subsite-specific dietary risk factors for colorectal cancer: a review of cohort studies

8/10/2019 Subsite-specific dietary risk factors for colorectal cancer: a review of cohort studies

1/15

Hindawi Publishing CorporationJournal o Oncology Volume , Article ID , pageshttp://dx.doi.org/ . / /

Review ArticleSubsite-Specific Dietary Risk Factors for Colorectal Cancer: A Review of Cohort Studies

Anette Hjartker, 1 Bjarte Aagnes, 2 Trude Eid Robsahm, 2 Hilde Langseth, 2

Freddie Bray, 3 and Inger Kristin Larsen 4

Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, P.O. Box Blindern, Oslo, Norway Department of Etiological Research, Cancer Registry of Norway, Institute of Population-Based Cancer Research, Oslo, Norway Section of Cancer Information, International Agency for Research on Cancer, Lyon, FranceDepartment of Registration, Cancer Registry of Norway, Institute of Population-Based Cancer Research, Oslo, Norway

Correspondence should be addressed to Anette Hjart aker; [email protected]

Received October ; Accepted November

Academic Editor: Anne-Kathrin Illner

Copyright Anette Hjartaker et al. Tis is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Objective. A shif in thetotalincidence rom lef- to right-sidedcolon cancer hasbeen reportedand raisesthe questionas towhetherli estyle risk actors are responsible or the changing subsite distribution o coloncancer. Te present studyprovides a review o thesubsite-speci c risk estimates or the dietary components presently regarded as convincing or probable risk actors or colorectalcancer: red meat, processed meat, ber, garlic, milk, calcium, and alcohol. Methods. Studies were identi ed by searching PubMedthrough October , and by reviewing re erence lists. Tirty-two prospective cohort studies are included, and the estimates arecompared by sex or each risk actor.Results. For alcohol, there seems tobe a stronger association with rectal cancer than with coloncancer, and ormeata somewhat strongerassociation withdistal colon andrectal cancer, relative to proximalcolon cancer. For ber,milk, and calcium, there were only minor differences in relative risk across subsites. No statement could be given regarding garlic.Overall, many o the subsite-speci c risk estimates were nonsigni cant, irrespective o exposure. Conclusion. For some dietary components the associations with risk o cancer o the rectum and distal colon appear stronger than or proximal colon, but not

or all.

1. Introduction

Global estimates or indicate that colorectal cancer isthe third most common cancer in the world [ ]. Reports inseveral countries have described diverging incidence rates incolorectal cancer by subsite, including, in relative terms, anincreasing proportion o proximal tumors [ ], and thusa shif in absolute incidence rom lef- to right-sided coloncancers.

Te reasons or this trend are not well understood; thesubsites differ in physical unction, artery supply, histology,andinnervation, andthey also derive rom differentsegmentsin the primitive intestinal tract in the embryo [ ]. Teproximalcolonoriginates rom themidgut, whereas thedistalcolon and the rectum derivate originate rom the hindgut.

Comparisons have also shown that proximal colon tumorstend to have differentmolecular characteristics, with a higherproportion o microsatellite instability, and are more likely tohave CpG islandmethylator phenotype andKi-ras mutationsthan distal colon and rectal tumors [ ].

It has been estimated that percent o all colorectalcancer cases can be prevented in high-risk populationsthrough modi cations o diet, physical activity habits, andweight control [ ]. According to the recent report romthe World Cancer Research Fund/American Institute orCancer Research (WCRF/AICR), there is convincing evi-dence that dietary ber protects against colorectal cancerand that red and processed meat and alcohol (particularly in men) increase the risk o the disease [ ]. Further, itis stated that garlic, milk, and calcium probably protect


2/15

Journal o Oncology

: Search results as o October , .

Queries Result Search colorectal neoplasms Search risk factors Search diet

Search nutrition Search alcohol Search cohort study Search OR OR Search AND Search AND

Search AND Search case control study Search review Search NO NO Search Limits: Human, English

againstcolorectal cancer. Nodistinction is, however, made orthe different subsites o the colorectum [ ]. Also, althoughmeta- [ ] and pooled analyses [ ] have providedquantitative synthesis or several o the dietary risk actors,little emphasis has been placed on subsite risks.

Based on the biological differences in the colorectalsegments and the reported differences in incidence, we may suggest differences across the segments in their associationto li estyle actors, such as diet. Te aim o the present paperis to give an updated overview summarizing the etiologicaldifferences between the colorectal subsites with regard to thedietary actors considered to be convincing or probable risk

actors or colorectal cancer.

2. Material and Methods

Te speci c risk actors studied were red meat, processedmeat, ber, garlic, milk, calcium, and alcohol, selected givenan a priori assessment o their importance in colorectalcancer etiology ollowing the WCRF/AICR report in [ ], and or which their modi cation could lead to areduction in rates o colorectal cancer. Te outcome was therisk o primary colorectal cancer according to subsite.

A search or cohort studies published as original arti-cles was conducted via a search o PubMed (http://www .ncbi.nlm.nih.gov/ ), using a search strategy that combinedthe term colorectal neoplasms with the terms risk actorsand cohort study with either the term diet, nutrition,or alcohol. Te search was restricted to studies publishedor available online as o October , , in the Englishlanguage. A detailed description o the search strategy andthe resulting papers retrieved is given in able , and theprocedure is described in Figure . A total o articles wereidenti ed and reviewed according to title and abstract. Teinitial evaluation yielded articles in the study databaseandunderwent a secondevaluation based on ull-text review.A similar PubMed search or case-control studies nestedwithina cohortidenti ed articles,o which oneunderwent

ull-text review. Inaddition, urther articles wereidenti ed

by scanning the re erence lists o retrieved articles, reviews,meta-, and pooled analyses and underwent ull-text review (Figure ).

Studies were included i theyprovidedrisk estimates(andcorresponding con dence intervals) or both proximal anddistal colon cancer. Proximal colon (right sided) includes

ceacum, ascending and transverse colon, while distal colon(lef-sided) includes descending colon, sigmoid exure, andsigmoideum. Some studies have not ollowed the above-mentioned classi cation [ ], and this is speci ed in

able . Further, to be included, the cohorts had to beeither population based, registry based, or obtained romcensuses. Studies on speci c subpopulations (e.g., hospital-based cohorts) were not included norwere studies examiningsecond cancers, metastasis, survival, or mortality. Followingthe ull-text evaluation, articles were excluded or thereasons indicated in Figure , and articles were includedin the review. O these, seven gave data on (red) meat [

], ve on processed meat [ ], eight on ber [ ],one on garlic [ ], three on milk [ ], seven on calcium[ , , ], and ten on alcohol [ , , , ].

Te number adds up to more than as several papersgive data or more than one risk actor. I several papers onthesame risk actorwere published ora given cohort, all datawas retrieved rom the most recent paper. Te relative risk (RR), hazard rate ratio (HRR), hazard ratio (HR), incidencerate ratio (IRR) or odds ratio (OR), and corresponding

% con dence intervals ( % CI) or each risk actor arepresented (as RR) in Figures , sorted in ascending ordero magnitude by sex. For one study the risk estimates weretabulated according to the lowest versus highest exposurecategory in the original paper [ ] and are presented as theinverse o the value in the corresponding gure (Figure ).

Meta-analyses were not per ormed as a consequence o the heterogeneity in actors such as exposure measurement,the categorization o risk actor levels, and the con oundersadjusted or in the studies [ ].

3. Results

Overall, data rom cohorts with in ormation on one ormore o the risk actors were included in this review. able gives a detailed description o the studies, with in ormationon cohort size, number o cases, sex, age distribution, ollow-up time, and actors adjusted or in the risk analyses. Te

European Prospective Investigation into Cancer and Nutri-tion (EPIC) study consists o subcohorts rom Europeancountries. Te other cohorts were rom the USA ( cohorts),Sweden ( ), Denmark ( ), Japan ( ), Korea ( ), UK ( ), andthe Netherlands ( ). Te subsite-speci c results (Figures ) are described according to the individual risk actors, as

presented in the ollowing.

. . Red Meat. Seven cohort studies were included in thereview, o which ve reported on red meat [ , ],oneonbee and pork [ ], and one on meat (not urther speci ed)[ ].Te vestudieson redmeat reported an increasedrisk o colorectalcancerwith increasing intake [ , ] ( able ),


3/15

Journal o Oncology

341 publications were identied from the literature searchLimited to prospective cohort studies on humans, in English,

published before October 8, 2012

46 publicationsidentied from

additional search

62 publicationsincluded fromother sources

233 publicationsexcluded after title and

abstract review

45 publicationsexcluded after

title andabstract review

1 publication108 publications

171 publications full-text reviewed

32 publications included in the review

(76) no analyses/risk estimates by colonic subsites(26) not a relevant risk factor

(2) data on mortality or survival only (5) outcome adenomas(6) updated data available

Not an appropriate design:

(1) reviews(5) meta-analysis or pooled analysis(2) randomised trials

(16) case-control studies

Number o publications:

(139 total)Reasons or exclusion afer ull-text review

F : Flow diagram o the study selection.

although most o the risk estimates were not statistically di -erent rom unity. In Figure , there appears to be somewhat

more consistency with the observation o an increased risk o cancers o the rectum and distal colon than there is orproximal colon cancer. Tis picture remains when restrictingthe evaluation to studies with adequate statistical power. In aSwedish study, women consuming or more grams o redmeat per day had an increased risk o distal colon cancero . ( % CI . . ) compared to women consumingless than grams/day, and there was a signi cant trendo increasing risk with increasing consumption ( trend 1250 versus 500 mg/day (total calcium) 0.58 (0.3 2, 1.05 )JPHC Ishihara, 74 661 versus < 33 7 mg/day (dietary calcium) 0.60 (0.3 5, 1.01 )COSM Larsson, 79 1445 versus 988 versus 1255 versus 8 3 0 versus 1270 versus


9/15

Journal o Oncology

Cohort Reference Age Exposure RR ( 95% CI)Proximal

EPIC Ferrari, 2007 [56] 3 570 >60 versus 0.14.9 g/day 0.92 (0.51 , 1.66 )CCPPS Pedersen, 200 3 [54] 2 3 95 >41 drinks/week versus nondrinkers 1.00 (0.40 , 2.40 )UKDCC Park, 2010 [58] 26 84 3 0 g/d versus >0011 versus 1.8 g/day 1.08 (0.82 , 1.44 )NLCS Bongaerts, 2008 [57] 55 69 3 0 g/day versus abstainers 2.28 (1.12 , 4.62 )NLCS Bongaerts, 2008 [57] 55 69 3 0 g/day versus abstainers 1.19 (0.69 , 2.06 )KNHS Shin, 2011 [ 3 6] 3 080 Almost everyday versus none 1.20 (0.90 , 1.70 )MCS Akhter, 2007 [55] 40 64 45.6 g/day versus never 1.40 (0.72 , 2.75 )UKDCC Park, 2010 [58] 26 84 3 0 g/d versus >02.0 versus 41 drinks/week versus nondrinkers 1.00 (0.40 , 2.3 0)NLCS Bongaerts, 2008 [57] 55 69 3 0 g/day versus abstainers 1.41 (0.94 , 2.11 )NCKPS Klatsky, 1988 [27] Missing 3 drinks/day versus never 1.52 (0.40 , 5.71 )EPIC Ferrari, 2007 [56] 3 570 >60 versus 0.14.9 g/day 1.68 (1.08 , 2.62 )UKDCC Park, 2010 [58] 26 84 3 0 g/d versus >011 versus 1.8 g/day 0.98 (0.73 , 1.33 )NLCS Bongaerts, 2008 [57] 55 69 3 0 g/day versus abstainers 1.3 2 (0.52 , 3 .3 7)RCLA Wu, 1987 [52] Missing 3 1 mL/day versus nondaily 1.66 (0.80 , 3 .60)UKDCC Park, 2010 [58] 26 84 3 0 g/d versus >002.0 versus 060 versus 0.14.9 g/day 1.93 (1.3 5, 2.78 )CCPPS Pedersen, 200 3 [54] 2 3 95 >41 drinks/week versus nondrinkers 2.20 (1.00 , 4.60 )NCKPS Klatsky, 1988 [27] Missing 3 drinks/day versus never 3 .17 (1.05 , 9.57)NLCS Bongaerts, 2008 [57] 55 69 3 0 g/d versus abstainers 0.72 (0.16 , 3 .10)UKDCC Park, 2010 [58] 26 84 3 0 g/d versus >00


10/15

Journal o Oncology

Pro essionals Follow-Up Study that studied colon canceronly, non-signi cant inverse associations were seen betweentotal calcium and distal colon cancer or both men andwomen, and a pooled analysis o the two cohorts gave asigni cant inverse trend or distal colon cancer ( trend =0.01 ) [ ]. Likewise, a signi cant inverse trend was reported

between total calcium and distal colon cancer in the SwedishMammography Cohort ( trend = 0.02 ) [ ], while ina US emale cohort, a signi cant inverse trend was seenbetween dietary calcium intake and proximal colon cancer( trend = 0.01 ) [ ]. No association was seen or totalcalcium. In a US study examining both sexes, total calciumintake was signi cantly inversely associated with proximalcolon cancer among men ( trend = 0.04 ), but not dietary calcium, and there were no signi cant associations seen orwomen (estimatesnotgiven) [ ]. Ina study o Swedish men,a signi cant inverse relationwas ound between total calciumand rectal cancer ( trend = 0.02 ), whereas nonsigni cantinverse associationswere seen orcancer o thecolon [ ].Nosigni cant associations between dietary calcium and cancero any colorectal subsite were reported in a Japanese study either or men or or women (estimates not given or women)[ ]. An earlier study o Hawaiian-Japanese men reported asigni cant inverse association between dietary calcium andsigmoid colon cancer ( trend = 0.02 ) [ ]. In addition tothe eight cohorts included in this paper, an additional cohortstudyamongAmerican women statesthat their data providedlittle support or a protective effect o total calcium at eithertumor subsite, but does not present risk estimates [ ].

. . Alcohol. en articles [ , , , ] were includedin the review ( able ). Tree analyses or both sexes com-bined consistently showed a higher risk o rectal cancerwith increasing alcohol consumption and no signi cantassociations or any o the colon subsites [ , , ]. Inthe EPIC study [ ] an increased risk was reported both orrectal and distal colon cancer, whereas in the UK dietary cohort consortium (part o which is included in the EPICstudy) [ ] a signi cantly increased risk was ound or distalcolon cancer only (Figure ). Several sex-speci c analyseshave been done. Among men, two older studies with limitedstatistical power reported that alcohol consumption waspositively related to proximal colon cancer only [ , ],whereas our more recent studies reported non-signi cantassociations with proximal colon cancer [ , , , ]. Astudy among Japanese men reported increased risk or distalcolon cancer, but the con dence interval or the risk estimatewas very wide (consumption o . grams alcohol or moreper day compared to never drinkers HR . , % CI .

