PESTICIDES as an Important Cause of Lung Cancer in Young Nonsmokers

ContributorsProf. Dr. Mustafa ÖzdoğanAuthor · Medical review · Editorial approval

PublishedMay 13, 2026

UpdatedOct 16, 2026

18 min read

PESTICIDES as an Important Cause of Lung Cancer in Young Nonsmokers

In this article

For decades, oncology's central public health message was clear: eat plenty of fruit, vegetables and whole grains; reduce meat and processed foods; stop smoking. These three measures were globally accepted cornerstones of cancer prevention. However, in April 2026 in San Diego, at the American Association for Cancer Research (AACR) Annual Meeting, a study presented by Jorge Nieva and his team from the University of Southern California (USC) Norris Comprehensive Cancer Center — Abstract 5039 — raised a counterintuitive question about this established narrative. In an American cohort of 187 people, all diagnosed with lung cancer before age 50 and most never having smoked, consumption of vegetables, fruit and whole grains was markedly above the national average. The “Healthy Eating Index-2015” score, measured using validated food frequency questionnaires, was above the American reference value (58/100) in all three mutation groups — 64,9 in those with EGFR pathway mutations; 65,5 in fusion-positive cases (ALK/ROS1/RET/NTRK); and 63,5 in other cases with mixed mutations. In other words, these patients really were eating “healthily.” The team's proposed hypothesis is as follows: Pesticide/herbicide residues accumulating on fruit, vegetables and whole grains grown by conventional, nonorganic methods may, counterintuitively, be an overlooked environmental risk factor in the epidemiology of EGFR pathway and fusion-positive molecular types of lung cancer — especially in young women. The study generates a hypothesis rather than proof — but the high lung cancer rates observed for years in agricultural workers, and the fact that EGFR-mutant tumor biology is entirely distinct from smoking, lend weight to the finding. This article will examine the substance of the study, the biological rationale for the hypothesis and the limitations that matter.

— BACKGROUND: WHY IS LUNG CANCER INCREASING IN YOUNG WOMEN? —

A rising paradox in a world of declining smoking

In the mid 1980s, after cigarette consumption peaked in the US, lung cancer incidence in men began to fall markedly. This was one of the most visible public health successes in the fight against cancer: tobacco control measures (advertising bans, high taxes, bans in public places, warning labels, the era before vaping) had worked. However, the same decline was not observed in women. The gap in incidence between men and women gradually closed; in the 2010s, the rates were almost equal; in the 2020s, lung cancer became more common in women than in men among people under 50. This reversed a historical pattern that had persisted for decades.

In addition, this new, young, predominantly female lung cancer cohort consists mostly of people who have never smoked or had very low exposure. They also differ pathologically: the tumors are mostly adenocarcinomasand are biologically distinct from smoking-related small-cell carcinoma or squamous-cell carcinoma. Their molecular signatures are also different: in a substantial proportion of these young lung cancer cases, EGFR (epidermal growth factor receptor) mutations, ALK/ROS1/RET/NTRK fusions (gene fusions) or MET exon-14 skipping are among the actionable, targetable genomic alterations that characterize the disease. This is therefore a form of lung cancer arising from a different biological mechanism, rather than the typical tumor biology of smoking (TP53 + KRAS + high TMB).

The environmental factors behind this situation are among the most intensively studied oncology puzzles of the past decade. Hypotheses include air pollution and exposure to PM2.5 particles, cooking fumes (a known risk particularly in Asian cohorts), secondhand smoke, occupational exposures, radon, viral infections (HPV subtypes are disputed), estrogen-related biological pathways (the reason for female predominance) and, finally — as proposed by Nieva's team — dietary pesticide/herbicide residues.

— STUDY DESIGN AND COHORT —

USC Norris and the Epidemiology of Young Lung Cancer cohort (NCT04640259)

Gorbatov and colleagues examined participants enrolled in the Epidemiology of Young Lung Cancer study, a total of 187 patients , in a retrospective analysis. All patients had been diagnosed with lung cancer at age 50 or younger. Most were women (157, %84). Patients were divided into three molecular groups according to shared biological mechanisms:

MOLECULAR GROUPING — YLC COHORT

EGFR Pathway: Patients with EGFR mutations + ERBB2 (HER2) mutations. The classic young, female, nonsmoking phenotype of lung adenocarcinoma.
Fusion-Positive: Patients harboring ALK, ROS1, RET and NTRK gene fusions. Classic “actionable” target genomic alterations.
Other/Mixed Mutations: MET exon 14 skipping, TP53, KRAS, BRAF and other alterations. This group is the most biologically heterogeneous.

