Tel.: +450-687-5010, Fax: +450-686-5599
Letter to the Editor
Comments on a recent meta-analysis: Obesity and lung cancer
Article first published online: 17 OCT 2012
Copyright © 2012 UICC
International Journal of Cancer
Volume 132, Issue 8, pages 1962–1963, 15 April 2013
How to Cite
El-Zein, M., Parent, M.-E. and Rousseau, M.-C. (2013), Comments on a recent meta-analysis: Obesity and lung cancer. Int. J. Cancer, 132: 1962–1963. doi: 10.1002/ijc.27854
- Issue published online: 13 FEB 2013
- Article first published online: 17 OCT 2012
- Accepted manuscript online: 18 SEP 2012 05:39AM EST
- Manuscript Accepted: 13 SEP 2012
- Manuscript Received: 22 AUG 2012
The role of body mass index (BMI) in relation to lung cancer risk has been the object of considerable interest over the years. For the most part, epidemiologists' primary concern has been to determine whether BMI has an effect on lung cancer that is independent of smoking, a key confounder. We, therefore, read with great interest the quantitative appraisal of the epidemiological evidence on obesity and incidence of lung cancer by Yang et al.1 Their meta-analysis revealed a significant inverse relationship between BMI, used as a proxy measure of overweight and obesity, and lung cancer based on results from 20 cohort and 11 case–control studies. We would like to comment on some inter-related methodological and interpretational aspects of the analyses presented, namely, heterogeneity across studies, methodological quality of original studies and confounding by smoking.
An overall estimate of association of 0.79 [95% confidence interval (CI): 0.73–0.85] was reported for excess weight (BMI ≥ 25 kg/m2) compared with normal weight (BMI = 18.5–24.9 kg/m2) based on 31 studies. However, the inconsistency coefficient I2 statistic, an estimate of the proportion of total variation in study estimates due to heterogeneity, was 82.9%. In the presence of such heterogeneity, a pooled estimate, even from a random-effects model, may not be justifiable as it does not provide a fair representation of overall results. Visual inspection of the Forest plot (Fig. 2), where study-specific risk estimates and their CIs are plotted, also indicates a heterogeneous set of studies. In particular, heterogeneity could be attributed to six cohort2–7 and five case–control8–12 studies. It would have been informative to identify which aspects of the individual studies contributed to heterogeneity. This could have entailed (1) identifying distinctive characteristics of individual studies; (2) pinpointing sources of variability between studies and thereby (3) providing guidance for presenting a valid quantitative pooled estimate of association between obesity and lung cancer.
There was still a strong indication of heterogeneity in most stratified analyses across several important features defined by methodological characteristics. Results varied substantially within studies defined by study design, gender, study population, body size assessment method and smoking status. Of note, in analyses restricted to histological subtypes, heterogeneity might have tapered off as a result of the small number of studies included.
Methodological Quality of Original Studies
The synthesis of results from individual studies did not include a formal evaluation of study quality. Analyses stratified according to a score for methodological quality could have provided a way to explore the sources of heterogeneity between studies. If results of this meta-analysis were to be significantly affected by the quality of original studies, then its conclusion may be less meaningful. Moreover, although the lack of symmetry in the funnel plot (Fig. 3) could be related to publication bias and/or the presence of heterogeneity, asymmetry might have also resulted from the overestimation of exposure effects in smaller studies of lower methodological quality. Could it be that study quality differed considerably between larger and smaller studies, or between case–control and cohort studies?
Numerous appraisal tools are currently available for evaluating the methodological quality of studies, with the flexibility of formulating additional study-specific criteria. For instance, one can adapt a given quality assessment instrument and assign a higher quality score to studies that included more than a single aspect of an individual's smoking history such as intensity, duration and time since cessation; variables conventionally adjusted for in BMI–lung cancer association studies.
Confounding by Smoking
Most fundamentally, controversy still centers on the effect of smoking in studies investigating the relationship between BMI and lung cancer. Subgroup analyses according to smoking found that excess weight is a protective factor against lung cancer, irrespective of smoking status. While the inverse association was indeed strengthened in current and former smokers compared with the overall meta-analysis, a high degree of heterogeneity was still detected among former smokers (I2 = 77.0%), although less pronounced among current smokers (I2 = 34.6%). Figure 4 revealed an inverse association in female non-smokers, but still based on significant statistical heterogeneity (I2 = 74.1%).
The authors were most likely restricted in their exploration of smoking effects by the variety of approaches and metrics used in individual studies to account for smoking. This being said, future efforts may be most beneficial if a more comprehensive approach to represent lifetime smoking behavior were to be embraced. As an illustrative example, Tarnaud et al. recently examined the association between BMI and lung cancer separately by smoking status and by further adjusting for lifetime tobacco smoking exposure using a comprehensive smoking index that incorporated several smoking-related characteristics (i.e., smoking status, intensity, duration and time since cessation).13
Furthermore, the general tendency in previous studies has been to present results along the following classification of BMI categories: underweight, normal weight, overweight and obese. Given the inter-relationship between smoking and body weight, analysis of data pertaining to the lowest BMI category in relation to cancer risk might have provided further insights. In fact, the authors seem to have extrapolated the findings from their meta-analysis to suggest a higher risk of lung cancer among smokers with low BMIs, although no pooled lung cancer risk estimates were presented for underweight subjects compared to those with normal BMI. In this respect, it is unclear which findings warranted the article's concluding remark that “smokers should improve their nutritional status and maintain a suitable body weight.”
In conclusion, while conducting a meta-analysis on the relation between BMI and lung cancer represents a highly valuable contribution, we believe that caution should be exercised when interpreting the results presented by Yang et al.1 Consideration of the above-mentioned concerns would indeed provide reassurance in the validity of the meta-estimates presented.