. ) [ ]. In terms o consumption o alcohol and rectumcancer among men, two o our studies have reported apositive association [ , ].

For women, ewer signi cant associations have beenreported. In the Iowa Womens Health Study, alcohol wasnot signi cantly associated with either proximal colon ordistal colorectal cancer, nor did urther separation intodistal colon and rectal cancer reveal any signi cant associ-ations (estimates notgiven) [ ]. Furthermore, no signi cantassociationswereseen inanearlier studyo Americanwomen

[ ] or in the UK dietary cohort consortium [ ]. In TeNetherlands Cohort Study, however, women consuming grams or more o alcohol perday had HRR o proximal coloncancer o . ( % CI . . ) compared to abstainers,whereas analyses or the other subsites were not sufficiently robust to draw any conclusions [ ]. In a Korean study,

women who requently consumed alcohol or who consumedgreater amounts o alcohol had a higher risk o rectal cancer[ ].

4. Discussion

Tis review provides an overall and updated synthesis o the results rom cohort studies examining the associationbetween dietary actors that are convincingly or probably related to the risk o colorectal cancer and subsite-speci ccolorectal cancer. Our study indicates that consumption o alcohol is more strongly related to the risk o rectal cancerthan to colon cancer, also that meat consumption tends tobe somewhat more strongly related to the risk o distal coloncancer and rectal cancer than proximal colon cancer, thatthere are only minor differences in relative risk or colorectalcancer across the major subsites or ber, milk, and calcium,and that no statement can be given or garlic due to limiteddata. It should be noted that or all exposures the majority o the analyses showed nonsigni cant associations with cancerrisk, the exception being the positive association betweenalcohol consumption and rectal cancer.

Te pathway o colorectal cancer is complex. Te subsiteetiology is rather poorly understood, and the mechanism orthe various dietary actors is likely to differ. Even or red andprocessed meat, both established risk actors or colorectalcancer, the underlying mechanisms are not well de ned. Onesuggested mechanism or the somewhat stronger association

or rectum and distal colon relative to proximal colon cancerrelates to the enhanced endogenous ormation o carcino-genic N-nitroso compounds with a high intake o meat [ ].Te level o markers o N-nitroso compounds appears to behigher in tissue rom distal colon and rectum than in thato the proximal colon [ ]. Tis nding is in keeping witha previously meta-analyses which implicated processed meatconsumption as a stronger risk actor or cancer occurrenceat the distal colon relative to the proximal colon [ ].

A suggested mechanism by which dietary ber may decrease the risk o colorectal cancer is linked to the er-mentation o ber. Te ermentation produces short-chain

atty acids and, in particular, acetic, propionic, and butyricacids. Butyrate is particularly o interest, as it has beenshown to induce apoptosis and to be cytotoxic to bothcolorectaladenoma andcarcinomacells [ ]. Studies on micehave shown that the concentration o butyrate is highestin the distal colon [ ]. In humans, ber is ermented inthe proximal colon, and the total amount o short-chain

atty acids has been estimated to be considerably higher inthe proximal site compared to the distal [ ]. One study on proximal and distal colonocytes indicates, however, thatbutyrate is a more important source o energy or the distal,than or the proximal, colonic mucosa [ ]. I so, this


11/15

Journal o Oncology

could be a relevant biological mechanism explaining any differences in risk between the sites. In addition, ber may dilute the concentration o carcinogenic substances in thedistal colon. Despite these ndings, a pooled analysis on berreported no convincing differences in colorectal risk or the various anatomical subsites [ ], in line with our own results

published here.Te reduced risk o colorectal cancer with increasingconsumption o milk is likely to be at least partly mediated by calcium, which is thought to have a protective effect throughits ability to bind bile acids, and its growth-restraining anddifferentiation- and apoptosis-inducing effect on colorectalcells [ ]. No convincing site-speci c associations regardingmilk and calcium were seen in our review. A ormer pooledanalysis reported an inverse association or milk limited tocancer o the distal colon and rectum [ ], while the results

rom two meta-analysis on site-speci c impact o calciumhave been con icting [ , ].

Te consumption o alcohol is associated with increas-ing risk o cancer in several organs in the digestive tract,including the colorectum [ ]. Alcohol is not a carcinogenitsel , but acts as a tumor promoter and possibly as a co-carcinogen. Alcohol also acts as a solvent and thus mightincrease the exposure to other carcinogens by enhancing thepenetration o carcinogens into the cell [ ]. Acetaldehydeis a metabolite o alcohol and may be the most importantagent responsible or the carcinogenic effect as it is highly toxic, mutagenic, and carcinogenic [ ]. A pooled analyseson alcohol reported similar risk across all areas o the largebowel [ ]. However, a stronger association with alcohol

or rectal cancer compared with colon cancer as seen inour paper could possibly be related to a higher degree o epithelial hyperregeneration in rectum [ ]. Te numbero sex-speci c analysis is presently too low to suggest any signi cant interaction by sex at the subsite level.

