Association Between Green Tea and Colorectal Cancer Risk: A Meta-analysis of 13 Case-control Studies

Colorectal cancer (CRC) is the third most common cancer worldwide and the leading cause of cancer mortality in western conutries (Siegel et al., 2012). China, Japan, and South Korea have experienced an increase of two to four times in the incidence of CRC during the past few decades (Sung et al., 2005), and more than 700,000 new CRC cases occur in ASEAN (the Association of Southeast Asian Nations) countries (Kimman et al., 2012). The definite mechanism of CRC development is still unclear, both environmental factors and genetic susceptibility are considered as risk factors. The incidence rate and mortality rate of CRC in many asian countries are still considerably lower than in western countries (Edwards et al., 2010), suggesting that there maybe some potential protective factors play a role in risk for CRC in this population. Tea is a widely consumed beverage worldwide, generally consumed in the forms of green, oolong, and black tea, all of them originated from the dried leaves of plant Camellia sinensis. Of them, green tea constitutes about 20% of the world tea production, mainly consumed in China and Japan. Green tea is produced by steaming


Introduction
Colorectal cancer (CRC) is the third most common cancer worldwide and the leading cause of cancer mortality in western conutries (Siegel et al., 2012). China, Japan, and South Korea have experienced an increase of two to four times in the incidence of CRC during the past few decades (Sung et al., 2005), and more than 700,000 new CRC cases occur in ASEAN (the Association of Southeast Asian Nations) countries (Kimman et al., 2012). The definite mechanism of CRC development is still unclear, both environmental factors and genetic susceptibility are considered as risk factors. The incidence rate and mortality rate of CRC in many asian countries are still considerably lower than in western countries (Edwards et al., 2010), suggesting that there maybe some potential protective factors play a role in risk for CRC in this population.
Tea is a widely consumed beverage worldwide, generally consumed in the forms of green, oolong, and black tea, all of them originated from the dried leaves of plant Camellia sinensis. Of them, green tea constitutes about 20% of the world tea production, mainly consumed in China and Japan. Green tea is produced by steaming Xue-Jun Wang 1& , Xian-Tao Zeng 2& , Xiao-Li Duan 3 , Huan-Chao Zeng 1 , Rui Shen 4 , Ping Zhou 5 * or pan-frying tea leaves, which inactivates the enzymes and prevents the oxidation of tea constituents. Green tea polyphenols have been extensively studied as cancer chemopreventive agents. The catechins are major consisted of (-)-epigallocatechin-3-gallate (EGCG), (-)-epigallocatechin (EGC), (-)-epicatechin-3-gallate (ECG), and (-)-epicatechin (EC), EGCG is the most abundant and active compound that can block cancer progression (Jankun et al., 1997;Kanwar et al., 2012).
This possible cancer preventive mechanism of green tea has caught much attention in the past three decades. Many animal models have been demonstrated that the green tea catechins against carcinogenesis at different organ sites (Yang et al., 2011), and this conclusion is supported by many epidemiological studies. However, there are also some published studies had come to the meaningless or opposite conclusions, and individual studies may be underpowered to detect the effect of different tumor site of CRC. Given the inconsistent associations between consumption of green tea and the potential protection implications for CRC, we conducted a meta-analysis for deriving a more precise estimation of this association.

Literature search
We initially identified published studies that concerned green tea consumption in relation to CRC risk by searching the PubMed up to May 10th, 2012. The following search terms were used: (1) "colorectal" or "colonic" or "rectal" or "colon" or "large bowel"; (2) "neoplasm" or "cancer"; (3) "green tea" or "catechin" or "tea"; (4) "case-control" or "case control" or "case". These search themes were combined using "and" without restrictions. Additionally, we also checked the reference lists of retrieved papers and recent reviews.

Study selection
We included studies that met all of the following criteria: (1) case-control study; (2) tested the association between green tea and CRC risk; (3) the cancer type did not contain adenocarcinoma; (4) the diagnoses of CRC was confirmed either histological, pathologically or cytological; (5) the site of cancer included colon, rectum, or colorectum; (6) the adjusted odds ratios (OR) and relevant corresponding 95% confidence intervals (CIs) were reported, for highest versus non/lowest level of green tea intake. Two investigators reviewed the eligibility of all studies according to the predetermined selection criteria independently, disagreements were resolved by consultation with the third one.

