Annals of Oncology Advance Access originally published online on July 27, 2006
Annals of Oncology 2006 17(9):1459-1463; doi:10.1093/annonc/mdl166
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 European Society for Medical Oncology
epidemiology |
Body size and laryngeal cancer risk
1 Istituto di Ricerche Farmacologiche "Mario Negri", Milan
2 Clinica Otorinolaringoiatrica, DNTB, Università Milano-Bicocca, Monza (MI)
3 Unità di Epidemiologia e Biostatistica, Centro di Riferimento Oncologico, Aviano (PN)
4 Divisione di Otorinolaringoiatria, Azienda Ospedaliera "S. Maria degli Angeli", Pordenone, Italy
5 International Agency for Research on Cancer, Lyon Cedex, France
6 Istituto di Statistica Medica e Biometria, Università degli Studi di Milano, Milan, Italy
*Correspondence to: Dr C. Bosetti, Istituto di Ricerche Farmacologiche "Mario Negri", Via Eritrea 62 – 20157 Milan, Italy; Tel: +39 0239014.526; Fax: +39 0233200231; E-mail: bosetti{at}marionegri.it
 |
Abstract |
|---|
 Top Abstract introductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Background: A few studies have analyzed the role of lifetime anthropometric measures on laryngeal cancer risk.
Patient and methods: This relation was investigated using a multicentre case-control study from Italy, conducted between 1992 and 2000, and including 460 incident, histologically confirmed laryngeal cancer cases, and 1088 controls admitted to the same network of hospitals as cases for acute, non neoplastic condition. Odds ratios (OR) and corresponding 95% confidence intervals (CI) were obtained from multiple logistic regression, including terms for major confounding factors, such as physical activity and energy intake.
Results: An inverse association with laryngeal cancer risk was found for body mass index (BMI) in both sexes (OR for the lowest compared to the highest quintile was 1.47, 95% CI 0.93–2.33 in men and 8.11, 95% CI 1.38–47.66 in women) and for BMI at age 50 years (OR=1.65, 95% CI 0.88–3.11) in men and 7.84, 95% CI 0.69–88.58 in women). An inverse association was also observed with waist-to-hip ratio (WHR) at diagnosis in men only (OR=4.56, 95% CI 2.62–7.95 for the lowest compared to the highest quintile).
Conclusions: This study supports the existence of a relation between leanness and laryngeal cancer risk. In particular, men with less abdominal fat (characterized by a lower WHR) had an increased risk of laryngeal cancer.
Key words: body mass index, case-control study, laryngeal cancer, risk factors
|
introduction |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Tobacco smoking is, together with alcohol drinking, the major risk factor for laryngeal cancer [1, 2]. Beyond these established risk factors, a role for diet and nutrition has also been suggested, including in particular the favourable effect of fruit and vegetables [3, 4].
With reference to body mass index (BMI), a cohort study of American veterans including 515 laryngeal cancer cases among Whites and 67 among Blacks reported an inverse association, with a relative risk of 0.8 (95% confidence interval, CI, 0.7–0.8) for obese men as compared to non-obese ones among whites and of 0.5 (95% CI: 0.4–0.6) among blacks [5]. In a case-control study from Italy including 388 cases, an inverse association between BMI and laryngeal cancer risk was shown (odds ratio, OR, 2.0 for the lowest quartile as compared to the highest, 95% CI, 1.4–2.9) [6]. A subsequent case-control study from Italy, including 68 female cases, found a non significant inverse association between BMI and laryngeal cancer (OR = 0.4, 95% CI 0.2–1.1 for the highest versus the lowest tertile) [7]. None of these studies, however, assessed lifetime weight history.
