Extended Data Fig. 2: PB2 and EMT6 tumor models for the evaluation of immune cell infiltrates.
(a) Correlations between p95HER2 and both CD3 protein levels and tumor size. Protein levels determined by Western densitometry (arbitrary units). Simple linear regressions with 95% confidence bands are shown. Pearson correlations (R) with associated one-tailed p values are listed in the plot. (b) Example of pancytokeratin staining of a PB2 tumor using the Akoya PhenoImager HT, with staining presented as a Pathview pseudochromogenic IHC image. Corresponding example of tumor/stromal mask, generated based on the pancytokeratin staining. (c) Quantification of Ly6G, CD161, F4/80, and CD19 as determined by mIF analysis of PB2 tumors described in Fig. 2. No significant difference in the infiltration of cells expressing any of these markers was seen between p95HER2-low and p95HER2-high expressing tumors, within either cancer cell dominated ‘tumor’ or stromal regions. Each data point indicates marker quantification from a single ROI. Data represent mean ± SEM. Statistical analyses by two-tailed Mann-Whitney U test. P values are indicated within the plots. (d) Schematic depicting the strategy for constitutive expression of human FL-HER2 in EMT6 cells, along with dox-induced mouse p95HER2. Flow cytometric analysis demonstrating cell-surface expression of FL-HER2 in EMT6 cells. (e) Growth of orthotopic EMT6 tumorgrafts in syngeneic, immunocompetent BALB/c mice. The expression of human FL-HER2 in EMT6 cells slows tumor growth but not as dramatically as when expressed in the PB2 – C57BL/6 model depicted in Fig. 2c. N = 10 per group. Data represent mean ± SEM. Statistical analysis by unpaired, two-tailed Student’s t test with Welch’s correction and two-stage Benjamini-Hochberg correction for multiple comparisons. P values are indicated within the plot.