Data extraction
Two reviewers extracted first author's last name, year of publication, country, site of cancer, source of controls, number of cases and controls, age, gender, exposure, adjusted OR and 95% CI, adjusted estimates of risk, independently, any disagreements were resolved by consensus.

Statistical analysis
The Comprehensive Meta-Analysis software, version 2.2 (Biostat, Englewood, New Jersey) (Borenstein et al., 2005) was used to computed pooled ORs and 95% CIs, generate forest plots, determine whether there was a statistical association, and assess heterogeneity. If heterogeneity existed, the random effects model was used, or the fixed effects model was used.
In addition, we investigated the influence of a single study on the overall risk estimate by removing each study in each turn, to test the robustness of the main results. Subgroup analysis was also performed according to source of control, country, and site of cancer.

Green tea and risk of CRC
There was significant heterogeneity across the studies (p<0.001, I 2 =76.9%), so the random effects was used. The overall results showed that high green tea consumption could decrease 5% risk of CRC, but there was not a statistically significant compared with non/lowest level of green tea intake (OR=0.95, 95% CI:0.81-1.11, p=0.49); when we switched to fixed model, the results showed
When we omited one study in each turn, the ORs between 0.90 to 0.97, the p value between 0.14 to 0.74, that indicated the main result was robustness (Figure 3).

Publication bias
Based on visualization of the funnel plot ( Figure  4), it was symmetrical, that indicated there was no publication bias existed. This was confirmed by Egger linear regression (intercept =-1.94, p=0.06).

Discussion
This meta-analysis evaluated the association between green tea consumption and CRC risk, based on 13 published case-control studies. The overall result showed that indicated that high green tea consumption could weakly reduction the risk of CRC, but the association without statistically significant. Sensivity analysis by omiting individual studies and switching effect models were both supported the overall result was robust. The subgroup analyses results by stratifying the studies according to site of cancer, source of control, and country were consistent with overall result. However, the studies performed in Asia (China and Japan) indicated a weakly reduction trend, while in America (Argentina, USA, and Canada) and in Europe (Sweden and Italy) showed a weakly increase trend. That may indicated green tea is benefit for Asia people, mainly in China and Japan.
Compared with the previous meta-analysis published in 2006 by Sun et al (Sun et al., 2006), our meta-analysis included 6 eligible case-control studies before 2006 (Baron et al., 1994;Tavani et al., 1997;Munoz et al., 1998;Slattery et al., 1999;Woolcott et al., 2002;Il'yasova et al., 2003) and 3 after 2006 Wu et al., 2011;. In addition, their results showed that high green tea consumption had a statistically significant reduction risk of CRC in overall result (OR=0.74, 95%CI= 0.63-0.86) and in conlon cancer (OR=0.74, 95%CI= 0.60-0.93), but not consistent in rectal cancer (OR=0.98, 95%CI= 0.61-1.60). The trend is similar with our result, but statistically significant was disappeared. The major strength of our study was that we used adjusted ORs instead of primary data, as we know, that can provide more precise and credible result.
Does green tea can decrease the CRC risk? Results from a human experimental randomized controlled trail support the hypothesis of a protective role of green tea for the chemoprevention of metachronous colorectal adenomas (Shimizu et al., 2008). Another randomized, placebo controlled, multicentre trial to investigate the effect of green tea extract nutriprevention of metachronous colon adenomas in the elderly population is undergoing (Stingl et al., 2011), whether it can obtain a significant result is still unkown. For colorectal adenomas is unlike CRC, so the high quality andomized, placebo controlled trials for CRC are necessary to performed.
There were also some limations of our meta-analysis. First of all, heterogeneity cannot be ruled out, neither in overall or subgroup analyses, and the protective effect of green tea was changed into susceptible factor in fixed model. If there were no heterogeneity existed, we could not kown how the result would be. Second, although no publication bias detected, there were six trails beyonded the guidelines (Figure 4), we could not find the reason and to explore it. And the non-Englishstudies could not be reviewed because of the language barrier. Lastly, for we could not extract the data of dose of green tea consumption and risk of CRC, that a dose-respone analysis could not perform to assess the relationship more precisely.
In summary, there is insufficient information from case-control studies to conclude that green tea can be linked to the prevention of CRC in humans.