Enumeration and targeted analysis of KRAS, BRAF and PIK3CA mutations in CTCs captured by a label-free platform: Comparison to ctDNA and tissue in metastatic colorectal cancer

Abstract

Treatment of advanced colorectal cancer (CRC) requires multimodal therapeutic approaches and need for monitoring tumor plasticity. Liquid biopsy biomarkers, including CTCs and ctDNA, hold promise for evaluating treatment response in real-time and guiding therapeutic modifications. From 15 patients with advanced CRC undergoing liver metastasectomy with curative intent, we collected 41 blood samples at different time points before and after surgery for CTC isolation and quantification using label-free Vortex technology. For mutational profiling, KRAS, BRAF, and PIK3CA hotspot mutations were analyzed in CTCs and ctDNA from 23 samples, nine matched liver metastases and three primary tumor samples. Mutational patterns were compared. 80% of patient blood samples were positive for CTCs, using a healthy baseline value as threshold (0.4 CTCs/mL), and 81.4% of captured cells were EpCAM+ CTCs. At least one mutation was detected in 78% of our blood samples. Among 23 matched CTC and ctDNA samples, we found a concordance of 78.2% for KRAS, 73.9% for BRAF and 91.3% for PIK3CA mutations. In several cases, CTCs exhibited a mutation that was not detected in ctDNA, and vice versa. Complementary assessment of both CTCs and ctDNA appears advantageous to assess dynamic tumor profiles.

Keywords: Vortex; circulating tumor DNA; circulating tumor cells; colorectal cancer; liquid biopsy.

Figure 1. Study Workflow and aims

A. Blood was collected from 15 patients with colorectal cancer (CRC) metastatic to the liver (2 tubes from each patient at several time points, respectively). B. One tube was used for CTC analysis: CTCs were enriched using the Vortex Gen1 Chip (Vortex BioSciences). Collected CTCs were fixed, stained and enumerated. Then, DNA was extracted for mutational analysis of CTC DNA by PCR based Sanger sequencing (3 genes). C. Another tube of whole blood was used for analysis of ctDNA: After centrifugation, plasma was collected and DNA was extracted for mutational analysis of plasma circulating tumor DNA (ctDNA). ctDNA mutational analysis was performed by panel sequencing (4 genes) after prior enrichment of targeted mutants using the SCODA mutation enrichment technology. D. While standard treatments are currently used for certain patient cohorts, E. our aim is to promote tailored cancer patient treatment.

Figure 2. Microfluidic device design and performance

A. Blood sample processing consists of the following steps: 1) Priming of the device with wash solution (PBS) to eliminate air bubbles. At high flow rates (4 mL/min), laminar microvortices develop in the rectangular cavities. 2) Sample Infusion: The larger cancer cells get trapped in the vortices, while smaller red blood cells (RBCs) and white blood cells (WBCs) either pass through or transiently enter vortices. 3) Wash by switching to the wash solution at the same flow rate, thereby removing RBCs and WBCs, while the CTCs remain in the vortices. 4) Release of captured cells by lowering the flow rate to dissipate the vortices. Cells are collected into wells of a 96-well plate for further downstream analysis. B. The device consists of molded PDMS bonded to glass and contains one wash inlet, one sample inlet, and one outlet. C. Capture Purity in blood is evaluated by spiking ~500 HCT116 cells into 10x-diluted blood collected from a healthy donor. After processing through Vortex device, cells were stained with DAPI, EpCAM and CD45, respectively, for visualization of the nucleus and to identify blood cells. Average capture purity (i.e. percentage of contamination with WBC) is shown for cell spiking into blood.

Figure 3. CTC immunostaining and enumeration

A. Representative images of CTCs and WBCs stained with EpCAM, CD45, and DAPI. Scale bar represents 20 um. B. The number of CTCs per ml blood was determined for each patient. The number of CTCs in the preoperative blood draws are depicted by green circles for patients that received or did not receive neoadjuvant treatment. The number of CTCs in samples from age-matched healthy donors (HD) are depicted by red triangles. CTC numbers were higher in CRC patients compared to HD. A threshold was set at 0.4 CTCs per ml blood (mean HD + 2 standard deviations). C. The number of EpCAM⁺/DAPI⁺/CD45^- CTCs compared to the number of EpCAM^-/DAPI⁺/CD45^- CTCs is shown for all patients at different time points (pre-, postoperatively and during follow-up visits) by green and blue stacked bars, respectively. Orange dotted lines represent the threshold of 0.4 CTCs/ml, above which positive cells are counted as CTCs.

Figure 4. Longitudinal CTC enumeration results

Graphs A-I showthe evolution of CTC numbers during the clinical course and treatment of a subset of CRC patients with hepatic metastases. Orange colored bars represent Chemotherapy. Black stars behind patient numbers indicate status post neoadjuvant therapy prior to resection of liver metastases. Blue arrows depict computer-tomography scans. Orange dotted lines represent the HD threshold of 0.4 CTCs/ml, above which positive cells are counted as CTCs. NED = no evidence of disease. PD = progressive disease.

Figure 5. Longitudinal CTC enumeration results in comparison to ctDNA and CEA levels

Graphs A and B showthe development of CTC numbers alongside ctDNA and CEA levels during the clinical course and treatment of two CRC patients with hepatic metastases. Orange colored bars represent chemotherapy. Black stars behind patient numbers indicate status post neoadjuvant therapy prior to surgical resection of liver metastases. Blue arrows depict computer-tomography scans. Orange dotted lines represent the HD threshold of 0.4 CTCs/ml, above which positive cells are counted as CTCs. NED = no evidence of disease. PD = progressive disease.