Fig 1.
Pre-clinical characterization of the viral probe’s efficacy in melanoma cells.
(A) Melanoma and glioma cells (Mel624 and U251, respectively, are shown in the top and bottom rows as representative examples) were incubated with the probe, followed by immunofluorescence staining for Melan-A. Robust GFP expression indicated the efficacy of the probe for melanoma was comparable to that for glioma. The identity of melanoma cells is confirmed by Melan-A expression that coincided with GFP. Each row shows the same cells, with DAPI as a nuclear stain in the left-most panel, and the rightmost panel (Merge) show merged images of all three fluorescent channels. Bar, 30 um. (B) Flow cytometry analysis of BRAF WT (MeWo) and BRAF mutated (Mel624) cells indicated that for both cell lines, GFP signal peaked after 48 hours of exposure to the viral probe, with no further increase after 72 hours.
Fig 2.
Integration of the viral probe with a semi-automated, computer-driven image acquisition and analysis system.
(A) Melanoma cells (A375P cells shown here as a representative example) incubated with the probe for 24 hours were visualized under fluorescent microscopy. Tiled images were taken for each well. Hoechst dye was added before image acquisition to delineate nucleated cells, which take up the dye and emit blue fluorescent signal under UV light. Bar, 30um. (B) Scatter plots showing individual cells, plotting size (X-axis) and fluorescent intensity (standard deviations (SD) above background) (Y-axis) identified by the imaging program. Control blood studies from a single subject with spiked melanoma cells (middle graph) or without spiked cells (left graph) demonstrated that a stringent GFP intensity cutoff (black dotted line) excluded false positive signals and contributed to greater specificity for patient samples. The size and fluorescent intensify cutoffs were applied to the blood samples of patients with metastatic melanoma (data from one representative patient is shown in the right graph). Color choices are for aesthetic purposes only. Melanoma cells in culture (middle panel, showing 60% of the total cells identified by the imaging program, not meant to reflect recovery rate of spiked melanoma cells) rapidly adhere to the culture surface and hence show greater variability in size compared to the CTCs from patients with melanoma (right panel). (C) Confirmation of melanoma origin of CTCs. Patient samples were fixed after exposure to probe, and immunofluorescence then performed for Melan-A and DAPI. The DAPI+/GFP+/Melan-A+ (arrows) cells confirm the identification of melanoma CTCs while the adjacent DAPI+ only cells (arrowheads) consist of white blood cells. Bar, 30 um.
Fig 3.
ROC curve demonstrating favorable telomerase-based CTC assay test characteristics for identifying patients with and without melanoma.
A CTC count threshold of 1.1 CTCs/mL results in sensitivity of 90.0% and specificity of 91.7%. Analysis was performed using blood samples from 10 melanoma patients and 6 controls.
Table 1.
Pilot study patient demographics, disease course characteristics at time of the CTC analysis.
Table 2.
Results of one-way analysis of variance for single variables and linear regression model for multivariate analysis.
Fig 4.
Detection of BRAF mutations in isolated melanoma cells in culture and in CTCs from melanoma patients.
(A) PCR results indicating the presence or absence of the BRAFV600E mutation. Cells exposed to the probe and rendered fluorescent (left inset) were isolated via capillary-based methods. Whole genome amplification (WGA) was performed on the DNA extracted from each cell, followed by quantitative polymerase chain reaction (qPCR) analysis using primers specific for the BRAFV600E mutation. The presence of the mutation results in signal (Delta Rn, Y-axis) detectable by the 28th cycle and a curve of the characteristic “s” shape (left, arrow points to 28th cycle). The absence of the mutation results in no characteristic curve by the 28th cycle (right, arrow points to 28th cycle). (B) Isolation and genetic analysis for mutated BRAFV600E in melanoma cells in culture and spiked in control blood. A375P (homozygous BRAFV600E mutated), Mel624 (heterozygous BRAFV600E mutated), and MeWo (homozygous BRAF WT) cells were isolated in culture and after being spiked into control blood using the capillary-based technique described. The DNA was extracted from each cell and subject to WGA, followed by qPCR analysis with primers specific for the BRAFV600E mutation. In each case, the qPCR analysis accurately confirms the presence or absence of the BRAFV600E mutation for each cell line. (C) Isolation of CTCs from patients and subsequent genetic analysis for BRAF mutation status. CTCs were isolated from an additional cohort of patients via capillary-based methods followed by DNA extraction, WGA, and qPCR analysis for BRAFV600E. In each case, the BRAF mutation status of the isolated CTCs corresponded to that of the primary tumor.