Tere are some methodological issues in this studywhichmay have impacted on the ndings. Tere is considerable variability in several key characteristics o the assembledcohort studies: the ollow-up time across studies includedin this paper varied rom to years, while the totalnumber o colorectal cancer cases analyzed ranged rom

to . Short-term ollow-up studies tend to accrue alower number o cases, and the resultingestimatesare subjectto greater uncertainty. Te substrati cation o cases by sex,tumor location, and exposure categories, as presented here,inevitably leads to smaller numbers and a greater degreeo imprecision in the estimates, even or relatively largestudies.Statistically signi cantassociations may thusbe morespurious at the subsite-speci c level. However, restrictingthe evaluation to studies with adequate statistical power didnot change the overall picture. Long-term ollow-up studieswill commonly accrue a greater number o cases, but aremore prone to measurement error given that an increasing

ollow-up time raises the possibility that exposure status o the participants will change, leading to misclassi cation o exposure and under- or overestimation o the risk estimates.However, thenatural historyo colorectal canceris onaverageo long duration, and exposure in the more distant past may be the most relevant when estimating subsequent risk.

Another issue is the difference in risk actor dosagesbetween studies and that the categories compared sometimes vary considerably between studies. For instance, or calcium,in the Swedish study by Larsson et al. [ ] daily intakeo mg or more is compared to an intake o less than

mg/day, whereas in the Japanese study by Ishihara et al.

[ ] daily intake o mg ormoreis compared with an intakeo less than mg/day. Care ul reading o the exposurecategories (given in the gures) is there ore necessary whenevaluating the ndings. In addition, the speci c con ounding

actors adjusted or at the analysis stage differ betweenstudies.

O the studies that were ull-text reviewed, wereexcluded as they did not reveal in ormation at the level o subsite location. Given the high proportion o cohort studies

ailing to report subsite-speci c estimates, the potentialpublication bias prohibited a ormal meta-analysis [ ]. A

urther rationale or this decision is the lack o uni ormity inexposure categories within each risk actor.

In summary, the strength o the association betweendietary components and cancer o the large bowel may partially dependon theanatomic location within thecolorec-tum. Te most consistent nding is the stronger associationbetween alcohol and rectal cancer, compared with alcoholand proximal and distal colon cancer. Meat (red and pro-cessed) is possibly more strongly associated with risk o distalcolon cancerandrectalcancer than therisk o proximalcoloncancer. For ber, milk, and calcium there seem to be only minordifferencesin relativeriskacross thesubsites. However,caution is required as the number o papers presenting risk estimatesby colorectal subsite is limited, particularly ormilk and garlic. Also, most o the subsite-speci c analyses reportnon-signi cant ndings.

Conflict of Interests

Te authors declare that they have no con ict o interests.

Acknowledgment

Te authors thank LenaLeder orhercontribution to thedataextraction and paper preparation.

References

[ ] J. Ferlay, H. R. Shin, F. Bray, D. Forman, C. Mathers, and D.M. Parkin, GLOBOCAN : Cancer Incidence and Mortality Worldwide, IARC Cancer Base, no. , International Agency orResearch on Cancer, Lyon, France, .

[ ] F. Kee, R. H. Wilson, R. Gilliland, J. M. Sloan, B. J. Rowlands,and R. J. Moorehead, Changing site distribution o colorectalcancer, British Medical Journal , vol. , no. , p. , .

[ ] G. I. Slater, R. H. Haber, and A. H. Au ses Jr., Changingdistribution o carcinoma o the colon and rectum, Surgery,Gynecology & Obstetrics, vol. , pp. , .

[ ] Y. oyoda, . Nakayama, Y. Ito, A. Ioka, and H. sukuma, rends in colorectal cancer incidence by subsite in Osaka,Japan, Japanese Journal of Clinical Oncology , vol. , no. , pp.

, .


12/15

Journal o Oncology

[ ] C. Cucino, A. M. Buchner, and A. Sonnenberg, Continuedrightward shif o colorectal cancer, Diseases of the Colon and Rectum, vol. , no. , pp. , .

[ ] S. L. Saltzstein and C. A. Behling, Age and time as actors in thelef-to-right shif o the subsite o colorectal adenocarcinoma:a study o , cases rom the Cali ornia Cancer Registry, Journal of Clinical Gastroenterology , vol. , no. , pp. ,

.[ ] R. W. Beart, L. J. Melton, M. Maruta, M. B. Dockerty, H. B.

Frydenberg, and W. M. OFallon, rends in rightandlef-sidedcolon cancer, Diseases of the Colon & Rectum, vol. , pp.

