Etiology and risk factors for lung cancer in female Asian never smokers: a systematic review

Abstract

Background:

Lung cancer remains the leading cause of cancer-related mortality worldwide, with a growing proportion of cases occurring in female Asian never-smokers (FANS). Although tobacco exposure remains the predominant risk factor, emerging evidence highlights the substantial role of non-tobacco determinants. However, the relative contributions of these factors remain poorly defined.

Objectives:

This systematic review aimed to synthesize existing evidence on biological, clinical, and environmental risk factors for lung cancer among FANS to identify consistent and potentially modifiable exposures to inform risk-stratified prevention strategies. Genetic and molecular determinants were excluded.

Design:

Systematic review.

Data sources and methods:

Following PRISMA-P guidelines, we systematically searched PubMed, Embase, CINAHL, and Web of Science for articles published between January 2000 and July 2025 to identify peer-reviewed observational studies reporting non-tobacco risk factors for lung cancer in FANS. Due to the heterogeneity in study designs, a narrative synthesis was performed on the final 42 studies.

Results:

Indoor air pollution, such as high-temperature cooking oil fumes, prolonged solid-fuel use, and inadequate household or workplace ventilation, consistently conferred elevated risks (2–4× higher). Risk estimates increased up to 12-fold for high cumulative exposures. This risk was also seen in occupational exposures such as commercial cooking and International Agency for Research on Cancer-classified industrial jobs. Family history of lung cancer, especially among first-degree relatives, nearly doubled risk in several large cohorts. Prior non-malignant lung diseases such as tuberculosis, asthma, and lymphangioleiomyomatosis were also significantly associated with increased risk.

Conclusion:

Lung cancer in FANS may be driven by largely modifiable non-tobacco exposures. These findings underscore the need to expand risk assessments beyond smoking history to incorporate detailed household, occupational, and clinical exposures, and to develop targeted, culturally informed prevention and screening strategies for this high-risk population.

Trial Registration:

The protocol was registered in PROSPERO (CRD420251111993).

Keywords

Asian females lung cancer never-smokers risk factors systematic review

Introduction

Lung cancer is the leading cause of cancer-related mortality worldwide.¹ Traditionally, cigarette smoking has been recognized as the primary risk factor. The United States Preventive Task Force (USPSTF) currently recommends lung cancer screening only for individuals with a ⩾20 pack-year smoking history who either currently smoke or have quit within the past 15 years.² However, over recent decades, there has been a growing incidence of lung cancer among never-smokers, steadily increasing from 8% in the 1990s to 13% in 2018.^3,4 These trends have prompted greater investigation into understanding the etiology of lung cancer in never-smokers.

Asian women represent a demographic subgroup with a disproportionately high risk, with recent studies showing lung cancer incidence rates of 57% among female Asian never-smokers (FANS) and up to 86% in certain subgroups, such as Asian Indians.⁵ The alarming magnitude of disease burden has led to calls for expanding lung cancer screening eligibility to include FANS.⁶ The underlying causes remain poorly understood, though current evidence suggests a multifactorial etiology involving both genetic and environmental influences.⁷

Prior observational and mechanistic research has suggested that lung cancer in never-smokers arises from a constellation of non-tobacco determinants that differ substantially from smoking-related disease.⁸ Existing literature discusses risk factors such as chronic exposure to household and occupational air pollutants (such as cooking oil fumes, solid-fuel combustion products, and poor ventilation), occupational inhalants in commercial and industrial settings, reproductive and hormonal factors influencing lifetime estrogen exposure, familial susceptibility and shared environmental exposures, prior inflammatory or infectious lung disease, and dietary patterns affecting oxidative stress and immune regulation. Many of these exposures are highly prevalent in Asian populations and often underrecognized in routine clinical assessment. However, the relative contribution and consistency of these risk factors across diverse populations of FANS remain incompletely defined, creating a critical gap in prevention, screening, and risk stratification efforts.

To address this gap, this systematic review synthesizes current evidence on clinical and non-tobacco environmental risk factors for lung cancer among FANS that can be identified through patient history and routine clinical evaluation. Genetic and molecular determinants were excluded to focus on factors that are clinically recognizable without laboratory or genomic testing, as these are the most applicable to real-world risk assessment and early detection.

Methods

We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols (PRISMA-P) Statement and registered this study protocol in the PROSPERO database. Specifically, we aimed to examine and synthesize evidence on the associations between biological, clinical, and environmental non-tobacco risk factors and lung cancer incidence among FANS, including occupational exposures, hormonal and reproductive factors, dietary and vitamin intake, infectious agents, indoor air pollution (IAP), family history, and acquired or hereditary lung diseases. Risk factors were operationalized according to each study’s definitions. Outcomes measures included lung cancer incidence, histologic subtypes, and where reported, relative risk estimates such as odds ratios, hazard ratios, and dose–response relationships.

Study selection

The databases PubMed/MEDLINE, Embase, CINAHL, and Web of Science were used, and searches included articles published prior to July 24, 2025. Key concepts for search strategy included: “lung cancer,” “risk factor,” “never-smoker,” “female.” The complete search strategy is available in the Appendix S1.

During the title/abstract screening and full-text screening phases, two reviewers independently reviewed each study. The reason for exclusion of each article was identified during the full-text screening. If there was a disagreement during the title/abstract or full-text screening, it was resolved by a group consensus. The Covidence software was used to generate a PRISMA diagram to track the studies at each stage of the review.

We included peer-reviewed, published manuscripts written in English and published after January 2000. To be eligible, studies needed to specifically examine non-tobacco, non-genetic risk factors for lung cancer among FANS. Never-smoking was defined as any individual that had smoked 100 or fewer cigarettes in their lifetime. Studies were individually reviewed if the term “non-smoker” was used to ensure alignment with this definition.

We excluded studies that were not peer-reviewed original articles (e.g., conference abstracts, editorials), as well as systematic reviews, meta-analyses, case reports, and small case series. Additional exclusions included studies not focused on Asian female populations, those limited to passive smokers, smokers, or mixed populations without stratification by both smoking status and sex, and studies reporting solely lung cancer mortality without analysis of risk factors. With respect to secondhand smoke, studies in which passive smoking was the primary exposure of interest were excluded to maintain focus on non-tobacco determinants. However, studies that adjusted for secondhand smoke as a confounder while evaluating other exposures of interest were retained, reflecting real-world epidemiologic practice and strengthening causal inference for the non-tobacco risk factors examined.

To preserve clinical relevance and applicability to individual patient risk assessment, this review deliberately restricted eligible exposures to individual-level, potentially modifiable factors that can be reasonably ascertained through routine clinical history or occupational and household exposure assessment. Accordingly, we excluded studies examining population-level or community-ambient exposures (e.g., ambient pollution, radiation, particulate matter) that are not readily modifiable at the patient level and are typically measured using geospatial or environmental monitoring rather than individual exposure histories. In contrast, household and workplace ventilation practices, cooking fuel type, and cumulative cooking exposure were included because these represent modifiable, patient-specific exposures. Genetic susceptibility factors (e.g., biomarkers, gene polymorphisms) were also excluded to preserve focus on individual-level, modifiable, clinically actionable exposures ascertainable through patient history.

These exposure boundaries were prespecified and applied consistently throughout screening and full-text review.

The eligibility criteria are specified through the Population/Patient, Intervention, Comparison, Outcome, Time, and Setting (PICOTS) framework shown in Table 1.

Table 1.

PICOTS framework.

Domain	Eligibility criteria
Population	Inclusion • Never smoking female Asian patients with lung cancer. Exclusion • Non-Asian patients • Patients without lung cancer • Solely male patients or cohorts without stratification by sex • Current or former smokers
Intervention	Inclusion • Non-tobacco risk factors Exclusion • Tobacco smoking • Population-level exposures assessed via community ambient metrics • Genetic risk factors
Comparator	• None
Outcomes	• Lung cancer (any type)
Timing	• Any follow-up period will be included
Setting	• Any care setting

Data collection

During the screening process, we developed a structured data abstraction form through an iterative approach to systematically capture key study characteristics and exposure–outcome data (Appendix S2). The form recorded the following variables: study design and population characteristics, category of non-tobacco risk factors, control groups, sample size, and relevant effect estimates. The primary outcomes of interest included lung-cancer incidence, histologic subtype, odds ratios, hazard ratios, and 95% confidence intervals. This standardized framework ensured consistent and comprehensive extraction of data across all included studies to facilitate the synthesis of non-tobacco risk factors for lung cancer in FANS.

