Technologies for the isolation of circulating tumor cells

Abstract

Metastasis, the spread and growth of tumor cells from the primary site to distant organs, is arguably the most devastating and deadly attribute of cancer, and is ultimately responsible for 90% of cancer-related deaths. Circulating tumor cells (CTCs) are exceedingly rare cells found in the whole blood of cancer patients which have the potential to serve as a 'blood biopsy'. The intricate characterization of these cells could result in an entire new class of therapies directly targeting metastasis. Present technologies enable only a susbset of potential analyses to be conducted, principally due to sub-optimal cell isolation sensitivity, purity, throughput, or handling method. Here, we present two novel technologies to address the challenge of CTC isolation. First, we build on affinity-based microfluidic cell capture platforms by developing sacrificial hydrogel coatings to enable the innocuous release of captured cells; we demonstrate that model CTCs captured from whole blood remain viable and proliferative following release and are compatible with downstream immunostaining and FISH analysis. Second, we present a novel cell sorting system that interrogates over 10 million individual events each second, resulting in a high throughput, ultra-efficient rare cell sorter that delivers enriched cells in a vial, readily compatible with virtually any downstream assay. This is the first system combining the high sensitivity and single cell resolution that is characteristic of FACS with the practicality of MACS at a throughput and specificity afforded by inertial focusing, enabling operation in both 'positive selection' and 'negative depletion' modes. We find greater than 90% cell isolation efficiencies with over 2.5 log depletion of contaminating WBCs. Furthermore, the system is applied to clinical patient samples, and proof-of-concept is demonstrated in a cohort of breast, lung and prostate patients. Working in a negative depletion mode to isolate target cells in an unbiased fashion, we used the system to assess single putative CTCs isolated from an endogenous pancreatic mouse model for gene expression of tumor markers. Initial data confirms CTC heterogeneity at the single cell level, and positions us to move forward with single cell transcriptome sequencing, which may reveal a broad array of CTC phenotypes including metastatic precursors.