When "Healthy Eating" Gets a Cancer Diagnosis: The USC Study That Rattled Nutrition Science

A small observational study from USC suggesting that fruit-and-vegetable-heavy diets may raise lung cancer risk in young non-smokers has generated outsized headlines — but the weight of prior evidence, and the study's own limitations, tell a more complicated story.

The Study That Sparked the Headlines

In April 2026, Dr. Jorge Nieva, a medical oncologist at the USC Norris Comprehensive Cancer Center, presented findings from the Epidemiology of Young Lung Cancer Project at the annual meeting of the American Association for Cancer Research (AACR) [1]. The study surveyed 187 patients diagnosed with lung cancer before age 50, collecting data on their demographics, dietary habits, smoking history, and tumor molecular characteristics [2].

The central finding: young, non-smoking lung cancer patients reported eating healthier diets than the general U.S. population. Their average Healthy Eating Index (HEI) score — a standardized metric that rates overall diet quality on a scale of 0 to 100 — was 65, compared to a national average of 57 [1]. These patients consumed an average of 4.3 daily servings of dark green vegetables and legumes versus the U.S. average of 3.6, and 3.9 servings of whole grains versus 2.6 nationally [2]. Women in the cohort reported higher HEI scores than men and were diagnosed with lung cancer more frequently than men in the same age group [3].

Nieva's team proposed a mechanism: pesticide residues. Commercially produced fruits, vegetables, and whole grains carry higher levels of pesticide residue than meat, dairy, or processed foods, and these pesticides contain bioactive compounds capable of inducing mutations, oxidative stress, and inflammatory processes in lung tissue [2]. The hypothesis drew support from occupational data showing that agricultural workers exposed to pesticides have disproportionately higher lung cancer rates [1].

What the Study Did Not Do

The limitations are substantial, and Nieva's team has acknowledged several of them.

First, the study did not directly measure pesticide levels in patients' blood, urine, or food [2]. Instead, it used published estimates of average pesticide residues across broad food categories to infer exposure — a significant inferential leap.

Second, this was not a controlled comparison. There was no matched control group of young non-smokers who did not develop lung cancer. The study compared lung cancer patients' diets against population-wide averages from the general U.S. population, which includes smokers, older adults, and people across all health statuses [1]. Without a proper control group of demographically matched healthy individuals, the finding that cancer patients ate more vegetables than the average American may reflect nothing more than the demographics of who gets surveyed.

Third, the results have not yet been published in a peer-reviewed journal [4]. The findings were presented at a conference, which is a preliminary step in the scientific process.

Dr. David Yashar, a hematologist-medical oncologist at MemorialCare Cancer Institute who was not involved in the research, offered a measured response: "In general, these are fascinating findings; however, I think we must be careful in the way we analyze the data." Yashar emphasized that the results do not suggest fruits, vegetables, or whole grains themselves increase lung cancer risk [4].

The Funding and Conflict-of-Interest Picture

The study was funded by a mix of nonprofit and industry sources: the Addario Lung Cancer Medical Institute, GO2 for Lung Cancer, the Beth Longwell Foundation, and Upstage Lung Cancer, alongside grants from the National Institutes of Health (R25CA225513) and the National Cancer Institute (P30CA014089) [2]. AstraZeneca and Genentech also provided funding, and Dr. Nieva disclosed receiving consulting payments from both companies [2]. While pharmaceutical funding does not automatically invalidate research, it is relevant context for a study that generated headlines suggesting healthy foods may cause cancer — a finding that, if accepted uncritically, could discourage produce consumption without benefiting any patient population.

What 40 Years of Prior Research Actually Shows

The USC study's finding runs counter to a large and consistent body of evidence. A 2025 systematic review and dose-response meta-analysis published in the Journal of Cancer Prevention analyzed 41 prospective cohort studies on fruit and vegetable intake and lung cancer risk [5]. The results were clear: higher consumption of total fruits and vegetables was associated with a summary risk ratio of 0.81, meaning a 19% reduction in lung cancer risk. Fruits alone showed a risk ratio of 0.78 (22% reduction), vegetables 0.87 (13% reduction), and cruciferous vegetables 0.82 (18% reduction) [5].

The European Prospective Investigation into Cancer and Nutrition (EPIC), one of the largest cohort studies ever conducted with more than 478,000 participants across 10 European countries, found that a 100 g/day increase in fruit and vegetable consumption was associated with a reduced lung cancer risk (hazard ratio 0.94, 95% CI 0.89–0.99) [6]. The protective effect was most pronounced among current smokers who ate more vegetables [7].

A separate 2019 meta-analysis of prospective cohort studies found that higher fruit consumption was negatively associated with lung cancer risk among both current and former smokers [8].

Source: American Cancer Society / CDC

Data as of Jan 1, 2025CSV

To put the relative magnitudes in perspective: smoking increases lung cancer risk by roughly 15 to 30 times over baseline [9]. The incidence rate among current smokers is approximately 132.5 per 100,000, compared to 12.4 per 100,000 among never-smokers — a tenfold difference [9]. Any effect of diet, whether protective or harmful, operates within a much narrower band. The USC study did not calculate a hazard ratio or absolute risk change at all; it simply observed that its 187 cancer patients ate more produce than the average American.

The Residual Confounding Problem

The most likely explanation for the USC finding, according to several epidemiologists, is residual confounding — the possibility that unmeasured or poorly controlled variables are driving the apparent association.

Lung cancer in never-smokers is a growing area of concern. The CDC estimates that 10% to 20% of U.S. lung cancers — between 20,000 and 40,000 cases annually — occur in people who never smoked or smoked fewer than 100 cigarettes in their lifetime [9]. Radon exposure is the leading cause of lung cancer among non-smokers, responsible for an estimated 21,000 deaths per year according to the EPA [10]. Air pollution is another established risk factor: a 2022 study published in Nature identified compelling genomic evidence linking air pollution to lung cancer development in never-smokers [11].

Other known risk factors for lung cancer in non-smokers include secondhand smoke exposure (contributing to roughly 7,300 lung cancer deaths annually), occupational carcinogens such as asbestos, and genetic predisposition — particularly EGFR mutations, which are more common in women and people of East Asian ancestry [9][11].

Health-conscious individuals who eat more fruits and vegetables also tend to exercise more, live in certain geographic areas, and have different occupational profiles. Any of these correlated behaviors or exposures could account for the pattern the USC study observed. Without controlling for radon exposure, residential air quality, occupational history, and genetic markers, the dietary association remains uninterpretable.

The Beta-Carotene Precedent

Nutrition science has been here before. In the 1990s, two major randomized controlled trials — the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study (ATBC) and the Beta-Carotene and Retinol Efficacy Trial (CARET) — tested whether beta-carotene supplements could prevent lung cancer in smokers. Both trials found the opposite: beta-carotene supplementation increased lung cancer risk. The CARET trial recorded 28% more lung cancers and 17% more deaths among participants taking beta-carotene and vitamin A [12]. A later meta-analysis of 109,394 subjects confirmed a 24% increase in lung cancer risk among smokers who received high-dose beta-carotene supplements [12].

The beta-carotene episode is instructive for two reasons. First, it demonstrates that isolated nutrients can behave differently from the whole foods that contain them — a principle that cuts against drawing conclusions about pesticide exposure from dietary patterns alone. Second, it illustrates how preliminary findings in nutritional epidemiology can lead to premature public health conclusions. The beta-carotene finding was robust (randomized, large-sample, replicated), and it still took years to properly contextualize. The USC study, by contrast, is observational, small, and unreplicated.

The Broader Crisis in Nutritional Epidemiology

The USC study arrives at a moment when nutritional epidemiology faces serious questions about its reliability. A 2022 review in Critical Reviews in Food Science and Nutrition documented an "irreproducibility crisis" in the field, noting that many prominent epidemiologic associations — including studies on vitamin C, vitamin D, selenium, calcium, and low-fat diets — have not been corroborated by large randomized trials [13]. The authors identified core structural problems: tiny effect sizes for nutritional risk factors, massive confounding among densely correlated variables, and researchers making causal claims from associational data [13].

Source: OpenAlex

Data as of Jan 1, 2026CSV

The volume of published research on diet and lung cancer has grown substantially, with more than 201,000 papers published to date and a peak of 28,092 in 2023 alone [14]. This flood of literature makes it easy for any single counterintuitive finding to gain disproportionate media attention, particularly when it challenges widely held beliefs about healthy eating.

A large pooled analysis that specifically examined whether pesticide residues on fruits and vegetables were associated with cancer risk found no association. The hazard ratios were 0.99 for high-pesticide-residue produce and 1.01 for low-pesticide-residue produce per serving per day [15]. The authors noted that these findings do not rule out associations with specific chemicals, but the broad-strokes claim that pesticides on produce drive cancer risk was not supported.

What Patients and Clinicians Should Take Away

No major oncology or public health body has issued guidance changes based on the USC study. The World Cancer Research Fund continues to recommend diets rich in fruits, vegetables, and whole grains as protective against multiple cancer types, including lung cancer [16]. The American Cancer Society's dietary guidelines remain unchanged.

For never-smokers concerned about lung cancer, the evidence points to actionable steps that have nothing to do with avoiding produce: testing homes for radon (particularly in high-risk geographic areas), reducing exposure to secondhand smoke, and discussing screening options with a physician if there is a family history of lung cancer [9][10].

For current and former smokers, the existing evidence base strongly supports increasing fruit and vegetable consumption as one component of reducing lung cancer risk — alongside, of course, smoking cessation [5][6].

The USC study raises a legitimate question about whether pesticide exposure through diet warrants further investigation. But the study did not measure pesticides, did not use a control group, and did not calculate a risk estimate. It is, as Nieva himself described it, "an early yet critical step" — not a basis for dietary recommendations, and certainly not a reason to eat fewer fruits and vegetables [2].

The Standard That Matters

A single observational study of 187 patients, without peer review, without a control group, and without direct exposure measurement, does not overturn decades of convergent evidence from hundreds of thousands of participants across dozens of prospective cohort studies. The threshold for changing dietary guidance is — and should be — far higher than a conference presentation.

The real story here is not that fruits and vegetables cause lung cancer. The real story is the rising incidence of lung cancer among young never-smokers, particularly women, and the urgent need to identify the actual causes — whether radon, air pollution, genetic factors, or something not yet measured. That investigation deserves rigorous science, not headlines that tell people to put down the broccoli.