To provide further information on the issue, we analysed data from a large case-control study on laryngeal cancer conducted in Italy, including detailed information on several anthropometric measures and weight at various ages. Female subjects included in the present case-control study were also considered (for BMI at diagnosis only) in a previous report on major risk factors for laryngeal cancer in women [7].
|
materials and methods |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Between January 1992 and November 2000, a case-control study on cancer of the larynx was conducted in the greater Milan area and the province of Pordenone, northern Italy [4]. Cases were 460 patients (415 men, 45 women) aged 30–80 years (median age 61 years) with incident, histologically-confirmed squamous-cell carcinoma of the larynx, admitted to the major teaching and general hospitals in the study areas. Controls were 1088 patients (863 men, 225 women) aged 31–79 years (median age 61 years) admitted to the same hospitals as cases for acute non-neoplastic conditions. Hospital admissions due to diseases related to tobacco smoking or alcohol drinking were excluded, as well as all chronic conditions which may have led to dietary modifications. Controls were frequency-matched with cases by quinquennia of age and area of residence, with a control-to-case ratio of about five for women, and of two for men. Control patients belonged to the following diagnostic categories: traumas, mostly fractures and sprains (26%); other orthopedic disorders, such as low back pain and disc disorders (31.8%); acute surgical conditions (22.6%); eye diseases (8.5%); and other miscellaneous diseases, such as skin and dental conditions (11.1%). Less than 5% of both cases and controls contacted refused to participate.
All interviews were carried out in a hospital setting by ad hoc trained interviewers using a standard structured questionnaire. This included information on socio-demographic characteristics, lifetime history of smoking and alcohol-drinking, physical activity at work and in the leisure time at different ages, and a validated food frequency questionnaire. Study subjects were asked to report their height and habitual weight in the two years before cancer diagnosis (or hospital admission for controls). Additional anthropometric measures were obtained, including weight at 30 and 50 years; and the circumferences of waist (measured 2 cm above the umbilicus) and hip (measured at the maximal protrusion) measured at the time of interview. Waist-to-hip ratio (WHR) was calculated as the ratio between these measures; BMI was computed as weight in kilograms divided by height in square meters; changes between BMI before cancer diagnosis (or hospital admission for controls) and weight at 30 or 50 years were also computed. The distribution of height, weight, WHR and BMI at several ages was different in men and women, hence sex-specific cut-off points were used.
OR, and the corresponding 95% CI were computed using unconditional multiple-logistic regression models [8]. Regression equations included terms for age (in quinquennia), study centre, years of education, tobacco smoking habit (never smoker, former smoker, current smoker of <15 cigarettes per day, 15–24 cigarettes per day, 
25 cigarettes per day), alcohol drinking (never, current or former drinker of less than 15 drinks per week, 15–41 drinks per week and 42 drinks per week), occupational physical activity at 30–39 years of age (very active, moderately active, and inactive), vegetable, fruit and non-alcohol energy intake (quintiles).
|
results |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Table 1 shows the distribution of laryngeal cancer cases and controls according to selected characteristics. By design, the proportion of women was higher in controls than in cases, and cases and controls had similar age distribution. Cases reported higher consumption of tobacco and alcohol, but did not differ from controls in terms of education.
|
The distribution of cases and controls, with the corresponding ORs, according to various body size measurements are shown in Table 2 for all subjects together, and by sex. No difference was observed between laryngeal cancer cases and controls with reference to height. An inverse association with laryngeal cancer risk was found for weight and BMI, with ORs of 1.45 (95% CI 0.94–2.24) and 1.66 (95% CI 1.08–2.57) for the lowest as compared to the highest quintile of weight and BMI, respectively. Corresponding values were 1.32 (95% CI 0.83–2.08) and 1.47 (95% CI 0.93–2.33) among men, and 5.17 (95% CI 0.89–29.89) and 8.11 (95% CI 1.38–47.66) among women. Despite the associations appearing stronger in women, these were based on small numbers of cases, and were consequently less precise. A stronger inverse association was observed for WHR in men (OR = 4.56, 95% CI 2.62–7.95 for a WHR <0.92 as compared with
0.99), with a significant trend in risk. No clear pattern of risk in relation to WHR was, however, found in women.
|
Table 3 gives the distribution of laryngeal cancer cases and controls and the corresponding ORs, according to BMI at different ages. BMI at age 30 was not associated to laryngeal cancer risk. An inverse association was observed for BMI at age 50, with ORs for the lowest quintile versus the highest of 1.65 (95% CI 0.88–3.11) in men, 7.84 (95% CI 0.69–88.58) in women, and 1.82 (95% CI 1.01–3.29) in all subjects. However, the trends in risk were not significant. No consistent relation was found with BMI change since age 30 or 50 (data not shown).
