Combination of CT and Plasma Long fragment cfDNA concentration improves diagnosis of pulmonary nodules

Technical Summary: Combination of CT and Plasma Long-Fragment cfDNA Concentration Improves Diagnosis of Pulmonary Nodules

Problem Statement
Lung cancer remains the leading cause of cancer-related mortality globally, with early diagnosis being critical for prognosis. While Low-dose Computed Tomography (LDCT) has improved the detection of pulmonary nodules (PNs), it suffers from a high false-positive rate (approximately 96.4%), particularly regarding indeterminate pulmonary nodules (IPNs). Current imaging systems, such as Lung-RADS, have limitations in specificity and can lead to unnecessary follow-ups, radiation exposure, and patient anxiety. Although circulating cell-free DNA (cfDNA) has emerged as a promising non-invasive liquid biopsy marker, previous research has largely focused on total cfDNA levels or specific mutations/methylation. There is a lack of systematic research regarding long-fragment cfDNA concentration as a specific biomarker for differentiating benign from malignant PNs, especially in the context of multimodal diagnostic models that integrate imaging and molecular data.

Methodology
This study enrolled 154 participants from January to December 2020, categorized into four groups: 15 healthy controls, 19 patients with benign PNs, 81 patients with indeterminate pulmonary nodules (IPNs), and 39 patients with lung cancer.

• Sample Collection & Processing: Venous blood was collected in EDTA tubes. Plasma was separated via double centrifugation (1600g and 16000g) and stored at −80°C. cfDNA was extracted from 200 μL of plasma using QIAamp DNA Blood Mini Kits.
• Quantification: Quantitative PCR (qPCR) was performed on a LightCycler LC480 machine. The study targeted LINE-1 repetitive elements to measure:
  • Short-fragment concentration: Serving as the proxy for total cfDNA concentration.
  • Long-fragment concentration: Defined by specific primers amplifying larger DNA fragments.
  • cfDNA Integrity (cfDI): Calculated as the ratio of long-fragment to short-fragment concentrations.
• Statistical Analysis: Non-parametric Kruskal-Wallis tests were used for group comparisons due to non-normal data distribution. Pearson correlation analysis examined relationships between cfDNA markers and clinical features (age, nodule diameter). A stratified binary logistic regression model was constructed to evaluate independent predictors of malignancy, incorporating demographic data, imaging features (nodule diameter, Lung-RADS classification), and cfDNA biomarkers. Diagnostic efficacy was evaluated using Receiver Operating Characteristic (ROC) curves and Area Under the Curve (AUC) analysis.

Key Results

• cfDNA Distribution: Contrary to the assumption that malignancy always yields the highest cfDNA, the benign nodule group exhibited significantly higher total cfDNA and long-fragment cfDNA concentrations compared to healthy controls and IPNs. However, the lung cancer group showed significantly higher long-fragment concentrations compared to the non-cancer group (healthy controls + IPNs).
• Correlations: Long-fragment cfDNA concentration was strongly correlated with cfDI (r=0.77, p<0.001). Age showed a weak positive correlation with both maximum nodule diameter and total cfDNA concentration.
• Diagnostic Thresholds:
  • The optimal diagnostic range for long-fragment cfDNA concentration was identified as 1.27–3.79 ng/ml. Within this range, specificity reached 90.43%, though sensitivity was 41.03% (AUC = 0.712 for long-fragment alone).
  • Imaging indicators alone showed good performance: Maximum nodule diameter (AUC = 0.835) and Lung-RADS classification (AUC = 0.831).
• Multimodal Model: The integration of long-fragment cfDNA concentration, maximum nodule diameter, and Lung-RADS classification into a single model yielded an AUC of 0.911 (95% CI: 0.864–0.958), with a sensitivity of 87.18% and specificity of 85.22%. This significantly outperformed single indicators.

Significance and Claims
The authors claim that this study provides direct comparative evidence across four clinical groups, challenging the simplistic assumption that "higher cfDNA equals cancer." They posit that benign inflammatory and necrotic processes can drive total cfDNA release more than early-stage malignancies, making total cfDNA a potential "rule-in" marker for benign disease rather than a standalone cancer detector.

The primary significance of the work lies in the identification of a specific concentration range (1.27–3.79 ng/ml) for long-fragment cfDNA that offers high specificity for distinguishing cancer from non-cancer states, addressing the biological heterogeneity of cfDNA dynamics. The study concludes that a multimodal diagnostic model combining long-fragment cfDNA concentration with standard imaging indicators (nodule diameter and Lung-RADS) markedly improves early lung cancer diagnostic accuracy. This approach offers a reliable basis for clinical application to reduce false positives and unnecessary invasive follow-ups for indeterminate nodules. The authors emphasize that while the results are promising, future validation in large, multi-center prospective cohorts is necessary before widespread clinical translation.