Genome-wide cfDNA Methylation Profiling of Pleural Effusion Reveals an Immuno-epigenetic Signature for Differentiating Malignant from Benign Cases.

BACKGROUND

Distinguishing malignant pleural effusion (MPE) from benign pleural effusion (BPE) remains clinically challenging because conventional cytology has limited sensitivity. We investigated whether cfDNA methylation profiling of pleural effusion supernatant could provide a non‑invasive diagnostic alternative.

METHODS

Patients with pleural effusion were consecutively and prospectively recruited from Beijing Chest Hospital between November 2022 and March 2024. We conducted genome-wide cfDNA methylation profiling on qualified pleural effusion samples. A mechanistically relevant diagnostic signature was developed using a knowledge-driven approach, integrating probes involving immune pathways and immune cell deconvolution data from the training set. The final signature's performance was subsequently tested in an internal validation cohort and benchmarked against conventional cytology.

RESULTS

A total of 101 participants were enrolled, including 45 MPE and 56 BPE according to the composite reference standard. We observed a distinct methylation landscape between MPE and BPE, identifying 3,119 hypermethylated and 232 hypomethylated DMPs in MPE that were enriched in immune‑related pathways (such as T cell activation). Deconvolution indicated significant differences in immune cell-derived cfDNA (including CD8+ T cells, B cells, eosinophils) between groups. A 7‑CpG-model based on LASSO classifier achieved an AUC of 0.980 (95% CI: 0.954-1.000) in the training set and 0.963 (95% CI: 0.896-1.000) in the internal validation set. The classifier substantially improved MPE detection in cytology‑negative or indeterminate cases and identified MPEs missed by conventional methods.

CONCLUSIONS

A 7‑CpG-model from pleural effusion cfDNA reliably discriminates MPE from BPE and shows promise as a reflex test for resolving cytologic uncertainty.