Data abstraction for the included studies was conducted following the same approach as the screening process: two of the three reviewers (S.G., K.L., and E.L.) independently extracted data from each article, and any discrepancies were resolved through discussion until consensus was reached. The extraction phase was conducted with dual review and consensus adjudication as mentioned earlier.

The Risk Of Bias In Non-randomized Studies of Exposures (ROBINS-E⁹) tool was used to assess methodological quality during the data extraction phase. Two reviewers independently evaluated the risk of bias for each included study across the following domains: confounding, selection of participants, classification of exposures, deviations from intended exposures, missing data, measurement of outcomes, and selection of the reported result. Discrepancies were resolved through discussion until consensus was achieved. Overall risk of bias judgments was assigned using the ROBINS-E algorithm, which integrates domain-level assessments and applies prespecified decision rules such that the overall risk corresponds to the highest level of bias identified in any critical domain. The final risk of bias judgment for all included studies is presented in Appendix S3.

Certainty of evidence and reporting bias assessment

The overall certainty of evidence for each major risk-factor category was evaluated using a qualitative adaptation of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework.¹⁰ For each domain of exposure (IAP, occupational exposures, family history, hormonal and reproductive factors, non-malignant lung disease, infectious agents, dietary/vitamin intake, and other), three authors (S.G., K.L., and E.L.) assessed evidence across five dimensions: risk of bias, inconsistency of findings, indirectness of evidence, imprecision of effect estimates, and potential publication bias. Categories were downgraded 1 or 2 levels if there was high risk of bias, inconsistency of results, indirectness of evidence, imprecision, or publication bias. Categories were upgraded 1 or 2 levels if there was a large magnitude of effect, plausible confounding reduced demonstrated effect, or dose–response gradient. An overall certainty rating (high, moderate, low, or very low) was then assigned for each risk-factor category.

Data synthesis

We performed a narrative synthesis because of substantial heterogeneity across the included studies, particularly in the risk factors evaluated, exposure definitions, study designs, and outcome measures reported. Studies differed widely in methodology, covariate adjustment, and the reporting of effect estimates, which precluded meaningful statistical pooling. We synthesized findings across studies to highlight statistically significant associations and, where available, clinically meaningful differences in lung-cancer incidence and risk stratification by exposure intensity or duration. This approach enabled us to identify consistent patterns of risk across heterogeneous studies and to inform the potential public health and clinical relevance of modifiable non-tobacco exposures in FANS.

Studies evaluated lung cancer in various categories mainly including adenocarcinoma and non-adenocarcinoma, with some studies also evaluating squamous cell carcinoma, small cell carcinoma, large cell carcinoma, and undifferentiated carcinoma. Some studies also evaluated Lung Imaging Reporting and Data System (Lung- RADS) distribution with likelihood of malignancy.

In order to evaluate the risk factors, eight categories were developed: IAP, occupational exposures, hormonal and reproductive factors, family history, non-malignant lung disease, infectious agents, dietary/vitamin intake, and multiple factors, as shown in Table 2 and visually represented in Figure 1.

Table 2.

Summary of risk factor categories.

Category	Definition	Key factors
Indoor air pollution	Airborne contaminants produced inside the household environment.	High-temperature cooking (deep-frying, stir-frying, pan-frying)
		Fumes from heating vegetable or animal oils
		Smoke and particulate matter from burning solid fuels (coal, wood, biomass) for cooking/heating
		Household ventilation (presence/absence of exhaust fans, chimneys, range hoods)
Occupational exposures	Inhaled chemicals, fumes, or particulates encountered in work settings that raise lung-cancer risk.	Chronic exposure to cooking oil fumes in commercial kitchens
		Industrial hazards recognized by IARC (solvents, industrial dusts, combustion by-products)
		Occupations: commercial cooks, painters, rubber and manufacturing workers
Hormonal and reproductive factors	Biological and external influences on lifetime estrogen and progesterone exposure.	Parity (number of live births)
		Breastfeeding duration
		Age at menarche and menopause, reproductive span, menstrual patterns
		Exogenous hormones (oral contraceptives, hormone-replacement therapy)
		Surgical interventions affecting hormones (oophorectomy, tubal ligation)
Family history	Presence of lung cancer or other malignancies in biological relatives (especially first-degree relatives).	Parents, siblings, or children with lung cancer
		Inherited genetic predispositions
		Shared household/environmental exposures
		Differences by number of affected relatives or by maternal vs paternal lineage
Non-malignant lung disease	Pre-existing or prior respiratory conditions that may predispose to lung cancer.	Tuberculosis
		Asthma
		Chronic bronchitis
		Pneumonia
		Pulmonary fibrosis
		Lymphangioleiomyomatosis
Infectious agents	Microbial or viral infections implicated in lung-cancer development.	Chlamydia pneumoniae seropositivity
		Human papillomavirus
Dietary and vitamin intake	Food consumption patterns, preparation methods, and micronutrient/supplement use that influence risk.	Fried or high-fat foods
		Fruit and vegetable consumption
		Soy products and isoflavones
		Antioxidants and micronutrients (tocopherols, riboflavin, methionine)
		Pharmacologic vitamin/mineral supplements
Multiple	Studies evaluating two or more risk factors.	Any analysis involving multiple overlapping exposures or categories

IARC, International Agency for Research on Cancer.

Figure 1.

Lung cancer risk factors in FANS.

Results

Literature selection

A total of 4074 records were identified through database searching, of which 3212 were excluded based on title/abstract. The full-text of the remaining articles was assessed for eligibility, and 820 were subsequently excluded. In total, 42 studies met inclusion criteria (Figure 2).

Figure 2.

PRISMA flow diagram showing number of records screened, excluded, and included.

The reasons for excluding 820 studies during full-text review: not focused on an Asian female population (n = 401); published before January 2000 (n = 140); focused on gene polymorphisms or genetic biomarkers (n = 87); not a peer-reviewed full-length published original article (e.g., conference abstract, editorial; n = 54); did not report on non-tobacco-related risk factors (n = 35); focused solely on passive smoking (n = 31); not lung cancer risk factor (n = 13); not lung cancer (n = 13); systematic reviews or meta-analyses (n = 12); solely focused on risk-model development (n = 11); focused exclusively on smokers or mixed populations without stratification by smoking status and sex (n = 10); not written in English (n = 9); solely reported on environmental exposure (n = 3); or case reports or small case series (n = 1).

The characteristics of the included studies are shown in Table 3. The exposure type and cohort size are shown in Table 4.

Table 3.

Study characteristics (n = 42).

