Inflammatory-metabolic alterations in asthma-COPD overlap compared with COPD: evidence from NHANES and a hospital-based cohort

Background: Asthma-chronic obstructive pulmonary disease overlap (ACO) shares clinical characteristics with COPD but may be associated with a distinct inflammatory-metabolic profile. This study aimed to investigate inflammatory-metabolic biomarker profiles associated with ACO and to identify biomarkers that differentiate ACO from COPD. Methods: A two-cohort study was conducted. The National Health and Nutrition Examination Survey (NHANES) served as the primary cohort, while a hospital-based cohort from Chengdu was included as a complementary hospital-based clinical cohort. In NHANES, participants were classified into non-asthma/COPD, asthma-only, COPD-only and ACO groups. Weighted comparisons, pairwise analyses and multinomial logistic regression were performed to evaluate associations between 19 inflammatory-metabolic biomarkers and disease phenotypes. Variance inflation factor (VIF) analysis and least absolute shrinkage and selection operator (LASSO) regression were used to identify candidate biomarkers distinguishing ACO from COPD. Similar analyses were subsequently performed in the hospital-based cohort. Results: Participants with ACO exhibited a greater inflammatory-metabolic burden than individuals with asthma alone, COPD alone or neither condition. Pairwise comparisons showed that several metabolic biomarkers, particularly CTI, RCII, RFM, and LAP, remained significantly elevated in ACO compared with COPD alone. In multinomial logistic regression analyses, higher levels of inflammatory biomarkers (CLR, NLR, ELR, MLR, SII, and SIRI) and metabolic biomarkers (TyG and its derived indices, RFM, LAP, CTI, and RCII) remained significantly associated with ACO, whereas CALLY was inversely associated with ACO. In the NHANES cohort, LASSO regression identified ELR, MLR, CTI, and CALLY as the candidate biomarkers. The combined model (AUC = 0.580, 95% CI 0.550-0.611) showed the highest discriminatory performance among the evaluated biomarker combinations for distinguishing ACO from COPD. In the hospital-based cohort, LASSO regression selected TyG, AIP, CTI, MLR, and CLR. The combination of TyG, AIP, CTI, and MLR achieved the highest AUC (0.583, 95% CI 0.551-0.615). Conclusion: ACO was associated with inflammatory-metabolic alterations compared with COPD. Although the discriminatory performance of biomarker models was modest, findings from both cohorts consistently suggest an association between inflammatory-metabolic disturbances and the biological heterogeneity of ACO.