AJCN North Carolina Research Campus
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental data
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Google Scholar
Right arrow Articles by Gallicchio, L.
Right arrow Articles by Alberg, A. J
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gallicchio, L.
Right arrow Articles by Alberg, A. J
Agricola
Right arrow Articles by Gallicchio, L.
Right arrow Articles by Alberg, A. J
American Journal of Clinical Nutrition, Vol. 88, No. 2, 372-383, August 2008
© 2008 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Carotenoids and the risk of developing lung cancer: a systematic review1,2,3

Lisa Gallicchio, Kristina Boyd, Genevieve Matanoski, Xuguang (Grant) Tao, Liwei Chen, Tram K Lam, Meredith Shiels, Edward Hammond, Karen A Robinson, Laura E Caulfield, James G Herman, Eliseo Guallar and Anthony J Alberg

1 From the The Prevention and Research Center, Mercy Medical Center, Baltimore, MD (LG); the Departments of Epidemiology (LG, KB, GM, XT, TKL, MS, EH, EG, and AJA) and International Health, Center for Human Nutrition (LC and LEC), Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; the Departments of General Internal Medicine (KAR), Oncology (JGH and AJA), and Occupational and Environmental Medicine (XT), Johns Hopkins School of Medicine, Baltimore, MD; and the Cancer Prevention and Control Program, Hollings Cancer Center and the Department of Biostatistics, Bioinformatics, and Epidemiology, The Medical University of South Carolina, Charleston, SC (AJA)

2 Supported by the World Cancer Research Fund.

3 Reprints not available. Address correspondence to AJ Alberg, Hollings Cancer Center, Medical University of South Carolina, PO Box 250955, Charleston, SC 29425. E-mail: alberg{at}musc.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Carotenoids are thought to have anti-cancer properties, but findings from population-based research have been inconsistent.

Objective: We aimed to conduct a systematic review of the associations between carotenoids and lung cancer.

Design: We searched electronic databases for articles published through September 2007. Six randomized clinical trials examining the efficacy of β-carotene supplements and 25 prospective observational studies assessing the associations between carotenoids and lung cancer were analyzed by using random-effects meta-analysis.

Results: The pooled relative risk (RR) for the studies comparing β-carotene supplements with placebo was 1.10 (95% confidence limits: 0.89, 1.36; P = 0.39). Among the observational studies that adjusted for smoking, the pooled RRs comparing highest and lowest categories of total carotenoid intake and of total carotenoid serum concentrations were 0.79 (0.71, 0.87; P < 0.001) and 0.70 (0.44, 1.11; P = 0.14), respectively. For β-carotene, highest compared with lowest pooled RRs were 0.92 (0.83, 1.01; P = 0.09) for dietary intake and 0.84 (0.66, 1.07; P = 0.15) for serum concentrations. For other carotenoids, the RRs comparing highest and lowest categories of intake ranged from 0.80 for β-cryptoxanthin to 0.89 for {alpha}-carotene and lutein-zeaxanthin; for serum concentrations, the RRs ranged from 0.71 for lycopene to 0.95 for lutein-zeaxanthin.

Conclusions: β-Carotene supplementation is not associated with a decrease in the risk of developing lung cancer. Findings from prospective cohort studies suggest inverse associations between carotenoids and lung cancer; however, the decreases in risk are generally small and not statistically significant. These inverse associations may be the result of carotenoid measurements’ function as a marker of a healthier lifestyle (higher fruit and vegetable consumption) or of residual confounding by smoking.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Whereas a causal relation between cigarette smoking and lung cancer was firmly established >4 decades ago, not until the 1970s did the potential role of dietary factors in the development of lung cancer garner widespread interest. This interest was spurred by the pioneering work of Bjelke (1) and Bjelke et al (2) exploring the potential protective role of vitamin A and by a seminal review by Peto et al (3) discussing the chemopreventive potential of β-carotene, a provitamin A carotenoid. Evidence from observational epidemiologic studies rapidly accumulated and tended to support an inverse association of lung cancer incidence with β-carotene intake and with serum concentrations of β-carotene. This evidence led to the initiation of several large-scale randomized chemoprevention trials to test the hypothesis that β-carotene supplements protected against lung cancer, but those trials had disappointing results. Indeed, β-carotene supplementation actually was found to increase the risk of lung cancer in high-risk populations (4, 5).

Whereas high-dose β-carotene supplementation is ineffective in reducing lung cancer risk in randomized trials, many questions remain about the potential benefits of the intake of lower doses of β-carotene over prolonged periods. Furthermore, there is substantial interest in the potential role of other carotenoids in lung cancer prevention. Previous reviews of this topic (6-9) were not performed systematically. Given the public health importance of clarifying the potential role of carotenoids in lung carcinogenesis and given the extensive and diverse body of evidence available, we conducted a systematic and quantitative review of the evidence, derived from randomized clinical trials (RCTs) and from prospective observational studies, for the associations between carotenoids and the risk of lung cancer.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
The present study was carried out as a part of a project funded by the World Cancer Research Fund (WCRF) and the American Institute for Cancer Research to develop a report providing a global perspective on food, nutrition, physical activity, and the prevention of cancer (10). All of the work funded under this project was conducted by using a standardized protocol developed by the WCRF (Internet: http://www.wcrf.org/research/second_wcrf_aicr_report.lasso).