|
The effect of BMI and WHR in strata of selected covariates on laryngeal cancer risk is considered in Table 4. All the estimates for subjects in the lowest as compared to the highest quintiles of BMI and WHR were above unity, in the absence of significant heterogeneity. No clear pattern of risk was observed across strata of age, education, and alcohol drinking for both BMI and WHR. With reference to tobacco smoking, the highest point estimate was observed in smokers for BMI, but in non smokers for WHR.
|
|
discussion |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
The results of the present study – the first to our knowledge which allowed an assessment of laryngeal cancer risk in relation to lifetime history of a large number of body size measurements – indicate that leanness is an indicator of laryngeal cancer risk. In particular, men with less abdominal fat (characterized by a lower WHR) had an increased risk of laryngeal cancer. The inverse associations with BMI and WHR were similar – although apparently less marked – to those reported for cancers of the oral cavity, pharynx and oesophagus [6, 9, 10].
These associations were not explained by major recognized risk factors for laryngeal cancer, including tobacco smoking and alcohol drinking, as well as by socioeconomic indicators, dietary factors and physical activity. Although all these potential confounding factors were only weakly correlated with anthropometric measures before cancer diagnosis, we included them in our analyses.
Although the stronger associations were observed for anthropometric measures, the inverse relation between BMI and laryngeal cancer was already observed at age 50, i.e. about a decade on average prior to cancer diagnosis. Thus, it is unlikely that this association is due exclusively due to weight loss following early cancer symptoms. Further, leanness was not associated to a general reduction in food intake, since cases reported, if anything, a higher energy intake [11]. Since waist and hip measurements were made at interview and not self-reported, no information was available on WHR in the past.
It is possible that leanness in the few years before diagnosis is associated with decreased intake of protective nutrients or with increased intake of unfavorable ones. However, the intake of vegetables and fruit (two major favorable dietary correlates of laryngeal cancer) [3, 12] was not appreciably correlated with BMI and WHR, and these factors showed no significant confounding or modifying effect on any of the estimates.
The inverse relations between BMI and WHR and laryngeal cancer were consistent across strata of major covariates, the apparent variation observed being not significant, and hence attributable to the play of chance. Thus, in contrast to what has been observed for other cancers of the upper aerodigestive tract [6, 9, 10], tobacco smoking and alcohol drinking did not appear to meaningfully modify the associations observed. However, it is possible that the high consumption of alcohol and tobacco – which characterize most patients with laryngeal cancer – could have contributed in reducing their weight. Tobacco and alcohol may in fact affect energy balance and reduce body weight [13–17]. Further, in a Japanese study of 177 healthy workers, BMI was inversely associated to 8-hydroxydeoxyguanosine, a marker of oxidative DNA damage [18].
A potential source of bias is the use of hospital controls, since some of the diagnoses of controls have been associated with either reduced (e.g. fractures) or increased BMI (e.g. arthritis, cataract) [19, 20]. However, the association was observed across various diagnostic categories of controls, and the distributions of height and weight in the comparison group were similar to those from population surveys conducted in Italy [21]. In order to minimize recall bias, due to the onset of or treatment for cancer, we asked body weight in the two years prior to interview. Since, however, waist and hip measurements were made at interview, a possible reduction of WHR following cancer diagnosis cannot be excluded. The similar hospital setting may have in any case increased the comparability of information between cases and controls.
Among the strengths of the study are the large dataset, the high participation rate of cases and controls, the comparable catchment areas of the study subjects, the strict control for tobacco and alcohol as well as for other potential confounding factors, including total energy intake.
|
Acknowledgements |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
The authors thank Mrs I. Garimoldi for editorial assistance.
|
Footnotes |
|---|
Contract grant sponsor: Italian Association for Cancer Research, Milan, the Italian League against Cancer, and the Italian Ministry of Education (PRIN 2005).
