Wieslaw J edrychowski*, K aren Steindorf**, Tadeusz Popiela***, J urgen Wahrendorf**, Beata Tobiasz-Adamczyk*, J an K ulig*** and Agnieszka Penar*.

RISK OF COLORECTAL CANCER DUE TO ALCOHOL CONSUMPTION AT THE DEFICIENT LEVEL OF VITAMIN INTAKES. A HOSPITAL-BASED CASE-CONTROL STUDY IN POLAND.

From *) Chair of Epidemiology and Preventive Medicine, Coll.Med. J agiellonian University in K rakow, 7, Kopernika street,Poland.

**)German Cancer Research Center, Unit of Environmental Epidemiology, Heidelberg, Germany.

***) First Surgical Clinic, University Hospital, Coll.Med. J agiellonian University, 40, Kopernika street, K rakow, Poland.

Acknowledgements:

The project has been financially supported bythe grant from the State Committee forResearch in Poland (No 4-PO5D-025-11), andpartly by the German Federal Ministry of Health(program for the German-Polish co-operation inthe field of health and medicine).

Food frequency questionnaire together with the computer program used in the study was the adapted version of the German food questionnaire, kindly obtained from Dr Heiner Boeing

Alcohol drinking habits in the last 30 years:

•A great variability of alcohol consumption across countries,• A big increase in alcohol consumption between 1960-1975,• The range of drinking habits has expanded and people are more tolerant of alcohol,• Increase in alcohol-related mortality and morbidity and drunken driving.

Needs for future research:

1. Very little information is available on some high-risk groups,2.The clustering of potentially harmful habits such as drinking, smoking, taking pills, andmalnutrition (interaction with other psychoactive sunstances),3. Feasibility, methods of measurements, and priorities.

Alcohol, in addition to tobacco and diet, is amajor environmental factor. Numerousepidemiological studies have shown that thereis a strong association of alcohol consumptionwith cancers of upper aerodigestive tract andliver, pancreas and breast. Someepidemiological studies found also theassociation between current and pastalcohol drinking and colorectal cancer,but the strength of association at this siteis not very strong.

It is disputable whether alcohol drinkers runthe higher risk of colon cancer only becausethey eat less vegetables and fruits that containprotective antioxidant vitamins or whether thecancerogenic effect from alcohol isindependent of dietary deficiencies. Thus, thepurpose of this study was to assess ahypothetical interaction that involvessimultaneous exposure to alcohol andconsumption of micro-nutrients that may haveprotective property against colorectal cancer.

Objective of the study

The purpose of the study was to assess the relation betweensimultaneous exposure to alcoholand consumption of micronutrientsthat may have protective propertyagainst colorectal cancer.

Methods The hospital based case- control study of colorectal cancer hasbeen carried out between January 1998and November 1999 in Krakow, Poland. Intotal, 180 incident cases of colorectalcancer confirmed by histopathology wererecruited from the University Hospital inKrakow. The equal number of controlsindividually matched by gender and age(+/-5 years) were chosen amongstpatients with no history of cancer fromthe same hospital.

An interviewer-administered foodfrequency questionnaire for 148 fooditems combined with quantity of foodseaten was used to assess the usualdietary pattern.

Results The data confirmed the consistent inverse association between intake of retinol, thiamine or antioxidant micronutrients(carotene, vitamin C and E) and the occurrence of colorectalcancer when the analysis has been carried out on quartile intakedata. Multivariate analyses done on micronutrients as continuousvariables, showed for retinol intake only a significant OR trend(OR=0.54; 95%CI: 0.38-0.77).Alcohol intake appeared to be an important risk factor for thiscancer site and the risk increased with the amount of pure alcoholintake. The group with multivarious deficient intakes of retinol,caroten and vitamin C and E, but with higher consumption ofalcohol run a noticeably higher risk of colorectal cancer (OR=6.79;95%CI: 2.08-22.18). However, the risk was markedly lower amongthose alcohol drinkers who reported high intake ofmicronutrients (OR=1.35; 95% CI: 0.39-4.67).

An interviewer-administered food frequency questionnaire (FFQ)combined with quantity of foods eaten was used to assess the usualdietary pattern for 148 beverage and food items. For each food orbeverage item, a commonly used unit or portion size was specified,and participants were asked how often on average over the one yearthey consumed that amount of each food. The dietary interview focusedon the reference period that was defined as one year prior to 5 yearsbefore diagnosis for cases or prior to corresponding date of hospitaladmission for the controls.

Nutrient intakes were calculated by multiplying the consumptionfrequency of each unit of food by the nutrient content of the specifiedportion, using composition values from the Polish Institute of Nutrition.