Risk factor category	GRADE	Author and year	Study period (follow-up)	Country	study design
Occupational exposure (n = 2)	High	Jang et al.¹¹	2007–2011	Korea	Nationwide, retrospective, population-based
		Pronk et al.¹²	1996–2000 (2005)	China	Single city, prospective, population-based
Hormonal/Reproductive (n = 7)	Moderate	Elbasheer et al.¹³	2004–2008 (2018)	China	Multiregion, prospective, population-based
		Gallagher et al.¹⁴	1989–2000	China	Single city, prospective, population-based
		Jeon et al.¹⁵	2009–2014 (2016)	Korea	Nationwide, retrospective, population-based
		Tan et al.¹⁶	1994–1997 (2011)	Singapore	Nationwide, prospective, population-based
		Weiss et al.¹⁷	1996–2000 (2005)	China	Single city, prospective, population-based
		Wilunda et al.¹⁸	1990–2013	Japan	Nationwide, prospective, population-based
		Yin et al.¹⁹	1984–2007	Japan, Korea, China, Singapore	Pooled Analysis
Indoor air pollution (n = 8)	High	Chen et al.²⁰	2002–2010	Taiwan	Multicenter hospital-based case–control
		Fang et al.²¹	2003–2010	China	Multiregion, population-based, prospective case–control
		Kim et al.²²	1996–2000	China	Multiregion, prospective, population-based
		Kim et al.²³	2022 (2023)	China	Single-center, prospective, hospital-based
		Mu et al.²⁴	2005–2007	China	Single-center, hospital-based, case–control
		Wang et al.²⁵	2002–2004	China	Single center, retrospective, case–control
		Wong et al.²⁶	2006–2013	China	Multicenter, retrospective, population-based case–control
		Yu et al.²⁷	2002–2004	Hong Kong	Population-based, prospective, case–control
Family history (n = 6)	High	Lee et al.²⁸	2011–2015	South Korea	Single-center, retrospective, hospital-based
		Tse et al.²⁹	2002–2004	Hong Kong	Single-center, retrospective, community-based case–control
		Wang et al.³⁰	2013–2019 (2020)	China	Multicenter, prospective, population-based
		Wu et al.³¹	1992–2002	Taiwan	Multicenter, retrospective, population-based case–control
		Yin et al.³²	2005–2008	Singapore	Multicenter, retrospective, hospital-based case–control
		Zhang et al.³³	1996–2000 (2004)	China	Single-center, prospective, population-based
Infectious agents (n = 4)	High	Cheng et al.³⁴	NR	Taiwan	Multicenter, retrospective, hospital-based case–control
		Cheng et al.³⁵	NR	Taiwan	Multicenter, retrospective, hospital-based case–control
		Chiou et al.³⁶	NR	Taiwan	Multicenter, retrospective, hospital-based case–control
		Liu et al.³⁷	2000–2009	China	Single center, hospital-based case–control
Non-malignant lung disease (n = 3)	High	Liang et al.³⁸	2004–2007	China	Multicenter, retrospective, hospital-based case–control
		Torasawa et al.³⁹	2001–2022	Japan	Single-center, retrospective, hospital-based
		Wang et al.⁴⁰	2002–2004	Hong Kong	Single-center, retrospective, population-based, case–control
Dietary/Vitamin intake (n = 8)	Moderate	Butler et al.⁴¹	1993–1998 (2007)	Singapore	Prospective, population-based
		Li et al.⁴²	1996–2000	China	Prospective, population-based
		Lim et al.⁴³	2005–2008	Singapore	Multicenter, hospital-based case–control
		Lim et al.⁴⁴	2005–2008	Singapore	Multicenter, hospital-based case–control
		Seow et al.⁴⁵	1996–1998	Singapore	Population-based prospective cohort
		Shimazu et al.⁴⁶	1995–1999 (2005)	Japan	Multicenter, prospective population-based cohort
		Takata et al.⁴⁷	1997–2000 (2009)	China	Population-based prospective cohort
		Wu et al.⁴⁸	1997–2000 (2010)	China	Single city, population-based, prospective
Multiple (n = 4)	High	Liang et al.⁴⁹	2015–2017	China	Multicenter, retrospective, population-based matched case–control
		Seow et al.⁵⁰	1993–1998 (2005)	Singapore	Single-center, prospective, population-based
		Wu et al.⁵¹	2013–2018 (2021)	China	Multicenter, prospective, population-based
		Zhou et al.⁵²	1991–1995	China	Multicenter, retrospective, population-based, case–control

GRADE, Grading of Recommendations Assessment, Development and Evaluation.

Table 4.

Exposure type and cohort size.

Author and year	Exposure definition	FANS cohort total sample size	Measurement	Comparator	Endpoint
Butler et al.⁴¹	Intake was assessed for five fried meat items (stir-fried pork slices or cubes, deep-fried and stir-fried chicken, and deep-fried and stir-fried fish). The assessment captured the usual diet over the past year.	~31,000	Self-reported	Three tertiles of intake	Incidence of any lung cancers
Chen et al.²⁰	Cooking time years: This index measured cumulative exposure to cooking oil fumes (involved average daily times of cooking, years of cooking in age range)	2604	Self-reported	Age, sex, smoking status, education level matched healthy controls without cancer	Diagnosis of any lung cancer
	Fume extractor use ratio: This was the proportion of fume extractor use over the total lifetime of cooking time, calculated as (total number of years of fume extractor use)/(total number of cooking years)
Cheng et al.³⁴	Presence of high risk HPV 16/18 DNA in lung tumor and normal tissue	56	Laboratory testing	Non-cancer control subjects with lung disease	Detection of HPV 16/18 DNA
Cheng et al.³⁵	Presence of HPV 6/11 DNA in tissue	45	Laboratory testing	Non-cancer patients with lung disease	Detection of HPV 6/11 DNA
Chiou et al.³⁶	Presence of HPV 16/18 DNA in blood	138	Laboratory testing	Non-cancer control subjects with no history of cancer	Detection of HPV DNA and diagnosis of lung cancer
Elbasheer et al.¹³	Age of menarche, oral contraceptive use/duration, average breastfeeding duration, number of pregnancies, parity, age at first birth, menopausal status/age, total reproductive period, history of oophorectomy/hysterectomy	282,558	Self-reported	Within cohort	Incidence of lung cancer
Fang et al.²¹	Poor ventilation, coal used for cooking, hot cooking oil, solid fuel use for heating, environmental tobacco smoking integrated for an overall score	2758	Self-reported	Controls with no cancer history	Incidence of lung cancer
Gallagher et al.¹⁴	Parity, menstrual factors, contraception, procedures as fixed variables	~259,378	Self-reported	Within cohort	Incidence of lung cancer
Jang et al.¹¹	Occupation as school cook	23,449	Administrative records	Female clerk workers	Incidence of lung cancer
Jeon et al.¹⁵	Age at menarche, age at menopause, reproductive period, parity number, total lifetime breastfeeding history, duration of hormone replacement therapy, duration of oral contraceptive use	4,335,259	Self-reported	Within cohort	Incidence of lung cancer
Kim et al.²²	Kitchen ventilation condition and cooking fuel by exposure duration	71,320	Self-reported	Within cohort	Incidence of lung cancer
Kim et al.²³	Culinary workers years of service, cooking duties, cooking duration at home, recipe type	203	Self-reported and monitored	Nutrition teachers	Lung-RADS positivity
Lee et al.²⁸	Presence of absence of family history of lung cancer	10,151	Self-reported	Patients without family history of lung cancer	Prevalence of ground glass nodules and incidence of lung cancer
Li et al.⁴²	Phytoestrogen exposure based on dietary intake of isoflavones	956	Self-report and monitored	Individually matched controls	Incidence of lung cancer
Liang et al.³⁸	history of physician-diagnosed nonmalignant respiratory conditions (NMRCs) (pulmonary tuberculosis, chronic bronchitis, emphysema, asthma, and bronchiectasis), age, and year in which each condition was first diagnosed	505	Self-reported	Female matching population controls	Incidence of lung cancer
Liang et al.⁴⁹	Participants’ basic information (including age, sex, educational level, and occupation), exposure to environmental tobacco smoke, condition of living quarters, cooking habits, physical exercise habits, stress level, eating habits, alcohol consumption, occupational exposure, history of chronic diseases, and family history of cancer	2070	Self-reported	Non-smoking matching population controls by sex and age	Incidence of lung cancer
Lim et al. ⁴³	Diet (fruit, vegetables, soy products, meats)	970	Self-reported	matching population controls by sex and age	Incidence of lung cancer
Lim et al.⁴⁴	Regular use (defined as use at least twice a week for a period of a month or more) of aspirin, non-aspirin Nonsteroidal Anti-Inflammatory Drugs (NSAID) and COX-2 selective NSAID, paracetamol, steroid cream, and steroid pills	818	Self-reported	Matching population controls by sex and age	Incidence of lung cancer
Liu et al.³⁷	Chlamydia pneumoniae IgG antibody levels	282	Laboratory testing	Matching population controls by sex and non-smoking status	Incidence of lung cancer
Mu et al.²⁴	Home particulate matter measurement	382	Laboratory testing	Matching population controls by sex and age	Incidence of lung cancer
Pronk et al.¹²	Lifetime occupational history	71,067	Self-reported	Non-smoking women who worked outside the home	Incidence of lung cancer
Seow et al.⁴⁵	Reproductive factors and dietary intake of fruit, vegetables and soy foods	839	Self-reported	matching population controls by sex and age	Incidence of lung cancer
Seow et al.⁵⁰	Parity, age at menarche/menopause, cycle length, and use of exogenous hormones, and dietary soy and soy isoflavonoid intake	31,057	Self-reported	Singapore Chinese women	Incidence of lung cancer
Shimazu et al.⁴⁶	soy foods or isoflavone intake	38,211	Self-reported	Japanese men and women	Incidence of lung cancer
Takata et al.⁴⁷	Daily intakes of total calories and nutrients, excluding vitamin B6, vitamin B12, and methionine (estimated using the Chinese Food Composition Tables)	74,941	Self-reported	Non-smoking women in Shanghai	Incidence of lung cancer
Tan et al. ¹⁶	Reproductive history (ages at menarche and menopause, pregnancy and delivery histories, ages at first and last delivery, breastfeeding history, and the use of exogenous hormones).	26,473	Self-reported	Singapore Chinese women	Incidence of lung cancer
Torasawa et al. ³⁹	Diagnosis of LAM; patients with LAM followed for lung cancer incidence	418	Clinical diagnostic records/registry confirmation	General Japanese population	Incidence of lung cancer during follow-up
Tse et al. ²⁹	Family cancer history defined as ⩾1 first-degree relative diagnosed with any cancer and separate indicator for family history of lung carcinoma	505	Self-reported	No family cancer history vs family history categories	Lung cancer case status (presence vs absence of lung cancer)
Wang et al.²⁵	Defined as frequency or regular exposure to cooking with fuels	505	Self-reported	Lower vs higher categories (e.g., never/low vs regular/longer exposure)	Incidence of lung cancer
Wang et al.⁴⁰	Self-reported history of pulmonary diseases and family history of lung/any cancer before lung cancer diagnosis	504	Self-reported	Absence vs presence of reported conditions	Lung cancer case status
Wang et al.³⁰	Family history of cancer, especially lung cancer in first-degree relatives	547,218	Self-reported	No family history vs ⩾1 affected relative	Incidence of lung cancer
Weiss et al.¹⁷	Duration between menarche and menopause; age at menopause categorized into groups	71,314	Self-reported	Categories of reproductive span/age at menopause	Incidence of lung cancer
Wilunda et al.¹⁸	Age at menopause minus age at menarche	42,615	Self-reported	Within cohort	Incidence of lung cancer
Wong et al.²⁶	Lifetime use of bituminous coal for household heating or cooking categorized by duration and type of coal used	1524	Self-reported	Non-bituminous coal users vs bituminous coal users	Incidence of lung cancer
Wu et al.³¹	Family cancer aggregation among first-degree relatives of female lung cancer probands	216	Self-reported	Relatives without family cancer aggregation vs those with aggregation	Incidence of lung cancer
Wu et al.⁴⁸	Intake of vitamin E from diet and supplements	65,066	Self-reported	Intake quartiles or categories of vitamin E intake	Incidence of lung cancer
Wu et al.⁵¹	Baseline characteristics including age, sex, lifestyle, family history, comorbidity	546,382	Self-reported	Reference categories for each exposure variable	Incidence of lung cancer
Yin et al.³²	Self-reported family history of lung cancer	944	Self-reported	No family history vs positive history	Lung cancer risk incidence
Yin et al.¹⁹	Harmonized family cancer history data across >308,000 women in Asia Cohort Consortium	308,949	Self-reported	No family history vs presence of history	Incidence of lung cancer and mortality
Yu et al.²⁷	“Dish-years” defined as cooking dishes using deep-frying/frying/stir-frying once per day for 1 year; cumulative measure	485	Self-reported	Non-exposed or lower dish-years vs higher dish-years categories	Incidence of lung cancer
Zhang et al.³³	Number of first-degree relatives with any cancer	71,571	Self-reported	Patients with no family members with any cancer	Incidence of lung cancer
Zhou et al.⁵²	Cooking fumes, family history of lung cancer, age at menarche, length of menstrual cycle, number of live births, and intake of vitamin C, E, beta-carotene	144	Self-reported	Matched healthy female controls	Incidence of lung adenocarcinoma