, .[ ] F. Levi, L. Randimbison, and C. La Vecchia, rends in subsite

distribution o colorectal cancers and polyps rom the VaudCancer Registry, Cancer , vol. , no. , pp. , .

[ ] E. Mitry, A. M. Benhamiche, C. Couillault et al., Effect o age,period o diagnosis and birth cohort on large bowel cancerincidence in a well-de ned French population, ,European Journal of Cancer Prevention, vol. , no. , pp.

, .

[ ] H. Singh, A. A. Demers, L. Xue, D. urner, and C. N. Bern-stein, ime trends in colon cancer incidence and distributionand lower gastrointestinal endoscopy utilization in Manitoba, American Journal of Gastroenterology , vol. , no. , pp.

, .[ ] H. akada, . Ohsawa, S. Iwamoto et al., Changing site

distribution o colorectal cancer in Japan, Diseases of the Colon& Rectum, vol. , pp. , .

[ ] M. Torn, R. Bergstrom, U. Kressner, P. Spar en, M. Zack, andA. Ekbom, rends in colorectal cancer incidence in Sweden

by gender, localization, time period, and birth cohort,Cancer Causes and Control , vol. , no. , pp. , .

[ ] J. R. Jass, Subsite distribution and incidence o colorectal

cancer in New Zealand, , Diseases of the Colon &Rectum, vol. , pp. , .[ ] R. Scheiden, P. Pescatore, Y. Wagener, N. Kieffer, and C.

Capesius,Coloncancer in Luxembourg:a nationalpopulation-based data report, , BMC Cancer , vol. , article ,

.[ ] I. K. Larsen and F. Bray, rends in colorectal cancer incidence

in Norway : an interpretation o the temporal pat-terns by anatomic subsite, International Journal of Cancer , vol.

, no. , pp. , .[ ] F. Y. Li and M. D. Lai, Colorectal cancer, one entity or three,

Journals of Zhejiang University-Science B, vol. , pp. ,.

[ ] M. L. Slattery, K. Curtin, R. K. Wolff et al., A comparison o colon andrectal somatic DNAalterations, Diseases of the Colon& Rectum, vol. , no. , pp. , .

[ ] World Cancer Research Found and American Institute orCancer Research, Food, Nutrition, Physical Activity, and thePrevention of Cancer: A Global Perspective, Washington, DC,USA, .

[ ] World Cancer Research Found and American Institute orCancer Research, Food, Nutrition, Physical Activity, and thePrevention of Colorectal Cancer. Continuous Update Project,Colorectal Cancer Report , Summary , Washington, DC,USA, .

[ ] M. Huncharek, J. Muscat, and B. Kupelnick, Colorectal cancerrisk and dietary intake o calcium, vitamin D, and dairy

products: a meta-analysis o , cases rom observationalstudies, Nutrition and Cancer , vol. , no. , pp. , .

[ ] J.A. Bergsma-Kadijk, P. vant Veer, E. Kampman, andJ. Burema,Calcium does not protect against colorectal neoplasia, Epi-demiology , vol. , no. , pp. , .

[ ] S. C. Larsson and A. Wolk, Meat consumption and risk

o colorectal cancer: a meta-analysis o prospective studies,International Journal of Cancer , vol. , no. , pp. ,.

[ ] D. S. Chan, R. Lau, D. Aune et al., Red and processed meatand colorectal cancer incidence: meta-analysis o prospectivestudies, PLoS ONE, vol. , no. , article e , .

[ ] E. Cho, S. A. Smith-Warner, D. Spiegelman et al., Dairy oods,calcium, and colorectal cancer: a pooled analysis o cohortstudies, Journal of the National Cancer Institute , vol. , pp.

, .[ ] Y. Park, D. J. Hunter, D. Spiegelman et al., Dietary ber

intake and risk o colorectal cancer: a pooled analysis o prospectivecohort studies, Te Journalof theAmerican Medical Association, vol. , pp. , .

[ ] E. Cho, S. A. Smith-Warner, J. Ritz et al., Alcohol intake andcolorectal cancer: a pooled analysis o cohort studies, Annalsof Internal Medicine, vol. , pp. , .

[ ] A.L. Klatsky, M. A.Armstrong, G. D.Friedman, andR. A. Hiatt,Te relations o alcoholic beverage use to colon and rectalcancer, American Journal of Epidemiology , vol. , no. , pp.

, .[ ] G. N. Stemmermann, A. Nomura, and P. H. Chyou, Te

in uence o dairy and nondairy calcium on subsite large-bowelcancer risk, Diseases of the Colon and Rectum, vol. ,no. ,pp.

, .[ ] A. A. Razzak, A. S. Oxentenko, R. A. Vierkant et al., Alcohol

intake and colorectal cancer risk by molecularly de ned sub-types in a prospectivestudy o older women, Cancer PreventionResearch, vol. , pp. , .

[ ] E. Giovannucci, E. B. Rimm, M. J. Stamp er, G. A. Colditz, A.Ascherio, and W. C. Willett, Intake o at, meat, and ber inrelation to risk o colon cancer in men, Cancer Research, vol.

