Detection of KRAS and BRAF Hotspot Mutations on Cytology Smears in Metastatic Colorectal Carcinomas Using Pyrosequencing
Ly Ma, Lorna Ogden, Jianli Dong, Vicki Schnadig, Mahmoud Eltorky, Fangling Xu, Gengming Huang, Ranjana Nawgiri. University of Texas Medical Branch, Galveston, TX
Background: Anti-EGFR therapy is hindered when KRAS or BRAF mutations are present. Currently, KRAS/BRAF testing is conducted on formalin-fixed paraffin-embedded (FFPE) tissue from primary colorectal cancer (CRC) which may not be readily available. In this study, we examined KRAS/BRAF mutations utilizing fine needle aspiration (FNA) smears of metastatic CRC as a more cost-effective, less invasive approach and investigated the KRAS/BRAF results between primary CRC and subsequent metastases.
Design: We received institutional funding to investigate 15 patients with CRC diagnosed on colectomy specimens and with subsequent metastases diagnosed on FNA cytology. All cases were reviewed by cyto- and surgical pathologists. Genomic DNA from marked areas was extracted and purified from 5-10 um FFPE sections and was extracted with and without purification from air-dried FNA smears. Qiagen® pyrosequencing assays were used to detect KRAS codons 12, 13, and 61 and BRAF codon 600 mutations.
Results: All primary specimens (100%) were successfully sequenced at all codons. KRAS mutations were detected in 6 of 15 (40%) primaries, and BRAF mutation in 2 of 15 (13%) cases. Sequencing was possible for all metastases (100%) but was dependent on extent of purification. Twelve of 15 (80%) FNA smears were successful using both crude and purified DNA. Two smear cases, 300 (BRAF only) and 500 cells (KRAS/BRAF), failed with crude DNA but were successful with purified DNA. The third case (400 cells) failed all assays with purified DNA but was successful with crude DNA. Discordance was demonstrated in one metastatic case (200 cells) which had wild-type KRAS, whereas the primary tumor had mutations in codons 13 (GGT>GAT) and 61 (CAA>CAT) at 26% and 21% mutant allele frequencies, respectively. No additional mutations were detected in the metastatic specimens.
Conclusions: Our KRAS and BRAF mutation rates in primary CRC specimens and our concordance rate were similar to the most recent literature. Of particular interest is the case that detected both codons 12 and 61 KRAS mutations in the primary tumor but showed wild-type results in FNA metastatic tumor cells. Importantly, when DNA was extracted by both crude extraction and further purification, KRAS and BRAF hotspot codons were successfully genotyped with few amplification inhibitors in all of our FNA specimens, supporting the use of FNA smear preparations when limitations of using primary CRC specimens exist.
Tuesday, March 5, 2013 1:00 PM
Poster Session IV # 65, Tuesday Afternoon