Study search
For the WCRF report, we sought all evidence on the associations between dietary intake, physical activity, or anthropometric measures and lung cancer that were reported in epidemiologic study designs, including RCTs, cohort studies, case-control studies, and ecologic studies. Case series and case reports were excluded. The search strategy used was the same as that described in a previous publication from our group (11), except that outcome search terms for lung cancer were used instead of those for nasopharyngeal carcinoma. Specifically, we searched for "lung neoplasm" as a MESH heading and, in the title or abstract, for the term "lung" in combination with "carcinoma," "neoplasm," or "tumor." The following electronic databases were searched: PubMed, Embase, Pascal, ISI Web of Science, NIAAA Alcohol and Alcohol Problems Science, The Cochrane Library, Biological Abstracts, Cumulative Index to Nursing and Allied Health Literature, Index Medicus for the World Health Organization (WHO) Eastern Mediterranean Region, Index Medicus for South East Asian Region, and Latin American and Caribbean Center on Health Sciences Information. The WCRF search included all studies published up to April 2006; there were no language restrictions. In addition, the study team hand-searched the references cited in the 1997 WCRF report (12), in the articles chosen for data abstraction, and in the relevant review articles or meta-analyses identified in the PubMed search. After the original WCRF search, we updated the PubMed search through September 2007.

Study selection and data abstraction
The following exclusion criteria were applied to the abstracts identified in the literature search: 1) studies had no original data (ie, reviews, editorials, and meta-analyses); 2) studies did not address the association between dietary intake, physical activity, or anthropometric measures and lung cancer; and 3) studies were not conducted in humans. The full-text articles of all references selected after application of these criteria were reviewed by using the same criteria. For the full WCRF report, full-text articles describing RCTs, cohort studies, case-cohort studies, case-control or nested case-control studies, and ecological studies of the association between carotenoids (including those that assessed biomarkers of dietary intake) and lung cancer were selected and analyzed. For the present report, references detailing RCTs of the efficacy of supplements or prospective investigations (cohort, case-cohort, or nested case-control studies) of the association between carotenoids (including those that assessed biomarkers of dietary intake) and lung cancer were analyzed. Full-text references examining these specific associations among these types of studies were excluded if measures of association or variability were not reported or could not be calculated by using the data provided. If separate reports from the same study were published, the report with the most updated data was selected for inclusion; in the case of duplicate publication, only one publication was included. The eligibility of each abstract or full-text article was assessed independently in a standardized manner by 2 reviewers; these duplicate reviews were performed by pair-wise combinations of many of the authors in addition to trained research assistants.

Data abstraction for selected articles was performed serially by 2 reviewers, using an electronic abstraction database created by WCRF. Disagreements between reviewers were resolved by consensus.

Statistical analysis
In the publications identified for inclusion into this systematic review, evidence was reported on total carotenoids, β-carotene, {alpha}-carotene, β-cryptoxanthin, lycopene, and lutein-zeaxanthin. Data examining carotenoid supplements, dietary intake of carotenoids, and serum carotenoid concentrations were analyzed separately. RCTs were analyzed separately from prospective observational studies.

For all studies, relative risks (RRs) and their 95% CIs were abstracted or derived from data reported in the publications. When several RR estimates were reported in a study, we selected the RR adjusted for the most covariates. Pooled RR estimates were obtained by using inverse-variance weights in random-effects models. Statistical heterogeneity was assessed by using the DerSimonian and Laird Q statistic and the I2 statistic.

For RCTs of β-carotene supplements, the meta-analysis was based on intention-to-treat comparisons. For prospective studies of dietary carotenoid intake or serum carotenoid concentration, we pooled the RRs for lung cancer and compared the highest with the lowest category of carotenoid intake or concentration reported in each study. Because cigarette smoking is a strong confounder of the association of carotenoid intake and serum carotenoid concentrations with lung cancer, pooled analyses were restricted to those studies that adjusted for cigarette smoking status. For the dietary intake analyses, the associations of the intake of the specific types of carotenoids and total carotenoids with lung cancer were also conducted in strata defined by smoking status (current, former, or never) to examine whether the associations differed among the smoking subgroups. Smoking-stratified analyses were not conducted for serum concentrations of carotenoids because few studies (n = 3) reported smoking-stratified results, and the disparate categories used to report the results in these studies precluded pooling (13-15).

Sensitivity analyses to examine the influence of each individual study were conducted by excluding each study from the meta-analysis and comparing the point estimates without and with that study. Publication bias was examined by using funnel plots.

Finally, dose-response meta-analyses were conducted by using the methods of Greenland and Longnecker (16) for studies reporting ≥3 exposure categories. Because the consumption levels and serum concentrations of carotenoids included in the reference categories differed across studies, we calculated the pooled RRs associated with a relative change in carotenoid intake from each study's own reference category. All analyses were conducted by using STATA software (version 9.2; StataCorp, College Station, TX).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Search results
The original WCRF search on dietary intake, anthropometric measures, physical activity, and lung cancer yielded 22 994 references, of which 21 385 were excluded after abstract review (Figure 1Go). Among the remaining 1609 articles, 50 publications reported RCT or prospective study investigations of the association between carotenoids and lung cancer. On the basis of the full-text review, we excluded 14 articles in which the same or more-updated data were presented in a separate publication (17-30) and 1 article that did not report a measure of association or variability (31). Therefore, 35 references, reporting the results of 5 distinct RCTs and 24 cohort studies, that were identified in the original WCRF search up to April 2006 met the inclusion criteria and were included in the systematic review. An additional 2 articles were identified in the supplemental PubMed search through September 2007 (32, 33). Two of the RCTs [the Carotene and Retinol Efficacy Trial (CARET) and the {alpha}-Tocopherol, β-Carotene Cancer Prevention Study (ATBC)] were subsequently analyzed as cohort studies, and, thus, these studies are included in the analyses of both types of studies.


Figure 1
View larger version (36K):
[in this window]
[in a new window]

  		FIGURE 1. Study selection process. *Other databases: NIAAA Alcohol and Alcohol Problems Science Database (n = 48), Cochrane Library (n = 101), CINAHL-EBSCOhost (n = 270), Agricola (n = 82), Index Medicus for WHO Eastern Mediterranean Region (n = 4), Index Medicus for South East Asian Region (n = 5), and Latin American and Caribbean Center on Health Sciences Information (n = 114). **Shaten et al (31): only the direction of association and P value were reported. ***Some of the references (and studies) examined multiple exposures; therefore, the numbers listed below do not add up to the number of distinct references (37) and studies (29) included in this systematic review.

