Associations Between Sex-Specific Reproductive Factors and Risk of New-Onset Lung Cancer Among Female Patients

BackgroundSeveral characteristics distinguish lung cancer in female patients from that in male patients, with adenocarcinoma being more prevalent in female patients and occurring more frequently in female patients who do not smoke. Uncertainty surrounds the relationship between female-specific reproductive factors and lung cancer risk.Research QuestionAre sex-specific reproductive factors associated with risk of lung cancer in different genetic risk groups and histologic types?Study Design and MethodsA Cox proportional hazard model was used to evaluate the association between multiple reproductive factors and the risk of lung cancer developing in a prospective cohort study involving 273,190 female individuals from the UK Biobank. Subgroup analyses stratified by age, smoking status, BMI, genetic risk, and histologic subtype were conducted to emphasize the modification effects further.ResultsA total of 1,182 cases of lung cancer in female patients were recorded over a median follow-up period of 12.0 years in the cohort study. In multivariable-adjusted models, early menarche (age ≤ 11 years: hazard ratio [HR], 1.22; 95% CI, 1.03-1.46), early menopause (age ≤ 46 years: HR, 1.49; 95% CI, 1.19-1.86), a shorter reproductive span (≤ 32 years: HR, 1.42; 95% CI, 1.18-1.71; and 33-35 years: HR, 1.24; 95% CI, 1.00-1.53), and early age at first birth (age ≤ 20 years: HR, 1.63; 95% CI, 1.33-2.01) were associated with a higher risk of lung cancer. Stratified analysis revealed that several reproductive factors, including early age at menopause, shortened reproductive span, and early age at first birth, showed a substantially stronger relationship with an elevated risk of lung cancer, particularly of lung adenocarcinoma, in populations with high genetic risk and more detrimental behaviors.InterpretationEarly age at menopause, a shortened reproductive life span, and early age at first birth were associated with higher risks of lung cancer, particularly of lung adenocarcinoma, in a subpopulation with higher genetic susceptibility and detrimental behaviors. The evidence provided by this study emphasizes the significance of screening for multiple reproductive factors to prevent lung cancer among female individuals. Several characteristics distinguish lung cancer in female patients from that in male patients, with adenocarcinoma being more prevalent in female patients and occurring more frequently in female patients who do not smoke. Uncertainty surrounds the relationship between female-specific reproductive factors and lung cancer risk. Are sex-specific reproductive factors associated with risk of lung cancer in different genetic risk groups and histologic types? A Cox proportional hazard model was used to evaluate the association between multiple reproductive factors and the risk of lung cancer developing in a prospective cohort study involving 273,190 female individuals from the UK Biobank. Subgroup analyses stratified by age, smoking status, BMI, genetic risk, and histologic subtype were conducted to emphasize the modification effects further. A total of 1,182 cases of lung cancer in female patients were recorded over a median follow-up period of 12.0 years in the cohort study. In multivariable-adjusted models, early menarche (age ≤ 11 years: hazard ratio [HR], 1.22; 95% CI, 1.03-1.46), early menopause (age ≤ 46 years: HR, 1.49; 95% CI, 1.19-1.86), a shorter reproductive span (≤ 32 years: HR, 1.42; 95% CI, 1.18-1.71; and 33-35 years: HR, 1.24; 95% CI, 1.00-1.53), and early age at first birth (age ≤ 20 years: HR, 1.63; 95% CI, 1.33-2.01) were associated with a higher risk of lung