High Throughput Isolation and Analysis of Circulating Tumor Cells for Monitoring Cancer

Abstract

Although tissue biopsies are an excellent diagnostic and prognostic tool, they are highly invasive and therefore are performed with cause. “Liquid biopsies” are a possible alternative to traditional tissue biopsies that are less invasive and lower risk, so they can be performed more routinely. Liquid biopsies allow for the detection and analysis of circulating biomarkers such as circulating tumor cells (CTCs). Although prognostically informative, CTCs are extremely rare with around 10 CTCs per mL of blood compared to 106 white blood cells in the same blood volume. Isolating and analyzing a larger number of CTCs will allow for more informative analysis to be conducted.

To isolate a larger number of CTCs, a high throughput continuous wearable system was developed. This system uses an inertial microfluidic device, the CTCKey™, in series with three herringbone graphene oxide (HBGO) devices. The inertial device enriches CTCs 5-fold at 2.4 mL/min from whole blood based on cell size, then directs the CTC enriched streams to HBGO devices where CTCs are captured on the chips surface. After being processed through the system, the blood is then returned to the patient, increasing the volume of blood that can be processed from approximately 10 mL using traditional blood draws to 240 mL over a two-hour period. The CTCKey™ was able to recover 99% of MCF7s and CellSearch post CTCKey™ recovered 71% of MCF7s.

Another part of this work was to isolate CTCs from hepatocellular carcinoma patients before, during, and after radiation treatment from 29 unique patients. CTCs were enumerated based on the expression of Ck18, asialoglycoprotein receptor, and EpCAM. The differences in number of CTCs and different sub-populations of CTCs between patients and time points were compared. It was determined that an increase in CTC numbers before and during treatment was highly prognostic of disease progression (p=0.0173) while that same trend did not hold for changes in CTC numbers before and after treatment. RNA analysis was also performed on these patients using microarrays. A small number of genes were identified as being differentially regulated in progressed patients when compared with stable patients. This data shows promise in predicting HCC patient outcomes, but it would be beneficial to verify the findings in a larger patient cohort.

In addition to characterizing bulk CTCs, analysis from individual CTCs at the single cell level could provide information about tumor heterogeneity that is missed in bulk analysis. The workflow used to isolate cells from HCC patient samples was modified to prepare samples for the DEPArray, a single cell isolation technology. After process modification, single CTCs were able to be isolated from samples processed using the Labyrinth then fixed using either PFA or alcohol fixation. It was demonstrated that the optimized methods yielded single cell DNA/RNA and the samples have been successfully prepared for copy number variation analysis; however, the developed approach can enable other analysis such as transcriptomic expression.

Overall, innovative microfluidic technologies developed in the thesis work provides the ability to isolate more CTCs through a high throughput wearable system which allows for large blood volumes to be processed. Downstream analysis of HCC samples showed that changes in CTCs are correlated with patient outcomes. Now that single cells can be isolated, heterogeneity between single CTCs can be determined. Collectively, these advancements in CTC isolation and analysis will lead to improved patient outcomes.