One sample fits all: a microfluidic-assisted methodology for label-free isolation of CTCs with downstream methylation analysis of cfDNA in lung cancer

Abstract

Lung cancer (LC) is a major cause of mortality. Late diagnosis, associated with limitations in tissue biopsies for adequate tumor characterization contribute to limited survival of lung cancer patients. Liquid biopsies have been introduced to improve tumor characetrization through the analysis of biomarkers, including circulating tumour cells (CTCs) and cell-free DNA (cfDNA). Considering their availability in blood, several enrichment strategies have been developed to augment circulating biomarkers for improving diagnostic, prognostic and treament efficacy assessment; often, however, only one biomarker is tested. In this work we developed and implemented a microfluidic chip for label-free enrichment of CTCs with a methodology for subsequent cfDNA analysis from the same cryopreserved sample. CTCs were successfully isolated in 38 of 42 LC patients with the microfluidic chip. CTCs frequency was significantly higher in LC patients with advanced disease. A cut-off of 1 CTC per mL was established for diagnosis (sensitivity = 76.19%, specificity = 100%) and in patients with late stage lung cancer, the presence of ≥5 CTCs per mL was significantly associated with shorter overall survival. MIR129-2me and ADCY4me panel of cfDNA methylation performed well for LC detection, whereas MIR129-2me combined with HOXA11me allowed for patient risk stratification. Analysis of combinations of biomarkers enabled the definition of panels for LC diagnosis and prognosis. Overall, this study demonstrates that multimodal analysis of tumour biomarkers via microfluidic devices may significantly improve LC characterization in cryopreserved samples, constituting a reliable source for continuous disease monitoring.