cancer. Stratified analysis revealed that several reproductive factors, including early age at menopause, shortened reproductive span, and early age at first birth, showed a substantially stronger relationship with an elevated risk of lung cancer, particularly of lung adenocarcinoma, in populations with high genetic risk and more detrimental behaviors. Early age at menopause, a shortened reproductive life span, and early age at first birth were associated with higher risks of lung cancer, particularly of lung adenocarcinoma, in a subpopulation with higher genetic susceptibility and detrimental behaviors. The evidence provided by this study emphasizes the significance of screening for multiple reproductive factors to prevent lung cancer among female individuals. Take-home PointsStudy Question: Are sex-specific reproductive factors associated with the risk of lung cancer in different genetic risk groups and histologic types?Results: Early menopause, shortened reproductive span, and early age at first birth are associated with a higher risk of lung cancer, particularly of lung adenocarcinoma, in those populations with a high genetic risk and more detrimental behaviors.Interpretation: This study emphasizes the significance of screening for multiple reproductive factors for the prevention of lung cancer in female individuals. Study Question: Are sex-specific reproductive factors associated with the risk of lung cancer in different genetic risk groups and histologic types? Results: Early menopause, shortened reproductive span, and early age at first birth are associated with a higher risk of lung cancer, particularly of lung adenocarcinoma, in those populations with a high genetic risk and more detrimental behaviors. Interpretation: This study emphasizes the significance of screening for multiple reproductive factors for the prevention of lung cancer in female individuals. Lung cancer is the second most common cancer and the leading cause of cancer-related mortality in the world.1Ferlay J. Colombet M. Soerjomataram I. et al.Cancer incidence and mortality patterns in Europe: estimates for 40 countries and 25 major cancers in 2018.Eur J Cancer. 2018; 103: 356-387Abstract Full Text Full Text PDF PubMed Scopus (1669) Google Scholar,2Dyba T. Randi G. Bray F. et al.The European cancer burden in 2020: incidence and mortality estimates for 40 countries and 25 major cancers.Eur J Cancer. 2021; 157: 308-347Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar In 2020, 2.21 million new cases of lung cancer (11.4% of all cancer cases) and 1.80 million deaths (18.0% of all deaths resulting from cancer) were estimated by the Global cancer statistics 2020,3Sung H. Ferlay J. Siegel R.L. et al.Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.CA Cancer J Clin. 2021; 71: 209-249Crossref PubMed Scopus (48824) Google Scholar thus causing a tremendous burden on socioeconomic and medical resources. Compared with lung cancer in male patients with a history of smoking, a subtype of adenocarcinoma occurs more frequently in female patients who have never smoked.4Zang E.A. Wynder E.L. Differences in lung cancer risk between men and women: examination of the evidence.J Natl Cancer Inst. 1996; 88: 183-192Crossref PubMed Scopus (629) Google Scholar Although smoking exhibits a dose-response relationship with all major subtypes of lung cancer, this relationship is attenuated in adenocarcinoma, indicating that other risk factors may play a significant role in this cell type.5Kabat G.C. Aspects of the epidemiology of lung cancer in smokers and nonsmokers in the United States.Lung Cancer. 