Resolving the Controversy on EGFR/KRAS Mutations in Pulmonary Squamous Cell Carcinoma Via Comprehensive Pathologic Assessment Incorporating Immunohistochemistry
Natasha Rekhtman, Paul K Paik, Maria E Arcila, Laura J Tafe, Geoffrey R Oxnard, Andre L Moreira, Travis D William, Maureen F Zakowski, Kris G Mark, Marc Ladanyi. Memorial Sloan-Kettering Cancer Center (MSKCC), New York, NY; Dartmouth Hitchcock Medical Center, Lebanon, NH; Dana-Farber Cancer Institute, Boston, MA
Background: There is a persistent controversy as to whether EGFR/KRAS mutations occur in pulmonary squamous cell carcinoma (SQCC). We hypothesized that the reported variability may reflect diagnoses rendered on poorly-sampled tumors and the inherent difficulties in morphologic distinction of poorly differentiated SQCC from adenocarcinoma (ADC). The recent development of a robust immunohistochemical (IHC) approach that aids in this critical distinction provides an opportunity to reassess EGFR/KRAS and other targetable kinase mutation frequencies in a pathologically homogeneous series of SQCC.
Design: Ninety-five resected SQCC verified by IHC as p63+/TTF-1- were tested for activating mutations in EGFR, KRAS, BRAF, PIK3CA, NRAS, AKT1, ERBB2/HER2, and MAP2K1/MEK1. In addition, all tissue samples from rare patients with the diagnosis of “SQCC” harboring EGFR/KRAS mutations encountered during 5 years of routine clinical testing at MSKCC were reassessed pathologically.
Results: The screen of 95 IHC-verified SQCC revealed no EGFR/KRAS mutations (0%; 95% CI 0-3.8%), but a low rate of PIK3CA (4%; 95% CI 1-10%) and AKT1 (1%; 95% CI 0-5.7%) mutations. Detailed morphologic and IHC reevaluation of EGFR/KRAS-mutant “SQCC” identified during routine clinical testing (n=16) resulted in reclassification of 10 (63%) cases as adenosquamous carcinoma (AD-SQC) and 5 (31%) cases as poorly-differentiated ADC morphologically mimicking SQCC (i.e. ADC with “squamoid” morphology). One (6%) case had no follow-up.
Conclusions: We conclude that EGFR/KRAS mutations do not occur in pure pulmonary SQCC, and samples diagnosed as “SQCC” harboring these mutations represent pitfalls in pathologic diagnosis of AD-SQC and ADC, which can largely be resolved by comprehensive pathologic assessment utilizing IHC. Our findings 1) highlight the value of IHC in the diagnosis of SQCC, which clarifies conflicting molecular data, 2) suggest a sharp biological divide in the patterns of oncogenic driver mutations between lung ADC (pure or combined) vs pure SQCC, and 3) establish the rate of several potentially targetable mutations in a pathologically homogeneous set of SQCC.
Monday, March 19, 2012 2:00 PM
Platform Session: Section D, Monday Afternoon