For 166 of the 187 patients (%88,8), a validated Food Frequency Questionnaire (FFQ) was completed. Diet quality was assessed with the Healthy Eating Index-2015 (HEI-2015) — a standardized measure of American diet quality. Reference values came from the American National Health and Nutrition Examination Survey (NHANES).

— MAIN FINDINGS —

Healthy eating — yet still cancer

64,9

HEI-2015 score in patients with EGFR pathway mutations (US reference: 58)

65,5

HEI-2015 score in fusion-positive patients

63,5

HEI-2015 score in patients with other/mixed mutations

%84

Proportion of women in the cohort (157/187)

There was no marked difference between the three molecular groups; all were approximately 6-7 points above the US reference. Distribution by sex:

Figure 1. Distribution of HEI-2015 dietary quality scores in the USC Norris Young Lung Cancer cohort. All molecular subgroups and both sexes were above US reference values. Young women had the highest score (65,6).

Looking at specific food categories:

YLC COHORT vs US REFERENCE — FOOD GROUP CONSUMPTION

Vegetables: 4,2 vs 3,5 (US ref)
Fruit: 3,3 vs 2,5
Whole grains: 3,9 vs 2,6

In other words, compared with the average American population, the young lung cancer cohort consumes substantially more fruit, vegetables and whole grains. This is the “ideal” dietary profile sought by public health campaigns. Yet the same cohort has lung cancer.

Our research shows that young nonsmokers who eat more healthy foods than the general population are more likely to develop lung cancer. This counterintuitive finding raises important questions about an unknown environmental risk factor associated with foods that are otherwise beneficial in relation to lung cancer — a factor that needs to be addressed. — Jorge J. Nieva, MD, medical oncologist and lung cancer specialist, USC Norris Comprehensive Cancer Center (AACR 2026 presentation press release)

— HYPOTHESIS: PESTICIDE/HERBICIDE RESIDUES —

A hidden side effect of healthy food?

The biological rationale for the hypothesis proposed by Nieva's team rests on three pillars:

First pillar — pesticide/herbicide residues are present at higher concentrations in fruit, vegetables and whole grains than in meat and dairy products. The reason is simple: plants — especially tomatoes, peppers, leafy greens with broad leaf surfaces and fruit eaten with the skin — often retain detectable pesticide/herbicide residues even after washing. Animal products are affected indirectly by agricultural chemicals (through feed), whereas plant products are subject to primary exposure. Each year, the US Department of Agriculture's Pesticide Data Program (PDP) reports that residues of 30+ different pesticides are present at detectable levels in fruit and vegetables in American stores. The Environmental Working Group's (EWG) “Dirty Dozen” list (the 12 products with the highest pesticide burden — strawberries, spinach, cabbage groups, grapes, peaches, pears, peppers, cherries) is a popular summary of this literature.

Second pillar — lung cancer incidence is high in agricultural workers. This is an occupational epidemiology finding known for decades. Agricultural workers applying pesticides, herbicides and fungicides (especially those spraying from the air, but also those applying them at ground level) have higher lung cancer incidence, NHL (non-Hodgkin lymphoma) and other cancer risks than comparison cohorts. The Agricultural Health Study (AHS) — a large cohort following 50.000+ agricultural workers in the US — has documented this relationship. The pesticide–lung cancer relationship is therefore already accepted in the occupational setting; Nieva's team asks whether ingesting the same substances through the diet, at much lower doses but chronically, also poses a risk.

Third pillar — EGFR-mutant adenocarcinoma biology is independent of tobacco/asbestos. EGFR mutations (particularly exon 19 deletion and exon 21 L858R) are enriched in young, nonsmoking women of East Asian ethnicity; they are biologically distinct from the typical tobacco-induced KRAS/TP53 mutation pattern. Which environmental factors contribute to EGFR-mutant tumor formation is one of modern oncology's greatest puzzles. A significant signal from pesticide/herbicide exposure in the EGFR-mutant cohort — as shown by Nieva's team — supports this hypothesis biologically.

WHICH PESTICIDE/HERBICIDE MOLECULES ARE SUSPECTED?

Nieva's team did not focus on specific molecules; they did not directly measure pesticides in blood or urine samples from people in their cohort. This is the study's most important limitation and was highlighted as a priority for subsequent molecular epidemiology studies. The most frequently mentioned suspected candidate molecules include: glyphosate (classified by IARC in 2015 as a “possible carcinogen for humans”, Group 2A); chlorpyrifos (an organophosphate insecticide whose agricultural use was banned in the US in 2021); paraquat (a herbicide associated with Parkinson's disease and pulmonary fibrosis); atrazine (one of the most widely used herbicides, banned in the EU but permitted in the US); and nicotinoid insecticides (particularly imidacloprid). The study does not establish which molecule or combination is the actual risk factor — that is the task of subsequent studies.

