KRAS Mutation Analysis in Colonic Adenocarcinomas: Correlation between Biopsy and Resection Specimens
JF Schmidt, G Soucy, RD Odze, L Dejesa-Jamanila, K Arnold, Q Yang, C Kuslich, RH Lash. Caris Diagnostics, Irving, TX; Brigham & Women's Hospital, Boston, MA; Caris Diagnostics, Phoenix, AZ
Background: The KRAS gene is mutated in 30-40% of colorectal adenocarcinomas (CRCs) resulting in a constitutively active protein. Recent clinical trials that demonstrated EGFR inhibitor resistance in CRCs with activating KRAS mutations led to recommendations for KRAS mutation analysis in the pretreatment workup of all patients with stage IV CRC. Unfortunately, in some circumstances only biopsy material is available for analysis prior to treatment. The aim of this study was to determine the correlation between biopsy and resection tissue regarding the presence and type of KRAS mutations analyzed by sequencing.
Design: Pretreatment biopsy and resection specimens from 30 CRC patients were selected (1 block from biopsies and 2 tumor blocks from resections). For each tissue block, the areas with the highest percentage of viable tumor nuclei were chosen for molecular analysis. In the biopsy specimens with sufficient separately identifiable intramucosal carcinoma, the intramucosal and invasive areas were analyzed separately (12 patients). The selected areas were macrodissected, DNA was isolated, and KRAS mutations in codons 12, 13, and 61 were identified by Sanger sequencing using primers flanking the codons of interest.
Results: Overall, KRAS mutations were identified in 11/30 (37%) of the tumors. A 100% correlation was noted regarding the presence/absence of KRAS mutations between biopsy and resection specimens (both blocks) for all 30 patients. The same point mutation was identified in biopsy and resection specimens in 11/11 (100%); interestingly, in 2 cases, 2 different point mutations were identified within the biopsy, 1 mutation in the invasive component and a different mutation in the intramucosal component.
Conclusions: Our finding of perfect correlation between KRAS mutation status in biopsy and resection specimens from individual patients suggests that biopsy material is sufficient for clinical KRAS mutation analysis. Specific KRAS point mutation differences can be identified within CRC biopsies.
Tuesday, March 23, 2010 1:00 PM
Poster Session IV # 79, Tuesday Afternoon