A model workflow for microfluidic enrichment and genetic analysis of circulating melanoma cells

Abstract

Circulating melanoma cells (CMCs) are responsible for the hematogenous spread of melanoma and, ultimately, metastasis. However, their study has been limited by the low abundance in patient blood and the heterogeneous expression of surface markers. The FDA-approved CellSearch platform enriches CD146-positive CMCs, whose number correlates with progression-free survival and overall survival. However, a single marker may not be sufficient to identify them all. The Parsortix system allows enrichment of CMCs based on their size and deformability, keeping them viable and suitable for downstream molecular analyses. In this study, we tested the strengths, weaknesses and potential convergences of both platforms to integrate the counting of CMCs with a protocol for their genetic analysis. Samples run on Parsortix were labeled with a customized melanoma antibody cocktail, which efficiently labeled and distinguished CMCs from endothelial cells/leukocytes. The capture rate of CellSearch and Parsortix was comparable for cell lines, but Parsortix had a higher capture rate in real-life samples. Moreover, double enrichment with both CellSearch and Parsortix succeeded in removing most of the leukocyte contamination, resulting in an almost pure CMC sample suitable for genetic analysis. In this regard, a proof-of-concept analysis of CMCs from a paradigmatic case of a metastatic uveal melanoma patient led to the identification of multiple genetic alterations. In particular, the GNAQ p.Q209L was identified as homozygous, while a deletion in BAP1 exon 9 was found hemizygous. Moreover, an isochromosome 8 and a homozygous deletion of the CDKN2A gene were detected. In conclusion, we have optimized an approach to successfully enrich and retrieve viable CMCs from metastatic melanoma patients. Moreover, this study provides proof-of-principle for the feasibility of a marker-agnostic CMC enrichment followed by CMC phenotypic identification and genetic analysis.Kindly check and confirm the processed contributed equally is correctly identify We confirm.

Keywords: CMC genetic analysis; CMC phenotypic analysis; Circulating melanoma cells (CMCs); Liquid biopsy; Melanoma; Parsortix PC1 system.

Fig. 1

Workflow for CMC enrichment and genetic characterization. Blood from melanoma patients was processed through the CellSearch platform and/or the Parsortix system equipped with a customized antibody cocktail. DNA derived from harvested CMCs, superamplified or not, was analyzed with a customized NGS panel, ddPCR and MLPA to identify tumor genetic alterations. Paired cfDNA was used to assess/validate the genetic alterations identified in CMCs. The figure was created with BioRender (https://biorender.com/).

Fig. 2

Representative images of live-cell immunofluorescence assays (A) and Parsortix experiments (B-D) of melanoma cells. (A-B) Melanoma cells (SK-MEL-28 or A375-MA2) were stained with antibodies against melanoma markers CD146/HMW-MAA/MART-1/GP100 (in green) and with DAPI for cell nuclei (in blue). (C) Bright-field images of cells captured in Parsortix cassette. (D) Endothelial cells (red arrows) can be discriminated from CMCs (white arrow) when positive for both CD146 (in green) and CD34 (in pink). Scale bar (lower right) represents 10 μm. Abbreviations: APC, allophycocyanin; FITC, fluorescein isothiocyanate; MEL, melanoma markers.

Fig. 3

Representative images of captured melanoma cells identified by Parsortix and CellSearch systems. (A) Melanoma cells (A375-MA2) were enriched and stained with Parsortix system using melanoma markers (in green, CD146/MART-1/HMW-MAA/GP100), leukocyte and endothelial markers (in pink, CD45/CD16/CD34), and DAPI for cell nuclei (in blue). Scale bar (lower right) represents 10 μm. (B) Melanoma cells (A375-MA2) enriched and stained with CellSearch platform. Cells were immunomagnetically captured by a ferrofluid antibody against CD146, and were stained with HMW-MAA melanoma marker (MEL-PE), CD45/CD34 (APC) leukocyte and endothelial markers, respectively, and DAPI for cell nuclei. Abbreviations: APC, allophycocyanin; FITC, fluorescein isothiocyanate; MEL-FITC, melanoma markers; PE, phycoerythrin.

Fig. 4

Evaluation of the performance of the Parsortix system. Comparison of the captured cell line melanoma cells (A-B) or CMCs (E-F) by Parsortix vs. CellSearch in terms of number (A/E) or capture rate (B/F). Graphs C and D display the harvest rate of melanoma and NSCLC cells from the Parsortix cassette without (C) and with (D) the supplement of 1.8% BSA. The performance of CellSearch was set as 100% and indicated by the dotted line. Experiments #1 - #23 refer to spiking experiments of SK-MEL-28, A375-MA2, and H441 cells into healthy donor blood (A-D); experiments #01 - #04 refer to four blood samples belonging to 3 metastatic melanoma patients (E-F). The graphs were performed using GraphPad version 8.0 for Windows (GraphPad Software Inc., San Diego, CA, USA).

Fig. 5

Representative images of CMCs from uveal and cutaneous melanoma patients captured with Parsortix system. CMCs derived from a uveal melanoma patient at the progression (A) and from a cutaneous melanoma patient at T0 (B). Cells were immunostained with melanoma markers (in green, CD146/MART-1/HMW-MAA/GP100), leukocyte/endothelial markers (in pink, CD45/CD16/CD34) and DAPI for cell nuclei (in blue). Scale bar (lower right) represents 10 μm. Abbreviations: APC, allophycocyanin; FITC, fluorescein isothiocyanate.