were also associated with increased CRC risk (24). Chillies have carcinogenic properties due to their capsaicin compound and might explain the positive association between spices and CRC risk (24). Furthermore, the unprocessed tapioca contains toxins such as linamarin and cyanide derivatives. These toxins could directly act on the bowel mucosa to cause CRC (24). and alcohol use were associated with increased risk, but regular fruit eating was associated with decreased risk of CRC. According to Castelló et al. (56), high and low adherence to Mediterranean and westernised dietary patterns respectively may decreased CRC risk among the Spain population. According to Fung et al. (57), westernised eating habits were associated with colon cancer in United States (US) women. Even while the majority of foods were comparable, there were others such tapioca and dry fish, that were only available in Asia and had ties to the illness. Thus, investigating the food habits of the Asian population is essential.
There were several strengths and limitations of the current scoping review. Strength-wise, the current and updated evidence provided in this review are useful to develop dietary intervention programmes for the mitigation and prevention of CRC. Limitation-wise, first there was insufficient interventional study designs with the majority were observational studies (28 case-control and six prospective cohort studies) which are known to have bias. Secondly, only 11 studies were conducted on dietary patterns as compared to 24 studies that were on single foods and nutrients. Dietary patterns are preferred because this approach captures the overall diet which also includes variation of foods and combination of nutrients in a studied population. Thirdly, most of the included studies were not stratified based on the disease outcome or anatomical subsites such as proximal colon, distal colon and rectum.

Conclusion
In conclusion, single foods and nutrients and dietary patterns such as meats, alcohol and westernised diet have been found to be associated with the increase risks of CRC while fruits, vegetables and traditional decrease the risks of CRC among Asian population. The limitations or gaps identified from this review will assist and guide health professionals, researchers and policymakers to design suitable study designs and topics related to diet and CRC for future studies.
caffeine (33), selenium (42) and INQ of folate, riboflavin, calcium and vitamin C (41). Folate influences the availability of methyl groups in the CRC pathway, which enhances deoxyribonucleic acid (DNA) synthesis and repair (35). Besides that, vitamin E and caffeine act as an antioxidant to fight against free radicals and effectively scavenge reactive oxygen molecules and hydroxyl radicals (33) while selenium which also act as an antioxidant protects against oxidative stress (42). On the other hand, Riboflavin (vitamin B2) plays an important role in DNA methylation, stability, synthesis and repair (43). For calcium, it can modulate the CRC-related cell signalling pathways and inhibit oxidative DNA damage while vitamin C, has antioxidative and proapoptosis effect, inhibit cell proliferation and reduce inflammation (43).
The CRC risk factors and protective factors were not due only to single foods and nutrients, but it was also influenced by dietary patterns. According to Park et al. (44), westernised dietary patterns characterised by high intake of meats, carbohydrates (fast foods), oil and sugar were associated with enhanced risk of CRC especially among Korean women. A positive association was also observed between the western patterns and CRC in Iran (45). Cho et al. (46) reported that higher consumption of C-reactive protein (CRP)-dietary pattern score (inflammatory diet) derived from reduced rank regression (RRR) was associated with increased CRC risk and a stronger risk was observed in rectal cancer than colon cancer. Inflammatory diet patterns not only could be obtained from RRR but they could also be derived from an index-based method such as dietary inflammatory index (DII). All studies included inflammatory diet patterns using DII (47-51) (n = 5) and showed a positive association with CRC risk. Another six dietary patterns including traditional (44), prudent (44), healthy (45), Mediterranean (52), CHFP (53) and HEI-2010 (52) dietary patterns were found to be protective against CRC.
This review found that findings of diet relationship with CRC between Asia and Western countries had similarities. The similarities could be seen in several studies conducted by Bradbury et al.  (54) indicated that red meat and alcohol enhanced the risk while fibre was found protective against CRC among United Kingdom (UK) population. Another study done in the UK population by Sanjoaquin et al. (55) revealed that white bread