FANS, female Asian never-smokers; HPV, human papillomavirus; LAM, lymphangioleiomyomatosis.

Risk of bias

Across the 42 included studies, the overall risk of bias was judged as some concerns, primarily driven by residual confounding and limitations in exposure assessment, while most other ROBINS-E domains were consistently rated as low risk. Only one study was assessed as having a high risk of bias; this study was retained because it addressed a unique population or exposure not captured in other studies, and its inclusion contributed valuable data for the overall synthesis. For occupational exposures, confounding was the main source of bias, while exposure measurement and all other domains were largely at low risk; the overall direction of bias was considered toward the null. Hormonal factors similarly demonstrated some concerns in confounding and exposure measurement, with low risk across remaining domains and an overall bias direction toward the null. In the IAP category, most studies had some concerns related to confounding and exposure assessment, with one study rated at high risk in exposure measurement and another at high risk for selective reporting; consequently, the overall risk of bias was judged as some concerns, with mixed potential bias toward or away from the null. Studies evaluating family history showed some concerns related to confounding and participant selection, while other domains were consistently low risk; bias was considered variably toward or away from the null, with a tendency toward overestimation of harm due to shared genetic and environmental factors. For infectious exposures, confounding remained the primary concern, with all other domains rated as low risk, resulting in an overall judgment of some concerns and bias toward harm of exposure. Studies of pre-existing lung disease exhibited some concerns across confounding, exposure measurement, and participant selection, with low risk in remaining domains; the overall bias was judged as some concerns, with potential bias away from the null due to reverse causation. Dietary intake studies demonstrated some concerns in confounding and exposure measurement but low risk across other domains, with an overall bias direction toward the null. Finally, studies assessing multiple risk factors showed consistent patterns of some concerns related to confounding and exposure assessment, with low risk across remaining domains, leading to an overall judgment of some concerns.

Certainty of evidence

GRADE Certainty of Evidence ratings are presented in the GRADE column of Table 3, and the complete breakdown of these judgments is summarized in Appendix S4. IAP, occupational exposures, and multiple categories were downgraded by two levels due to imprecision and inconsistency, and upgraded by two levels due to a large magnitude of effect and evidence of a dose-dependent response. Hormonal and reproductive factors and dietary and vitamin intake were downgraded by two levels due to imprecision and inconsistency, and upgraded by one level due to evidence of a dose-dependent response. Non-malignant lung disease and infectious agents were downgraded by two levels due to imprecision and inconsistency, and upgraded by two levels due to a large magnitude of effect.

GRADE certainty of evidence definitions

High certainty: High confidence that the true effect lies close to that of the estimate of the effect.

Moderate certainty: Moderate confidence in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low certainty: Limited confidence in the effect estimate; the true effect may be substantially different from the estimate of the effect.

Very low certainty: Low confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

Occupational exposure

Pronk et al.¹² demonstrated that while International Agency for Research on Cancer (IARC) A- or B-list carcinogenic occupations were not significantly associated with overall lung cancer risk, elevated risks were observed among specific subgroups, specifically painters (Relative Risk (RR) = 2.0) and rubber workers (RR = 1.7; p ⩽ 0.1). In a nationwide cohort of Korean school cooks, Jang et al.¹¹ found that long-term exposure to cooking oil fumes was associated with a significantly elevated risk of lung cancer, particularly adenocarcinoma, highlighting the risks of high-temperature cooking environments (HR = 4.23; 95% CI: 2.36–7.58). Together, these findings emphasize occupational exposure for lung cancer risk in FANS.

Indoor air pollution

Across studies, IAP, including poor ventilation, solid-fuel use, and cumulative cooking exposure, was consistently linked to increased lung cancer risk in FANS. Zhou et al.⁵² found that lung adenocarcinoma risk rose with increasing exposure to cooking fumes (OR = 1.32, 0.73, 2.71; p_trend = 0.027) and that cooking practices generating frequent or occasional fumes conferred a markedly elevated risk (OR = 4.53; 95% CI: 2.09–9.94). Yu et al.²⁷ showed that >200 dish-years of cumulative cooking exposure was strongly associated with lung cancer (adjusted OR (aOR) = 34.0; 7.16–161.4), particularly for deep-frying, frying, and stir-frying. Chen et al.²⁰ added further evidence from Taiwan, demonstrating that cumulative cooking oil fume exposure >160 years was linked to a threefold increase in risk (aOR 3.17, 95% CI: 1.34–7.68, p_trend = 0.02). Similarly, Wang et al.²⁵ reported that ⩾101 dish-years increased lung cancer odds by approximately 2.7- to 4.2-fold, with adenocarcinoma risk rising further at >150 dish-years (OR ≈ 3.2).²⁵

Kim et al.²² found poor ventilation (HR = 1.49; 95% CI: 1.15–1.95) and poor ventilation plus ⩾20 years of coal use (HR = 2.03; 1.35–3.05) to be significant, though coal or soybean/vegetable oil use alone were not. Similarly, Kim et al.²³ reported markedly higher risks with inadequate workplace ventilation (fan-or-window only vs both; OR = 14.89; 3.30–67.23; p < 0.001) and electric appliances at home vs liquefied-gas fuel (OR = 4.59; 1.06–19.89; p = 0.041) for RADS-positive nodules. A deeper dive into coal use by Wong et al.²⁶ reported lifetime use of bituminous coal raised lung-cancer risk up to 12.6 times (95% CI: 6.51–24.24; p ≈ 4 × 10⁻¹⁴) compared with never users, with risk further amplified by greater cumulative coal use and strongest when exposure began at birth.

