Fig 1.
Flow cytometry reveals many CD11b+CD45- cells are also pCK+.
PBMCs isolated from patients with NSCLC and depleted of CD45+ cells were stained for flow cytometric evaluation. (A) Cells were gated first on live cell populations, then on cells which were either CD11b+CD45lo or CD11b-CD45-. (B) Subsequent plots displayed previously gated CD11b+CD45lo cells (red) over all other live cells (black) on X and Y axes of FSC vs. SSC, (C) CD45 vs. pCK, or (E) CD45 vs. the isotype control for pCK, IgG1. (D) This staining and gating strategy was performed over three different patient samples to evaluate staining intensity, or mean fluorescence intensity (MFI), of pCK or (F) the isotype control. Significance was evaluated by one-tailed, paired T-tests with significance considered at p<0.05 (*).
Fig 2.
CD11b+ cells are present after antibody-based CTC capture.
Cells were isolated from NSCLC patient samples using an antibody against MUC1. (A) Cells were then stained and imaged with a fluorescence microscope and crops of representative 10x images were given as examples of CD11b+CD45lo or CD11b-CD45+ cells. (B) Cells were identified and analyzed in JEX, then graphed in R with each cell represented as one point. Cells were displayed on an X-axis of mean fluorescence intensity (MFI) of CD45 staining, and a Y-axis of cell size. CD45+ cells (red points) were excluded by manual gating on clustered populations. (C) Subsequent gating in R displayed previously excluded cells (blue) with newly excluded cells (red) and positively selected cells (green) on X- and Y-axes of CD45 and CD11b MFI, respectively. Gating in this plot was based on exclusion of the population of cells which emerged from additional CD11b staining. (D) Cells were isolated from one NSCLC patient sample in three parallel experiments using either an antibody against EpCAM, MUC1, or Vim. The average frequency of CD11b+ cells isolated from different antibodies was 22, 27 and 47%, respectively. (E) Results from five different patients were compared using one-tailed, paired T-tests with significance considered at p<0.05 (*).
Fig 3.
CD11b staining increases specificity of CTC identification.
CTCs were captured from CD45 depleted buffy coats with MUC1 or EpCAM labelled PMPs, then stained and identified in JEX and R based on CD45 and pCK staining. (A) pCK-CD45+ cells were excluded in the first gate (red dots), while pCK+CD45- cells were positively selected (green dots) as traditionally identified CTCs. (B) Subgated plots enabled further exclusion of CD11b+ cells (red dots), from CD11b- cells positively selected as CTCs (green dots) displayed over previously excluded cells (blue dots). Red dots in this subgate represent false-positives. (C) Example 40x images were shown for a WBC (CD11b-CD45+pCK-), a CTC false-positive (CD11b+CD45lopCK+), and a true CTC (CD11b-CD45-pCK+). (D) Data from multiple samples were graphed as individual data points to demonstrate the frequency of false-positives using either EpCAM or MUC1 capture antibodies. Results from four different patients were compared using a two-tailed, paired T-test.
Fig 4.
High specificity in CTC identification affects PD-L1 evaluation.
(A) CTCs captured with MUC1 and identified from patient 324 as either pCK+CD45- or CD11b-pCK+CD45- were graphed as individual data points and compared for average PD-L1 MFI and (C) frequency of PD-L1+ CTCs. (B) Data from seven different patient samples were represented as seven single data points and compared by CTC identification criteria, either with or without CD11b exclusion criteria, for average PD-L1 expression and (D) frequency of PD-L1 expression on CTCs. Results were compared using two-tailed T-tests; unpaired for each individual sample, and paired for overall comparisons.