 
Randomized controlled trials of β-carotene supplements and lung cancer
We identified 6 RCTs that tested the efficacy of β-carotene supplements in lung cancer prevention (Table 1Go). Three trials were conducted in the United States (4, 34, 35), and one trial each was conducted in Finland (5), Australia (36), and China (33). Three trials enrolled only persons at high risk of lung cancer (4, 5, 36). Three trials tested β-carotene compared with placebo (5, 34, 35), one trial tested β-carotene plus retinol compared with placebo (4), one trial tested β-carotene compared with retinol (36), and one trial tested β-carotene plus {alpha}-tocopherol plus selenium compared with placebo (33).


View this table:
[in this window]
[in a new window]

  		TABLE 1 Randomized clinical trials examining the association between β-carotene supplements and lung cancer1

 
The RRs of lung cancer associated with β-carotene supplements among the RCTs are shown in Figure 2Go. The pooled RR for the 3 studies comparing β-carotene supplements with placebo (5, 34, 35) was 1.10 (95% CL: 0.89, 1.36). β-carotene supplements were associated with statistically significant increases in lung cancer risk in placebo-controlled studies conducted in high-risk populations (4, 5).


Figure 2
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 2. Forest plot of relative risks (RR) and 95% confidence limits (CL) for β-carotene supplement trials. ATBC, {alpha}-Tocopherol, β-Carotene Cancer Prevention Study; CARET, Carotene and Retinol Efficacy Trial; Australia W Perth, Australia Western Perth. *Pooled RR estimates were obtained by using inverse-variance weights in a random-effects model.

 
Carotenoid intake and lung cancer
A total of 16 cohort studies reported an association between dietary intake of carotenoids and lung cancer (Table 2Go). Ten studies were conducted in the United States, 4 in Europe, and 1 each in Canada and Singapore. Eight studies included only men, 3 included only women, and 5 included both men and women. The cohorts varied in size from 878 (Zutphen Study; 53) to 121 700 (Nurses’ Health Study; 41), and the number of cases ranged from 33 (48) to 1787 (37). Information on diet was collected by using a food-frequency questionnaire in 12 studies, a dietary history questionnaire or interview in 2 studies, and dietary recall in 2 studies. Validation of the dietary instrument was mentioned in 7 studies (15, 37-39, 41, 42, 46). No study included supplement use in the estimates of dietary carotenoid intake.


View this table:
[in this window]
[in a new window]

  		TABLE 2 Prospective studies examining the association between dietary carotenoid intake and lung cancer1

 
Nine studies reported total carotenoid intake; 8 of those studies reported smoking-adjusted estimates. All of the 8 studies reported an RR of <1, although only 2 of these RRs were statistically significant (15, 45) (Figure 3Go). The pooled RR for lung cancer comparing the highest with the lowest category of total carotenoid intake for the studies reporting smoking-adjusted estimates was 0.79 (95% CL: 0.71, 0.87) (Figure 3Go), and there was little evidence of statistical heterogeneity [Q(7 df) = 3.79, P = 0.80] (Table 3Go). The dose-response meta-analysis estimate among 4 studies with sufficient dose data (37, 41, 48) showed a 2% decrease in risk for a 1000-µg/d increase in intake (RR: 0.98; 95% CL: 0.97, 0.99).


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 3. Forest plot of relative risks (RR) and 95% confidence limits (CL) for highest-versus-lowest category of dietary total carotenoid intake and lung cancer risk among prospective studies; only risk estimates adjusted for smoking are included. ATBC, {alpha}-Tocopherol, β-Carotene Cancer Prevention Study; CARET, Carotene and Retinol Efficacy Trial; Finnish MCES, Finnish Mobile Clinic Health Examination Survey; HPFS, Health Professionals Follow-up Study; NHANES, National Health and Nutrition Examination Survey; NHS, Nurses’ Health Study; NR, not reported; NY Cohort, New York State Cohort; US Lutheran BIS, United States Lutheran Brotherhood Insurance Society. *Pooled RR estimates were obtained by using inverse-variance weights in a random-effects model.

 

View this table:
[in this window]
[in a new window]

  		TABLE 3 Dietary carotenoids: dose-response and highest-versus-lowest category meta-analyses of smoking-adjusted lung cancer estimates reported by cohort studies1

 
Among the specific carotenoids, β-carotene was the most frequently studied. The comparison of the highest with the lowest category of β-carotene intake was associated with an 8% statistically nonsignificant reduction in lung cancer risk among those studies reporting smoking-adjusted estimates (pooled RR: 0.92; 95% CL: 0.83, 1.01) (Table 3Go and Figure 4Go). Each 500-µg increase in daily β-carotene intake was associated with a 1% decrease in lung cancer risk (dose-response pooled RR: 0.99; 95% CL: 0.98, 1.00) (Table 3Go). When the analyses were restricted to never smokers in the 5 studies that reported smoking-stratified data, the RR comparing the highest and the lowest category of β-carotene intake was 0.97 (95% CL: 0.61, 1.52) (Table 4Go). The interactions between the individual carotenoids and smoking status with respect to the development of lung cancer were not statistically significant.


Figure 4
View larger version (26K):
[in this window]
[in a new window]

  		FIGURE 4. Forest plot of relative risks (RR) and 95% confidence limits (CL) for highest-versus-lowest category of dietary β-carotene intake and lung cancer risk among prospective studies; only risk estimates adjusted for smoking are included. ATBC, {alpha}-Tocopherol, β-Carotene Cancer Prevention Study; Canadian NBSS, Canadian National Breast Screening Study; CARET, Carotene and Retinol Efficacy Trial; Finnish MCES, Finnish Mobile Clinic Health Examination Survey; HPFS, Health Professionals Follow-up Study; NHS, Nurses’ Health Study; NR, not reported; US Lutheran BIS, United States Lutheran Brotherhood Insurance Society. *Pooled RR estimates were obtained by using inverse-variance weights in a random-effects model.

