Enhancing detection of circulating tumor cells with activating KRAS oncogene in patients with colorectal cancer by weighted chemiluminescent membrane array method.

Background: In 2008, the National Comprehensive Cancer Network suggested conducting a KRAS mutations test in metastatic colorectal cancer (mCRC) patients prior to administering therapy that uses anti-epidermal growth factor receptor (EGFR) monoclonal antibody. However, tests of KRAS mutations have been limited when traditional molecular techniques, such as polymerase chain reaction (PCR) combined direct sequencing, are used to obtain and analyze patients' cancer tissues. If the primary tumor or metastatic tissues of patients with mCRC is unavailable, then such analysis will not be feasible. Our laboratory has successfully established a colorimetric membrane array analysis platform that could detect activating KRAS mutations from the peripheral blood of patients with various malignancies.

Methods: The current research aims to improve the above-mentioned technique not only by using chemiluminescence detection to replace color development, but also to add scores weighted according to the relevance of each gene to activating KRAS mutations.

Results: Our results show that the described weighted chemiluminescent membrane array (WCHMA) can detect circulating tumor cells (CTCs) harboring activating KRAS mutations in the peripheral blood in CRC. The sensitivity, specificity, and accuracy were 90.2, 94.9, and 93.5%, respectively, and the detection limitation was three colon tumor cells per millimeter of blood. The current study would significantly improve the detection sensitivity and accuracy over that of our previously designed membrane array method.

Conclusions: These findings also highlight the need to prompt further prospective studies on more cases of CRC to further establish the clinical relevance of activating KRAS mutation detection from peripheral blood in anti- EGFR-based chemotherapy that uses activating KRAS detection chips and the WCHMA analysis method.