22PD - Body mass index and lung cancer survival: Results from the ICARE study

Date 17 April 2015
Event ELCC 2015
Session Epidemiology, early stage NSCLC and surgery
Topics Cancer Aetiology, Epidemiology, Prevention
Lung and other Thoracic Tumours
Presenter Harinakshi Sanikini
Citation Annals of Oncology (2015) 26 (suppl_1): 6-9. 10.1093/annonc/mdv044
Authors H. Sanikini1, G. Clain2, M. Sanchez2, A.V. Guizard3, A.S. Woronoff4, B. Tretarre5, F. Molinie6, P. Delafosse7, D. Luce8, I. Stücker2
  • 1Cancer Epidemiology: Gene, Environment, Hopital Paul Brousse, 94807 - Villejuif/FR
  • 2Cancer Epidemiology: Gene, Environment, Hopital Paul Brousse, Villejuif/FR
  • 3Registre Des Cancers Du Calvados, Centre Francois Baclesse, Caen/FR
  • 4Registre Des Cancers Du Doubs, CHU Besançon, Hôpital Jean Minjoz, Besançon/FR
  • 5Registre Des Cancers De L'hérault, ICM Regional Cancer Institute of Montpellier, Montpellier/FR
  • 6Registre Des Cancers De Loire-atlantique Et De Vendée, CHU de Nantes, Nantes/FR
  • 7Registre Des Cancers De L'isère, CH Mutualiste de Grenoble, Grenoble/FR
  • 8Santé Et Environnement, Inserm UMRS 1085, Pointe-à-Pitre/FR



Several epidemiological studies have shown that lower body mass index (BMI) is associated with increased risk of lung cancer. However, studies investigating the association between BMI and lung cancer survival are limited. We investigated whether BMI 2 years before interview and BMI at interview is associated with lung cancer survival in the ICARE study (Investigation of occupational and environmental causes of respiratory cancers).


A population-based case-control study was conducted in ten departments of France between 2001 and 2007. Analyses were based on 2,285 lung cancer cases. Data on anthropometric measurements (height, weight 2 years before interview and weight at interview) were collected using standardized questionnaires through face-to-face interviews, and these data were used to calculate BMI 2 years before interview and BMI at interview. BMI (kg/m2), was classified into 4 categories: (<18.5), (18.5-25), (25-30) and (≥30). Cox proportional hazard models were used to estimate the hazard ratios (HR) and corresponding 95 % confidence interval (CI). Models were adjusted for age, gender, area of residence, education level and comprehensive smoking index. BMI (18.5-25 kg/m2) was used as a reference group.


Of the 2,285 lung cancer cases that were included in this study, a total of 1,988 subjects died. No significant association was observed between BMI 2 years before interview and lung cancer mortality (Table 1). For BMI at interview and lung cancer mortality, we found a significant positive association among subjects with BMI <18.5 kg/m2, while a significant inverse association was observed among subjects with BMI 25-30 kg/m2 and BMI ≥30 kg/m2 (Table 1).

Adjusted hazard ratios for lung cancer mortality according to BMI categories, the ICARE study 2001-2007.

Deaths (N) HR (95% CI)
BMI 2 years before interview (kg/m2)
<18.5 56 0.96 (0.73-1.27)
18.5-25 891 Reference
25-30 647 0.98 (0.88-1.09)
≥ 30 288 1.00 (0.87-1.14)
Missing 106 -
BMI at interview (kg/m2)
< 18.5 184 1.39 (1.19-1.64)
18.5-25 1071 Reference
25-30 514 0.78 (0.70-0.87)
≥ 30 166 0.82 (0.69-0.97)
Missing 53 -


These data suggests that BMI 2 years before interview is not associated with lung cancer mortality. However, lower BMI (<18.5 kg/m2) at interview is associated with increased lung cancer mortality.


All authors have declared no conflicts of interest.