Cytology Samples Are Comparable to Histological Samples for EGFR Mutation Testing in Non-Small Cell Lung Cancer (NSCLC)
Ian Bosdet, Rola H Ali, Sean Young, Aly Karsan, Diana N Ionescu. BC Cancer Agency, Vancouver, BC, Canada
Background: Bronchial washings/brushings and fine-needle aspirations are often used for early screening and cytological diagnosis of lung cancer. The reliability of the different cytology samples for the molecular testing of NSCLC is controversial. Some authorities prefer to restrict molecular testing to histology samples. However, a large number of the diagnostic procedures for NSCLC are performed in community hospitals. In this study we discuss the feasibility of EGFR testing on cytology samples with available cell blocks (CB), as compared to core biopsy samples and resection specimens.
Design: 586 advanced stage NSCLC samples with non-squamous histology were referred by Medical Oncologists for EGFR mutation testing at BCCA Clinical Cancer Genetics Lab within a 14-month period. The formalin fixed paraffin embedded tissue samples were analyzed specifically for the in-frame deletions in exon 19 and the point mutation (L858R) in exon 21 and extensive data on sample characteristics were collected.
Results: Samples were received from 23 centers: 316 samples were from the lung, 78 from lymph nodes and 151 from metastatic sites, most commonly pleura (60), bone (35) and brain (25). A total of 115 cytology CB, 323 biopsy samples and 108 resection samples were analyzed over a 14 months period with an average intralaboratory turn around time of 7.4 calendar days. We analyzed 483 adenocarcinomas and 58 NSCLC not otherwise specified (NOS). Prior to analysis each sample was assessed by a pathologist for adequacy and tumor content and 38 test requests were cancelled (5.11%). Of the 546 tested cases 70 showed exon 19 deletion, 39 exon 21 mutation, 399 were wild type and 37 failed. The EGFR mutation rate was 21.45%. The average sample tumor cellularity and median DNA content for cytology samples (N=112) compared to biopsies (N= 301) and resections (N= 99) were 60%, 76% and 81% and 2.5 ug, 4.6 ug and 34ug. Only 8 of 37 failed tests were cytology samples. Reasons for test failure included low cellularity (16), decalcification (5) and fixation (2). The failure rate for cytology samples (8/115) was almost identical to that of histology samples (29/431) at 6.95% and 6.72% respectively.
Conclusions: Cytology samples are comparable to histology samples in terms of DNA quantity and rate of failure. Therefore, material from cell blocks is suitable for EGFR mutation testing and cytology samples should be preserved as cell blocks whenever possible.
Tuesday, March 20, 2012 1:00 PM
Poster Session IV # 300, Tuesday Afternoon