 

View this table:
[in this window]
[in a new window]

  		TABLE 4 Random effects estimates for highest-versus-lowest category of dietary intake of specific carotenoids and lung cancer stratified by smoking status1

 
There was an overall statistically significant inverse association with lung cancer risk in comparison of the highest with the lowest intake of the other carotenoids except lutein [pooled RR: 0.89 (0.79, 1.00); 0.80 (0.72, 0.89), 0.86 (0.77, 0.97), and 0.89 (0.79, 1.00) for {alpha}-carotene, β-cryptoxanthin, lycopene, and lutein-zeaxanthin, respectively] (Table 3Go). (Forest plots of these analyses can be found in Figures S1–S4 under "Supplemental data" in the current online issue.) Results for lutein intake were in the direction of a decrease in lung cancer risk; however, the pooled RR was not statistically significant (0.90; 0.74, 1.10) (Table 3Go; also, see Figure S5 under "Supplemental data" in the current online issue).

Two cohort studies reported nonsignificant associations between dietary β-carotene supplements and lung cancer; both studies were detailed in Michaud et al (41). In one of the cohorts, a nonsignificant association in the direction of increased risk was seen (1.23; 0.55, 2.76), whereas, in the other cohort, a nonsignificant inverse association was observed (0.82; 0.36, 1.85), but the CIs were wide, and they overlapped unity.

Serum carotenoid concentrations and lung cancer
Of the 26 cohort studies included, 15 investigated the association between serum or plasma carotenoid concentrations and lung cancer (Table 5Go). Eleven of these were nested case-control studies, and 4 (15, 32, 62, 63) were prospective cohort studies. Six studies were conducted in the United States, 4 in Asia, 4 in Europe, and 1 in Australia. Seven of the studies included only men. The number of cases ranged from 31 (Japan; 62) to 278 (CARET; 54). Lung cancer incidence was the outcome studied for all but 4 studies, in which lung cancer mortality was the endpoint (52, 61-63). Carotenoids were measured in serum samples by HPLC for all but one of the identified studies (63).


View this table:
[in this window]
[in a new window]

  		TABLE 5 Prospective studies examining the association between serum carotenoids and lung cancer1

 
Of the studies examining the smoking-adjusted association between total serum carotenoids and lung cancer risk, only one (62) reported an RR of >1; however, only one of the RRs reported was statistically significant (61) (Figure 5Go). In studies reporting a smoking-adjusted risk estimate, an inverse association that was not statistically significant was observed for the pooled RR in comparison of the highest with the lowest category of total carotenoid concentrations (0.70; 0.44, 1.11) (Table 6Go). Ten studies reported data on serum β-carotene concentration and lung cancer (Figure 6Go); the pooled RR comparing the highest with the lowest category of β-carotene concentrations was compatible with a 16% lower risk but was not statistically significant (0.84; 0.66, 1.07). Nonsignificant inverse associations were also observed for {alpha}-carotene (0.89; 0.59, 1.33), β-cryptoxanthin (0.82; 0.40, 1.68), and lutein-zeaxanthin (0.95; 0.67, 1.36) (Table 6Go; also, see Figures S6–S8 under "Supplemental data" in the current online issue). Only the results for lycopene were statistically significant (pooled RR: 0.71; 95% CL: 0.51, 0.98) (Table 6Go; also, see Figure S9 under "Supplemental data" in the current online issue).


Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		FIGURE 5. Forest plot of relative risks (RR) and 95% confidence limits (CL) for highest-versus-lowest category of serum total carotenoid concentrations and lung cancer risk among prospective studies; only risk estimates adjusted for smoking are included. Japan Collab Cohort, Japan Collaborative Cohort Study; NR, not reported. *Pooled RR estimates were obtained by using inverse-variance weights in a random-effects model.

 

View this table:
[in this window]
[in a new window]

  		TABLE 6 Serum carotenoids: highest-versus-lowest category meta-analysis of smoking-adjusted lung cancer estimates reported by cohort studies1

 

Figure 6
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 6. Forest plot of relative risks (RR) and 95% confidence limits (CL) for highest-versus-lowest category of serum β-carotene concentrations and lung cancer risk among prospective studies; only risk estimates adjusted for smoking are included. ATBC, {alpha}-Tocopherol, β-Carotene Cancer Prevention Study; CARET, Carotene and Retinol Efficacy Trial; Finnish MCES, Finnish Mobile Clinic Health Examination Survey; NR, not reported; UK, BUPA study, United Kingdom, British United Provident Association Study. *Pooled RR estimates were obtained by using inverse-variance weights in a random-effects model.

 
Dose-response associations between serum or plasma concentrations of carotenoids and lung cancer were examined in 6 of the 15 studies (Table 6Go). Pooled results from these 6 studies showed that each 0.1-µmol/L increase in β-carotene was associated with a 5% decrease in lung cancer risk, although the pooled RR was not statistically significant (dose-response pooled RR: 0.95; 95% CL: 0.87, 1.03). Pooled results for the other carotenoids, including total carotenoids, were based on ≤2 studies; statistically significant dose-response associations were observed for total carotenoids (0.64; 0.46, 0.88) and lycopene [(1 study) 0.82; 0.68, 0.98].


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The role that carotenoids may play in protecting against lung cancer was studied by systematically evaluating a large, diverse body of epidemiologic evidence. Enough evidence was obtained from high-quality prospective studies to perform meta-analyses of both dietary assessment studies and biomarker studies and to do so for 5 individual carotenoids and total carotenoids. For β-carotene, substantial evidence has also been generated from RCTs of supplements, which permitted a comparison of observational evidence with experimental evidence.

