Virtual Library
Start Your Search
Tricia L. Larose
Author of
-
+
P2.06 - Epidemiology/Primary Prevention/Tobacco Control and Cessation (ID 707)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Epidemiology/Primary Prevention/Tobacco Control and Cessation
- Presentations: 1
- Moderators:
- Coordinates: 10/17/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
-
+
P2.06-001 - Circulating Cotinine Concentrations, Self-Reported Smoking, and Lung Cancer Risk in the Lung Cancer Cohort Consortium (LC3) (ID 7509)
09:30 - 16:00 | Presenting Author(s): Tricia L. Larose
- Abstract
Background:
Tobacco exposure is the main determinant of lung cancer. Data on tobacco exposure is often collected via self-reported smoking which is prone to misclassification. In the current study we sought to evaluate the extent to which cotinine, an objective measure of recent tobacco exposure, can inform on lung cancer risk.
Method:
We used data from 20 international cohorts and designed a nested case-control study including 5364 incident lung cancer cases and 5364 controls. Cases and controls were individually matched by age, sex, cohort, and smoking status (never, former, current). Cases from never and former smokers were intentionally oversampled. Centralised biochemical analysis was performed to measure circulating cotinine concentrations as an objective indicator of recent tobacco exposure. The association between cotinine and lung cancer risk was evaluated using conditional logistic regression. Discrimination between future lung cancer cases and controls was evaluated using receiver operating characteristic (ROC) curves.
Result:
Higher circulating cotinine concentrations were strongly associated with lung cancer risk among current smokers (OR~ per doubling in cotinine concentrations ~[~log2~]: 1.22, 95% confidence interval [CI]: 1.17-1.28, especially women (OR ~log2~: 1.35, 95% CI: 1.22-1.50). Among former smokers, positive associations between higher cotinine concentrations and lung cancer risk were driven by those participants with cotinine levels consistent with active smoking (OR ~log2~: 1.08, 95% CI: 1.05-1.11). We saw no association between cotinine concentrations and risk in never smokers, with the exception of men from the United States whose cotinine levels may be indicative of active smoking (OR: 1.17, 95% confidence interval [CI]: 1.02-1.34). Having cotinine levels consistent with active smoking (>115 nmol/L) was more common among self-reported former smokers (cases: 14.6%, controls: 9.2%) and less common among self-reported never smokers (cases: 2.7%, controls: 0.8%). A risk prediction analysis in current smokers indicated that circulating cotinine combined with self-reported smoking can provide a small improvement in discrimination between future lung cancer cases and controls in comparison to self-reported smoking alone (improved area under the curve for men: 2%, women: 4%).
Conclusion:
Cotinine as an objective measure of tobacco exposure was consistently associated with lung cancer risk among current smokers, especially women. Former and never smokers with cotinine levels consistent with active smoking also showed increased risk. For lung cancer risk prediction modelling among current smokers, cotinine combined with self-reported smoking performed slightly better than cotinine or self-reported smoking alone.