Mu et al.²⁴ observed increased odds with cooking in the living room (OR = 2.62; 95% CI: 1.35–5.08), no kitchen ventilator (OR = 1.78; 1.09–2.90), ⩾15–21 meals/week (OR = 3.30; 1.32–8.22), and solid-fuel cooking only (OR = 4.08; 2.17–7.67), as well as heating exposures (coal furnace OR = 2.00; 1.24–3.23; coke oven OR = 3.20; 1.40–7.32; heatable brick bed OR = 8.13; 2.20–30.08). Fang et al.²¹ reported that females were more vulnerable, with sex modifying the IAP–lung cancer association to yield higher odds in women. Collectively, these findings highlight that high-temperature cooking fumes, inadequate ventilation, prolonged solid-fuel use, and cumulative cooking intensity are major determinants of lung cancer risk in women, with never-smokers disproportionately affected.

Family history

Six studies evaluated the family history of cancer and lung cancer risk among FANS. Five studies reported that a family history of lung cancer, particularly in first-degree relatives, was significantly associated with increased lung cancer risk, with some evidence of higher risk when the affected relative was female or multiple relatives were affected. Wu et al.³¹ reported that family history of lung cancer was a strong predictor of lung cancer risk (adjusted OR 5.7, 95% CI: 1.9–16.9), particularly when the affected relative was female (OR 14.4, 95% CI: 2.7–75.5). Zhou et al.⁵² found that family history of lung cancer among first-degree relatives had an OR of 7.65 (95% CI: 0.90–169.84). Wang et al.³⁰ observed significantly elevated risk for first-degree relatives (HR 1.50, 95% CI: 1.29–1.75, p < 0.001), with the risk increasing with multiple affected relatives (HR: 2.43, 95% CI: 1.21–4.91, p = 0.013 for ⩾3 relatives). Yin et al.³² likewise reported increased risk among Southeast FANS with a family history of lung cancer (aOR 2.78, 95% CI: 1.57–4.90). Lee et al.²⁸ found that family history was an independent predictor of lung ground glass nodules (GGN) growth (HR 2.14, 95% CI: 1.03–4.44, p = 0.04). Only one study, Zhang et al.,³³ did not find a significant association (RR 0.89, 95% CI: 0.42–1.91).

Beyond lung cancer, four studies have also highlighted family history of other, non-lung malignancies was linked to elevated lung cancer risk. Tse et al.²⁹ found that any family cancer history in first-degree relatives was associated with higher odds of lung cancer (OR 2.20, 95% CI: 1.32–3.67).²⁹ Yin et al.³² reported similar risks (aOR 1.67, 95% CI: 1.19–2.35). Zhang et al.³³ found significant associations with family history of colorectal (RR, 2.38; 95% CI: 1.21–4.70), stomach (RR, 2.16; 95% CI: 1.01–4.65), and pancreatic cancer (RR, 4.19; 95% CI: 1.04–16.95). Wang et al.³⁰ also observed an increased risk among first-degree relatives with any family history of cancer, with the strongest association observed for paternal history (HR 2.15, 95% CI: 1.66–2.77), followed by siblings (HR 2.02, 95% CI: 1.30–3.12, p = 0.002) and maternal history (HR 1.33, 95% CI: 1.14–1.57, p < 0.001).³⁰

Non-malignant lung disease

Three studies evaluated non-malignant lung disease. Liang et al.³⁸ found that a history of any non-malignant lung disease was associated with a two-fold higher risk of lung cancer (aOR 2.0, 95% CI: 1.2–3.4), and pulmonary tuberculosis had a markedly elevated risk (aOR 4.7, 95% CI: 1.6–13.2). Asthma was also associated with increased risk, especially for small cell lung cancer (aOR 6.2, 95% CI: 1.5–25.8), but chronic bronchitis did not demonstrate a significant association (aOR 1.7, 95% CI: 0.8–3.7). The odds ratios were adjusted for age, marital status, years of schooling, ethnicity, 5 years ago BMI, passive smoking exposure, coal use, exposure to coal smoke, and cooking fumes. Similarly, Wang et al.⁴⁰ reported that asthma, tuberculosis, and pneumonia were linked to increased lung cancer risk, with risk compounding in individuals who had more than one prior pulmonary condition. The study also highlighted that the effects of non-malignant lung disease and family cancer history appeared additive, suggesting these act as independent risk factors.⁴⁰ Torasawa et al.³⁹ reported a significantly elevated incidence of lung cancer among non-smoking women with lymphangioleiomyomatosis (LAM), with a standardized incidence ratio of 13.6 (95% CI: 6.2–21.0).

Infectious agents

By contrast, evidence for infectious agents was more heterogeneous. Liu et al.³⁷ reported a nearly fourfold increased risk associated with Chlamydia pneumoniae seropositivity (aOR 3.92, 95% CI: 2.27–6.67, p < 0.001). Viral infections, particularly HPV, were also implicated: Cheng et al.³⁴ found HPV16/18 infection to be associated with an adjusted OR of 8.82 (95% CI: 2.28–34.16). Similarly, Chiou et al.³⁶ found HPV-16 OR = 13.6 (95% CI: 5.3–35.3) and HPV-18 OR = 7.1 (95% CI: 1.9–26.5). Meanwhile, for HPV-6, Cheng et al.^34,35 found no clear association with lung-cancer risk in FANS.

Hormonal/reproductive factors

Breastfeeding has consistently been associated with reduced lung cancer risk. In the Asia Cohort Consortium, Elbasheer et al.¹³ found that women who breastfed >12 months per child had a 14% lower risk than those who breastfed for 7–12 months, supporting the hypothesis that prolonged breastfeeding lowers lifetime estrogen exposure and confers protection. Parity findings are more mixed. Several studies reported a protective effect of childbirth, with parous women having lower risks than nulliparous women.^14,16,19,50 For example, Tan et al.¹⁶ observed decreasing risk with increasing number of deliveries (HRs 0.56, 0.55, 0.45 for 1–2, 3–4, ⩾5 deliveries, respectively), while Gallagher et al.¹⁴ noted increased risk among nulliparous women.

The use of oral contraceptives has generally been associated with elevated risk. Elbasheer et al.¹³ found an overall increased risk with oral contraceptives (HR = 1.16; 95% CI: 1.02–1.33). Reproductive span and timing of menopause may also contribute. Wilunda et al.¹⁸ reported increased adenocarcinoma risk with longer fertility spans and later menopause, whereas Weiss et al.¹⁷ observed decreased risk among lifetime non-smokers, and Seow et al.⁵⁰ found no association. Furthermore, Jeon et al.¹⁵ did not identify any significant association between lung cancer and reproductive factors. Such discrepancies suggest potential heterogeneity across populations and histology.

Other hormonal exposures may be relevant. Gallagher et al.¹⁴ reported elevated risks following oophorectomy and reduced risk with tubal ligation.¹⁴ In addition, Seow et al.⁵⁰ identified an inverse relationship between soy isoflavonoid intake and lung cancer incidence, suggesting dietary phytoestrogens may modulate risk. Overall, reproductive and hormonal factors appear to influence lung cancer risk in women, but findings vary across studies. Differences in study design, population composition, histologic subtypes, and endpoints (incidence vs mortality) may explain inconsistencies. Further mechanistic and population-based work is needed to clarify the interplay between hormonal exposures and lung carcinogenesis.

Dietary/vitamin intake

Several dietary factors have been investigated in relation to lung cancer risk, particularly among never-smoking populations. Fried and high-fat foods appear to increase risk, especially for adenocarcinoma. Butler et al.⁴¹ found higher fried meat and fried fish intake associated with elevated risk, with stronger associations in never smokers (fried fish HR 1.57, 95% CI: 1.06–2.33).

Antioxidants and vitamins show mixed findings. Wu et al.⁴⁸ reported that higher dietary tocopherol intake was protective (HR 0.78; 95% CI: 0.60–0.99), with stronger effects among passive smokers (HR 0.53; 95% CI: 0.29–0.97). However, vitamin E supplementation was associated with increased risk, particularly adenocarcinoma (HR 1.79; 95% CI: 1.23–2.60), highlighting potential differences between dietary sources and high-dose supplements. Takata et al.⁴⁷ found higher riboflavin and methionine intake were inversely associated with lung cancer, while other B vitamins were not.

