Validation of an assay to detect programmed death-ligand 1 (PD-L1) expression on circulating tumor cells
PD-L1 status is usually read from a tissue biopsy, where spatial and temporal heterogeneity can obscure it; this TriCon 2020 poster reports the analytical validation of the RarePlex PD-L1 CTC Panel Kit, an immunofluorescence assay that enumerates circulating tumor cells and scores PD-L1 on each cell individually.
Presented by RareCyte and University of Washington.
Read or download to see:
- Analytical performance was measured cell line by cell line against a single MFI threshold of 500. On the PD-L1-negative HAP-1 line specificity was 0.974; on the PD-L1-high SW900 line sensitivity reached 0.987 and accuracy 0.980; on the PD-L1-low H1650 line sensitivity fell to 0.694 and accuracy to 0.834 — the cost a single fixed threshold carries on dim expressers.
- Nucleated blood cells go onto the slide intact, and 4 fluorescence channels decide what each one is. Blood collected in AccuCyte Blood Collection Tubes is density-separated onto slides, stained for a nuclear dye, CK/EpCAM, CD45 and PD-L1, then scanned on a CyteFinder instrument; candidate CTCs are rank-scored on several hundred image-based features by a machine-learning model trained on more than 1000 clinical CTCs, then confirmed by blinded certified reviewers.
- In patient blood, PD-L1-positive and PD-L1-negative CTCs turned up in the same draw. One breast, one lung and one prostate cancer sample each yielded PD-L1-positive CTCs, with representative single-cell MFIs of 1689, 2153 and 823 against patient-matched negative cells at 295, 78 and 128 — PD-L1 status varying from cell to cell inside a single patient’s sample.