, no. , pp. , .[ ] A. Chao,M. J.Tun, C. J.Connell et al., Meat consumption and

risk o colorectal cancer, Te Journal of the American Medical Association, vol. , pp. , .

[ ] S. C. Larsson, J. Rafer, L. Holmberg, L. Bergkvist, and A. Wolk,Red meat consumption and risk o cancers o the proximalcolon, distal colon and rectum: the Swedish Mammography Cohort, International Journal of Cancer , vol. , no. , pp.

, .[ ] . Norat, S. Bingham, P. Ferrari et al., Meat, sh, andcolorectal

cancer risk: the European Prospective Investigation into cancerand nutrition, Journal of the National Cancer Institute , vol. ,no. , pp. , .

[ ] Y. Sato, N. Nakaya, S. Kuriyama, Y. Nishino, Y. subono, andI. suji, Meat consumption and risk o colorectal cancer inJapan: the Miyagi cohort study, European Journal of Cancer Prevention , vol. , no. , pp. , .

[ ] A. J. Cross, L. M. Ferrucci, A. Risch et al., A large prospectivestudyo meat consumptionand colorectal cancerrisk: an inves-tigation o potential mechanisms underlying this association,Cancer Research, vol. , no. , pp. , .


13/15

Journal o Oncology

[ ] A. Shin, J. Joo, J. Bak et al., Site-speci c risk actors orcolorectal cancer in a Korean population, PLoS One, vol. ,article e , .

[ ] K. A. Steinmetz, L. H. Kushi, R. M. Bostick, A. R. Folsom, andJ. D. Potter, Vegetables, ruit, and colon cancer in the Iowawomens health study, American Journal of Epidemiology , vol.

, no. , pp. , .[ ] C. S. Fuchs, E. L. Giovannucci, G. A. Colditz et al., Dietary ber

and the risk o colorectal cancer and adenoma in women, NewEngland Journal of Medicine, vol. , no. , pp. , .

[ ] S. C. Larsson, E. Giovannucci, L. Bergkvist, and A. Wolk,Whole grain consumption and risk o colorectal cancer: apopulation-based cohort o women, British Journal of Cancer , vol. , no. , pp. , .

[ ] A. M. Nomura, J. H. Hankin, B. E. Henderson et al., Dietary ber and colorectal cancer risk: the multiethnic cohort study,

Cancer Causes and Control , vol. , no. , pp. , .[ ] A. Schatzkin, . Mouw, Y. Park et al., Dietary ber and whole-

grain consumption in relation to colorectal cancer in the NIH-AARP Diet and Health Study, American Journal of Clinical

Nutrition , vol. , no. , pp. , .[ ] G. C. Kabat, J. M. Shikany, S. A. Beres ord et al., Dietary

carbohydrate, glycemic index, and glycemic load in relation tocolorectal cancer risk in the Womens Health Initiative, Cancer Causes and Control , vol. , no. , pp. , .

[ ] R. Egeberg, A. Olsen, S. Lof et al., Intake o wholegrainproducts and risk o colorectal cancers in the Diet, Cancer andHealth cohort study, British Journal of Cancer , vol. , no. ,pp. , .

[ ] N. Murphy, . Norat, P. Ferrari et al., Dietary bre intakeand risks o cancers o the colon and rectum in the EuropeanProspective Investigation into Cancer and Nutrition (EPIC),PLoS One, vol. , article e , .

[ ] M. L. McCullough, A. S. Robertson, C. Rodriguez et al.,Calcium, vitamin D, dairy products, and risk o colorectalcancer in the Cancer Prevention Study II Nutrition Cohort(United States), Cancer Causes and Control , vol. , no. , pp.

, .[ ] S. C. Larsson, L. Bergkvist, and A. Wolk, High- at dairy ood

and conjugated linoleic acid intakes in relation to colorectalcancer incidencein the SwedishMammographyCohort, Amer-ican Journal of Clinical Nutrition, vol. , no. , pp. ,

.[ ] S. C. Larsson, L. Bergkvist, J. Rutegard, E. Giovannucci, and A.

Wolk, Calcium and dairy ood intakes are inversely associatedwith colorectal cancer risk in the Cohort o Swedish Men, American Journal of Clinical Nutrition, vol. , no. , pp.

, .[ ] P. erry, J. A. Baron, L. Bergkvist, L. Holmberg, and A. Wolk,Dietary calcium and vitamin D intake and risk o colorectalcancer: a prospective cohort study in women, Nutrition and Cancer , vol. , no. , pp. , .

[ ] K. Wu, W. C. Willett, C. S. Fuchs, G. A. Colditz, and E.L. Giovannucci, Calcium intake and risk o colon cancer inwomen and men, Journal of the National Cancer Institute, vol.

, no. , pp. , .[ ] A. Flood,U. Peters, N. Chatterjee, J.V. Lacey Jr., C. Schairer, and

A. Schatzkin, Calcium romdiet and supplements is associatedwith reduced risk o colorectal cancer in a prospectivecohort o women, Cancer Epidemiology Biomarkers and Prevention, vol.

