AIM: To investigate the associations between dietary intake of polyphenols and colorectal cancer.

METHODS: The study subjects were derived from the Fukuoka colorectal cancer study, a community-based case-control study. The study subjects were 816 cases of colorectal cancer and 815 community-based controls. The consumption of 148 food items was assessed by a computer-assisted interview. We used the consumption of 97 food items to estimate dietary intakes of total, tea and coffee polyphenols. The Phenol-Explorer database was used for 92 food items. Of the 5 foods which were not listed in the Phenol-Explorer Database, polyphenol contents of 3 foods (sweet potatoes, satoimo and daikon) were based on a Japanese study and 2 foods (soybeans and fried potatoes) were estimated by ORAC-based polyphenol contents in the United States Department of Agriculture Database. Odds ratios (OR) and 95%CI of colorectal cancer risk according to quintile categories of intake were obtained by using logistic regression models with adjustment for age, sex, residential area, parental history of colorectal cancer, smoking, alcohol consumption, body mass index 10 years before, type of job, leisure-time physical activity and dietary intakes of calcium and n-3 polyunsaturated fatty acids.

RESULTS: There was no measurable difference in total or tea polyphenol intake between cases and controls, but intake of coffee polyphenols was lower in cases than in controls. The multivariate-adjusted OR of colorectal cancer according to quintile categories of coffee polyphenols (from the first to top quintile) were 1.00 (referent), 0.81 (95%CI: 0.60-1.10), 0.65 (95%CI: 0.47-0.89), 0.65 (95%CI: 0.46-0.89) and 0.82 (95%CI: 0.60-1.10), respectively (P_trend = 0.07). Similar, but less pronounced, decreases in the OR were also noted for the third and fourth quintiles of total polyphenol intake. Tea polyphenols and non-coffee polyphenols showed no association with colorectal cancer risk. The site-specific analysis, based on 463 colon cancer cases and 340 rectal cancer cases, showed an inverse association between coffee polyphenols and colon cancer. The multivariate-adjusted OR of colon cancer for the first to top quintiles of coffee polyphenols were 1.00 (referent), 0.92 (95%CI: 0.64-1.31), 0.75 (95%CI: 0.52-1.08), 0.69 (95%CI: 0.47-1.01), and 0.68 (95%CI: 0.46-1.00), respectively (P_trend = 0.02). Distal colon cancer showed a more evident inverse association with coffee polyphenols than proximal colon cancer. The association between coffee polyphenols and rectal cancer risk was U-shaped, with significant decreases in the OR at the second to fourth quintile categories. There was also a tendency that the OR of colon and rectal cancer decreased in the intermediate categories of total polyphenols. The decrease in the OR in the intermediate categories of total polyphenols was most pronounced for distal colon cancer. Intake of tea polyphenols was not associated with either colon or rectal cancer. The associations of coffee consumption with colorectal, colon and rectal cancers were almost the same as observed for coffee polyphenols. The trend of the association between coffee consumption and colorectal cancer was statistically significant.

CONCLUSION: The present findings suggest a decreased risk of colorectal cancer associated with coffee consumption.

Colorectal cancer is one of the most common cancers in the world and the second most common cause of cancer death in developed countries[1]. In Japan, the incidence rate of colorectal cancer, especially of colon cancer, has increased dramatically until the 1990s[2]. Lifestyle factors such as physical inactivity, obesity and alcohol use are known to confer increased risk of colorectal cancer. Of dietary factors, high intake of red meat has been related to increased risk of colorectal cancer, and fiber-containing plant foods and calcium have been considered to be protective[3].

Polyphenols are the most abundant antioxidants in foods and have drawn much interest as potentially anticarcinogenic compounds. The flavonoids have been identified as potential cancer-preventive components of fruits and vegetables[4]. Catechins and tea polyphenols inhibit tumorigenesis at initiation, promotion, and progression stages in animal experiments[5]. Coffee is the major source of polyphenol intake in populations in which coffee is commonly consumed[6], and phenolic compounds in coffee such as chlorogenic acids also have anticarcinogenic effects in animal models[7]. Epidemiological studies have suggested that polyphenol-rich foods such as fruits and vegetables[3,8] and beverages such as tea[9-13] and coffee[14,15] may be protective against colorectal cancer. Experimental studies on colorectal cancer cell lines also support a protective role of polyphenols in carcinogenesis by exerting antioxidant and anti-proliferative effects and by inducing cell cycle arrest[16]. In the study reported here, we addressed the association of dietary intake of polyphenols with colorectal cancer risk using data from a community-based case-control study in Japan[17].

The present data were derived from the Fukuoka colorectal cancer study, a community-based case-control study in Fukuoka, Japan. The research protocol was approved by the ethics committee of Kyushu University and collaborating hospitals. Details of the design and conduct of the study have been described elsewhere[17].

Demographic and lifestyle characteristics of colorectal cancer cases and controls were previously described[25-29]. The two groups did not differ much with respect to sex, age, residence area and most other factors. Body mass index was greater and a history of parental colorectal cancer was slightly more frequent in the cases than in the controls. There was no measurable difference in total or tea polyphenol intake between cases and controls, but intake of coffee polyphenols was lower in cases than in controls (Table 1). In the whole sample, intake of total polyphenols derived from tea and coffee was 38% and 34%, respectively.

The associations of intakes of total, tea and coffee polyphenols with colorectal cancer are shown in Table 2. Statistically significant decreases in the OR were observed in those with high intake of coffee polyphenols (the third and fourth quintile categories) compared with the lowest quintile, but the OR for the top quintile did not decrease further, showing an upward tendency. Similar, but less pronounced, decreases in the OR were also noted for the third and fourth quintiles of total polyphenol intake. Tea polyphenols and non-coffee polyphenols showed no clear association with colorectal cancer risk regardless of adjustment for the dietary factors.

Multivariate-adjusted ORs in the subsite-specific analysis are summarized in Table 3. Cases of colon and rectal cancers numbered 463 and 340, respectively. There were 188 cases who had proximal colon cancer only and 272 cases with distal colon cancer alone. The OR of colon cancer decreased with increasing intakes of coffee polyphenols; although the individual OR for the third to fifth quintiles were not significantly different from unity, a decreasing trend was statistically significant. Distal colon cancer showed a more evident inverse association with coffee polyphenols than proximal colon cancer. The association between coffee polyphenols and rectal cancer was U-shaped, with significant decreases in the OR at the second to fourth quintile categories. Again, there was a tendency for the OR of colon and rectal cancer to decrease in the intermediate categories of total polyphenols. The decrease in the OR in the intermediate categories of total polyphenols was most pronounced for distal colon cancer. Intake of tea polyphenols was not associated with either colon or rectal cancer.

Additionally, we analyzed the associations of coffee consumption with colorectal cancers (Table 4). The associations of coffee consumption with colorectal, colon, and rectal cancers were almost the same as observed for coffee polyphenols. Nonetheless, the trend of the association between coffee consumption and colorectal cancer was statistically significant. The trend of an inverse association with coffee was also statistically significant for distal colon cancer. The OR of proximal colon cancer tended to decrease with increasing consumption of coffee, but the trend failed to reach statistical significance.

The present study showed protective associations of coffee polyphenols and coffee consumption with the risk of colorectal cancer, especially of colon cancer. Non-coffee polyphenols showed no measurable association with the overall or site-specific risk of colorectal cancer. Thus, a modestly decreased risk of colorectal cancer associated with total polyphenol intake was probably ascribed to coffee polyphenols.

The association between coffee consumption and colorectal cancer has been examined in many studies in different populations. A meta-analysis of 12 cohort studies found a relative risk (RR) of 0.91 (95%CI: 0.81-1.02) of colorectal cancer for the highest versus lowest coffee consumption[15]. A slightly pronounced decrease in the RR of colorectal cancer for the highest versus lowest consumption was noted in a combined analysis of 3 Japanese cohort studies (RR = 0.83, 95%CI: 0.62-1.10)[15]. On the other hand, a decrease in the risk of colorectal cancer associated with coffee use was more pronounced in a meta-analysis of 24 case-control studies[14]. The pooled OR for the highest versus non/low consumption was 0.70 (95%CI: 0.60-0.81) for colorectal cancer, 0.75 (95%CI: 0.60-0.81) for colon cancer and 0.87 (95%CI: 0.75-1.00) for rectal cancer[14]. A weaker association in the cohort studies is probably ascribed to a different time of exposure under consideration[14]. Coffee consumption in the recent past was assessed in most case-control studies. In the cohort studies, a follow-up shorter than 10 years was more likely to result in an inverse association between coffee consumption and colorectal cancer risk than a longer follow-up[15]. In a recent large cohort study[30], coffee consumption was associated with a decreased risk of proximal colon cancer, but not of distal colon cancer. The present study showed no such site-specific association with coffee.

