# Using gene methylation detection in bronchoscopic samples combined with morphology-based pathology for lung cancer diagnosis

## TL;DR

Combining gene methylation detection with traditional pathology in bronchoscopic samples improves lung cancer diagnosis accuracy.

## Contribution

Demonstrates that combining SHOX2 and RASSF1A methylation detection with morphology-based pathology significantly increases diagnostic sensitivity for lung cancer.

## Key findings

- Combined methylation detection in alveolar lavage fluid achieved 94.1% sensitivity, significantly higher than morphology-based pathology.
- The AUC values for combined detection were 0.9431 for alveolar lavage and 0.9828 for TBNA samples, showing strong diagnostic performance.
- Strong methylation detection results had a high positive predictive value for lung cancer.

## Abstract

Lung cancer is a common malignant tumor with high incidence and mortality. Bronchoscopy is key for its diagnosis, but traditional morphology-based pathology has limitations like missed diagnoses. This study explored the auxiliary diagnostic value of gene methylation detection combined with morphology-based pathology in bronchoscopic samples.

A retrospective analysis was conducted on 654 patients who underwent bronchoscopy at Zhangjiakou First Hospital between March 2023 and May 2024. The cohort included 571 alveolar lavage fluid samples and 83 transbronchial needle aspiration (TBNA) samples. Methylation-specific PCR (MS-PCR) was employed to detect the methylation status of the SHOX2 and RASSF1A genes in these samples. The diagnostic value was evaluated by comparing the results with the final clinical diagnosis.

For alveolar lavage fluid samples, the sensitivity of combined SHOX2 and RASSF1A methylation detection reached 94.1%, which was significantly higher than the 30.0% sensitivity of morphology-based pathology (P < 0.001). For TBNA needle tract washing fluid samples, the sensitivity of combined methylation detection was 94.8%, higher than the 86.2% sensitivity of morphology-based pathology, though the difference was not statistically significant (P = 0.075). When gene methylation detection was combined with morphology-based pathology, the area under the curve (AUC) values were 0.9431 for alveolar lavage fluid samples and 0.9828 for TBNA samples, indicating favorable diagnostic performance. Additionally, strong positive results from methylation detection exhibited a high positive predictive value for lung cancer.

Gene methylation detection in alveolar lavage fluid and TBNA needle tract washing fluid, when combined with morphology-based pathology, can enhance the diagnostic efficiency of lung cancer and is worthy of further research.