, no. , pp. , .

[ ] J. Ishihara, M. Inoue, M. Iwasaki, S. Sasazuki, and S. sugane,Dietary calcium, vitamin D, and the risk o colorectal cancer, American Journal of Clinical Nutrition, vol. , no. , pp.

, .[ ] A. H. Wu, A. Paganini-Hill, R. K. Ross, and B. E. Henderson,

Alcohol, physical activity and other risk actors or colorectalcancer: a prospective study, British Journal of Cancer , vol. ,no. , pp. , .

[ ] E. Giovannucci, E. B. Rimm, A. Ascherio, M. J. Stamp er, G.A. Colditz, and W. C. Willett, Alcohol, low-methioninelow-

olate diets, and risk o colon cancer in men, Journal of theNational Cancer Institute , vol. , no. , pp. , .

[ ] A. Pedersen, C. Johansen, andM. Grnbk, Relationsbetweenamount and type o alcohol and colon and rectal cancer in aDanish population based cohort study, Gut , vol. , no. , pp.

, .[ ] M. Akhter, S. Kuriyama, N. Nakayaet al., Alcoholconsumption

is associated with an increased risk o distal colon and rectalcancer in Japanese men: the Miyagi Cohort Study, European Journal of Cancer , vol. , no. , pp. , .

[ ] P. Ferrari, M. Jenab, . Norat et al., Li etime and baselinealcohol intake and risk o colon and rectal cancers in theEuropean prospective investigation into cancer and nutrition(EPIC), International Journal of Cancer , vol. , no. , pp.

, .[ ] B. W. C. Bongaerts, P. A. van den Brandt, R. A. Goldbohm, A. F.

P. M. de Goeij, and M. P. Weijenberg, Alcohol consumption,type o alcoholic beverage and risk o colorectal cancer atspeci c subsites, International Journal of Cancer , vol. , no.

, pp. , .[ ] J. Y. Park, C. C. Dahm, R. H. Keogh et al., Alcohol intake and

risk o colorectal cancer: results rom the UK Dietary CohortConsortium, British Journal of Cancer , vol. , no. , pp.

, .[ ] M. Egger, G. D. Smith, and J. A. Sterne, Uses and abuses o

meta-analysis, Clinical Medicine, vol. , pp. , .[ ] J. Lin,S. M.Zhang, N.R. Cook, J. E.Manson,I. M.Lee, and J. E.

Buring,Intakes o calcium andvitamin D andrisk o colorectalcancer in women, American Journal of Epidemiology , vol. ,no. , pp. , .

[ ] S. A. Bingham, R. Hughes, and A. J. Cross, Effect o white versus red meat on endogenous N-nitrosation in the humancolon and urther evidence o a dose response, Journal of Nutrition , vol. , no. , pp. S S, .

[ ] A. C. Povey, C. N. Hall, A. F. Badawi, D. P. Cooper, and P.J. OConnor, Elevated levels o the pro-carcinogenic adduct,O( )-methylguanine, in normal DNA rom the cancer prone

regions o the large bowel,Gut , vol. , no. , pp. , .[ ] A. Hague, A. M. Manning, K. A. Hanlon, L. I. Huschtscha, D.Hart, and C. Paraskeva, Sodium butyrate induces apoptosisin human colonic tumour cell lines in a p -independentpathway: implications or the possible role o dietary bre inthe prevention o large-bowel cancer, International Journal of Cancer , vol. , no. , pp. , .

[ ] W. Scheppach and F. Weiler, Te butyrate story: old winein new bottles? Current Opinion in Clinical Nutrition and Metabolic Care, vol. , no. , pp. , .

[ ] J. M. Wong, R. de Souza, C. W. C. Kendall, A. Emam, and D.J. Jenkins, Colonic health: ermentation and short chain atty acids, Journal of Clinical Gastroenterology , vol. , no. , pp.

, .


14/15

Journal o Oncology

[ ] W. E. Roediger, Role o anaerobic bacteria in the metabolicwel are o the colonic mucosa in man, Gut , vol. , no. , pp.

, .[ ] S. A. Lamprecht and M. Lipkin, Chemoprevention o colon

cancer by calcium, vitamin D and olate: molecular mecha-nisms, Nature Reviews Cancer , vol. , no. , pp. , .

[ ] G. Poschl and H. K. Seitz, Alcohol and cancer, Alcohol and Alcoholism, vol. , no. , pp. , .[ ] U. A. Simanowski, F. Stickel, H. Maier, U. Gartner, and H. K.

Seitz, Effect o alcohol on gastrointestinal cell regenerationas a possible mechanism in alcohol-associated carcinogenesis, Alcohol , vol. , no. , pp. , .


15/15

Submit your manuscripts athttp://www.hindawi.com

subsite-specific dietary risk factors for colorectal cancer: a review of cohort studies

Documents