— MAIN OBJECTIONS TO THE HYPOTHESIS —

A single AACR abstract does not change practice: why the limitations matter

Important warning. This study is a conference abstract (AACR Abstract 5039), rather than a peer-reviewed full-text publication. The findings do not provide sufficient evidence to change clinical practice or public health recommendations. The consumption of vegetables, fruit and whole grains has robustly demonstrated benefits for cancer prevention at the level of cumulative evidence (in colorectal, prostate, stomach, head and neck, and many other cancers). For only a specific subgroup (young nonsmokers, particularly women who have developed EGFR-mutant adenocarcinoma), an environmental factor hypothesis is put forward by this study.

The study's methodological limitations are as follows:

STRENGTHS

A specific clinical phenotype (gene-based), a validated food frequency questionnaire (HEI-2015), a robust national reference such as NHANES, a prestigious academic center such as USC Norris, a biologically plausible hypothesis for environmental lung cancer and the ability to exclude the classic smoking hypothesis in the EGFR-mutant cohort.

LIMITATIONS

No control group (the cohort was compared with the general population using reference values). No blood/urine pesticide measurements. Small cohort (187 patients, only 28 men). A single center. Association rather than causation. Patients' retrospective food questionnaires were administered after diagnosis (risk of recall bias). Perhaps the “healthy” eaters were also a subpopulation with different (familial? genetic? hormonal?) characteristics.

The most critical limitation — because pesticides were not measured — is this paradox: the study shows an association between a “healthy diet” and “lung cancer,” but does not prove that pesticides are actually the causal agent underlying that association. Alternative explanations: high carotene intake (in the CARET and ATBC studies of the 1990s, beta-carotene supplementation paradoxically increased lung cancer risk in men who smoked — beta-carotene may have reappeared through a different mechanism in this modern cohort); phytoestrogen intake (high consumption of soy/legumes may affect estrogen-dependent pathways); microbiome changes; or entirely confounding variables (socioeconomic status, education, differences between urban/rural life).

— WHAT SHOULD THE PRACTICAL RECOMMENDATION BE, AND WHAT SHOULD IT NOT BE? —

From this study, the conclusion that should never be drawn is: “I should not eat fruit, vegetables or whole grains.” This interpretation is dangerous and wrong. Plant foods in the diet are among the strongest foundations of cancer prevention, confirmed by decades of epidemiological evidence. This level of evidence does not change with a single conference abstract.

The practical conclusion that can be drawn from this study: If you are increasing your fruit and vegetable consumption, you can also consider strategies to reduce the burden of pesticide residues.

REDUCING PESTICIDE EXPOSURE — PRACTICAL RECOMMENDATIONS

Washing: Wash all fruit and vegetables, including the skin, under running water with a brush for 20-30 seconds. Washing with water removes %50-70 of surface pesticide residues.
Washing with vinegar or baking soda: Soaking for 5-10 minutes in water + %1 vinegar or water + a dessertspoonful of sodium bicarbonate was found to be additionally effective for chlorpyrifos and organophosphate pesticides (Yang and colleagues 2017, JAFC).
Peeling: Peel when possible — especially products on the EWG “Dirty Dozen” list (except strawberries, which cannot be peeled).
Choosing organic selectively: Buying products with the highest pesticide burden (strawberries, spinach, grapes, cherries, peaches) organically, and those with the lowest burden (avocados, corn, pineapples, melons) conventionally, is a budget-friendly strategy.
Eating in season: Local produce in season rather than imported produce — the pesticide burden is generally lower.
Variety: Avoid excessive reliance on one type of produce; variety balances both nutritional value and potential residue diversity.

— PERSPECTIVE FROM TURKEY —

The pesticide residue landscape in Turkey. Although Turkey is formally aligned with EU pesticide residue limits, enforcement is uneven. Products purchased domestically — especially from grocery stores and public markets — sometimes have levels above maximum residue limits (MRLs), as detected in the Turkish Ministry of Agriculture and Forestry's annual monitoring reports. Products most frequently exceeding MRLs: tomatoes, peppers, cucumbers, grapes, apricots and peaches. Glyphosate was renewed for another 10 years of authorization in the EU in March 2024; its use is permitted in Turkey. Chlorpyrifos was banned in the EU in 2020 and remains permitted for some applications in Turkey. Turkish consumers' pesticide exposure is therefore probably higher than the European average.

Is the epidemiology of lung cancer in young people in Turkey (under 50, nonsmokers) increasing in parallel with the US? The answer is not yet clear. Case-based observations shared by oncology units at Memorial, Acıbadem, Anadolu Medical Center, Hacettepe and Istanbul University suggest a trend consistent with the US: in Turkey, too, the cohort of nonsmoking patients under 50 with EGFR-mutant lung adenocarcinoma, predominantly women, has grown markedly over the past 15 years. However, systematic epidemiological data at national level are limited. Joint studies by the Turkish Thoracic Society and Turkish Oncology Group are attempting to fill this gap.

For Turkish patients with EGFR mutations, following confirmation by genomic testing, the 2026 standard for exon 19 deletion / L858R is osimertinib (3rd-generation EGFR-TKI) as first-line treatment; for the rarer exon 20 insertion, amivantamab or similar approaches. In Turkey, osimertinib has been available under SGK coverage since 2018; reimbursement conditions are relatively broad. In this respect, Turkish patients with EGFR mutations are currently at an upper-middle global level for access to modern treatment.

— FUTURE RESEARCH DIRECTIONS —

During the presentation, Nieva's team clearly listed the follow-up work needed:

First step — Direct biomarker measurement. Measurement of specific pesticide metabolites in blood and urine from the same cohort (for example, glyphosate's AMPA metabolite and dialkyl phosphate metabolites of organophosphates). This takes the association one step closer to causation.

Second step — Animal models. Testing the effect of chronic low-dose pesticide exposure on EGFR-mutant lung adenocarcinoma development in transgenic mouse models (with a design similar to Tang's 2019 PNAS study of e-cigarette-related lung adenocarcinoma).

Third step — Validation in a large cohort. Reassessment of the relationship between pesticide intake (through food questionnaires or matching with residue databases) and lung cancer incidence in multicenter prospective cohort studies — for example, the Million Women Study, EPIC cohort, NIH-AARP cohort or similar registries.

Fourth step — Mechanistic molecular study. Investigation of which pesticide molecules can induce DNA damage, oxidative stress or epigenetic changes in epithelial cells using in-vitro cell culture and organoid models.

⚡ DROZDOGAN ACADEMY — CONCEPTUAL FRAMEWORK

⚡ WHY DOES IT MATTER?

The young (under 50), nonsmoking, predominantly female lung cancer cohort has grown markedly worldwide — in the US, EU and Asia — over the past 15 years. This cohort is biologically distinct from typical tobacco-induced pathology: it is characterized by actionable genomic alterations such as EGFR mutations and ALK/ROS1/RET/NTRK fusions. The reason for this increase is one of the most important puzzles modern oncology is trying to solve. Although hypothesis-generating, AACR 2026 Abstract 5039 raises the possibility that “plant-based diets regarded as healthy” may be an environmental risk factor through pesticide residues. This could affect the future of public health recommendations.

✅ STRENGTHS / EVIDENCE

A cohort from a prestigious academic center such as USC Norris (NCT04640259). A validated HEI-2015 food frequency questionnaire. NHANES reference values. Separate analyses of specific molecular subgroups (EGFR pathway, fusion-positive, other). A biologically plausible mechanism (lung cancer has already been established in agricultural worker cohorts). EGFR-mutant biology excludes the smoking hypothesis.

⚠️ LIMITATIONS

Only an AACR conference abstract (not peer-reviewed). No blood/urine pesticide measurements — no direct biomarker evidence. No control group (comparison only with NHANES reference values). Small cohort (187 patients, 28 men). A single center (USC Norris). Retrospective food questionnaire (recall bias). Association rather than causation. Confounding variables were not controlled (socioeconomic status, education, urban/rural life). Findings are insufficient to change practical public health recommendations; they cannot be turned into a “do not eat fruit and vegetables” message.

🩺 IMPLICATIONS FOR CLINICAL PRACTICE

Patients must never be told to “reduce fruit and vegetables.” Instead, consumption of plant products should continue, while strategies to reduce pesticide burden may be recommended (washing under running water for 20-30 seconds, washing with vinegar/baking soda for 5-10 minutes, peeling, choosing organic products from the Dirty Dozen list, eating local produce in season, variety). When taking the history of young women who develop EGFR-mutant adenocarcinoma — a profile frequently seen at Memorial Göztepe — questions about “living in an agricultural environment, pesticide use in a home garden, occupational agricultural exposure, rural life” should now be considered an additional layer of information. This does not change the treatment plan but enriches biological understanding.

❓ FUTURE QUESTIONS

Can this association be converted into causation through direct biomarker measurement? Which specific pesticide/herbicide molecules (glyphosate? chlorpyrifos? paraquat? atrazine?) are the actual risk factors? Can chronic low-dose pesticide exposure induce EGFR-mutant adenocarcinoma in animal models? What is the pesticide exposure profile in Turkey's young lung cancer cohort? Do the different molecular signatures of EGFR-mutant tumors contain other specific mutation patterns distinct from the smoking-related cohort (for example, the exon 19 vs 21 signal)? What are the long-term effects of differences in pesticide authorization policies between the EU and Turkey on cancer epidemiology? Do follow-up cohorts show a difference in lung cancer incidence between organic and conventional food consumption (over the long term)? Can standardized public health recommendations be made about the effectiveness of washing and preparation methods in reducing pesticide residues?

References

Gorbatov SD, Bittoni MA, Wu AH, Harper A, Virginia K, Florez N, Gitlitz BJ, Nieva JJ. Dietary patterns in young lung cancer: mutation-specific environmental associations [abstract]. Cancer Research 2026;86(7 Suppl):Abstract 5039. American Association for Cancer Research (AACR) Annual Meeting 2026, San Diego, CA, 17-22 April 2026. doi:10.1158/1538-7445.am2026-5039
ClinicalTrials.gov. Epidemiology of Young Lung Cancer Study. NCT04640259. (Registration page for the USC Norris cohort, current version 2025.)
National Cancer Institute. Healthy Eating Index (HEI). Accessed: 2026. (HEI-2015 methodology reference.)
Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey (NHANES). Atlanta, GA: CDC, NCHS. (Source of reference values.)
Tang Y, Salame TM, Eckschlager T, and colleagues. Pulmonary adenocarcinoma in mice exposed to e-cigarette smoke. Proceedings of the National Academy of Sciences USA 2019;116(43):21727-21731. (Transgenic mouse model of e-cigarettes and lung adenocarcinoma; methodologically similar reference.)
International Agency for Research on Cancer (IARC). Monograph Vol. 112: Some Organophosphate Insecticides and Herbicides. Lyon: WHO/IARC, 2017. (Carcinogenicity assessment of glyphosate and other pesticides.)
Yang T, Doherty J, Zhao B, Kinchla AJ, Clark JM, He L. Effectiveness of commercial and homemade washing agents in removing pesticide residues on and in apples. Journal of Agricultural and Food Chemistry 2017;65(44):9744-9752. (Effectiveness of vinegar/baking soda washing in reducing pesticide residues.)
Alavanja MCR, Hofmann JN, Lynch CF, and colleagues. Non-Hodgkin lymphoma risk and insecticide, fungicide and fumigant use in the Agricultural Health Study. PLoS ONE 2014;9(10):e109332. (Agricultural worker–pesticide–cancer cohort reference.)
Du H, Cao T, Lu X, Zhang T, Luo B, Li Z. Mediterranean diet patterns in relation to lung cancer risk: a meta-analysis. Frontiers in Nutrition 2022;9:844382. doi:10.3389/fnut.2022.844382 (Meta-analysis of the Mediterranean diet and lung cancer.)
The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. NEJM 1994;330(15):1029-1035. (ATBC study; the paradoxical increase in lung cancer with beta-carotene in men who smoked.)
Omenn GS, Goodman GE, Thornquist MD, and colleagues. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. NEJM 1996;334(18):1150-1155. (CARET study.)
Environmental Working Group. EWG's 2025 Shopper's Guide to Pesticides in Produce: Dirty Dozen and Clean Fifteen. Washington, DC: EWG, 2025.
Turkish Ministry of Agriculture and Forestry. Annual Reports of the Plant Protection Product Residue Monitoring Program. Ankara: TOB, current edition 2025.

This article is intended for medical news/information and does not provide individual treatment guidance. AACR 2026 Abstract 5039 is a conference abstract that has not undergone peer review; the results may be revised when it becomes a full publication. This article emphatically does not recommend reducing fruit and vegetable consumption. Plant foods are one of the robust epidemiological foundations of cancer prevention; a single conference abstract cannot change this established evidence. The article aims to share an environmental hypothesis that has entered the agenda of contemporary oncology, and to present methods for reducing pesticide exposure as practical recommendations. Treatment for EGFR-mutant adenocarcinoma should be planned individually by a multidisciplinary oncology board.

Reader comments

Share your comment

Your name and comment are published only after administrator approval. Your email address is never shown to visitors.

Be the first to comment.

DROZDOGAN AI

Explore this article together

Ask about the article and read answers alongside published sources from our site.

Menu

PESTICIDES as an Important Cause of Lung Cancer in Young Nonsmokers

A rising paradox in a world of declining smoking