For the individual carotenoids, the associations were consistently in the protective direction, but they tended to be weak and statistically nonsignificant. In dietary intake studies, the comparisons of the highest with the lowest category ranged from an 8% (β-carotene) to a 20% (lycopene) lower risk of lung cancer; for serum concentrations, the range was 5% (lutein-zeaxanthin) to 29% (lycopene). Only for lycopene were the results statistically significant in both dietary intake and serum studies. If carotenoids truly protect against lung cancer, greater levels of intake would be expected to confer greater protection. Statistically significant dose-response trends were observed only for β-carotene and β-cryptoxanthin in dietary intake studies and only for lycopene in serum studies. Thus, for individual carotenoids, the combined results do not provide compelling evidence of a protective association.

For total carotenoids, however, the results provided stronger and more consistent evidence of an inverse association. The comparisons of the highest with the lowest category of carotenoid intake showed that lung cancer risk in persons with high dietary intake was 21% lower (statistically significant) and that in those with high serum concentrations was 30% lower (statistically nonsignificant). Statistically significant dose-response trends were present in both the dietary intake and serum studies.

Thus, the evidence in favor of a genuine inverse association with lung cancer risk is stronger for the more global measure of total carotenoids than for any of the individual carotenoids. This pattern of results could emerge if measurement of total carotenoids captures the combined influence of the individual carotenoids. This interpretation would be more tenable if the protective association between total carotenoids and lung cancer were much stronger than the associations observed for the individual carotenoids, but, in fact, the associations for total carotenoids and the individual carotenoids did not differ markedly. The similarities in the associations may be due to interactions between the specific carotenoids in the competition for absorption (64). Alternatively, the various carotenoid measurements may be serving as markers of the same thing, such as fruit and vegetable consumption or a healthier lifestyle.

In contrast to the observational studies of β-carotene, several large-scale RCTs designed to determine the efficacy of β-carotene supplementation provided no evidence of a protective association. In fact, the results of the ATBC and CARET trials suggest that high-dose supplemental β-carotene increases lung cancer risk in current smokers (29, 65). Among the reasons postulated for the increase in risk observed in the RCTs is that cigarette smoke creates conditions of high oxidative stress in which β-carotene may express prooxidant activity (65). The RCT results would not be predicted from the results of prospective cohort studies, in which the strongest inverse association between β-carotene intake and lung cancer risk was observed among current smokers. However, the observational studies and RCTs address inherently different questions. The RCTs used purified β-carotene doses 5–10 times greater than normal dietary intake (65-67)—doses at which any anti-cancer properties of β-carotene may be lost or reversed. Furthermore, because antioxidant nutrients may exert their protective effect in the earlier stages of carcinogenesis, the RCTs may have administered β-carotene too late in the carcinogenic process. Indeed, β-carotene can favor the growth of already initiated cells (68). Except for the major divergence between the 2 forms of evidence seen in current smokers, the results of the prospective observational studies and the RCTs are not as conflicting as it once seemed. Even though the direction of the overall associations still varies, the associations for both prospective cohort studies and RCTs were compatible with a null association.

Cigarette smoking is the principal cause of lung cancer, and it tends to be closely associated with less healthful diets, including lower intake of fruit and vegetables (69). Cigarette smoking is also associated with depletion of circulating provitamin A carotenoid concentrations (70). Because cigarette smoking is so strongly associated with both lung cancer and carotenoid consumption, and also because smoking history may not be measured in enough detail, residual confounding by smoking remains a viable explanation for the associations between carotenoids and lung cancer seen in the observational studies. In fact, in analyses restricted to never smokers, the association between β-carotene intake and lung cancer was null. In contrast, the intake of carotenoids other than β-carotene tended to be inversely associated with lung cancer among both never and current smokers, but few studies reported the results with stratification by smoking status.

In contrast, if error in measuring dietary carotenoid intake is considered, it may be that the true associations are stronger than those reported in the studies included in the present report. In the setting of observational cohort studies, this measurement error would be expected to be nondifferential with respect to lung cancer risk. Therefore, the measurement error would bias the associations toward the null.

Our conclusions are independent of the WCRF report (10), which had 2 conclusions relevant to the present study. The first conclusion, that the evidence is "convincing" (highest possible evidence grade) that β-carotene supplements cause lung cancer in current smokers, was straightforward and was based on the consistent results of 2 high-quality randomized placebo-controlled trials (4, 5). The second relevant WCRF conclusion was that the evidence is "probable" that foods containing carotenoids protect against lung cancer. We have not attempted to rate the evidence in this way, but our evidence synthesis is compatible with a more conservative interpretation. The WCRF report includes other data and uses different criteria for judgment; the 2 key differences in approaches are the specific evidence included and the use of formal meta-analyses. The WCRF report considered all available epidemiologic evidence, whereas the present report selectively focused on prospective studies adjusted for cigarette smoking. The rationale for our approach is that focusing on the highest-quality evidence is most likely to lead to valid conclusions. Compared with the evidence synthesized in the present report, the evidence from case-control studies of this topic tended to more strongly favor a protective association (10). For example, retrospective case-control studies of circulating carotenoid concentrations represent a suboptimal approach, because the effects of lung cancer and its treatment on circulating carotenoid concentrations would strongly bias the associations in the protective direction. Another difference in approaches is that we meta-analyzed the data to provide a formal quantitative synthesis of the evidence, whereas the WCRF did so only for the assessment of dose-response trends. Quantifying the strength and statistical significance of associations by formally pooling results across studies provides useful guidance for drawing inferences across large and complex bodies of evidence, such as those considered here.

For individual carotenoids, the results of prospective observational studies do not provide compelling evidence that a higher intake or greater circulating concentrations reduce lung cancer risk. The strongest evidence of an inverse association between carotenoids and lung cancer was for total carotenoids. Even for this association, however, caution in drawing inferences is needed, because other alternative explanations cannot be ruled out. These alternative explanations include residual confounding by cigarette smoking and the possibility that carotenoid measurements are serving as a marker for a diet higher in fruit and vegetables or for a healthier lifestyle in general. These concerns are accentuated by the sobering results of the RCTs of β-carotene supplementation, which provide strong evidence that supplementation clearly does not protect against lung cancer and, in fact, increases lung cancer risk among current cigarette smokers.


