Correlation of Tumor Cell Percentage and Absolute Cell Count with EGFR and KRAS Analyses Using Cytological Specimens. A Retrospective Study of 118 Cases
Thomas J Gniadek, Delicia Munfus-McCray, Peter Illei, Ming-Tseh Lin, Michele Thiess, Edward Gabrielson, Frederic Askin, Qing K Li. Johns Hopkins University School of Medicine, Baltimore, MD
Background: Lung adenocarcinoma is the leading cause of cancer deaths in the United States. EGFR and KRAS mutational analysis is critical to guide treatment with tyrosine-kinase inhibitors. Fine-needle aspiration cytology is an established method for diagnosing and staging lung cancer. In this study, we correlated the EGFR/KRAS mutational status in primary and metastatic non-small cell lung cancer with the percentage and absolute number of tumor cells using cytological specimens.
Design: Using pathology archives from our academic center, 118 cases of cytology lung adenocarcinomas submitted for molecular tests were identified from July 2008 to Sept 2011. Cell block sections were reviewed by pathologists to confirm diagnoses. Formalin fixed paraffin embedded sections were used for EGFR and KRAS analysis. Mutational status was determined by sequencing exons 18 to 21 of EGFR and codons 12 and 13 of KRAS. Retrospective counting of tumor cells was done on the cytology H&E slides, using visual estimation and a computer-aided automated segmentation algorithm.
Results: Of the 118 cases, 97 had sufficient tumor material for EGFR and KRAS analysis and H+E slides. Of the 97 cases, 22 cases (22.7%) were primary lung lesions. Metastatic sites included lymph nodes (51/75, 68%), pleura (12/75, 16%), and others (12/75, 16%). EGFR and KRAS mutations were detected in 23/97 (23.7%) cases and 35/97 cases (36.1%), respectively. The correlation of percentage and absolute cell number count is summarized in the Table 1. The minimum number of tumor cells in cytological samples that had a KRAS analysis was approximately 100 cells, whereas, EGFR tests were done in specimens containing approximately 200 or more tumor cells.
Conclusions: In this case series, positive EGFR test results were generally detected in specimens with greater numbers of tumor cells than for positive KRAS test results. In addition to the variable sensitivities of analytic methods used for studying EGFR and KRAS mutations, it is well-known that both DNA quantity and quality are critical for molecular analysis. Our study suggests that the absolute tumor cell count (likely indicative of DNA quantity) is an important criterion, in addition to tumor cell percentage. Further study is needed to examine the quality of DNA in cytological material and correlation with test results.
|% Tumor||N Samples||Mean Cell Count||Range||Standard Error of the Mean||Pearson|