Are Cytology Blocks Adequate for EGFR Mutational Testing? An Institutional Experience
DL Aisner, C Deshpande, Z Baloch, CD Watt, LA Litzky, AR Sepulveda, C Langer, T Evans, VM Van Deerlin. Univ of Pennsylvania, Philadelphia, PA; Univ of Pennsylvania, Philadelphia
Background: EGFR mutation status has been shown to predict response to anti-EGFR tyrosine kinase inhibitors in non-small cell lung cancer (NSCLC). In patients with advanced stage NSCLC, surgical resection is not part of routine care, therefore evaluation of mutational status is increasingly requested on biopsy or fine needle aspiration (FNA) specimens, in which the available material is limited. There are limited data on the suitability of cytology cell blocks for EGFR mutation testing. In this study we report our institutional experience with cytology cell block material for EGFR mutation testing.
Design: We retrospectively reviewed EGFR mutation analyses performed on 135 surgical (SP) and cytology cell blocks (CB) from Oct. 2007-Sept. 2009. One hundred fifteen (115) SP and 20 CB specimens were evaluated for L858R and exon 19 in-frame deletions (analytic sensitivity ∼5%). Cytology cell block specimens were evaluated for overall specimen size (total cellularity) and percent tumor. Percent tumor was scored as <5%, 6-10%, 11-25%, 26-50%, >50%. Immunohistochemistry for TTF-1 and CK7 were used to assist in assessment of tumor percentage, when available. Demographic features such as gender and smoking status were evaluated, as EGFR mutations are more frequently seen in women and non-smokers.
Results: Of the 115 SP and 20 CB specimens, 19 (16.5%) and 7 (35%) were positive for EGFR mutation, respectively. The mutation rates were not statistically different between the surgical and cytologic specimens (p=0.065). Of the 20 CB, half were <2mm2; of the 7 cases with a mutation, 4 (57%) were <2mm2. Limited DNA (<25ng/uL) was obtained from 70% (14/20) of CB specimens, including 71% (5/7) of those which were mutation positive; additionally, 57% (4/7) of the positive FNA specimens had extremely low DNA yields (<6.25ng/uL). 20% (4/20) of all FNA specimens had <10% tumor, however all 7 of mutation positive cases had >10% tumor. There were no differences between the SP and CB specimens with regard to patient gender or smoking status (p=0.31 and p=0.27 respectively).
Conclusions: Targeted mutational testing was successfully performed in CB specimens, even when scant. These data indicate that CB specimens provide an alternative source for molecular evaluation of NSCLC, and that tumor percentage may be more important than DNA yield in determining the suitability of these specimens for testing.
Monday, March 22, 2010 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 229, Monday Morning