   ACKNOWLEDGMENTS
 
The authors’ responsibilities were as follows: LG and AJA: full access to all of the data in the study and responsibility for the integrity of the data and the accuracy of the data analysis; LG, GM, XT, KAR, and AJA: study concept and design; LG, KB, XT, LC, TL, MS, EH, KR, EG, and AJA: acquisition of the data; LG, KB, GM, XT, LC, LEC, JGH, EG, and AJA: analysis and interpretation of the data; LG, EG, and AJA: drafting of the manuscript; KB, GM, XT, LC, TKL, MS, EH, LEC, KAR, JGH, EG, and AJA: critical revision of the manuscript for important intellectual content; LG, XT, and EG: statistical analysis; KB, XT, LC, TKL, MS, EH, and AJA: administrative, technical, or material support; and KB, GM, XT, LEC, EG, and AJA: study supervision. None of the authors had a personal or financial conflict of interest.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Bjelke E. Dietary vitamin A and human lung cancer. Int J Cancer 1975;15:561–5.[Medline]
  2. Kvale G, Bjelke E, Gart JJ. Dietary habits and lung cancer risk. Int J Cancer 1983;31:397–405.[Medline]
  3. Peto R, Doll R, Buckley JD, Sporn MB. Can dietary beta-carotene materially reduce human cancer rates? Nature 1981;290:201–8.[Medline]
  4. Omenn GS, Goodman GE, Thornquist MD, et al. Risk factors for lung cancer and for intervention effects in CARET, the Beta-Carotene and Retinol Efficacy Trial. J Natl Cancer Inst 1996;88:1550–9.[Abstract/Free Full Text]
  5. Virtamo J, Pietinen P, Huttunen JK, et al. Incidence of cancer and mortality following alpha-tocopherol and beta-carotene supplementation: a postintervention follow-up. JAMA 2003;290:476–85.[Abstract/Free Full Text]
  6. Ruano-Ravina A, Figueiras A, Freire-Garabal M, Barros-Dios JM. Antioxidant vitamins and risk of lung cancer. Curr Pharm Des 2006;12:599–613.[Medline]
  7. Ziegler RG, Colavito EA, Hartge P, et al. Importance of alpha-carotene, beta-carotene, and other phytochemicals in the etiology of lung cancer. J Natl Cancer Inst 1996;88:612–5.[Free Full Text]
  8. Cooper DA, Eldridge AL, Peters JC. Dietary carotenoids and lung cancer: a review of recent research. Nutr Rev 1999;57:133–45.[Medline]
  9. Handelman GJ. The evolving role of carotenoids in human biochemistry. Nutrition 2001;17:818–22.[Medline]
  10. World Cancer Res Fund (WCRF) and the American Institute for Cancer Res (AICR). Food, nutrition, physical activity, and the prevention of cancer: a global perspective. Washington, DC: AICR, 2007.
  11. Gallicchio L, Matanoski G, Tao XG, et al. Adulthood consumption of preserved and nonpreserved vegetables and the risk of nasopharyngeal carcinoma: a systematic review. Int J Cancer 2006;119:1125–35.[Medline]
  12. World Cancer Res Fund (WCRF) and the American Institute for Cancer Res (AICR). Food, nutrition, and the prevention of cancer: a global perspective. Washington, DC: AICR, 1997.
  13. Yuan JM, Ross RK, Chu XD, Gao YT, Yu MC. Prediagnostic levels of serum beta-cryptoxanthin and retinol predict smoking-related lung cancer risk in Shanghai, China. Cancer Epidemiol Biomarkers Prev 2001;10:767–73.[Abstract/Free Full Text]
  14. Knekt P, Aromaa A, Maatela J, et al. Serum vitamin A and subsequent risk of cancer: cancer incidence follow-up of the Finnish Mobile Clinic Health Examination Survey. Am J Epidemiol 1990;132:857–70.[Abstract/Free Full Text]
  15. Holick CN, Michaud DS, Stolzenberg-Solomon R, et al. Dietary carotenoids, serum beta-carotene, and retinol and risk of lung cancer in the alpha-tocopherol, beta-carotene cohort study. Am J Epidemiol 2002;156:536–47.[Abstract/Free Full Text]
  16. Greenland S, Longnecker MP. Methods for trend estimation from summarized dose-response data, with applications to meta-analysis. Am J Epidemiol 1992;135:1301–9.[Abstract/Free Full Text]
  17. Bandera EV, Freudenheim JL, Marshall JR, et al. Impact of losses to follow-up on diet/alcohol and lung cancer analyses in the New York State Cohort. Nutr Cancer 2002;42:41–7.[Medline]
  18. Knekt P. Vitamin E and smoking and the risk of lung cancer. Ann N Y Acad Sci 1993;686:280–7.[Medline]
  19. Knekt P, Jarvinen R, Seppanen R, et al. Dietary antioxidants and the risk of lung cancer. Am J Epidemiol 1991;134:471–9.[Abstract/Free Full Text]
  20. Ocke MC, Bueno-de-Mesquita HB, Feskens EJ, van Staveren WA, Kromhout D. Repeated measurements of vegetables, fruits, beta-carotene, and vitamins C and E in relation to lung cancer. The Zutphen Study. Am J Epidemiol 1997;145:358–65.[Abstract/Free Full Text]
  21. Paganini-Hill A, Chao A, Ross RK, Henderson BE. Vitamin A, beta-carotene, and the risk of cancer: a prospective study. J Natl Cancer Inst 1987;79:443–8.[Medline]
  22. Speizer FE, Colditz GA, Hunter DJ, Rosner B, Hennekens C. Prospective study of smoking, antioxidant intake, and lung cancer in middle-aged women (USA). Cancer Causes Control 1999;10:475–82.[Medline]
  23. Comstock GW, Helzlsouer KJ, Bush TL. Prediagnostic serum levels of carotenoids and vitamin E as related to subsequent cancer in Washington County, Maryland. Am J Clin Nutr 1991;53(suppl):260S–4S.[Medline]
