Understanding key assay parameters that affect measurements of trastuzumab-mediated ADCC against Her2 positive breast cancer cells

Abstract

Use of the antibody trastuzumab to kill HER2+ breast cancer cells is an attractive therapy because of its specificity and minimal adverse effects. However, a large fraction of HER2+ positive patients are or will become resistant to this treatment. No other markers are used to determine sensitivity to trastuzumab other than HER2 status.Using the xCELLigence platform and flow cytometry, we have compared the ability of mononuclear cells (MNCs) from normal and breast cancer patients to kill different breast cancer cell lines in the presence (i.e., ADCC) or absence of trastuzumab. Image analysis and cell separation procedures were used to determine the differential contribution of immune cell subsets to ADCC activity. The assay demonstrated that ADCC activity is dependent on the presence of trastuzumab, the level of HER2 expression on the target, and the ratio of MNCs to tumor cells. There is a wide range of ADCC activity among normal individuals and breast cancer patients for high and low HER2-expressing tumor targets. Fresh MNCs display higher ADCC levels compared with cryopreserved cells. Natural killer cells display the highest ADCC followed by monocytes. T cells and B cells were ineffective in killing. A major mechanism of killing of tumor cells involves insertion of granzyme B and caspase enzymes via the antibody attached MNCs.

Figures

Figure 1. HER2 expression varies among three different breast cancer cell lines. Cell surface levels of HER2 were measured by indirect immunofluorescent staining and flow cytometry. Non-specific binding of the secondary antibody is shown in Panel A. HER2 levels are represented in panel B for the cell lines SKBR3 v (green), JIMT-1 (orange), BT474 clone 5 early passage (blue) and BT474 clone 5 later passage (violet). Non-specific staining of BT474 clone 5 early passage also is shown by the dashed line in panel B.

Figure 2. Trastuzumab alone does not affect the normalized cell index for BT474 clone 5 or SKBR3v cells as measured by the xCELLIgence system. BT474 clone 5 cells (panel A) or SKBR3v cells (panel B) were cultured for approximately 26 h and then were treated with increasing concentrations of trastuzumab. The black curve is derived from cells maintained in the absence of trastuzumab while the other colors represent increasing concentrations of trastuzumab as indicated. The cell index (CI) was determined every 15 min over the course of the experiment and normalized to 1 at the time of trastuzumab addition. Curves represent the average of three replicates. Because the standard deviations for each curve overlap, they are not shown in this figure, and growth differences were considered to be insignificant.

Figure 3. Cell index (CI) values are proportionally reduced with increasing effector to target (E/T) ratios in the presence of trastuzumab. BT474 clone 5 cells were maintained for 26 h and then were treated with media alone (control) or with media plus mononuclear cells isolated from human blood (panel A). Cells were treated in an identical fashion in Panel B except for the inclusion of 0.1 ug/ml of trastuzumab. Cell index values were normalized at the time of addition. Blue represents growth with no mononuclear cells (control) while green, orange, purple and red represents growth in the presence of MNCs at E/T ratios of 0.5/1, 1/1, 2:1, and 6/1, respectively. The vertical dashed lines indicate the 16 h window of time after treatment used to determine AUC values. Normalized cell index values are plotted in 15 min increments as the average of three replicate with the standard deviation.

Figure 4. ADCC activity varies widely among 23 individuals and target cell type. Mononuclear cells from 23 individuals or U937 cells were applied to BT474 clone 5 target cells (top panels) or SKBR3v cells (bottom panels). All E/T ratios were 6:1. Samples replicated three or more times are presented as the mean and standard deviation. The light bars and dark bars represent percent cell kill mediated by effectors with and without trastuzumab, respectively. The ADCC activity represents the difference in percent cell kill due to the addition of 0.1 ug/ml of trastuzumab and is represented in the charts of histograms to the right in the same sequence order of samples.

Figure 5. Natural killer cells exhibit the greatest ADCC activity among sub-populations of mononuclear cells. MNCs were tested for ADCC killing effect (panel A) or were separated into sub-populations and then tested. Panel B shows NK cells, Panel C shows monocytes, Panel D shows B cells, and Panel E shows T cells. Green lines represent the control, blue show 0.1 ug/ml of trastuzumab alone, violet shows MNCs or sub-populations at E/T of 6 to 1, and red shows MNCs or sub-populations at E/T of 6 to 1 in the presence of 0.1 ug/ml of trastuzumab. The flow cytometry results measuring the distribution of immune subtypes among purified cells and MNCs are given in the table insert at the bottom right.

Figure 6. Trastuzumab increases the number of CD16+ cells associated with BT474 clone 5 cells during ADCC. BT474 clone 5 cells were treated with MNCs in the absence (top panels) or presence of 0.1 ug/ml of trastuzumab (bottom panels) in a 6-well plate for analysis by immunofluorescence microscopy or in a 96-well plate for analysis by the xCELLIgence system. After significant ADCC activity was detected by the xCELLigence system, cells in the 6-well plate were analyzed by immunofluorescence microscopy. Phase contrast images are shown in the left panels, staining for CD16 in the middle panel, and staining for HER2 in the right panels.

Figure 7. The “kiss of death” assay shows increased granzyme B and caspase activity in target tumor cells in the presence of trastuzumab. Target cells labeled with the fluorescent dye,TFL4, alone (panel A) or effector cells not labeled with TFL4 (panel C) were analyzed by flow cytometry. TFL4-labeled target cells were mixed with effector cells at an E/T ratio of 14:1 alone (panel B) or with 0.1 ug/ml of trastuzumab (panel D). Values indicated for region R1 indicate the percentage of target cells with no green fluorescence (indicating no granzyme B and caspase activity). Values associated with region R2 identify the percentage of target cells (red fluorescence) with granzyme B/caspase activity (green fluorescence).