Inflammation does not amplify insulin resistance link to cancer history in NHANES study

A large, nationally representative analysis of United States adults has found that systemic inflammation does not strengthen the cross-sectional association between insulin resistance and a history of cancer, according to a study published in Cardiovascular Diabetology – Endocrinology Reports. Using data from the 2017–2018 National Health and Nutrition Examination Survey, or NHANES, researchers tested whether the neutrophil-to-lymphocyte ratio, a common blood marker of inflammation, modifies the relationship between insulin resistance and self-reported, physician-diagnosed cancer. The answer was a statistical null: the interaction between the two measures was not significant, and insulin resistance itself showed no independent association with cancer history in the adjusted model.

The study, led by Xiaorui Wang, analyzed 3,746 adults who had fasted for at least eight hours before examination. NHANES employs a complex, multistage probability sampling design, and all analyses accounted for the survey weights, strata and clusters needed to produce estimates representative of the non-institutionalized U.S. adult population. Insulin resistance was quantified with the Homeostasis Model Assessment index, known as HOMA-IR, calculated as fasting insulin in microunits per milliliter multiplied by fasting glucose in milligrams per deciliter, divided by 405. Systemic inflammation was measured as the neutrophil-to-lymphocyte ratio, or NLR, the absolute neutrophil count divided by the absolute lymphocyte count. Cancer history was defined as a self-reported, physician-diagnosed history of any cancer.

The weighted characteristics of the sample set the demographic stage. The mean age of participants was 44.2 years, with a standard error of 0.58. The weighted mean HOMA-IR was 3.80, with a standard error of 0.16, and the weighted mean NLR was 2.07, with a standard error of 0.04. The total prevalence of physician-diagnosed cancer was 8.1 percent, with a standard error of 0.6 percent. These figures describe a generally middle-aged population in which roughly one in twelve adults reported a prior cancer diagnosis, a distribution consistent with the epidemiology of cancer in the United States, where risk rises steeply with age.

In the primary weighted multivariable logistic regression model, adjusted for age, sex and body mass index, a one-unit increase in HOMA-IR was not associated with cancer history. The odds ratio was 1.00, with a 95 percent confidence interval of 0.98 to 1.02 and a P value of 0.941, a result indistinguishable from no association. Age was the strongest predictor in the model: each additional year of age was associated with an approximately 7 percent higher odds of reporting cancer history, with an odds ratio of 1.07, a 95 percent confidence interval of 1.06 to 1.09 and a P value below 0.001. Males had significantly higher cancer history than females, with an odds ratio of 1.72, a 95 percent confidence interval of 1.10 to 2.68 and a P value of 0.020. Body mass index showed no significant cross-sectional association with cancer history, with a P value of 0.399.

The central question of the study concerned effect modification. The researchers added an interaction term, HOMA-IR multiplied by NLR, to a model that included the main effects of both variables. The interaction coefficient was 0.00808, with a standard error of 0.00835 and a P value for interaction of 0.359. In plain terms, the data provided no evidence that the strength of the association between insulin resistance and cancer history varies meaningfully across levels of systemic inflammation. The direction and significance of the other variables in the model remained largely consistent with the primary analysis. All tests were two-sided with an alpha of 0.05, and missing covariate data, which affected less than 5 percent of participants, were handled by complete-case analysis without imputation.

The scientific motivation behind the question is well grounded. Insulin resistance, a hallmark of type 2 diabetes mellitus and metabolic syndrome, drives a state of hyperinsulinemia that can promote tumorigenesis through insulin and insulin-like growth factor-1 signaling pathways. At the same time, insulin resistance perpetuates chronic low-grade systemic inflammation, creating a cycle that aggravates metabolic dysfunction and tissue damage. The neutrophil-to-lymphocyte ratio has emerged as a robust and stable hematological marker of systemic inflammatory burden and has been independently associated with future cardiovascular events and overall mortality. The researchers hypothesized that if inflammation amplifies the carcinogenic risk associated with insulin resistance, that synergy should be detectable at the population level, including among the large reservoir of people with undiagnosed prediabetes or subclinical metabolic syndrome.

Why, then, was no interaction found? The authors offer several candidate explanations, each pointing to the boundaries of what a single cross-sectional analysis can resolve. First, overall cancer is a highly heterogeneous endpoint. Insulin resistance and chronic inflammation have been strongly linked to specific cancer types, including colorectal, liver, pancreatic and endometrial cancers. Pooling all cancers, which arise through vastly different pathological mechanisms, may dilute and mask interaction effects that exist only for cancers tied to the metabolic-inflammatory pathway. The authors call for future cancer type-stratified analyses, particularly for colorectal and liver cancers, to uncover potentially concealed specific associations.

Second, the choice of inflammatory biomarker may matter. NLR is a practical and stable indicator reflecting innate immune response and systemic inflammatory load, but the chronic low-grade inflammation associated with insulin resistance often involves more specific pathways, such as adipose tissue macrophage polarization and dysregulation of adipokines and cytokines including adiponectin, leptin, interleukin-6 and tumor necrosis factor-alpha. NLR may not fully capture these metabolism-specific inflammatory processes. The authors suggest that a multi-dimensional assessment incorporating markers such as high-sensitivity C-reactive protein and interleukin-6 might better elucidate potential interaction effects. Third, the synergistic effect may not follow a simple linear relationship but could emerge only above certain thresholds of metabolic dysfunction or inflammatory burden. In a general-population sample, most individuals may not reach those critical thresholds. This interpretation is consistent with external evidence showing that the independent effect of NLR on cancer risk is modest, with a hazard ratio of approximately 1.03, and the authors note that methods such as restricted cubic splines could formally test potential non-linear relationships in future work.

The study’s limitations deserve careful weight. Because the design is cross-sectional, HOMA-IR and NLR were measured at the time of the NHANES examination, whereas cancer history was based on self-reported prior physician diagnosis. For participants with a history of cancer, the measured values may reflect metabolic and inflammatory status after cancer diagnosis or treatment rather than before cancer onset. The observed associations therefore cannot be used to infer causality or temporal sequence, a critical caveat for interpreting the null findings. The cancer outcome relied on self-report, which is susceptible to recall bias and underreporting, especially for early-stage or asymptomatic cancers. Although key covariates were adjusted for, residual confounding by factors not fully measured, such as dietary quality, physical activity levels or specific medication history, remains possible. The study also lacks an external validation cohort and had no data on specific cancer types or a broader panel of inflammatory markers.

Against these limitations stand real strengths. The data come from the rigorously designed NHANES, providing a nationally representative sample that allows generalization to the non-institutionalized U.S. adult population. The key exposure variables were based on standardized laboratory measurements, ensuring objectivity and reliability, and all analyses followed survey guidelines and incorporated complex sampling weights to guarantee the accuracy of statistical inferences. The protocol was approved by the National Center for Health Statistics Research Ethics Review Board, and informed consent was obtained from all participants. For clinicians and public health planners, the practical message is measured: combining HOMA-IR and NLR for pan-cancer screening in the general population may have limited additive value, and it remains important to distinguish this null finding from NLR’s established prognostic value in patients with existing cancer. The authors conclude that future prospective studies with cancer type-stratified analyses and more comprehensive inflammatory marker panels are needed to further elucidate the relationship between insulin resistance, inflammation and cancer.

Subject of Research: Cross-sectional NHANES study of whether systemic inflammation modifies the association between insulin resistance and cancer history in U.S. adults

Article Title: The interplay of insulin resistance and systemic inflammation on cancer history: a nationally representative cross-sectional study from NHANES

Article References: Wang, X. (2026). The interplay of insulin resistance and systemic inflammation on cancer history: a nationally representative cross-sectional study from NHANES. Cardiovascular Diabetology – Endocrinology Reports, 12(1), Article 66. https://doi.org/10.1186/s40842-026-00335-y

Image Credits: AI Generated

DOI: 10.1186/s40842-026-00335-y

Keywords: insulin resistance, systemic inflammation, neutrophil-to-lymphocyte ratio, HOMA-IR, NHANES, cancer history, cross-sectional study, effect modification, type 2 diabetes, metabolic syndrome, epidemiology, logistic regression