1996; 15: 1-20Abstract Full Text PDF PubMed Scopus (69) Google Scholar Female individuals who do not smoke are approximately three times more susceptible to lung cancer than male individuals who do not smoke. Although the incidence of lung cancer in females has increased substantially in recent decades worldwide, particularly lung adenocarcinoma (LUAD),6Gabrielson E. Worldwide trends in lung cancer pathology.Respirology. 2006; 11: 533-538Crossref PubMed Scopus (112) Google Scholar the risk factors, with the exception of smoking and pollution, have not been identified fully. Hormonal and reproductive factors are thought to be involved in the occurrence of lung cancer in female patients5Kabat G.C. Aspects of the epidemiology of lung cancer in smokers and nonsmokers in the United States.Lung Cancer. 1996; 15: 1-20Abstract Full Text PDF PubMed Scopus (69) Google Scholar,7Schabath M.B. Wu X. Vassilopoulou-Sellin R. Vaporciyan A.A. Spitz M.R. Hormone replacement therapy and lung cancer risk: a case-control analysis.Clin Cancer Res. 2004; 10: 113-123Crossref PubMed Scopus (172) Google Scholar; however, these findings are far from consistent across studies. Numerous studies have attempted to investigate sex discrepancies by examining the relationship between reproductive factors and multiple cancers.8Lambertini M. Santoro L. Del Mastro L. et al.Reproductive behaviors and risk of developing breast cancer according to tumor subtype: a systematic review and meta-analysis of epidemiological studies.Cancer Treat Rev. 2016; 49: 65-76Abstract Full Text Full Text PDF PubMed Scopus (163) Google Scholar, 9Kvale G. Heuch I. Ursin G. Reproductive factors and risk of cancer of the uterine corpus: a prospective study.Cancer Res. 1988; 48: 6217-6221PubMed Google Scholar, 10Chiaffarino F. Pelucchi C. Parazzini F. et al.Reproductive and hormonal factors and ovarian cancer.Ann Oncol. 2001; 12: 337-341Abstract Full Text PDF PubMed Scopus (88) Google Scholar In addition to reproductive tumors, these reproductive factors and fluctuations in sex hormones have been linked to tumors stemming from nonreproductive organs and tissues, such as the colorectum, bladder, and lung.11Xu X. Mo Q. Shen H. Wang S. Liu B. Reproductive and hormonal factors and bladder cancer risk: a prospective study and meta-analysis.Aging (Albany NY). 2020; 12: 14691-14698Crossref PubMed Scopus (8) Google Scholar, 12Murphy N. Xu L. Zervoudakis A. et al.Reproductive and menstrual factors and colorectal cancer incidence in the Women's Health Initiative Observational Study.Br J Cancer. 2017; 116: 117-125Crossref PubMed Scopus (29) Google Scholar, 13Ben Khedher S. Neri M. Papadopoulos A. et al.Menstrual and reproductive factors and lung cancer risk: a pooled analysis from the international lung cancer consortium.Int J Cancer. 2017; 141: 309-323Crossref PubMed Scopus (28) Google Scholar The International Lung Cancer Consortium conducted a case-control study involving 2,064 patients with lung cancer and 5,342 control participants and discovered that late onset of menarche at 17 years of age or older, late onset of menopause at 55 years of age or older, and nonnatural menopause were linked to an elevated risk of lung cancer.14Jin K. Hung R.J. Thomas S. et al.Hormonal factors in association with lung cancer among Asian women: a pooled analysis from the International Lung Cancer Consortium.Int J Cancer. 2021; 148: 2241-2254Crossref PubMed Scopus (7) Google Scholar Conversely, another case-control study in Germany of 811 female participants with lung cancer and 912 female control participants found no association between lung cancer and reproductive factors regarding age at menarche and menopause, length of menstrual cycle, and number of live births and found a reduction in risk with the administration of hormone replacement therapy (HRT) and oral contraceptive pills (OCPs).15Kreuzer M. Gerken M. Heinrich J. Kreienbrock L. Wichmann H.E. Hormonal factors and risk of lung cancer among women?.Int J Epidemiol. 2003; 32: 263-271Crossref PubMed Scopus (148) Google Scholar Despite numerous case-control and cross-sectional studies conducted with relatively small sample sizes to assess the associations between multiple reproductive factors and lung cancer, the results have been inconclusive. Consequently, we conducted a large prospective cohort study based on the data of female individuals from the UK Biobank to investigate whether an association exists between sex-specific multiple reproductive factors and the incidence of lung cancer. This study was performed using data on females from the UK Biobank (application No. 84979). The UK Biobank collected medical information and genetic data from more than half a million UK participants between 2006 and 2010, which was approved by the National Information Governance Board for Health and Social Care and the North West Multicenter Research Ethics Committee. All the participants from UK Biobank provided informed consent for recruitment. In this study, we analyzed multiple self-reported reproductive factors, including age at menarche and menopause; reproductive life span; age at first live birth; number of live births, stillbirths, miscarriages, and abortions; history of hysterectomy or oophorectomy; as well as exposure to exogenous hormones such as OCPs and HRT. Early menarche was classified as the onset of menstruation before the age of 12 years. Early menopause was defined as the complete absence of menstruation before 47 years of age. The age difference between menarche and menopause was equated to the reproductive life span. Lung cancer incidence was determined according to the International Classification of Diseases, 10th Revision, codes (C33 and C34) from the UK Biobank. To determine the date of initial lung cancer diagnosis after the baseline assessment, we used inpatient data taken from hospitals in England, Scotland, and Wales, as well as national death registers. Patients with lung cancer were followed up from the date of diagnosis until the end of follow-up (October 2020), loss to follow-up, or death, whichever came first.16Xie J. Zhu M. Ji M. et al.Relationships between sleep traits and lung cancer risk: a prospective cohort study in UK Biobank.Sleep. 2021; 44: zsab089Crossref Scopus (24) Google Scholar In stratified analyses, lung cancer was categorized further into two main histologic subtypes: non-small cell lung cancer (NSCLC) and small cell lung cancer. Small cell lung cancer diagnosis was defined by International Classification of Diseases for Oncology, Third Edition, codes 8041, 8043, 8044, and 8045, and NSCLC diagnosis was characterized by squamous cell carcinomas (codes 8070, 8071, 8072, and 8083), adenocarcinomas (codes 8140, 8144, 8230, 8250, 8253, 8254, 8256, 8257, 8260, 8265, 8333, 8480, and 8551), and other remaining carcinoma codes.17Morgan E. Arnold M. Rutherford M.J. et al.The impact of reclassifying cancers of unspecified histology on international differences in survival for small cell and non-small cell lung cancer (ICBP SurvMark-2 project).Int J Cancer. 2021; 149: 1013-1020Crossref PubMed Scopus (8) Google Scholar,18World Health OrganizationInternational Classification of Diseases for Oncology (ICD-O)-3rd edition. 2013. World Health Organization. Accessed December 25, 2023.https://apps.who.int/iris/handle/10665/96612Google Scholar Continuous variables of baseline characteristics with normal distribution curves are shown as mean ± SD, and categorical variables are presented as percentages; these then were compared using Student t test and the χ2 test, respectively. Cox proportional hazards models with crude and fully adjusted models were used to estimate the associations between sex-specific reproductive factors and lung cancer incidence using hazard ratios (HRs) and 95% CIs. Adjusted covariates in the full model included age, ethnicity, BMI, smoking status, alcohol status, education level, and Townsend deprivation index (e-Appendix 1). The proportional hazards assumption was evaluated in accordance with Schoenfeld residuals and the Kaplan-Meier method, which was not violated by reproductive factors and covariates of interest in the multivariate models for any outcome. In addition, multiple adjusted restricted cubic splines19Gong J. Harris K. Peters S.A.E. Woodward M. Reproductive factors and the risk of incident dementia: a cohort study of UK Biobank participants.PLoS Med. 2022; 19e1003955Crossref Scopus (29) Google Scholar were constructed to demonstrate the shape of relationships between the risk of lung cancer and multiple reproductive factors. After eliminating 2.5% of the top and bottom distributions with poor precision, the median of these distributions was regarded as the reference value. Cumulative risk curves were used to determine the time intervals between various groups of reproductive factors. Moreover, to evaluate the impact of comprehensive reproductive factors thoroughly, we not only developed a model to estimate the effect of parturition-related factors containing the number of births, stillbirths, miscarriages, and abortions, but also analyzed the effect of multiple factors relevant to exogenous or endogenous sex hormones, including age at menopause, history of oophorectomy or hysterectomy, and administration of HRT and OCPs. Stratified analyses according to age, smoking status, BMI, 18 single nucleotide polymorphism-derived genetic risk scores (e-Appendix 1, e-Table 1),20McKay J.D. Hung R.J. Han Y. et al.Large-scale association analysis identifies new lung cancer susceptibility loci and heterogeneity in genetic susceptibility across histological subtypes.Nat Genet. 2017; 49: 1126-1132Crossref PubMed Scopus (353) Google Scholar and histologic subtypes were conducted to further explore the potential modifiable effects of some certain specific characteristics. To determine the strength of the results further, sensitivity analyses were performed with the exclusion of participants with histories of either oophorectomy or hysterectomy and multiple imputations with noncritical missing covariates such as age, BMI, race, smoking status, alcohol status, education level, and Townsend deprivation index. Participants who demonstrated lung cancer within the second or fifth year of follow-up also were excluded. To reduce the confounding effects of potential risk factors on reproductive factors, propensity score matching was conducted to match the participants with and without lung cancer. This was performed by generating two categories for each reproductive factor (reference categories consistent with those in the Cox regression) and using propensity score matching to exclude unmatched individuals based on the set of covariates, using a nearest neighbor ratio of 1 to 10 with replacement. All statistical analyses were conducted using R version 4.1.3 software (R Foundation for Statistical Computing) or STATA MP version 14.0 software (StataCorp). To reduce the type I error, the false discovery rate correction was applied for multiple comparisons, and a P value of < .05 was considered statistically significant (two-sided). As illustrated in e-Figure 1, 229,126 male patients were excluded initially, along with 175 participants with lung cancer at baseline. Finally, 273,190 female participants were enrolled in the study and 1,182 cases of new-onset lung cancer were recorded during a median follow-up period of 12.0 years. The baseline characteristics of the female participants included in the study were summarized in Table 1, which indicated that the participants with a lung cancer diagnosis were older, had lower educational backgrounds, and were more likely to currently smoke. In terms of reproductive factors, lung cancer was correlated with an earlier age at menarche and menopause, shorter reproductive span, earlier age at first live birth, lower likelihood of OCP use, and greater exposure to HRT. Information missing for individual reproductive factors of interest was provided in e-Table 2.Table 1Baseline Characteristics of 273,190 UK Biobank ParticipantsVariableNo Lung Cancer (n = 272,008)Lung Cancer (n = 1,182)Total (n = 273,190)P ValueAge, y56.32 ± 8.0061.43 ± 5.8756.34 ± 8.00< .001Race< .001 White256,101 (94.2)1,156 (97.8)257,257 (94.2) Other14,647 (5.4)22 (1.9)14,669 (5.4) Missing data1,260 (0.5)4 (0.3)1,264 (0.5)Education level< .001 College or university degree84,301 (31.0)193 (16.3)84,494 (30.9) Professional qualifications27,644 (10.2)149 (12.6)27,793 (10.2) A levels/AS levels or equivalent31,900 (11.7)86 (7.3)31,986 (11.7) O levels/GCSEs or equivalent77,395 (28.5)279 (23.6)77,674 (28.4) None of the above45,467 (16.7)439 (37.1)45,906 (16.8) Prefer not to answer2,831 (1.0)17 (1.4)2,848 (1.0) Missing data2,470 (0.9)19 (1.6)2,489 (0.9)Smoking status< .001 No smoking history161,785 (59.5)225 (19.0)162,010 (59.3) Past smoking history84,829 (31.2)506 (42.8)85,335 (31.2) Current smoking history23,892 (8.8)442 (37.4)24,334 (8.9) Prefer not to answer1,054 (0.4)8 (0.7)1,062 (0.4) Missing data448 (0.2)1 (0.1)449 (0.2)Alcohol use status< .001 Never15,910 (5.8)58 (4.9)15,968 (5.8) Former9,867 (3.6)99 (8.4)9,966 (3.6) Current245,406 (90.2)1,023 (86.5)246,429 (90.2) Prefer not to answer375 (0.1)1 (0.1)376 (0.1) Missing data450 (0.2)1 (0.1)451 (0.2)BMI, kg/m227.09 ± 5.2027.15 ± 4.9327.09 ± 5.20< .001Townsend deprivation index–1.34 ± 3.04–0.23 ± 3.43–1.33 ± 3.04< .001Age at menarche, y12.97 ± 1.6212.96 ± 1.7812.97 ± 1.62.850No. of children< .001 050,875 (18.7)194 (16.2)51,069 (18.7) 136,289 (13.3)149 (13.3)36,438 (13.4) 2118,637 (43.6)475 (39.8)119,112 (43.8) 348,021 (17.7)243 (21.2)48,264 (17.8) ≥ 417,362 (6.4)117 (9.5)17,479 (6.4)Age at first birth, y25.32 ± 4.6223.26 ± 4.2125.31 ± 4.62< .001No. of miscarriages.189 030,671 (35.5)142 (37.2)30,813 (35.5) 139,870 (46.3)161 (41.6)40,031 (46.2) ≥ 215,791 (18.3)85 (21.3)15,876 (18.3)No. of stillbirths.004 079,534 (92.0)345 (87.1)79,879 (92.0) 16,028 (6.9)41 (11.4)6,069 (6.9) ≥ 2955 (1.1)7 (1.5)962 (1.1)No. of abortions.109 048,188 (56.1)215 (54.2)48,403 (56.1) 130,379 (35.4)148 (39.9)30,527 (35.4) ≥ 27,326 (8.5)26 (5.9)7,352 (8.4)No. of reproductive years36.72 ± 5.3735.28 ± 6.2636.72 ± 5.37< .001Ever menopause163,451 (60.1)907 (94.4)164,358 (60.7)< .001Age at menopause, y49.67 ± 5.1348.30 ± 5.8449.68 ± 5.14< .001History of hysterectomy19,766 (7.3)131 (11.8)19,897 (8.1)< .001History of oophorectomy21,628 (8.0)148 (11.2)21,776 (8.2).002Exogenous hormone use Ever used OCPs219,377 (80.6)914 (78.8)220,291 (81.2).061 Ever used HRT103,158 (37.9)652 (55.1)103,810 (38.1)< .001 HRT duration, y6.92 ± 5.758.17 ± 6.696.92 ± 5.76< .001Data are presented as No. (%) or mean ± SD, unless otherwise indicated. AS = advanced subsidiary; GCSE = general certificate of secondary education; HRT = hormone replacement therapy; OCP = oral contraceptive pill. Open table in a new tab Data are presented as No. (%) or mean ± SD, unless otherwise indicated. AS = advanced subsidiary; GCSE = general certificate of secondary education; HRT = hormone replacement therapy; OCP = oral contraceptive pill. Multiple variables of reproductive lifespan showed a linear association (Fig 1) with risk of lung cancer (Table 2). Early menarche (age at menarche, ≤ 11 years) was linked to an increased incidence of lung cancer (HR, 1.22; 95% CI, 1.03-1.46; P = .025) compared with female participants who had experienced menarche at age 13. In comparison with participants who had reached menopause at a range of 50 to 52 years of age, early menopause was associated with a 49% and 42% increased risks of lung cancer in female participants who underwent menopause before 46 years of age (HR, 1.49; 95% CI, 1.19-1.86; P < .001) and between 47 and 49 years of age (HR, 1.42; 95% CI, 1.11-1.81; P = .005), respectively. In addition, we also analyzed the relationship between reproductive life span and risk of lung cancer and found that female participants with fewer reproductive years (≤ 32 years: HR, 1.42; 95% CI, 1.18-1.71; P < .001; 33-35 years: HR, 1.24; 95% CI, 1.00-1.53; P = .046) showed higher risks of lung cancer than those with a relatively normal reproductive life span ranging from 36 to 39 years. The separate restricted cubic splines of aforementioned factors demonstrated linear reverse associations between the risk of lung cancer and individual reproductive factors, such as age at menarche (Fig 1A), age at menopause (Fig 1B), and reproductive years (Fig 1C). Furthermore, a history of oophorectomy was analogous to an elevated risk of incident lung cancer (HR, 1.25; 95% CI, 1.04-1.49; P = .015), whereas a history of hysterectomy was not associated with risk of lung cancer compared with those who had never undergone either procedure.Table 2Associations Between Comprehensive Reproductive Factors and Incident Lung Cancer in Female ParticipantsReproductive FactorsModel 1aCrude.Model 2bAdjusted for age and BMI.Model 3cAdditionally adjusted for race, Townsend deprivation index, education level, smoking status, and alcohol use status.HR (95% CI)P ValueHR (95% CI)P ValueHR (95% CI)P ValueFDR-Adjusted P ValueNo. of children 00.95 (0.81-1.13).5731.21 (1.02-1.43).0271.10 (0.93-1.31).278.539 11.03 (0.85-1.23).7861.17 (0.97-1.41).0950.97 (0.80-1.17).720.813 2Reference…Reference…Reference…… 31.26 (1.08-1.47).0031.18 (1.01-1.82).0101.07 (0.92-1.25).392.715 ≥ 41.68 (1.37-2.06)< .0011.48 (1.09-1.11)< .0011.08 (0.88-1.34).458.729 P value (for trend)P < .001….365….662……Age at menarche, y ≤ 111.33 (1.12-1.58).0011.28 (1.07-1.52).0061.22 (1.03-1.46).025.097 121.07 (0.90-1.30).4311.05 (0.87-1.26).6101.07 (0.89-1.29).470.729 13Reference…Reference…Reference…… 141.08 (0.90-1.30).4151.02 (0.87-1.26).8040.96 (0.80-1.16).689.813 ≥ 151.29 (1.08-1.555).0061.22 (1.02-1.11).0341.03 (0.86-1.24).734.813 P value (for trend).586….429….088……Age at menopause, y ≤ 461.94 (1.56-2.41)< .0011.97 (1.58-2.45)< .0011.49 (1.19-1.86)< .001< .001 47-491.53 (1.20-1.95)< .0011.66 (1.30-1.11)< .0011.42 (1.11-1.81).005.031 50-52Reference…Reference…Reference…… 53-541.17 (0.89-1.53).2671.16 (0.88 -1.52).2921.19 (0.91-1.57).202.463 ≥ 550.98 (0.680-1.167).8480.85 (0.64-1.12).2460.84 (0.63-1.10).204.463 P value (for trend)< .001…< .001…< .001……Reproductive years ≤ 321.81 (1.51-2.17)< .0011.82 (1.52 -2.19)< .0011.42 (1.18-1.71)< .001< .001 33-351.35 (1.10-1.67).0041.40 (1.14-1.73).0011.24 (1.00-1.53).046.158 36-39Reference…Reference…Reference…… 40-420.97 (0.78-1.20).7680.926 (0.75-1.15).4781.01 (0.81-1.25).950.950 ≥ 431.03 (0.79-1.33).8490.904 (0.70-1.17).4460.96 (0.73-1.24).734.813 P value (for trend)< .001…< .001…< .001……Age at first birth, y ≤ 202.96 (2.43-3.60)< .0012.81 (2.30-3.42)< .0011.63 (1.33-2.01)< .001< .001 21-251.87 (1.56-2.24)< .0011.56 (1.31-1.89)< .0011.31 (1.09-1.58).004.031 26-30Reference…Reference…Reference…… 31-350.71 (0.50-1.00).0470.87 (0.62-1.24).4460.89 (0.63-1.26).494.729 ≥ 360.72 (0.37-1.41).3341.00 (0.51-1.96).9981.03 (0.53-2.01).933.950 P value (for trend)< .001…< .001…< .001……No. of miscarriages 0Reference…Reference…Reference…… 10.87 (0.70-1.10).2340.76 (0.61 -0.96).0020.98 (0.77-1.23).830.887 ≥ 21.16 (0.89-1.52).2741.07 (0.82-1.40).6351.21 (0.92-1.58).182.463 P value (for trend).485.097.958….268……No. of stillbirths 0Reference…Reference…Reference…… 11.57 (1.13-2.17).0071.22 (0.88-1.69).1691.14 (0.82-1.59).438.729 ≥ 21.69 (0.80-3.58).1671.44 (0.68-3.05).3391.19 (0.56-2.52).653.813 P value (for trend).003….134….682……No. of abortions 0Reference…Reference…Reference…… 11.09 (0.89-1.35).4101.35 (1.09-1.67).0061.06 (0.85-1.32).615.813 ≥ 20.80 (0.53-1.20).2731.15 (0.76-1.73).5760.79 (0.52-1.20).267.539 P value (for trend).761….034….393……Hysterectomy vs not1.70 (1.42-2.04)< .0011.10 (0.91 -1.32).3261.06 (0.88-1.28).559.788Oophorectomy vs not1.68 (1.42-2.00)< .0011.30 (1.09-1.11).0041.25 (1.04-1.49).015.078Ever used HRT2.02 (1.80-2.27)< .0011.25 (1.11-1.41)< .0011.08 (0.96-1.22).209.463Ever taken OCPs0.82 (0.72-0.95).0061.11 (1.10-1.46).0011.15 (1.00-1.33).054.167HRT duration, y1.03 (1.02-1.05)< .0011.02 (1.01-1.04)< .0011.02 (1.00-1.03).021.093HR = hazard ratio; FDR = false discovery rate; HRT = hormone replacement therapy; OCP = oral contraceptive pill.a Crude.b Adjusted for age and BMI.c Additionally adjusted for race, Townsend deprivation index, education level, smoking status, and alcohol use status. Open table in a new tab HR = hazard ratio; FDR = false discovery rate; HRT = hormone replacement therapy; OCP = oral contraceptive pill. In terms of parity-related factors, age at time of first live birth was identified as being associated with an increased risk of lung cancer in a linearly reverse relationship (Fig 1D), with female participants who were younger at the time of first live birth being at a greater risk of lung cancer developing (≤ 20 years: HR, 1.63; 95% CI, 1.33-2.01; P < .001; 21-25 years: HR, 1.31; 95% CI, 1.09-1.58; P = .004) (Table 2) than female participants whose first live birth occurred between 26 and 30 years of age. Neither stillbirth, miscarriage, or abortion were identified as risk factors for lung cancer incidence, nor was the number of children. In the fully adjusted model, a marginal relationship was found between OCP administration and the risk of lung cancer (HR, 1.15; 95% CI, 1.00-1.33; P = .054). Moreover, we identified a significant association between HRT administration and risk of lung cancer in the crude model (HR, 2.02; 95% CI, 1.80-2.27; P < .001); however, this association disappeared in the adjusted model after modifying for age, BMI, race, smoking status, alcohol status, educational level, and Townsend deprivation index. Interestingly, in the adjusted model, HRT was correlated with a higher risk of lung cancer in subpopulations with a baseline age of < 60 years in the stratified analysis (HR, 1.47; 95% CI, 1.17-1.84; P < .001) (e-Table 3). In addition, stratified analysis was performed according to age at baseline, smoking status, obesity status, genetic susceptibility, and histologic subtypes of lung cancer. We found that the associations of the main findings with multiple reproductive factors had a greater likelihood of significant HRs in subgroups with baseline age of < 60 years (e-Table 3), with a previous or current smoking history (e-Table 4, Fig 2A), and BMI of ≥ 25 kg/m2 (e-Table 5). Previous studies seldom have investigated the modified effect of genetic risks on the relationship between risk of lung cancer and its histologic subtypes, and we identified a stronger association between the initial findings in a subpopulation with a hi