Optimizing Endobronchial Ultrasound for Molecular Analysis. How Many Passes Are Needed?

Abstract

The current oncologic management of non-small cell lung cancer (NSCLC) requires pathologic differentiation between adenocarcinoma and squamous cell carcinoma. Furthermore, novel therapies for adenocarcinoma are clinically available for specific mutation profiles. Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has been shown to adequately obtain specimens for molecular profiling. However, it remains unclear what quantity of specimens is needed to provide suitable mutational genotyping for adenocarcinoma. The objective of this study was to determine the optimal number of aspirations per EBUS-TBNA procedure required in the presence of rapid on-site cytopathology evaluation (ROSE) for maximal diagnostic yield for molecular mutational analysis. From March 2010 to February 2012, cytopathologic data were collected from consecutive cases of adenocarcinoma or NSCLC not otherwise specified (NSCLC-NOS), diagnosed by EBUS-TBNA and ROSE. Samples of material obtained were air-dried and wet-fixed. Samples were tested for the KRAS, EGFR, and/or ALK mutations. Eighty-five patients who underwent EBUS-TBNA and were diagnosed with adenocarcinoma or NSCLC-NOS were identified. Of the 85 cases identified, 77 (90.6%) were classified as adenocarcinoma with the remaining 8 (9.4%) classified as NSCLC-NOS. Eighty-one of 85 (95.3%) were found to be adequate for molecular profiling. The median number of sites sampled was one. A median of four passes was needed to obtain adequate molecular profiling of 95.3%, using EBUS in conjunction with ROSE. With the use of EBUS-TBNA and ROSE, a minimum of four needle passes may provide an adequate amount of specimen for advanced molecular marker analysis.