Bronchoalveolar lavage fluid EGFR genotyping during bronchoscopic biopsy: concordance with tissue and additional molecular diagnostic yield in lung cancer.

BACKGROUND

Tissue-based EGFR genotyping during the initial diagnostic work-up for suspected lung cancer is often limited by inadequate bronchoscopic biopsy material, and the clinical utility of bronchoalveolar lavage fluid (BALF) EGFR genotyping remains unclear. This study aimed to assess concordance between tissue and BALF EGFR genotyping and to determine whether BALF EGFR testing provides additional molecular diagnostic yield beyond tissue-based testing.

METHODS

In this prospective single-center study, adults with suspected lung cancer underwent bronchoscopic biopsy with BALF collection for extracellular vesicle-derived DNA EGFR testing. The primary endpoint was concordance between tissue and composite BALF EGFR genotyping in pathologically confirmed lung adenocarcinoma. Secondary endpoints included positive percent agreement (PPA), negative percent agreement (NPA), and additional molecular detection yield. An exploratory stage-based concordance analysis compared early-stage (stage I-II) with advanced-stage (stage III-IV) disease.

RESULTS

Among 54 patients with histopathologically confirmed lung adenocarcinoma, 49 had evaluable tissue and composite BALF EGFR results and were included in the primary analysis. The overall concordance rate was 89.8% [44/49; 95% confidence interval (CI), 78.2-95.6%], with a PPA of 83.3% (20/24; 95% CI, 64.1-93.3%) and an NPA of 96.0% (24/25; 95% CI, 80.5-99.3%). Variant-level agreement among double-positive cases was 100.0% (20/20). In an exploratory subgroup analysis, concordance was similar between early-stage (stage I-II) and advanced-stage (stage III-IV) disease (89.2% vs. 91.7%; P>0.99). Additional molecular detection yield beyond tissue-based testing was observed in 3 of 30 patients (10.0%), including one case of genotyping rescue among tissue wild-type cases and two cases of EGFR mutation detection among patients with unavailable tissue genotyping results. Notably, both of the latter patients had non-diagnostic initial bronchoscopic biopsy results and were subsequently confirmed as having lung adenocarcinoma.

CONCLUSIONS

BALF EGFR genotyping showed high concordance with tissue EGFR results and provided additional molecular detection yield. BALF EGFR genotyping may complement tissue biopsy and improve molecular diagnostic yield in suspected lung cancer.