Cruciferous vegetables and soy isoflavones have been generally protective. Seow et al.⁴⁵ found soy food intake inversely associated with risk among non-smokers (OR ~0.56 for highest vs lowest tertile). More recent findings by Li et al.⁴² further support a protective effect of dietary isoflavones, with ~50% reduced risk across higher quartiles of intake. Shimazu et al.⁴⁶ observed trends toward reduced risk with isoflavone intake, particularly among never-smoking women (HR 0.67; 95% CI: 0.41–1.10), though results were not statistically significant.

Other dietary exposures have yielded variable associations. Lim et al.⁴³ found no link between processed meat or heterocyclic amines and lung cancer, but suggested fish consumption may be protective in never smokers, in contrast to Butler et al.⁴¹ finding of increased risk with fried fish, possibly reflecting differences in preparation methods. Medication use and anti-inflammatory exposures may also be relevant. Lim et al.⁴⁴ found no association between most analgesics or steroid use and lung cancer, but reported that aspirin use may reduce risk in Asian women, aligning with the proposed role of cyclooxygenase inhibition in lung carcinogenesis.

Multiple

Several studies identified additional risk and protective factors for lung cancer beyond traditional exposures. Liang et al.⁴⁹ reported multiple determinants, including cardiovascular disease (OR 2.05, 95% CI: 1.54–2.71), depression (OR 1.37, 95% CI: 1.06–1.77), high stress (OR 1.51, 95% CI: 1.05–2.18), family history of cancer (OR 1.91, 95% CI: 1.38–2.64), poor sleep quality (OR 1.75, 95% CI: 1.36–2.24), and cooking oil fume exposure (OR 2.11, 95% CI: 1.42–3.14), while frequent fruit consumption (OR 0.47, 95% CI: 0.34–0.64) and dietary supplements (OR 0.26, 95% CI: 0.11–0.59) were protective. Seow et al.⁵⁰ found that multiparity reduced lung cancer risk by 41%–51% compared with nulliparous women, and high soy isoflavonoid intake reduced risk by 25% among non-smokers. In contrast, Zhou et al.⁵² reported that increasing number of live births elevated lung cancer risk compared to 0–1 births (OR 1.11 for 2 births, 1.31 for 3 births, 2.03 for 4 births), while β-carotene (OR 0.34, 95% CI: 0.14–0.86) and vitamin C (OR 0.38, 95% CI: 0.16–0.94) intake were protective. Multivariate analyses identified eye irritation from cooking fumes (OR 11.4, 95% CI: 3.09–42.39), family history of lung cancer (OR 17.53, 95% CI: 1.63–188.89), lower economic status (Q4: OR 3.61, 95% CI: 1.25–10.44), ⩾4 live births (OR 2.32, 95% CI: 0.45–12.02), and vitamin E intake (OR 2.69, 95% CI: 1.19–6.36) as significant risk factors, and β-carotene remained protective (OR 0.28, 95% CI: 0.12–0.69).⁵² Wu et al.⁵¹ reported that male non-smokers had higher risk than females (HR 1.29, 95% CI: 1.20–1.38 after age adjustment; HR 1.38, 95% CI: 1.28–1.50 after full adjustment), yet most known factors could not explain this excess, and lifestyle (HR 1.33), comorbidities (HR 1.37), and family history (HR 1.37) were more harmful in women.

Review of study characteristics identified possible overlap in participant cohorts across several publications. The Shanghai Women’s Health Study was used in Kim et al.,²² Li et al.,⁴² Pronk et al.,¹² Takata et al.,⁴⁷ Wu et al.,⁴⁸ and Zhang et al.³³ The Singapore Chinese Health Study was used in Butler et al.,⁴¹ Seow et al.,⁵⁰ and Yin et al.³² The following study pairs were identified as using overlapping cohorts: Cheng et al.^34,35; Lim et al.^43,44; Wang et al.^25,40 Yin et al.¹⁹ used a pooled Asia Cohort Consortium with various cohorts.

Discussion

This review identified occupational, IAP exposures, and family history as the most consistent and well-supported non-tobacco risk factors for lung cancer in FANS. IAP, including high-temperature cooking oil fumes, prolonged solid-fuel use, and inadequate household or workplace ventilation, was consistently associated with elevated lung cancer risk, often 2–4-fold higher, and rising up to 12-fold with high cumulative exposures. Family history of lung cancer, especially in first-degree relatives, nearly doubled risk in several large cohort studies. In addition, prior non-malignant lung conditions such as tuberculosis, asthma, and LAM were significantly linked to increased susceptibility. Other contributors such as hormonal and reproductive factors, certain infectious agents, and dietary patterns showed mixed but notable associations. The age distribution across studies was relatively homogeneous, with most participants in their mid-50s to mid-60s. This narrow age range may reflect a cohort-specific pattern of risk, suggesting that certain exposures could differ in their relevance or magnitude among younger or older women.

The findings of this review underscore the importance of recognizing that lung cancer in FANS often arises from specific non-tobacco risk factors, particularly chronic exposure to high-temperature cooking oil fumes, prolonged solid-fuel use, inadequate household or workplace ventilation, and a strong family history of lung cancer. These exposures are common, largely preventable, and often under-assessed in routine clinical care. Clinicians should incorporate detailed household, occupational, and reproductive histories into risk assessments for Asian female patients. Counseling on practical mitigation strategies, such as improving kitchen ventilation, transitioning to cleaner cooking fuels, reducing deep-frying at high temperatures, and seeking earlier evaluation of respiratory symptoms, could meaningfully reduce disease burden. Additionally, identifying women with a strong family history of lung cancer or coexisting non-malignant lung diseases (e.g., tuberculosis, asthma, LAM) may support tailored screening strategies and early detection, even among never-smokers. By shifting focus beyond smoking-related risks, these insights can guide culturally relevant prevention measures, inform risk-stratified screening programs, and ultimately help close gaps in early diagnosis and outcomes for this vulnerable population.

Real-world evidence supports the effectiveness of such strategies. In an observational study, replacing smoky coal indoor stoves with portable stoves was associated with lower lung cancer mortality in both men and women, demonstrating that switching from highly polluting fuels can materially reduce cancer risk.⁵³ Additionally, a recent engineering-air quality study highlighted that well-designed kitchen ventilation systems (especially those optimized for gaseous pollutants) substantially lower indoor concentrations of volatile organic compounds and carcinogenic byproducts from cooking.⁵⁴ These intervention trials and observational comparisons convincingly demonstrate that air pollution exposures can be reduced via cleaner fuels, improved stoves, and household changes. However, none has yet demonstrated a direct reduction in lung cancer incidence, likely due to limitations including inadequate follow-up duration, insufficient sample sizes for rare outcomes, or incomplete exposure reductions. Future trials will require large populations, extended follow-up, high adherence to the intervention, and sufficient contrast in exposure to detect effects on lung cancer risk.

When assessing lung cancer risk in FANS, clinicians should incorporate targeted exposure-history questions beyond smoking status, as suggested in Table 5.

Table 5.

Exposure history questions for FANS.

Category	Questions
Household environment exposure:	• Does anyone in your living space smoke?• Can you describe your current living situation (housing type, ventilation, cooking methods)?
Family history:	• Do you have a family history of lung cancer? If so, which relatives?
Occupational exposure:	• Can you describe your current and past work environments?• Are you exposed to industrial fumes, dusts, or chemicals at work?
Combustion and fume exposure:	• Are you exposed to heated combustion products such as cooking oils, coal, charcoal, wood, or soot? How often?• How frequently are you exposed to fumes or smoke of any kind (cooking, heating, occupational, or environmental)?

FANS, female Asian never-smokers.