Coffee polyphenolic compounds include chlorogenic, caffeic, ferulic and cumaric acids, and diterpenes (cafestol and kahweol) have been shown to possess anticarcinogenic properties[7]. Possible mechanisms by which coffee polyphenols are protective against colorectal cancer include reduction in bile acid secretion[31], reduction in the synthesis of bile by down-regulation of the expression of bile acid homeostatic genes[32] and increase in colonic motility[33].

The present study did not provide any evidence for a decreased risk of colorectal cancer associated with dietary intake of tea polyphenols. Tea polyphenols have been shown to have anti-cancer properties in numerous in vitro and in vivo studies[34,35], but epidemiologic findings are inconsistent as regards tea and colorectal cancer risk. In China, regular green tea consumption was associated with a reduced risk of colorectal cancer in male non-smokers, but not in male smokers[9] and in women[10]. On the other hand, in cohort studies of Chinese in Singapore[11] and Japan[13], green tea consumption was unrelated to colorectal cancer risk in men and women combined. A pooled analysis of 13 cohort studies in Western countries showed an increased risk associated with tea consumption; the RR for a consumption of > 900 g in liquid/d) versus no consumption was 1.28 (95%CI: 1.02-1.61) in men and women combined[12]. A combined analysis of two Japanese prospective studies found no association between green tea consumption and colorectal cancer risk[13].

We examined the associations with polyphenol and coffee for proximal and distal colon cancer separately, because these two cancers have distinct molecular mechanisms in carcinogenesis[36]. It was recently suggested that molecular changes in colorectal cancer were continuous, rather than dichotomous, from rectum to ascending colon[37]. The number of colorectal cancer cases was not large enough to perform a detailed subsite analysis in the present study. It should be noted that the difference in the molecular changes among the detailed subsites of proximal colon was less marked as compared with the difference between proximal and distal colon cancer.

In addition to a large sample size and the use of community controls, the detailed dietary assessment was a notable strength in the present study. The dietary interview was conducted as an in-person interview with typical dishes and serving sizes shown on a computer display. The present study had some weaknesses to be mentioned. The retrospective assessment of diet is a problem inherent to case-control studies. Dietary intakes in the past year may not have captured a habitual consumption relevant to the development of colorectal cancer. Additionally, the participation rate was lower in the controls (60%) than in the cases (80%), and this may have caused a selection bias.

In conclusion, a case-control study in Japan showed protective associations of coffee polyphenols and coffee consumption with the risk of colorectal cancer, especially of colon cancer.

The authors acknowledge support from Emeritus Professors Keizo Sugimachi, Seiyo Ikeda, Sumitaka Arima, and Takayuki Shirakusa and from Drs. Motonori Saku, Yoichi Ikeda, Soichiro Maekawa, Kazuo Tanoue, Kinjiro Sumiyoshi, and Shoichiro Saito in conducting the survey of cases. The following physicians kindly supervised the survey of controls at their clinics: Drs. Hideaki Baba, Tomonori Endo, Hiroshi Hara, Yoichiro Hirokata, Motohisa Ikeda, Masayoshi Ishibashi, Fumiaki Itoh, Yasuhiro Iwanaga, Hideki Kaku, Shoshi Kaku, Minoru Kanazawa, Akira Kobayashi, Ryunosuke Kumashiro, Shinichi Matsumoto, Soukei Mioka, Umeji Miyakoda, Osamu Nakagaki, Nobuyoshi Nogawa (deceased), Nobuyuki Ogami, Toyoaki Okabayashi, Hironao Okabe, Nishiki Saku, Masafumi Tanaka, Masahiro Ueda, Bunichi Ushio, and Koheisho Yasunaga.