Specific risk factors for lung cancer in women: impact of lifetime hormonal exposure. Results from the WELCA (Women Epidemiology Lung Cancer)
GUEYE Y. 1, CORDINA DUVERGER E. 1, GUÉNEL P. 1, WISLEZ M. 2, TRÉDANIEL J. 3, BLONS H. 4, LAURENT-PUIG P. 4, ANTOINE M. 5, RADOÏ L. 1, CANONICO M. 1
1 Université Paris-Saclay, UVSQ, Inserm, Gustave Roussy, CESP, Inserm UMRS1018, Exposome Heredity Cancer and Health Team, Villejuif, France; 2 Thoracic Oncology Unit, Cochin Hospital, Institut du Cancer Paris Carpem, Assistance Publique Hôpitaux de Paris (AP-HP Center), Universite Paris Cité, Centre de Recherche des Cordeliers, UMRS 1138 Complement, Inflammation and Cancers, Paris, France; 3 Groupe Hospitalier Paris Saint Joseph, Université Paris Cité, Paris, France; 4 Institut du cancer Paris Carpem, Assistance Publique Hôpitaux de Paris (AP-HP Center), Universite Paris Cité, Department of Genomic Medicine for Tumors and Cancers, Centre de Recherche des Cordeliers, INSERM, CNRS SNC 5096, Sorbonne Universite, Universite, Paris, France; 5 Assistance Publique Hôpitaux de Paris (AP-HP Center), Tenon Hospital, Department of Pathology, Paris, France
Background
With nearly two million new cases annually worldwide, lung cancer is among the most frequent cancers and remains the leading cause of cancer-related death1. However, incidence trends differ by sex: they are decreasing among men but rising steadily among women, for whom a substantial proportion of cases remain unexplained, suggesting the existence of sex-specific risk factors2. Estrogen exposure has been proposed as a possible explanation, but the available evidence remains inconsistent. This study aimed to investigate the influence of lifetime exposure to hormonal factors on lung cancer risk in women.
Methods
WELCA is a multicentre population-based case-control study conducted in the Ile-de-France region between 2014 and 2017 among women aged 18 to 75 years. The study included 716 cases of incident primary lung cancer and 756 frequency-matched controls by age and administrative area. Detailed information was collected through standardized questionnaires on lifestyle factors, occupational history, menstrual and reproductive history, and use of exogenous hormonal treatments, including contraceptives, fertility treatments and menopausal hormone therapy, with information on type, timing and duration of use. Association between hormonal factors and lung cancer risk were estimated using multivariable logistic regression models adjusted for relevant confounders.
Results
The mean age was 60.9 years for cases and 61.9 years for controls. After adjustment for lifetime smoking consumption, educational level, body-mass index, occupational exposures and mutual adjustment for hormonal factors, a later age at last pregnancy (≥33 years vs ≤28 years) was associated with a lower risk of lung cancer (OR=0.56; IC95%: 0.39-0.80; p for trend<0.01). Ever use of hormonal contraceptives was also associated with an overall reduced risk (OR=0.55; IC95%: 0.37-0.82), both for oral contraceptives (OR=0.59; IC95%: 0.40-0.87) and for other types of contraception (OR=0.52; IC95%: 0.31-0.85). However, a later age at cessation of the pill use was associated with an increased risk (≥39 years vs ≤30 years: OR=1.43; IC95%: 1.04-1.98; p for trend=0.05). In addition, a longer reproductive lifespan was inversely associated with the lung cancer risk (≥39 years vs ≤36 years: OR=0.67; IC95%: 0.49-0.93; p for trend=0.01). No associations were observed for other hormonal factors (age at puberty, length and regularity of menstrual cycles, parity, breastfeeding, fertility treatments, age at onset and type of menopause, or menopausal hormone therapy) after fully adjustment. Results were consistent across major histological subtypes, including adenocarcinoma, which accounted for the majority of cases. Associations remained similar in analyses restricted to never-smokers, suggesting that findings were not driven by residual confounding by tobacco exposure.
Conclusion
These findings suggest that specific aspects of lifetime hormonal exposure, particularly reproductive timing, contraceptive use and duration of reproductive lifespan, may influence lung cancer risk in women. They support the hypothesis of a hormonally mediated pathway and highlight the need for further studies integrating hormonal history to better understand sex-specific lung cancer risk.
References
[1] Bray F and al., Global cancer statistics 2022. CA Cancer J Clin. 2024
[2] Overview of cancer in France 2025, 20 years special edition