Factors that influence idasanutlin sensitivity from the ex vivo patient sample screening data. (A) Schematic of the integration of patient clinical, demographic, whole exome sequencing (WES), RNA-seq, proteomics, and idasanutlin in vitro screening data to identify biomarkers that predict idasanutlin response. (B) The graph depicts a significant positive correlation between the venetoclax and idasanutlin AUC in primary patient samples (n = 336) from the Beat AML cohort as determined by a Spearman correlation coefficient test. (C) Correlation between the BCL2A1 gene expression level and the idasanutlin AUC (n = 202) in the Beat AML cohort as determined by a Spearman correlation coefficient test. (D) The graphs show the average difference in the idasanutlin AUC (x-axis; point with line indicating the 95% confidence interval) between patient samples with the specified events (y-axis) and those without the events. The colors indicates significance as determined by a Welch t test after false discovery rate (FDR) correction (FDR < 0.05). Both the full Beat AML cohort (left facet) and the cohort after removing samples with TP53, del(17p), or −17 events (right facet) are displayed. (E) The graph depicts a Reactome pathway hierarchy that contains all pathways that were significant in at least 1 of the 4 analyses (correlation between idasanutlin/venetoclax drug AUC and RNA-seq gene expression r ≥ 0.3 and correlation between idasanutlin/venetoclax drug AUC and the proteomics protein expression log ratio r ≥ 0.3 using Spearman correlation coefficient tests). (F) Idasanutlin AUCs from primary AML samples (n = 297 samples) were compared across different clinical characteristics (supplemental Table 1) in samples without a TP53 mutation or 17p deletion. Significance was determined using either a 2-tailed Mann-Whitney or a Kruskal-Wallis test (for categorical variables) or a 2-tailed Spearman correlation tests (for continuous variables). (G) The graph depicts the mean ± standard error of the mean (SEM) of the idasanutlin AUCs for leukemia samples with different FAB subtypes from the Beat AML cohort. Significant differences were determined using a 1-way analysis of variance (ANOVA) test. (H) The graph depicts the mean ± SEM of cell viabilities (3 biologic replicates) of leukemia blast cells, monocytes, and T cells isolated from 3 primary patient samples treated with dose gradients of idasanutlin for 3 days as determined by an MTS assay. (I) The graph depicts the percentage changes in leukemia blasts, monocytes, and T cells after treatment with gradient doses of idasanutlin (Ida), AMG232 (AMG), and DS3032b (DS), respectively, as determined by flow cytometry CD45 staining. (J) The graph depicts the normalized percentages of apoptotic cells (normalized to no drug treatment cells) determined by positive annexin V and/or propidium iodide (PI) staining. Idasanutlin treatment induced significantly less apoptosis in M4/M5 cells than in M0/M1 cells; at 300 nM (32.4% ± 16.6% vs 6.4% ± 5.8%; P = .03), 500 nM (39.8% ± 16.4% vs 6.5% ± 5.2%; P = .02), and 1 μM (54.1% ± 16.4% vs 20.1% ± 8.2%; P = .02). (K) Western blot images demonstrating the differences in the expression of p53, MDM2, and p53 downstream targets in M4/M5 and M0/M1 leukemia samples. The band areas were quantified using ImageJ software. Area ratios (target/vinculin for cytoplasm and target/histone 3 for nuclear protein) were determined. Significant differences were determined using Mann-Whitney tests. (L) Western blot images demonstrating the activation of the p53 pathway by idasanutlin treatment in non-M4/M5 (23-318 and 22-527) and M4/M5 (22-574, 22-555, 22-576, and 22-302) leukemia patient samples. GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Factors that influence idasanutlin sensitivity from the ex vivo patient sample screening data. (A) Schematic of the integration of patient clinical, demographic, whole exome sequencing (WES), RNA-seq, proteomics, and idasanutlin in vitro screening data to identify biomarkers that predict idasanutlin response. (B) The graph depicts a significant positive correlation between the venetoclax and idasanutlin AUC in primary patient samples (n = 336) from the Beat AML cohort as determined by a Spearman correlation coefficient test. (C) Correlation between the BCL2A1 gene expression level and the idasanutlin AUC (n = 202) in the Beat AML cohort as determined by a Spearman correlation coefficient test. (D) The graphs show the average difference in the idasanutlin AUC (x-axis; point with line indicating the 95% confidence interval) between patient samples with the specified events (y-axis) and those without the events. The colors indicates significance as determined by a Welch t test after false discovery rate (FDR) correction (FDR < 0.05). Both the full Beat AML cohort (left facet) and the cohort after removing samples with TP53, del(17p), or −17 events (right facet) are displayed. (E) The graph depicts a Reactome pathway hierarchy that contains all pathways that were significant in at least 1 of the 4 analyses (correlation between idasanutlin/venetoclax drug AUC and RNA-seq gene expression r ≥ 0.3 and correlation between idasanutlin/venetoclax drug AUC and the proteomics protein expression log ratio r ≥ 0.3 using Spearman correlation coefficient tests). (F) Idasanutlin AUCs from primary AML samples (n = 297 samples) were compared across different clinical characteristics (supplemental Table 1) in samples without a TP53 mutation or 17p deletion. Significance was determined using either a 2-tailed Mann-Whitney or a Kruskal-Wallis test (for categorical variables) or a 2-tailed Spearman correlation tests (for continuous variables). (G) The graph depicts the mean ± standard error of the mean (SEM) of the idasanutlin AUCs for leukemia samples with different FAB subtypes from the Beat AML cohort. Significant differences were determined using a 1-way analysis of variance (ANOVA) test. (H) The graph depicts the mean ± SEM of cell viabilities (3 biologic replicates) of leukemia blast cells, monocytes, and T cells isolated from 3 primary patient samples treated with dose gradients of idasanutlin for 3 days as determined by an MTS assay. (I) The graph depicts the percentage changes in leukemia blasts, monocytes, and T cells after treatment with gradient doses of idasanutlin (Ida), AMG232 (AMG), and DS3032b (DS), respectively, as determined by flow cytometry CD45 staining. (J) The graph depicts the normalized percentages of apoptotic cells (normalized to no drug treatment cells) determined by positive annexin V and/or propidium iodide (PI) staining. Idasanutlin treatment induced significantly less apoptosis in M4/M5 cells than in M0/M1 cells; at 300 nM (32.4% ± 16.6% vs 6.4% ± 5.8%; P = .03), 500 nM (39.8% ± 16.4% vs 6.5% ± 5.2%; P = .02), and 1 μM (54.1% ± 16.4% vs 20.1% ± 8.2%; P = .02). (K) Western blot images demonstrating the differences in the expression of p53, MDM2, and p53 downstream targets in M4/M5 and M0/M1 leukemia samples. The band areas were quantified using ImageJ software. Area ratios (target/vinculin for cytoplasm and target/histone 3 for nuclear protein) were determined. Significant differences were determined using Mann-Whitney tests. (L) Western blot images demonstrating the activation of the p53 pathway by idasanutlin treatment in non-M4/M5 (23-318 and 22-527) and M4/M5 (22-574, 22-555, 22-576, and 22-302) leukemia patient samples. GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal