Isolation of Circulating Tumor Cells from Patients with Metastatic Colorectal Cancer Using Label-Free Microfluidic Technology for Enumeration and Detection of KRAS, BRAF, PIK3CA Mutation

摘要

Background: Colorectal cancer (CRC) is the 3rd most common cancer diagnosed worldwide in both men and women. Only 39% of cancers are diagnosed at a localized stage, and 5-year survival rates decrease rapidly for patients with advanced and metastasized disease (stage III 61%, stage IV 8%). Better markers for detection of disease progression, therapeutic resistance and minimal residual disease are still needed. Liquid biopsies, such as Circulating Tumor Cells (CTCs) are emerging biomarkers shed by the tumor into the blood stream. The enumeration and mutational profiling of CTCs in a cancer patient's blood may be used to assess cancer prognosis, to evaluate treatment response, to guide tailored therapies as well as to detect cancer recurrence or minimal residual disease and also may be implemented for early cancer detection in the future. Method: For rapid label-free isolation of CTCs from peripheral blood we used the Vortex technology, a microfluidic device using inertia and laminar microvortices. From 15 patients with CRC metastatic to the liver that underwent liver metastatectomy with curative intent, we collected CTCs preoperatively, at the 5th postoperative day and during follow-up visits. Cells collected were immunostained for EpCAM, CD45, DAPI and enumerated using standardized classification criteria. For the mutation analysis, 24 CTC samples, 9 matched liver metastases and 3 primary FFPE tumor samples from 9 patients were analyzed. The hotspot regions of KRAS codons 12, 13, BRAF V600E, as well as PIK3CA codons 542, 545 (exon 9) and 1047 (exon 20) were amplified by PCR and subsequently subjected to Sanger sequencing at ELIM Biopharmaceuticals. CTC mutational patterns were compared to the ones in corresponding liver metastases tissue and when available, to the primary tumor. Results: 41 blood samples from 15 patients were collected at different time points prior to and after surgical resection of liver metastases. More CTCs were found in 41 patient samples (3.4 CTCs/mL, 0.1 - 29/mL) than in 10 samples from age-matched healthy donors (0.1 CTCs/mL, 0 - 0.4/mL). The purity obtained from the patient samples ranges from 0.3 to 63.5%, with 14.5% in average. 80% of all CRC samples were identified as positive for CTCs (based on a calculated threshold from healthy controls), with varying levels of EpCAM expression (81.4% of CTCs being EpCAM+). In 77.8% of our cohort (7 of 9 patients), we were able to detect at least one mutation. All 3 analyzed primary tumor samples showed the same mutation as the corresponding hepatic metastases. In 3 patients (P006, P009, and P013), our analysis revealed the presence of 2 different mutations concurrently (BRAF and PIK3CA in P006, or KRAS and PIK3CA in P009 and P013, respectively). There was a concordance of 77.8 % between hepatic metastatic tissue and CTC mutational analysis (7 of 9 patients, respectively). Conclusion: Vortex technology offers a fast and efficient label-free method to isolate CTCs with high purity and enable comparison of mutational profiles between CTCs and tumor tissue. This high purity enabled us to analyze the CTCs using Sanger sequencing on fixed and stained cells. This study confirms that Vortex technology is a reliable method to extract CTCs from CRC patient blood samples, and is compatible with downstream CTC immunostaining, enumeration, and detection of tumor mutations.