Emerging evidence also suggests that incorporating family history of lung cancer, presence of persistent ground-glass nodules, and prior pulmonary diseases into screening algorithms may improve the identification of never-smokers at elevated risk. For instance, the TALENT study in Taiwan explored low-dose computed tomography (LD-CT) screening in never-smokers and demonstrated that including family history as a criterion improved detection rates of early-stage adenocarcinomas that would have otherwise been missed under traditional smoking-based eligibility criteria.⁵⁵ In addition to TALENT, emerging screening initiatives in Western populations further underscore the clinical importance of improved risk stratification for FANS. The Female Asian Nonsmoker Screening Study (FANSS) in the United States is currently evaluating the feasibility and yield of low-dose CT screening in this high-risk group outside of Asia, addressing a critical evidence gap in non-smoking-related lung cancer prevention. Furthermore, Lee et al.²⁸ found that women with a family history of lung cancer had higher rates of persistent and growing ground-glass nodules on surveillance imaging, supporting the integration of family history into screening guidelines. Public health initiatives that link environmental risk reduction with accessible LD-CT screening programs tailored for high-risk never-smokers could bridge a critical gap in current lung cancer prevention efforts.

An important biological consideration in this population is the disproportionately high prevalence of epidermal growth factor receptor (EGFR)-mutated non-small cell lung cancer among FANS, particularly within adenocarcinoma histology. Several of the non-tobacco exposures identified in this review, including cooking oil fumes^56–58 and household fuel smoke,⁵⁹ are known to induce oxidative stress, DNA damage, and persistent airway inflammation, processes that may plausibly contribute to mutational events and selective pressures favoring EGFR-driven tumorigenesis. While the present review did not include molecular analyses and therefore cannot establish direct causal links between specific exposures and EGFR or other mutations, these converging epidemiologic and biologic observations support a model in which environmental and inflammatory exposures interact with underlying susceptibility to shape the molecular landscape of lung cancer in FANS. Future studies integrating detailed exposure assessment with tumor genomic profiling will be critical for clarifying these relationships and refining prevention and screening strategies in this high-risk population.

Although lung cancer in FANS is increasingly observed in populations residing outside of Asia, migration-related factors such as country of origin, duration of residence, acculturation, and environmental transitions are not reported with sufficient consistency across studies to permit meaningful synthesis. These variables are highly heterogeneous and confounded by occupational, socioeconomic, and healthcare-access factors and represent an important area for future investigation. These gaps highlight the need for prospective, multi-center cohort studies with standardized exposure assessments to clarify causal pathways and inform evidence-based prevention strategies for this high-risk population.

Interpretation of this review should be tempered by several limitations. First, the included studies varied widely in study design, exposure measurement, outcome definitions, and adjustment for confounders, precluding formal meta-analysis and limiting causal inference. Recall and selection biases likely affected many retrospective case–control studies, particularly those relying on self-reported lifetime cooking and fuel-use histories. Second, several exposures, such as infectious agents, reproductive factors, and dietary patterns, were assessed in relatively few studies with inconsistent endpoints (incidence vs mortality) or without stratification by histologic subtype, reducing generalizability. Additionally, regional heterogeneity, small sample sizes for rare exposures, and lack of standardized exposure metrics limited cross-study comparability. Inclusion criteria were limited to never-smokers and excluded non-smokers with previous smoking history. Risk factors such as broader environmental exposures and genetic biomarkers or gene polymorphisms were not included in this review, as the focus was on modifiable, targetable risk factors that can inform prevention and intervention strategies.

Conclusion

This systematic review reveals that lung cancer in FANS is driven by a complex interplay of non-tobacco and non-genetic factors, challenging the notion that lung cancer is primarily smoking-related. Prevention efforts must extend beyond tobacco control to target modifiable environmental and household exposures. Future large-scale, multi-center studies should integrate genetic risk profiling with standardized exposure assessments to guide effective risk-reduction strategies in this underrepresented population. Understanding these sex- and ethnicity-specific risk profiles is essential for advancing equitable lung cancer prevention, screening, and policy interventions.

Supplemental Material

sj-docx-1-tam-10.1177_17588359261446808 – Supplemental material for Etiology and risk factors for lung cancer in female Asian never smokers: a systematic review

Supplemental material, sj-docx-1-tam-10.1177_17588359261446808 for Etiology and risk factors for lung cancer in female Asian never smokers: a systematic review by Shreya Guha, Keza Levine, Elaine Liang, Alison S. Baskin, Michael Ou and Jeffrey B. Velotta in Therapeutic Advances in Medical Oncology

Supplemental Material

sj-docx-2-tam-10.1177_17588359261446808 – Supplemental material for Etiology and risk factors for lung cancer in female Asian never smokers: a systematic review

Supplemental material, sj-docx-2-tam-10.1177_17588359261446808 for Etiology and risk factors for lung cancer in female Asian never smokers: a systematic review by Shreya Guha, Keza Levine, Elaine Liang, Alison S. Baskin, Michael Ou and Jeffrey B. Velotta in Therapeutic Advances in Medical Oncology

Footnotes

Acknowledgements

The authors wish to acknowledge the valuable contributions of Eileen Chen, MLIS at UCSF, in helping design, refine, and validate the final search terms and the selection of databases.

Declarations

ORCID iDs

Shreya Guha

Elaine Liang

Alison S. Baskin

Jeffrey B. Velotta

Supplemental material

Supplemental material for this article is available online.

References

Zhang

Luo

Etxeberria

, et al. Global patterns and trends in lung cancer incidence: a population-based study. J Thorac Oncol 2021; 16(6): 933–944.

US Preventive Services Task Force. Screening for lung cancer: US Preventive Services Task Force Recommendation Statement. JAMA 2021; 325(10): 962–970.

Siegel

Fedewa

Henley

, et al. Proportion of never smokers among men and women with lung cancer in 7 US States. JAMA Oncol 2021; 7(2): 302–304.

Pinheiro

Callahan

Medina

, et al. Lung cancer in never smokers: distinct population-based patterns by age, sex, and race/ethnicity. Lung Cancer 2022; 174: 50–56.

DeRouen

Thompson

Canchola

, et al. Integrating electronic health record, cancer registry, and geospatial data to study lung cancer in Asian American, Native Hawaiian, and Pacific Islander ethnic groups. Cancer Epidemiol Biomarkers Prev 2021; 30(8): 1506–1516.

Kamtam

Shrager

JB.

We should be considering lung cancer screening for never-smoking Asian American females. J Thorac Cardiovasc Surg 2024; 168(1): 272–277.e1

Zhou

Lung cancer in never smokers—the East Asian experience. Transl Lung Cancer Res 2018; 7(4): 450–463.

Sun

Schiller

Gazdar

AF.

Lung cancer in never smokers—a different disease. Nat Rev Cancer 2007; 7(10): 778–790.

Higgins

JPT

Morgan

Rooney

, et al. A tool to assess risk of bias in non-randomized follow-up studies of exposure effects (ROBINS-E). Environ Int 2024; 186: 108602.

10.

Foroutan

Guyatt

Zuk

, et al. GRADE Guidelines 28: use of GRADE for the assessment of evidence about prognostic factors: rating certainty in identification of groups of patients with different absolute risks. J Clin Epidemiol 2020; 121: 62–70.

11.

Jang

Kim

, et al. Lung cancer risk in female school cooks: a nationwide retrospective cohort study in the Republic of Korea. Saf Health Work 2025; 16(1): 105–112.

12.

Pronk

Coble

, et al. Occupational risk of lung cancer among lifetime non-smoking women in Shanghai, China. Occup Environ Med 2009; 66(10): 672–678.

13.

Elbasheer

MMA

Bohrmann

Chen

, et al. Reproductive factors and risk of lung cancer among 300,000 Chinese female never-smokers: evidence from the China Kadoorie Biobank study. BMC Cancer 2024; 24(1): 384.

14.

Gallagher

Rosenblatt

Ray

, et al. Reproductive factors and risk of lung cancer in female textile workers in Shanghai, China. Cancer Causes Control 2013; 24(7): 1305–1314.

15.

Jeon

Shin

Han

, et al. Female reproductive factors and the risk of lung cancer in postmenopausal women: a nationwide cohort study. Br J Cancer 2020; 122(9): 1417–1424.

16.

Tan

Chow

, et al. Reproductive factors and lung cancer risk among women in the Singapore Breast Cancer Screening Project. Lung Cancer 2015; 90(3): 499–508.

17.

Weiss

Lacey

Shu

, et al. Menstrual and reproductive factors in association with lung cancer in female lifetime nonsmokers. Am J Epidemiol 2008; 168(11): 1319–1325.

18.

Wilunda

Sawada

Yamaji

, et al. Reproductive factors and lung cancer risk among never-smoking Japanese women with 21 years of follow-up: a cohort study. Cancer Epidemiol Biomarkers Prev 2021; 30(6): 1185–1192.

19.

