Extreme assay sensitivity in molecular diagnostics further unveils intratumour heterogeneity in metastatic colorectal cancer as well as artifactual low-frequency mutations in the KRAS gene

Sara Mariani1, Luca Bertero1, Simona Osella-Abate1, Cristiana Di Bello2, Paola Francia di Celle3, Vittoria Coppola1, Anna Sapino1, Paola Cassoni1,3,4 and Caterina Marchiò1,3,4

1Department of Medical Sciences, University of Turin, via Santena 7, Turin 10126, Italy
2Department of Molecular Biotechnology and Health Sciences, University of Turin, via Nizza 52, Turin 10126, Italy
3Pathology Unit, Azienda Ospedaliera Universitaria Città della Salute e della Scienza, via Santena 7, Turin 10126, Italy
4These authors contributed equally to this work.

Abstract:

Background:  Gene mutations in the RAS family rule out metastatic colorectal carcinomas (mCRCs) from anti-EGFR therapies.

Methods:  We report a retrospective analysis by Sequenom Massarray and fast COLD-PCR followed by Sanger sequencing on 240 mCRCs.

Results:  By Sequenom, KRAS and NRAS exons 2-3-4 were mutated in 52.9% (127/240) of tumours, while BRAF codon 600 mutations reached 5% (12/240). Fast COLD-PCR found extra mutations at KRAS exon 2 in 15/166 (9%) of samples, previously diagnosed by Sequenom as wild-type or mutated at RAS (exons 3-4) or BRAF genes. After UDG digestion results were reproduced in 2/12 analysable subclonally mutated samples leading to a frequency of true subclonal KRAS mutations of 1.2% (2.1% of the previous Sequenom wild-type subgroup). In 10 out of 12 samples, the subclonal KRAS mutations disappeared (9 out of 12) or turned to a different sequence variant (1 out of 12).

Conclusions:  mCRC can harbour coexisting multiple gene mutations. High sensitivity assays allow the detection of a small subset of patients harbouring true subclonal KRAS mutations. However, DNA changes with mutant allele frequencies <3% detected in formalin-fixed paraffin-embedded samples may be artifactual in a non-negligible fraction of cases. UDG pre-treatment of DNA is mandatory to identify true DNA changes in archival samples and avoid misinterpretation due to artifacts.

Read the complete pubblication at this link