Clinical Validation of a Circulating Tumor Cell Assay Using Density Centrifugation and Automated Immunofluorescence Microscopy
American Journal of Clinical Pathology. 2022;158(2):270-276. DOI 10.1093/ajcp/aqac040.
How to cite
AMA
Sabath DE, Perrone ME, Clein A, Tam M, Hardin M, Trimble S, et al. Clinical Validation of a Circulating Tumor Cell Assay Using Density Centrifugation and Automated Immunofluorescence Microscopy. Am J Clin Pathol. 2022;158(2):270-276. doi:10.1093/ajcp/aqac040APA
Sabath, D. E., Perrone, M. E., Clein, A., Tam, M., Hardin, M., Trimble, S., et al. (2022). Clinical Validation of a Circulating Tumor Cell Assay Using Density Centrifugation and Automated Immunofluorescence Microscopy. American Journal of Clinical Pathology, 158(2), 270-276. https://doi.org/10.1093/ajcp/aqac040BibTeX
@article{sabath2022clinical,
title = {Clinical Validation of a Circulating Tumor Cell Assay Using Density Centrifugation and Automated Immunofluorescence Microscopy},
author = {Sabath, Daniel E and Perrone, Marie E and Clein, Alisa and Tam, Michael and Hardin, Michael and Trimble, Sara and others},
journal = {American Journal of Clinical Pathology},
volume = {158},
number = {2},
pages = {270--276},
year = {2022},
doi = {10.1093/ajcp/aqac040}
}Circulating tumor cells shed from a tumor into the blood can signal recurrence or response to treatment, but they are rare, and the most widely used method captures them by pulling on an epithelial surface marker, EpCAM. Cells that carry little of that marker can slip through before they are ever counted.
This study validated a different approach for the clinical laboratory. Instead of selecting on a marker, it collects the whole buffy coat by density, spreads the nucleated cells onto glass slides, and finds the tumor cells by automated fluorescence imaging. The team tested it on blood spiked with known numbers of cultured cancer cells and on 47 paired clinical samples, measured head to head against an FDA-cleared reference method.
The assay was fully specific, linear, and reproducible, its counts tracked the reference, and where the two disagreed it was the density-based method that found more cells.
Key findings
- The assay detected spiked tumor cells with complete analytical specificity and stayed linear across the tested range. Across five cell-free control specimens the false-positive rate was 0%, giving 100% specificity, and over 25 to 125 spiked cells the response was linear (slope 0.99, R² = 0.96, P < .001) with recovery of 92% to 111%.
- Counts from the density-based assay tracked an FDA-cleared reference method across paired clinical samples. In 47 paired specimens from 19 patients, the two methods correlated with an R² of 0.8 after log transformation (P < .001) and agreed on adverse-prognosis classification in 87.5% of cases.
- Because it isolates cells by density rather than by EpCAM, the assay recovered tumor cells that marker-based capture missed. In the one markedly discordant specimen it counted 549 circulating tumor cells to the reference method’s 341, a gap the authors traced to cells expressing too little EpCAM to be caught by the marker-based method.
The AccuCyte–CyteFinder workflow in the methods
“For the AccuCyte/CyteFinder assay, the AccuCyte Sample Preparation System was used for the reproducible transfer of nucleated cells from blood to slides following the manufacturer’s instructions.12-14 Briefly, 7.5 mL of blood were transferred using a serologic pipet from the blood collection tube to a separation tube containing an internal float with a density similar to nucleated cells. After centrifugation (3000g for 25 minutes at room temperature), the blood was separated into 3 fractions: an RBC layer at the bottom of the tube, a buffy coat layer (containing nucleated cells, including CTCs, and platelets) in the space between the float and the tube wall, and a plasma layer above the float. An external ring was clamped around the tube using the CyteSealer device to create a physical barrier between the RBC layer and the buffy coat layer, and plasma was removed from above the float with a serologic pipet. A high-density displacement fluid was added above the float, and a collector was inserted that connects the contents of the separation tube to an isolation tube containing less dense fluid that further separates the blood components. After a second centrifugation (1000g for 20 minutes at room temperature), the buffy coat was buoyantly displaced into the isolation tube. A transfer fluid containing alcohol-based fixative was added, and the resulting mixture was spread onto 8 slides (Superfrost Plus slides, VWR International, LLC) using the CyteSpreader device.”
— Sabath et al., American Journal of Clinical Pathology (2022), Methods, “Sample Processing for RarePlex”
Disclosure: RareCyte is listed as an author affiliation on the publication cited above.
Why it matters for The AccuCyte–CyteFinder workflow users
If you run circulating tumor cell work in a clinical laboratory, the value of this study is that it holds the density-based approach to a formal validation and to the accepted predicate. The AccuCyte–CyteFinder workflow does not select cells on an epithelial marker; it separates the whole nucleated fraction by density, spreads it onto slides, and images every cell, so a tumor cell that has shed EpCAM is still on the slide to be found. Here that design returned complete analytical specificity, linear and reproducible counts, and results that matched an FDA-cleared reference method across 47 paired clinical samples. In the one specimen where the two diverged, it was the density-based workflow that caught the low-EpCAM cells the marker-based method missed. For a laboratory weighing whether to adopt it, the paper supplies what matters most: a marker-independent method, characterized against a validated standard, with the performance data a clinical workflow is judged on.