Yin

Kishida

Abe

, et al. Association between reproductive factors with lung cancer incidence and mortality: a pooled analysis of over 308,000 females in the Asia cohort consortium. Int J Cancer 2024; 154(12): 2090–2105.

20.

Chen

Fang

Chen

, et al. Impact of cooking oil fume exposure and fume extractor use on lung cancer risk in non-smoking Han Chinese women. Sci Rep 2020; 10(1): 6774.

21.

Fang

Zhou

Kim

, et al. The association between indoor air pollution and lung cancer risk in a Chinese Population. Indoor Air 2025; 2025(1): 9937960.

22.

Kim

Gao

Xiang

, et al. Home kitchen ventilation, cooking fuels, and lung cancer risk in a prospective cohort of never smoking women in Shanghai, China. Int J Cancer 2015; 136(3): 632–638.

23.

Kim

, et al. Cooking oil fume exposure and Lung-RADS distribution among school cafeteria workers of South Korea. Ann Occup Environ Med 2024; 36: e2.

24.

Liu

Niu

, et al. Indoor air pollution and risk of lung cancer among Chinese female non-smokers. Cancer Causes Control 2013; 24(3): 439–450.

25.

Wang

Chiu

Qiu

, et al. The roles of smoking and cooking emissions in lung cancer risk among Chinese women in Hong Kong. Ann Oncol 2009; 20(4): 746–751.

26.

Wong

JYY

Downward

, et al. Lung cancer risk by geologic coal deposits: a case-control study of female never-smokers from Xuanwei and Fuyuan, China. Int J Cancer 2019; 144(12): 2918–2927.

27.

ITS

Chiu

JSK

, et al. Dose-response relationship between cooking fumes exposures and lung cancer among Chinese nonsmoking women. Cancer Res 2006; 66(9): 4961–4967.

28.

Lee

Choi

Kim

, et al. Low-dose CT screening in East Asian women who have never smoked: association between family history of lung cancer and ground-glass nodule prevalence and growth. Radiology 2024; 313(3): e241286.

29.

Tse

sun, Rothman

, et al. Joint effects of environmental exposures and familial susceptibility to lung cancer in Chinese never smoking men and women. J Thorac Oncol 2014; 9(8): 1066–1072.

30.

Wang

Tan

, et al. Lung cancer risk in non-smoking females with a familial history of cancer: a multi-center prospective cohort study in China. J Natl Cancer Center 2021; 1(3): 108–114.

31.

Lee

Wang

, et al. Cancer aggregation and complex segregation analysis of families with female non-smoking lung cancer probands in Taiwan. Eur J Cancer 2004; 40(2): 260–266.

32.

Yin

Chan

CPY

Seow

, et al. Association between family history and lung cancer risk among Chinese women in Singapore. Sci Rep 2021; 11(1): 21862.

33.

Zhang

Shu

Gao

, et al. Family history of cancer and risk of lung cancer among nonsmoking Chinese women. Cancer Epidemiol Biomarkers Prev 2007; 16(11): 2432–2435.

34.

Cheng

Chiou

Sheu

, et al. The association of human papillomavirus 16/18 infection with lung cancer among nonsmoking Taiwanese women. Cancer Res 2001; 61(7): 2799–2803.

35.

Cheng

Chiou

Chen

, et al. Gender difference in human papillomarvirus infection for non-small cell lung cancer in Taiwan. Lung Cancer 2004; 46(2): 165–170.

36.

Chiou

Liaw

, et al. The presence of human papillomavirus type 16/18 DNA in blood circulation may act as a risk marker of lung cancer in Taiwan. Cancer 2003; 97(6): 1558–1563.

37.

Liu

, et al. Association of Chlamydia pneumoniae immunoglobulin G antibodies with the risk of lung cancer among non-smoking women in Liaoning, China. Thorac Cancer 2010; 1(3): 126–129.

38.

Liang

Guan

Yin

, et al. Risk of lung cancer following nonmalignant respiratory conditions among nonsmoking women living in Shenyang, Northeast China. J Womens Health (Larchmt) 2009; 18(12): 1989–1995.

39.

Torasawa

Shukuya

Uemura

, et al. Lymphangioleiomyomatosis as a potent lung cancer risk factor: insights from a Japanese large cohort study. Respirology 2024; 29(9): 815–824.

40.

Wang

ITS

Chiu

, et al. Previous pulmonary disease and family cancer history increase the risk of lung cancer among Hong Kong women. Cancer Causes Control 2009; 20(5): 757–763.

41.

Butler

Montague

Koh

, et al. Fried meat intake is a risk factor for lung adenocarcinoma in a prospective cohort of Chinese men and women in Singapore. Carcinogenesis 2013; 34(8): 1794–1799.

42.

Cai

Gao

, et al. Phytoestrogens and lung cancer risk: a nested case-control study in never-smoking Chinese women. Am J Clin Nutr 2022; 115(3): 643–651.

43.

Lim

Chuah

Eng

, et al. Meat consumption and risk of lung cancer among never-smoking women. Nutr Cancer 2011; 63(6): 850–859.

44.

Lim

Chuah

Eng

, et al. Aspirin and non-aspirin non-steroidal anti-inflammatory drug use and risk of lung cancer. Lung Cancer 2012; 77(2): 246–251.

45.

Seow

Poh

Teh

, et al. Diet, reproductive factors and lung cancer risk among Chinese women in Singapore: evidence for a protective effect of soy in nonsmokers. Int J Cancer 2002; 97(3): 365–371.

46.

Shimazu

Inoue

Sasazuki

, et al. Isoflavone intake and risk of lung cancer: a prospective cohort study in Japan. Am J Clin Nutr 2010; 91(3): 722–728.

47.

Takata

Cai

Beeghly-Fadiel

, et al. Dietary B vitamin and methionine intakes and lung cancer risk among female never smokers in China. Cancer Causes Control 2012; 23(12): 1965–1975.

48.

Xiang

Yang

, et al. Vitamin E intake and the lung cancer risk among female nonsmokers: a report from the Shanghai Women’s Health Study. Int J Cancer 2015; 136(3): 610–617.

49.

Liang

Wang

, et al. Lung cancer in never-smokers: a multicenter case-control study in North China. Front Oncol 2019; 9: 1354.

50.

Seow

Koh

Wang

, et al. Reproductive variables, soy intake, and lung cancer risk among nonsmoking women in the Singapore Chinese Health Study. Cancer Epidemiol Biomarkers Prev 2009; 18(3): 821–827.

51.

Tan

Yang

, et al. Sex disparity of lung cancer risk in non-smokers: a multicenter population-based prospective study based on China National Lung Cancer Screening Program. Chin Med J (Engl) 2022; 135(11): 1331–1339.

52.

Zhou

Wang

Guan

, et al. Indoor air pollution and pulmonary adenocarcinoma among females: a case-control study in Shenyang, China. Oncol Rep 2000; 7(6): 1253–1259.

53.

Hosgood

Chapman

Shen

, et al. Portable stove use is associated with lower lung cancer mortality risk in lifetime smoky coal users. Br J Cancer 2008; 99(11): 1934–1939.

54.

Yin

Pei

Liu

The effectiveness of kitchen ventilation for organic gaseous compound control in Chinese residential buildings. Build Environ 2022; 226: 109764.

55.

Chang

Chiu

, et al. Low-dose CT screening among never-smokers with or without a family history of lung cancer in Taiwan: a prospective cohort study. Lancet Respir Med 2024; 12(2): 141–152.

56.

Laursen

Christensen

Mulder

, et al. Airway and systemic biomarkers of health effects after short-term exposure to indoor ultrafine particles from cooking and candles – a randomized controlled double-blind crossover study among mild asthmatic subjects. Part Fibre Toxicol 2023; 20(1): 26.

57.

Tung

Liang

, et al. Cooking oil fume-induced cytokine expression and oxidative stress in human lung epithelial cells. Environ Res 2001; 87(1): 47–54.

58.

Lai

Jaakkola

JJK

Chuang

, et al. Exposure to cooking oil fumes and oxidative damages: a longitudinal study in Chinese military cooks. J Expo Sci Environ Epidemiol 2013; 23(1): 94–100.

59.

Mukherjee

Dutta

Roychoudhury

, et al. Chronic inhalation of biomass smoke is associated with DNA damage in airway cells: involvement of particulate pollutants and benzene. J Appl Toxicol 2013; 33(4): 281–289.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

2.73 MB

0.00 MB

3.20 MB