  24. Ratnasinghe DL, Yao SX, Forman M, et al. Gene-environment interactions between the codon 194 polymorphism of XRCC1 and antioxidants influence lung cancer risk. Anticancer Res 2003;23:627–32.[Medline]
  25. Stahelin HB, Rosel F, Buess E, Brubacher G. Cancer, vitamins, and plasma lipids: prospective Basel study. J Natl Cancer Inst 1984;73:1463–8.[Medline]
  26. Stahelin HB, Gey KF, Eichholzer M, et al. Plasma antioxidant vitamins and subsequent cancer mortality in the 12-year follow-up of the prospective Basel Study. Am J Epidemiol 1991;133:766–75.[Abstract/Free Full Text]
  27. Ito Y, Wakai K, Suzuki K, et al. Serum carotenoids and mortality from lung cancer: a case-control study nested in the Japan Collaborative Cohort (JACC) study. Cancer Sci 2003;94:57–63.[Medline]
  28. Omenn GS, Goodman GE, Thornquist MD, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. N Engl J Med 1996;334:1150–5.[Abstract/Free Full Text]
  29. Albanes D, Heinonen OP, Taylor PR, et al. Alpha-tocopherol and beta-carotene supplements and lung cancer incidence in the alpha-tocopherol, beta-carotene cancer prevention study: effects of base-line characteristics and study compliance. J Natl Cancer Inst 1996;88:1560–70.[Abstract/Free Full Text]
  30. Goodman GE, Thornquist MD, Balmes J, et al. The Beta-Carotene and Retinol Efficacy Trial: incidence of lung cancer and cardiovascular disease mortality during 6-year follow-up after stopping beta-carotene and retinol supplements. J Natl Cancer Inst 2004;96:1743–50.[Abstract/Free Full Text]
  31. Shaten BJ, Kuller LH, Kjelsberg MO, et al. Lung cancer mortality after 16 years in MRFIT participants in intervention and usual-care groups. Multiple Risk Factor Intervention Trial. Ann Epidemiol 1997;7:125–36.[Medline]
  32. Alfonso HS, Fritschi L, de Klerk NH, et al. Plasma vitamin concentrations and incidence of mesothelioma and lung cancer in individuals exposed to crocidolite at Wittenoom, Western Australia. Eur J Cancer Prev 2006;15:290–4.[Medline]
  33. Kamangar F, Qiao YL, Yu B, et al. Lung cancer chemoprevention: a randomized, double-blind trial in Linxian, China. Cancer Epidemiol Biomarkers Prev 2006;15:1562–4.[Abstract/Free Full Text]
  34. Cook NR, Le IM, Manson JE, Buring JE, Hennekens CH. Effects of beta-carotene supplementation on cancer incidence by baseline characteristics in the Physicians’ Health Study (United States). Cancer Causes Control 2000;11:617–26.[Medline]
  35. Lee IM, Cook NR, Manson JE, Buring JE, Hennekens CH. Beta-carotene supplementation and incidence of cancer and cardiovascular disease: the Women's Health Study. J Natl Cancer Inst 1999;91:2102–6.[Abstract/Free Full Text]
  36. de Klerk NH, Musk AW, Ambrosini GL, et al. Vitamin A and cancer prevention II: comparison of the effects of retinol and beta-carotene. Int J Cancer 1998;75:362–7.[Medline]
  37. Wright ME, Mayne ST, Stolzenberg-Solomon RZ, et al. Development of a comprehensive dietary antioxidant index and application to lung cancer risk in a cohort of male smokers. Am J Epidemiol 2004;160:68–76.[Abstract/Free Full Text]
  38. Yuan JM, Stram DO, Arakawa K, Lee HP, Yu MC. Dietary cryptoxanthin and reduced risk of lung cancer: the Singapore Chinese Health Study. Cancer Epidemiol Biomarkers Prev 2003;12:890–8.[Abstract/Free Full Text]
  39. Neuhouser ML, Patterson RE, Thornquist MD, Omenn GS, King IB, Goodman GE. Fruits and vegetables are associated with lower lung cancer risk only in the placebo arm of the beta-carotene and retinol efficacy trial (CARET). Cancer Epidemiol Biomarkers Prev 2003;12:350–8.[Abstract/Free Full Text]
  40. Rohan TE, Jain M, Howe GR, Miller AB. A cohort study of dietary carotenoids and lung cancer risk in women (Canada). Cancer Causes Control 2002;13:231–7.[Medline]
  41. Michaud DS, Feskanich D, Rimm EB, et al. Intake of specific carotenoids and risk of lung cancer in 2 prospective US cohorts. Am J Clin Nutr 2000;72:990–7.[Abstract/Free Full Text]
  42. Voorrips LE, Goldbohm RA, Brants HA, et al. A prospective cohort study on antioxidant and folate intake and male lung cancer risk. Cancer Epidemiol Biomarkers Prev 2000;9:357–65.[Abstract/Free Full Text]
  43. Knekt P, Jarvinen R, Teppo L, Aromaa A, Seppanen R. Role of various carotenoids in lung cancer prevention. J Natl Cancer Inst 1999;91:182–4.[Free Full Text]
  44. Yong LC, Brown CC, Schatzkin A, et al. Intake of vitamins E, C, and A and risk of lung cancer. The NHANES I epidemiologic followup study. First National Health and Nutrition Examination Survey. Am J Epidemiol 1997;146:231–43.[Abstract/Free Full Text]
  45. Bandera EV, Freudenheim JL, Marshall JR, et al. Diet and alcohol consumption and lung cancer risk in the New York State Cohort (United States). Cancer Causes Control 1997;8:828–40.[Medline]
  46. Steinmetz KA, Potter JD, Folsom AR. Vegetables, fruit, and lung cancer in the Iowa Women's Health Study. Cancer Res 1993;53:536–43.[Abstract/Free Full Text]
  47. Shekelle RB, Tangney CC, Rossof AH, Stamler J. Serum cholesterol, beta-carotene, and risk of lung cancer. Epidemiology 1992;3:282–7.[Medline]