Received for publication March 2, 2006. Revision received April 21, 2006. Revision received June 20, 2006. Accepted for publication June 20, 2006.
|
References |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
1. IARC. (1988) IARC Monographs on the evaluation of carcinogenic risks to humans. Alcohol drinking (International Agency for Research on Cancer, Lyon) Vol. 44:.
2. IARC. (2004) IARC Monographs on the evaluation of carcinogenic risks to humans. Tobacco smoke and involuntary smoking (International Agency for Research on Cancer, Lyon) Vol. 83:.
3. World Cancer Research Fund and American Institute Cancer Research. (1997) Food, nutrition and the prevention of Cancer: a global perspective. (World Cancer Research Fund, Washington, DC).
4. Talamini R, Bosetti C, La Vecchia C, et al. (2002) Combined effect of tobacco and alcohol on laryngeal cancer risk: a case-control study. Cancer Causes Control 13:957–964.[CrossRef][Web of Science][Medline]
5. Samanic C, Gridley G, Chow WH, et al. (2004) Obesity and cancer risk among white and black United States veterans. Cancer Causes Control 15:35–43.[CrossRef][Web of Science][Medline]
6. D'Avanzo B, La Vecchia C, Talamini R, Franceschi S. (1996) Anthropometric measures and risk of cancers of the upper digestive and respiratory tract. Nutr Cancer 26:219–227.[Web of Science][Medline]
7. Gallus S, Bosetti C, Franceschi S, et al. (2003) Laryngeal cancer in women: tobacco, alcohol, nutritional, and hormonal factors. Cancer Epidemiol Biomarkers Prev 12:514–517.
8. Breslow NE and Day NE. (1980) Statistical methods in cancer research. Vol. 1. The analysis of case–control studies. IARC Sci Publ No. 32.
9. Franceschi S, Dal Maso L, Levi F, et al. (2001) Leanness as early marker of cancer of the oral cavity and pharynx. Ann Oncol 12:331–336.
10. Gallus S, La Vecchia C, Levi F, et al. (2001) Leanness and squamous cell oesophageal cancer. Ann Oncol 12:975–979.
11. Bosetti C, La Vecchia C, Talamini R, et al. (2003) Energy, macronutrients and laryngeal cancer risk. Ann Oncol 14:907–912.
12. Bosetti C, La Vecchia C, Talamini R, et al. (2002) Food groups and laryngeal cancer risk: a case-control study from Italy and Switzerland. Int J Cancer 100:355–360.[CrossRef][Web of Science][Medline]
13. Klesges RC, Meyers AW, Klesges LM, La Vasque ME. (1989) Smoking, body weight, and their effects on smoking behavior: a comprehensive review of the literature. Psychol Bull 106:204–230.[CrossRef][Web of Science][Medline]
14. Perkins KA, Epstein LH, Marks BL, et al. (1989) The effect of nicotine on energy expenditure during light physical activity. N Engl J Med 320:898–903.[Abstract]
15. Hellerstedt WL, Jeffery RW, Murray DM. (1990) The association between alcohol intake and adiposity in the general population. Am J Epidemiol 132:594–611.
16. Colditz GA, Giovannucci E, Rimm EB, et al. (1991) Alcohol intake in relation to diet and obesity in women and men. Am J Clin Nutr 54:49–55.
17. Tavani A, Negri E, La Vecchia C. (1994) Determinants of body mass index: a study from northern Italy. Int J Obes Relat Metab Disord 18:497–502.[Web of Science][Medline]
18. Mizoue T, Kasai H, Kubo T, Tokunaga S. (2006) Leanness, smoking, and enhanced oxidative DNA damage. Cancer Epidemiol Biomarkers Prev 15:582–585.
19. Negri E, Pagano R, Decarli A, La Vecchia C. (1988) Body weight and the prevalence of chronic diseases. J Epidemiol Community Health 42:24–29.
20. Zang EA and Wynder EL. (1994) The association between body mass index and the relative frequencies of diseases in a sample of hospitalized patients. Nutr Cancer 21:247–261.[Web of Science][Medline]
21. Gallus S, Colombo P, Scarpino V, et al. (2006) Overweight and obesity in Italian adults 2004, and an overview of trends since 1983. Eur J Clin Nutr in press.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||