Establishing Quantitative Parameters in the Detection of Somatic Mutations of the Epidermal Growth Factor Receptor Gene in Cytology Samples of Non-Small Cell Lung Carcinomas
Elisa Brega, George Chong, Victor Cohen, Jason Agulnik, Goulnar Kasymjanova, Max Palayew, Bin Xu, David Small, Gerald Batist, Alan Spatz, Guilherme Brandao. Jewish General Hospital/McGill University, Montreal, QC, Canada
Background: 30-40% Asian and 10-15% Caucasians diagnosed with Non-Small Cell Lung Carcinoma have mutations in the tyrosine kinase domain of the Epidermal Growth Factor Receptor gene (EGFR). Two mutations, a short in-frame deletion in exon 19, and a point mutation in exon 21 account for 90% of EGFR mutations. Several NSCLC are diagnosed using only cytology, resulting in limited tumoral availability. This study goal is to establish minimal quantitative parameters in cytology for the detection of EGFR mutations.
Design: We prospectively analyzed 92 consecutive cases of non-small cell/non-squamous carcinoma (NSC/NSq) based on cytology and immunohistochemical profile. Exon 19 deletions and point mutation in exon 21 were tested using high resolution melting analysis and DNA sequencing. The samples include specimens from BAL/brush, effusions and trans-bronchial/thoracic/esophageal needle aspirations fixed in formalin 10%, centrifuged and aggregated with Histogel®, then classified as Q1 (<50 tumor cells/slide), Q2 (50-100 tumor cells/slide) and Q3 (>100 tumor cells/slide). DNA was extracted from 10x5µm-thick sections.
Results: Out of 92 NSC/NSq cases, 12 cases (13,04%) were positive for EGFR mutation (5 in exon 21, 7 in exon 19). Two cases had insufficient DNA for analysis, while in all others 78 cases no mutations were found. Among exon 19 positive cases, all but one were classified as Q3 (85.7%). One case positive for exon 19 was classified as Q1. Among exon 21 positive cases, 4 cases were classified as Q3 (80%), and only one as Q2 (20%). Among wild-type cases, 25 were classified as Q1 (27,1%), 14 as Q2 (15.2%) and 39 as Q3 (42,3%).
Conclusions: Only 1 out of 26 cases (3.84%) classified as Q1, and 1 out of 15 cases (6,66%) classified as Q2 had EGFR mutation detected. 49 cases were classified as Q3 (53.2% of cases). In this cohort, EGFR mutation was detected in 20.4% of cases. Our study suggests that despite being possible to detect EGFR mutations, less than 100 cells/slide appears to limit detection. Additional studies comparing detection in low cellular cytology samples with resected specimens could help establish minimal quantitative guidelines for optmization of EGFR mutation testing in cytology.
Tuesday, March 20, 2012 1:00 PM
Poster Session IV # 299, Tuesday Afternoon