Technical Summary: Smoking Status Stratifies Th1/Th2 Phenotypes in Pre-COPD High-Risk Individuals

Problem Statement

Chronic obstructive pulmonary disease (COPD) is a major global health burden, yet early intervention is hindered because the disease is typically diagnosed only after irreversible airflow limitation has occurred. The "pre-COPD" stage, characterized by early symptoms and structural changes without airflow obstruction, represents a critical window for prevention. However, the immune dysregulation occurring during this pre-clinical phase remains poorly characterized, particularly regarding how smoking status influences inflammatory phenotypes.

While Th1/Th2 polarization is well-documented in established COPD, it is unclear whether these distinct phenotypes emerge before airflow limitation occurs. Furthermore, conventional screening tools in primary care face limitations: spirometry is often unavailable or underutilized due to equipment and personnel shortages, and screening questionnaires (like the COPD-SQ) suffer from high false-positive rates when used alone. Additionally, existing research on inflammatory biomarkers has largely focused on established COPD, relied on conventional mean regression (which may miss subtle, distribution-specific abnormalities), and failed to systematically account for smoking status as a critical effect modifier in pre-COPD populations.

Methodology

This study employed a community-based cross-sectional design involving 149 participants aged ≥40 years from Minhang District, Shanghai. Participants were recruited via stratified cluster sampling and classified into three groups based on the 2024 Chinese Primary Care COPD Guidelines:

1. Healthy: COPD-SQ score < 16 and post-bronchodilator FEV1/FVC ≥ 0.7.
2. Pre-COPD High-Risk: COPD-SQ score ≥ 16 and post-bronchodilator FEV1/FVC ≥ 0.7.
3. Established COPD: Post-bronchodilator FEV1/FVC < 0.7.

Participants were further stratified by smoking status into a smoking subgroup (current and former smokers) and a never-smoking subgroup.

Data Collection:

• Clinical: Standardized questionnaires (including COPD-SQ), spirometry (pre- and post-bronchodilator), and smoking index calculation.
• Laboratory: Fasting peripheral blood was analyzed for complete blood count (CBC) parameters and 12 inflammatory cytokines (IL-5, IFN-α, IL-2, IL-6, IL-1β, IL-10, IFN-γ, IL-8, IL-17, IL-4, IL-12p70, TNF-α) using multiplex bead-based flow cytometry.

Statistical Analysis:

• Screening: Non-parametric tests (Kruskal-Wallis H, Mann-Whitney U) identified biomarkers with significant between-group differences.
• Interaction: Two-way nonparametric interaction models assessed the effect of smoking status on biomarker distributions.
• Core Analysis: Quantile regression was performed at the 25th (q=0.25), 50th (q=0.50), and 75th (q=0.75) percentiles. This approach was chosen to detect specific shifts in the biomarker distribution that conventional mean regression might overlook. Models were adjusted for age, sex, and smoking index (for the smoking subgroup).
• Correlation: Spearman's rank correlation analyzed the dose-response relationship between smoking index and core biomarkers.

Key Contributions

1. Stratified Phenotyping: The study demonstrates that smoking status acts as a key effect modifier, creating distinct Th1/Th2 inflammatory phenotypes specifically in the pre-COPD high-risk stage, which differ from those in established COPD.
2. Quantile-Specific Detection: By utilizing quantile regression, the study identified early immune dysregulation markers that are restricted to specific segments of the biomarker distribution (e.g., low-quantile or mid-quantile elevations), which would be masked by traditional mean-based analysis.
3. Primary Care Applicability: The research identifies low-cost, non-invasive peripheral blood biomarkers (routine CBC and cytokines) that could support stratified early screening strategies in primary care settings where advanced imaging or frequent spirometry is not feasible.

Results

Smoking Subgroup (Th1-Dominant Phenotype)

In smoking-related pre-COPD individuals, the study identified a Th1-dominant immune profile distinct from established COPD:

• Cellular Markers: Elevated neutrophil counts and percentages, and reduced lymphocyte percentages were observed across multiple quantiles (q=0.25, 0.50).
• Erythrocyte Disturbances: High-risk individuals showed reduced Hemoglobin (Hb), Mean Corpuscular Hemoglobin (MCH), and Mean Corpuscular Hemoglobin Concentration (MCHC), alongside elevated Red Blood Cell Distribution Width-Coefficient of Variation (RDW-CV). Notably, these erythrocyte abnormalities were more pronounced in the high-risk group than in established COPD, suggesting an inverse U-shaped trajectory.
• Cytokines: TNF-α was significantly elevated only at the lower quantile (q=0.25) in the pre-COPD group compared to healthy controls and COPD patients. In contrast, established COPD patients showed elevated IL-6 (q=0.50) and IL-8 (q=0.25).
• Correlation: In the smoking subgroup, cumulative smoking exposure (smoking index) was positively correlated with Th1-related cytokines (IL-6, TNF-α) but not with routine blood parameters, suggesting smoking specifically amplifies Th1 inflammatory responses.

Never-Smoking Subgroup (Th2-Skewed Phenotype)

Never-smoking pre-COPD individuals exhibited a Th2-skewed polarization:

• Cellular Markers: Unlike established COPD patients (who had significantly elevated eosinophils), pre-COPD high-risk individuals did not show significant eosinophil elevation.
• Erythrocyte Disturbances: A consistent and significant reduction in MCHC was observed across all three quantiles (q=0.25, 0.50, 0.75) in the high-risk group compared to healthy controls and COPD patients.
• Cytokines: IL-5 (a key Th2 cytokine) was significantly elevated only at the mid-quantile (q=0.50) in the pre-COPD group. Conversely, IL-12p70 and IFN-γ levels were reduced in both high-risk and COPD groups, indicating progressive Th1 suppression.

Established COPD Comparison

In established COPD, the distinct early-stage biomarkers (isolated low-quantile TNF-α in smokers, mid-quantile IL-5 in non-smokers) largely normalized or shifted. Smokers with COPD showed elevated IL-6/IL-8, while non-smokers showed increased eosinophils, suggesting that the early, transient immune signatures of pre-COPD evolve into different patterns as irreversible airflow limitation develops.

Significance and Claims

The authors claim that this study provides preliminary evidence that smoking status stratifies distinct Th1/Th2 phenotypes in pre-COPD high-risk individuals, a finding that is detectable only through quantile-specific analysis.

• Clinical Utility: The study proposes that routine blood parameters and inflammatory cytokines can serve as cost-effective, preliminary indicators for stratified screening in primary care. Specifically:
  • Smokers: May benefit from monitoring low-quantile TNF-α, neutrophil counts, and RDW-CV.
  • Never-Smokers: May require attention to mid-quantile IL-5 and universal MCHC reduction.
• Methodological Insight: The paper argues that quantile regression is a superior tool for identifying subtle, distribution-specific early abnormalities that conventional mean regression fails to capture.
• Limitations: The authors modestly frame these findings as exploratory. They acknowledge limitations including the small sample size, the cross-sectional nature (precluding causal inference), and the lack of airway-local validation (e.g., induced sputum). They emphasize that these biomarkers require validation in larger, prospective multicenter cohorts to confirm their predictive value for COPD progression and response to intervention.

In conclusion, the paper suggests that early immune dysregulation in pre-COPD is not uniform but is heavily dependent on smoking status, manifesting as Th1 dominance in smokers and Th2 skewing in non-smokers, with specific quantile-dependent biomarker signatures that could facilitate earlier, targeted intervention in primary care.