Estimates of the adjusted* odds ratios (ORs) and 95% Confidence Intrrvals for corelectal cancer by quartile intake of retinol mg/day

* adjusted by energy intake, total fibre intake and vocational actyvity

1 (<0.60) 2 (0.60-<0.83) 3 (0.84-<1.43) 4 (>=1.44)0

0.2

0.4

0.6

0.8

1

1.2

OR

Quartiles of intake of retinol mg/day

st nd rd th

Estimates of the adjusted* odds ratios (ORs) and 95% Confidence Intrrvals for corelectal cancer by quartile intake ofcarotene mg/day

1 (<2.10) 2 (2.10-<2.58) 3 (2.59-<3.28) 4 (>=3.29)0

0.2

0.4

0.6

0.8

1

1.2

OR

Quartiles of intake of carotene mg/day

* adjusted by energy intake, total fibre intake and vocational actyvity

Estimates of the adjusted* odds ratios (ORs) and 95% Confidence Intrrvals for corelectal cancer by quartile intake of vitamin E mg/day

1 (<7.34) 2 (7.34-<8.69) 3 (8.70-<10.59) 4 (>=10.60)0

0.2

0.4

0.6

0.8

1

1.2OR

Quartiles of intake of vitamin E mg/day

st nd rd th

* adjusted by energy intake, total fibre intake and vocational actyvity

Estimates of the adjusted* odds ratios (ORs) and 95% Confidence Intrrvals for corelectal cancer by quartile intake of vitamin B 1 mg/day

* adjusted by energy intake, total fibre intake and vocational actyvity

1 (<0.81) 2 (0.81-<0.94) 3 (0.95-<1.16) 4 (>=1.17)0

0.2

0.4

0.6

0.8

1

1.2OR

Quartiles of intake of vitamin B 1 mg/day

Estimates of the adjusted* odds ratios (ORs) and 95% Confidence Intrrvals for corelectal cancer by quartile intake of vitamin B 2 mg/day

1 (<1.05) 2 (<1.05-<1.27) 3 (1.28-<1.57) 4 (>=1.58)0

0.5

1

1.5

2

2.5

3

3.5OR

Quartiles of intake of vitamin B 2 mg/day

* adjusted by energy intake, total fibre intake and vocational actyvity

Estimates of the adjusted* odds ratios (ORs) and 95% Confidence Intrrvals for corelectal cancer by quartile intake of vitamin C mg/day

1 (<63.61) 2 (63.61-<79.81) 3 (79.82-<94.05) 4 (>=94.06)0

0.2

0.4

0.6

0.8

1

1.2

Quartiles of intake of vitamin C mg/day

* adjusted by energy intake, total fibre intake and vocational actyvity

OR

Vocational physical activity

Not working46.1%

Light work 36.1%

Moderate and17.8%

Not working41.1%

Light work in 32.8%

Moderate and 26.1%

CASES CONTROLS

Smoking

Never smokers48.3%

Ex-smokers34.4%

Current smokers17.2%

Never smokers41.1%

Ex-smokers35.6% Current smokers

23.3%

CASES CONTROLS

Leisure time activity (hours of watching TV per day, in tertiles)

Low (>= 2 hours/day)

38.3%

Medium 38.9%

High 22.8%

Low 27.8%

Medium39.4%

High 32.8%

CASES CONTROLS

Effect of alcohol intake on colorectal cancer

1 quartile (low) 2 3 4 quartile (high)

Alcohol intake (in quartiles)

0

2

4

6

8

10

12

14OR

st nd rd th

Adjusted* ORs and 95% confidence intervals of colorectal cancer by combined levels of micronutrients**

6.8

2.51.4

1.0 0.6 0.5

Low Medium High0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

OR

* Adjusted for education, total energy and total fibre intake, occupation phycical activity.

** Combined micronutrient level defined by median cut-off points:Low: subjects with each micronutriet intake values below the median for each micronutrient,High: subjects with an intake value above the median for each micronutrient,Medium: subjects with neither a low nor a high intake, i.e., mixed intake.

Median values for retinol= 0.83, vit.C= 80.0, vit. E= 8.7 and carotene= 2.6, alcohol= 0.56 g/day

High consumption alcohol (5.6> g/day)

Low consumption alcohol (0-5.6 g/day)

Micronutrients intake

(0.2-1.6) (0.12-2.29)

-

(2.1-22.2)

(0.9-7.0)

(0.4--4.7)

Conclusion The data support the

hypothesis that higher consumption of

alcohol, when combined with low intakes

of retinol and antioxidant vitamins may

considerably increase the risk of

colorectal cancer.