  48. Shekelle RB, Lepper M, Liu S, et al. Dietary vitamin A and risk of cancer in the Western Electric study. Lancet 1981;2:1185–90.[Medline]
  49. Shibata A, Paganini-Hill A, Ross RK, Henderson BE. Intake of vegetables, fruits, beta-carotene, vitamin C and vitamin supplements and cancer incidence among the elderly: a prospective study. Br J Cancer 1992;66:673–9.[Medline]
  50. Shibata A, Paganini-Hill A, Ross RK, Yu MC, Henderson BE. Dietary beta-carotene, cigarette smoking, and lung cancer in men. Cancer Causes Control 1992;3:207–14.[Medline]
  51. Chow WH, Schuman LM, McLaughlin JK, et al. A cohort study of tobacco use, diet, occupation, and lung cancer mortality. Cancer Causes Control 1992;3:247–54.[Medline]
  52. Connett JE, Kuller LH, Kjelsberg MO, et al. Relationship between carotenoids and cancer. The Multiple Risk Factor Intervention Trial (MRFIT) Study. Cancer 1989;64:126–34.[Medline]
  53. Kromhout D. Essential micronutrients in relation to carcinogenesis. Am J Clin Nutr 1987;45:1361–7.[Free Full Text]
  54. Goodman GE, Schaffer S, Omenn GS, Chen C, King I. The association between lung and prostate cancer risk, and serum micronutrients: results and lessons learned from beta-carotene and retinol efficacy trial. Cancer Epidemiol Biomarkers Prev 2003;12:518–26.[Abstract/Free Full Text]
  55. Ratnasinghe D, Forman MR, Tangrea JA, et al. Serum carotenoids are associated with increased lung cancer risk among alcohol drinkers, but not among non-drinkers in a cohort of tin miners. Alcohol Alcohol 2000;35:355–60.[Abstract/Free Full Text]
  56. Comstock GW, Alberg AJ, Huang HY, et al. The risk of developing lung cancer associated with antioxidants in the blood: ascorbic acid, carotenoids, alpha-tocopherol, selenium, and total peroxyl radical absorbing capacity. Cancer Epidemiol Biomarkers Prev 1997;6:907–16.[Abstract]
  57. Menkes MS, Comstock GW, Vuilleumier JP, Helsing KJ, Rider AA, Brookmeyer R. Serum beta-carotene, vitamins A and E, selenium, and the risk of lung cancer. N Engl J Med 1986;315:1250–4.[Abstract]
  58. Orentreich N, Matias JR, Vogelman JH, Salkeld RM, Bhagavan H, Friedman GD. The predictive value of serum beta-carotene for subsequent development of lung cancer. Nutr Cancer 1991;16:167–9.[Medline]
  59. Wald NJ, Thompson SG, Densem JW, Boreham J, Bailey A. Serum beta-carotene and subsequent risk of cancer: results from the BUPA Study. Br J Cancer 1988;57:428–33.[Medline]
  60. Nomura AM, Stemmermann GN, Heilbrun LK, Salkeld RM, Vuilleumier JP. Serum vitamin levels and the risk of cancer of specific sites in men of Japanese ancestry in Hawaii. Cancer Res 1985;45:2369–72.[Abstract/Free Full Text]
  61. Ito Y, Wakai K, Suzuki K, et al. Lung cancer mortality and serum levels of carotenoids, retinol, tocopherols, and folic acid in men and women: a case-control study nested in the JACC Study. J Epidemiol 2005;15(suppl):S140–9.[Medline]
  62. Ito Y, Kurata M, Hioki R, Suzuki K, Ochiai J, Aoki K. Cancer mortality and serum levels of carotenoids, retinol, and tocopherol: a population-based follow-up study of inhabitants of a rural area of Japan. Asian Pac J Cancer Prev 2005;6:10–5.[Medline]
  63. Eichholzer M, Stahelin HB, Gey KF, Ludin E, Bernasconi F. Prediction of male cancer mortality by plasma levels of interacting vitamins: 17-year follow-up of the prospective Basel study. Int J Cancer 1996;66:145–50.[Medline]
  64. van den Berg H. Carotenoid interactions. Nutr Rev 1999;57:1–10.[Medline]
  65. Mayne ST, Handelman GJ, Beecher G. Beta-carotene and lung cancer promotion in heavy smokers—a plausible relationship? J Natl Cancer Inst 1996;88:1513–5.[Medline]
  66. Greenwald P. Beta-carotene and lung cancer: a lesson for future chemoprevention investigations? J Natl Cancer Inst 2003;95:E1.[Free Full Text]
  67. Pietrzik K. Antioxidant vitamins, cancer, and cardiovascular disease. N Engl J Med 1996;335:1065–6.[Free Full Text]
  68. Rautalahti M, Albanes D, Virtamo J, Taylor PR, Huttunen JK, Heinonen OP. Beta-carotene did not work: aftermath of the ATBC study. Cancer Lett 1997;114:235–6.[Medline]
  69. Alberg A. The influence of cigarette smoking on circulating concentrations of antioxidant micronutrients. Toxicology 2002;180:121–37.[Medline]
  70. Alberg AJ, Chen JC, Zhao H, Hoffman SC, Comstock GW, Helzlsouer KJ. Household exposure to passive cigarette smoking and serum micronutrient concentrations. Am J Clin Nutr 2000;72:1576–82.[Abstract/Free Full Text]
Received for publication November 5, 2007. Accepted for publication April 28, 2008.





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental data
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Google Scholar
Right arrow Articles by Gallicchio, L.
Right arrow Articles by Alberg, A. J
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gallicchio, L.
Right arrow Articles by Alberg, A. J
Agricola
Right arrow Articles by Gallicchio, L.
Right arrow Articles by Alberg, A. J

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS