Main Menu            Health Library                Antioxidant - a Global Analysis

      VITAMIN A, BETA-CAROTENE & CANCER

Abstract:  Vitamin A used in amounts of about 5,000 IU appears usually to reduce the risk of most causes of cancer by about 34%.  This result is derived from a total of 46 comparisons from 34 studies.  A listing of results of this research follows.  A similar risk benefit was found for seven types of cancer, and this benefit may apply to all cancer. Beta Carotene is taken in the analysis as 3 IU equivalent for 1 IU of Vitamin A.   But Vitamin A in amounts of above about 5,000 IU appears to be harmful, users should not take more that this amount or of its equivalent of 15,000 IU of Beta-carotene.  Most research was for diet amounts, and typical diets include 2,500 IU of Vitamin A. Thus diet supplements of Vitamin A may have marginal benefit for those who have good dietary habits. and will produce minimal or no benefit for diet amounts much above 4,000 IU. 

 

See other papers of antioxidants and CV disease for Vitamin EVitamin AVitamin C; and Selenium ;papers of antioxidants and cancer for Vitamin EVitamin C; and Selenium. and Antioxidants, Death from All Causes and Antioxidants, a Global Analysis.

 

As described more completely for Vitamin A and Coronary Heart disease, Vitamin A designates a family of vitamins that are fat soluble. Most tables of vitamin values in foods are expressed in IU units of Vitamin A.  But most research is based on values of Beta-carotene, and most users take Vitamin A supplements as Beta-carotene. The body converts Beta-carotene to Vitamin A.  But it is claimed by some experts that the body converts Beta-carotene to Vitamin A only in amounts needed.  Typical conversion ratios from IU of Beta-carotene to IU equivalents in Vitamin A vary widely, and typically from from 1 to 2 to up to 1 to 6.  Life Ahead assumes a conversion value of 3 Beta-carotene to 1 Vitamin A for both the research results in Table A following and for user values in the program.  But it should be kept mind that Life Ahead values Beta Carotene IU's from supplements as of only 1/3rd the value of IU's of Vitamin A cited in usual tables food nutrient values. Values of Beta-carotene are sometimes expressed in mg, and 1 mg of Beta-carotene equals about 1600 IU of Beta-carotene.

The observation study research results found for Vitamin A and Beta-Carotene are summarized in Table B appended.  Differences in amounts of vitamin used are expressed in IU's of Vitamin A.  30 of 34 observation study measurements or about 90% of risk measurements showed a benefit.  An average risk measured for those taking Vitamin A from all 34 observation studies is 0.66. or a reduction in risk of 34%.  Three studies of risk for measurements of Vitamin A in blood averaged only 0.42 for a 58% reduction in risk.   These results were consistent for types of cancer as breast cancer 0.66 from 8 comparisons, 0.75 from 7 ovarian and cervical studies; and 0.66 for 9 comparisons of lung cancer.  See Table A for a summary of these results. An average risk value could be about 0.66 for a difference of about 4000 IU of Vitamin A. 

                  

                                           Table A

    Effect of Vitamin A in Reducing Risk of Types of Cancer

  Cancer Site  Comparisons      Avg Risk Ratio
 Breast 8 0.66
 Prostate 6 0.80
 Ovarian & Cervical 7 0.75
 Lung 9 0.66
 Colorectal 1 0.60
 Throat and Oral 1 0.63
 All Cancer 2 0.50
          Total 34 0.66

A concern is noted for the results from the large Physicians study.  This group that probably used a more healthful diet than usual probably were using reasonably adequate Vitamin A in diets, and this may be why the substantial supplement addition did not contribute. Life Ahead now would compute little advantage for taking Vitamin A supplements for those having much above 3000 IU in their diets.  Because nearly any antioxidant cause of cancer will be duration dependent a typical duration of use in the research to date probably was about 10  years.  This is an insufficient time for full potential benefit from an antioxidant to develop.

The results of the clinical studies show no beneft of vitamin A on risk of cancer  In fact, an average risk from 18 different measurememts of risk on various cancer causes showed negative benefit of 1.12.   A number of these clinical results on presumab ly healthy people were obtained on smokers in Finland in a single large study.  Antioxidants do not benefit smokers.   The short duration of the clinical studies could have been factor, and the Physicians in RA2 as before probably were getting enough Vitamin A in their diets.  As for the results of Vitamin A on heart disease, a possible reason for this lack of measured benefit is that use of the very high amounts of Vitamin A used in these studies that averaged 3 times the maximum levels used in the observation studies became counter-productive.  Observation study BC1 of Hunter comfirmed directly the much higher risks for use of very large amounts of vitamin A.  The amounts of Vitamin A used in these clinical studies of an average 12,000 equivalent IU for vitamin A are far above the 4-5,000 maximum for the RDA, and above the 5,000 IU level that can provide other health negatives such as bone loss. 

Data confirming this hypothesis comes from information in Hunter's large study BC1. Adding supplements of Vitamin A for those participants that had very low Vitamin A from food provided a benefit.  But supplements added to those having higher amounts of Vitamin A in in food provided no further benefit.  A maximum value of Beta-carotene benefit appears obtained at about 15,000 IU, or the equivalent of 5,000 IU of preformed Vitamin A as used herein. Risks for a total of Vitamin A beyond 10,000 IU actually reversed to higher values in this table of results.  This same maximum amount of Vitamin A also appeared indicated useful for heart disease.  The research to date shows that use of the very large amounts of Beta-carotene employed in the clinical studies probably not only was of no value but produced reverse or negative health benefit.

Life Ahead now values Vitamin A for both heart disease and cancer only for amounts up a total of 5,000 IU equivalent of preformed Vitamin A (or 15,000 IU of Beta-carotene) in a total of food plus supplements.  It thus provides a value for a Beta-carotene or Vitamin A supplement only when amounts in food are deficient vs. the present RDA values.  Again, Life Ahead values Vitamin A as one of 4 antioxidants, and will give no credit to any further amount of any of these antioxidants when a total limiting value of all antioxidants is attained.  As a general guide, vitamin supplements of Vitamin A or Beta-carotene may not produce much benefit in reducing risk of cancer because diet can supply much of the amount needed for best benefit.

Life Ahead values vitamin A and cancer as dependent on years of exposure as is done for other antioxidants. Study BC1 confirms the time related effect of vitamin A on cancer.  A formula derived for exposure and years of use with maximum value for Vitamin A of 5,000 IU is:

  Risk ratio of Vitamin A and cancer, with yrs of exposure to vitamin or its supplements and amt as amount of Vitamin A in IU

    risk ratio = exp( - 0.00038 * yrs * amt ^ 0.5 )           

  For use of beta carotene, at 3 units per unit of vitamin A  and maximum 15,000 IU this becomes         

   risk ratio - exp ( - 0.00022 * yrs * amt ^0.5 )

                                                                                          Table B

EFFECT of VITAMIN A and BETA-CAROTENE on CANCER

No

Study

Population

Sex

RR    

 

Error Margin

     Amt Diff,

Vit A IU

Avg Yrs

ratio per

yr

Notes

  OBSERVATION STUDIES      
ALL CANCER
ALC1 Wald, NJ, Brit J Cancer 1986, 57:428 271 cases vs. 533 controls of 22,000 in Britain M 0.6 All cancer from  top 2 quintiles vs bottom quintile
ALC2 Eichholzer, M EXS 1992,  62:398 204 cases of 3000 M&W 0.40 P<0.01 3,400E 12 0.92 From Mean Amount in Blood
BREAST CANCER

BC1

Hunter, DJ N Engl J Med 1993; 329

1439 of 89,000 ages 34-59 at start

W

0.69

0.84

1.23

1.63

0.82

0.68

0.68-0.95

0.48-1.45

0.78-1.95

0.68-3.95

0.50-1.34

0.36-1.26

<1530 to >7460

8000 suppl

8000-12000

13000-22000

 

11-14

 

 

 

2-4

5-7

0.98

0.98

 

 

0.94

0.94

 Suggests max benefit at 10,000

 

 

duration supplements

         same

BC2 Kushi LH, Cacner Causes Control 1993, 4:29 879 cases,  34,200 postmenopausal in Iowa W 0.90 no effect in trend, p-0.2 low vs high quintile
BC3 Freundenheim JL, J Natl Cancer Inst 1996, 88:340 297 cases and 311 controls in NY State

W

0.46

0.28-0.74

6,000 IU beta C

=2000 Vit A

Premenopausal women

quartiles

BC4

Franceschi, S; Eur J Cancer Prev 1997 6:535

2569 cases and 2588 controls, Italy

W

0.84

4,200E

7-10

0.98

Quintiles,diet

BC5 Zhang S, J Nat Cancer Inst 1999, 91.547

2697 events of 83,000 Nurses in US, dietary recall, quintiles

W

0.83

0.94

 

0.42

0.55

0.66-1.04

0.81-1.09

 

0.21-0.83

0.32-0.97

5932 IU dif in Beta Carotene 3050 diff Vitamin A

14

premenopausal

post menopausal

many variable adjustmt

Disease family history

Alcohol > 15 gm/day

BC6 Adzersen KH, Nutr Cancer 300, 46-131 310cases vs 353 controls in Germany   W 0.46 0.27-0.80

  BC7

Cho E, Cancer Eidemiol Biomarkers prev 2003, 12:713

704 cases from 90,000 premenopausal US nurses age 26-46

W

0.28

0.12-0.62

Vitamin A

8

 

Risk for Vitamin A for smokers only. No effects for other antiox on population

BC8 Cui Y, Am J Clin Nutr 2008,87:1009 2509 cases from 84,800 in Women's Hlt Initiative   W 0.78 0.66-0.94 7.6 High vs Low Quintile

 

 

  PROSTATE CANCER
P1 Enger SM, Cancer Epidemiol Biomarkers Prev 1996, 5:147 488 matched pairs with food frequency analysis. by quartiles M&W 0.6 0.41-1.00

p=0.04

Diet + Supplements     Substantially adjusted comparison.  For adenomas, not cancer
P2 Kristal, AR; Cancer Epidemiol Biomarker Prev 1999; 8;887 697 Cases and 656 controls M 0.59 0.32-1.04 5,000E 7 0.93 Supplement 7 times per week vs. none
P3 Schuurman AG, Cancer Causes Control 2002, 13:573 642 cases of 58,000 in Netherlands M 0.76     6.3    
P4 Stevens PL, Cancer Causes Control 2005,16:643 5571 deats on 445,000 US men M  1.07   0.99-1.15     MultiVitamin supplements only, not Just Vit E
P5 Kirsch VA, Natl Cancer Inst 2006, 98:245

1338 cases of 29,300 in US av age 63 at start

M

0.82

0.96

0.65-1.04

0.80-1.15

2500E

6,000

8

8

0.97

0.99

Supplement use

Diet only

         

 

 

 

 

 
  LUNG CANCER
L1 Menkes, MS N Engl J Med 1988; 315:1250 99Cases and 196 controls M&W 0.23

0.07-0.72

4,200 E

 

 

Quintiles of blood, Beta-carotene
L2 Mayne, ST J Natl Cancer Inst 1994, 86:33 413 Cases and controls, all non-smokers M&W 0.70

0.50-0.99

4,000E

7-10

0.96

Dose dependent diet
L3 Comstock GW, Cancer Epidemiol Biomarkers Prev 1997, 6:907 258 cases and 515 control in Maryland M&W 0,44 p=0.002       Quintiles in blood, avg = 12.4 ug/dl
L4 Stefani, ED, Nutr Cancer, 1999, 34:100 540 Cases vs. 541 controls M&W 0.43

0.29-0.64

3,400E

7

0.88

Quartiles of amount, food
L5 Zhou B, Oncol Rep 1999, 6:139 290 cases and controls in China  W 0.84 n/a       Intake of beta carotene
L6 Mannisto S, Cancer Biomarkers Prev 2004, 13:40

3155 events in 400,000  pooling 7 studies

M&W

0.65

0.98

 

0.58-0.73

0.82-1.06

 

 

 

age adjusted only

heavy adjustment

quintiles of beta car

L7 Cho, E, Int J Cancer 2006, 118:970 Pooled Analysis of 8 major studies M&W 0.96 0.83-1.01   6-16   Multivarient basis, results larger age adjusted
L8 Yuan FM, Cancer Epidemiol Biomarkers Prev 2001, 10:767 2009 lung cancer cases with 622 controls in China M&W 0.74 0.42-1.30 n/a 12   quintiles of beta carotene, avg 11.2 ug/ml
                   
  COLORECTAL CANCER              
CO1 Enger SM, Cancer Epidemiol Biomarkers Prev 1996, 5:147 488 matched pairs, age 50-74, food frequency , M&W 0.6 0.41-1.0       Ademonas only, case control
 
  OVARIAN CANCER
O1

Slattery, ML, Am J Epidemiol 1989, 130:497

85 cases, 492 controls    W 0.5

0.3-1.0

4,200E

7

0.91

 
O2

Fairfield, FM Cancer 2001; 92:2318

301 Events from 80,000 Nurses     W

1.04

 

0.72-1.51

 

4,200E

 

20

1.00

Vitamin A, Quintiles

 

O3

Pan SY, Cancer Epidemiol Biomarkers Prev 2004, 13:1521

441 cases and 2125 controls in Canada

 W

0.31

0.76

1.14

0.11-0.91

0.25-2.29

0.73-1.76

 

 15

7.5

<1 yrs

 

Beta carotene supplemt

note impressive effect of time of use on all supplement uses

O4

Silvera FA, Cancer Epidemiol Biomarkers Prev  2006, 15:395

502 cases from 49,600 in Canada age start 49,

W

0.79

 0.53-1.16

abt 5500 to 12,500 IU of Vitamin A

16.4

 

quartriles of diet and supplements

 

 

CERVICAL CANCER

 

 

 

 

 

 

 

 

CV1

Slattery, ML, Epidemiology 1:8

266 Cases, 408 controls

W

0.71

 

3,400E

 

 

Quartiles of Diet

 

 

BLADDER CANCER                      

B1

Zeeger, MP Br J Cancer 2001:85:977

569 Events of 3100 coholrt M&W 1.16   4,000E 10 1.01 Beta-carotene
B2

Castelao,JE, Int J Cancer, 2004, 110:417

1552 cases and controls in California M&W 0.58 p-0.001       Total Caratonoids
 
 

THROAT & ORALCANCER

OR1

Negri, E; Int J Cancer 2000, 86:122

754 Events vs. 1775 controls, Italy & Switzerland M&W 0.61   4,200E 10E 0.95 Quintiles
 

 CLINICAL STUDIES on HEALTHY PEOPLE

               ALL CANCER

RA1

Blot WJ, Am J Clin Nutr 1995, 62:1424S

2127 deaths of 29,600 population in Linxian China

M

W

0.93

0.79

0.77-1.12

0.64-0.98

45 IU Vit E + 5 mg selenium+ est 8,000 IU Vit A

5.3

 

Results on Healthy Persons.  See results on death from all causes

RA2

Hennekens, CH, N Engl J Med 1996, 334:1145``

22,000  Physicians, US, in two groups of 11,000

M

0.98

0.88

1.08

0.91-1.08

0.75-0.99

0.95-1.22

13,700

12

6

11

1.00

0.961.01

12 year trial

yrs 5-7

yrs 10-12

supplement

RA3

Lee, IM, J Natl Cancer Inst 1999, 91:2102

Groups of 19,300

W

1.11

0.78-1.58

10,700

5.5

1.02

Study 5-6 years

supplement

RA4 Coulter ID, J Gen Intern Med 2006, 21:735 29,000 smokers in Finland M 1.08 0.89-1.32 10,700 Vit A+ 75 IU Vit E 6.1   Note: RR for Vit E= 0.91 (0.74-1.12)
       

 

 

 
  COLORECTAL CANCER
RC1 Albanes, D, Cancer Causes Control 2000, 11:197

29,000 smokers in Finland, age 50-69

M

1.05

0.75-1.47

20 mg beta C

32,000 IU

5.5

1.01

Study 5-6 years

supplement

RC2 N Engl J Med 1994, 330:1029

149 Cases of 29,000 Smokers in Finland

M

1.04

 

20 mg beta C, 32,000 IU

6

1.01

 

 
  COLORECTAL ADENOMAS    

 

RCV1 Greenbert, ER, N Engl J, 1994 Med 331:141 215 each to 4 groups M&W 1.01 0.85-1.20

   13,700

4 1.00

4 year study

supplement

                   
  LUNG CANCER                
RL1 N Engl J Med 1994, 330:1029 878 Cases of 29,000 Smokers in Finland M 1.18 0.97-1.36 20 mg,  32,000 IU or 10,700 Vit A 5-8 1.03 high supplement amount
RL2 Omenn GS, N Engl J Med 1996, 334:1150 388 cases Lung Cancer on 18.300 smokers M 1.28 1.04-1.57 25,000 IU Vit A + 30 mg/day beta car 4 1.063 very high supplement amount includes 48,000 IU Beta Carotene
RL3 Virtamo J, JAMA 2003, 290:476 29,000 smokers in Finland M 1.15 0.91-1.45 75 IU Vit E 32,000 IU Beta Car = 10,700 Vit A 6.1 1.023 Note:  RR 75 IU Vit E = 0.93
                   
  PROSTATE CANCER                
RP1 N Engl J Med 1994, 330:1029 250 Cases of 29,000 Smokers in Finland M 1.23   20 mg, 32,000 IU 6 1.03 very high amount
RP2 Virtamo J, JAMA 2003, 290:476 29,000 smokers in Finland M 1.23   20 mg, 32000 IU 14   Followup fo 14 years, average results
                   
  BLADDER CANCER                
RB1 N Engl J Med 1994, 330:1029 155 Cases of 29,000 Smokers in Finland M 1.04   20 mg, 32,000 IU 6 1.01  
                   
RS1 STOMACH CANCER                
RS2 N Engl J Med 1994, 330:1029 126 Cases of 29,000 Smokers in Finland M 1.25   20 mg, 32,000 IU 6 1.05