Patritumab deruxtecan induces immunogenic cell death

Abstract

Antibody-drug conjugates (ADCs) enable targeted delivery of cytotoxic payload to cancer cells. Here, we characterized the mode of action of the ADC patritumab deruxtecan, which is a monoclonal antibody specific for Erb-B2 Receptor Tyrosine Kinase 3 (ERBB3, best known as HER3) coupled to the topoisomerase-I inhibitor DXd. Patritumab deruxtecan decreased viability and induced the relocation of calreticulin fused to green fluorescent protein (CALR-GFP) to the periphery of human osteosarcoma U2OS cells engineered to express HER3 but not in their parental counterparts only expressing the CALR-GFP biosensor. Patritumab deruxtecan as well as its payload DXd induced various traits of immunogenic cell death (ICD) including antibody detectable calreticulin membrane exposure, exodus of high mobility group protein B1 (HMGB1), as well as the release of ATP into cell culture supernatants. Moreover, DXd causes rapid inhibition of DNA-to-RNA transcription, which is a key predictor for ICD. Mouse cancer cells treated with DXd were able to vaccinate syngeneic immunocompetent mice against tumor challenge. Tumor-free mice developed immune memory that led to the rejection of syngeneic tumors after rechallenge. In conclusion, patritumab deruxtecan is equipped with a cytotoxic payload that induces hallmarks of ICD in vitro and elicits antitumor immunity in vivo.

Keywords: Antibody drug conjugate; anticancer immunity; cancer; immunotherapy.

Figure 1.

Patritumab deruxtecan decreases viability and induces CALR relocation in cells expressing HER3. (a) Scheme depicting the treatment and analysis workflow. (b-g) human osteosarcoma U2OS biosensor cells stably expressing calreticulin (CALR)-green fluorescent protein (GFP) and their derivatives overexpressing HER3 were treated for 5 days with increasing concentrations of non-targeting IgG-deruxtecan ADC control, anti-HER3 monoclonal antibody (U3–1287, patritumab) and patritumab deruxtecan (ADC) in a dose range covering 0.04; 0.16; 0.63; 2.5 to 10 µg/mL. Oxaliplatin (OXA; 50 µM) was used as a prototype inducer of immunogenic cell death. (b,c) cellular viability was assessed by microscopy and determined by determining the proportion of viable cells relative to untreated controls (d-g) calreticulin relocation was assessed by microscopy measuring GFP-signal heterogeneity as a surrogate marker for vesicular aggregation. Representative images (d,e) are depicted together with CALR relocation (f,g). Results are depicted as mean ± SD of quadruplicate assessments and were normalized to untreated controls. Data points depict single view fields. Significant increase was calculated using Welch’s t-test. Scale bars equal 10 µm.

Figure 2.

Patritumab deruxtecan induces hallmarks of immunogenic cell death in HER3 expressing derivatives but not in parental cells. Human osteosarcoma U2OS cells and their derivatives overexpressing HER3 were treated for 5 days with 0.16 µg/mL non-targeting IgG-deruxtecan ADC control, anti-HER3 monoclonal antibody (U3–1287, patritumab) or patritumab deruxtecan (ADC). Oxaliplatin (OXA; 5 µM) and mitoxantrone (MTX, 10 nM) were used as prototype inducers of immunogenic cell death. Parental cells were also treated for 5 days with the cytotoxic payload DXd (0.16 µg/mL). (a,b) calreticulin (CALR) exposure was assessed by surface immunostaining and flow cytometric analysis. The percent of viable DAPI⁻/CALR⁺ cells is depicted. (c,d) ATP secretion was measured in cell culture supernatant by detecting bioluminescence of the luciferase conversion assay. (e,f). HMGB1 exodus was assessed by ELISA in cell culture supernatants. Results are depicted as mean ± SD of triplicate assessments. ATP and HMGB1 data were normalized to the number of cells assessed by microscopy. Significant increase was calculated using Welch’s t-test.

Figure 3.

Patritumab deruxtecan induces bystander antitumor effect in parental cells cocultured with HER expressing derivatives. (a) Scheme depicting the coculture assay and analysis workflow. (b,c) human osteosarcoma U2OS cells and their derivatives overexpressing HER3 differentially tagged by fluorescent markers were cocultured at 1:1 ratio and treated for 5 days with 0.16 µg/mL non-targeting IgG-deruxtecan ADC control, anti-HER3 monoclonal antibody (U3–1287, patritumab); patritumab deruxtecan (ADC) or DXd. Oxaliplatin (OXA; 5 µM) was used as prototype cell death inducer. Cellular viability was assessed by fluorescent microscopy and determined by measuring the proportion of viable cells for each cell line relative to untreated controls. Results are depicted as mean ± SD of quadruplicate assessments. Data points depict single view fields. Significant decrease was calculated using Welch’s t-test.

Figure 4.

The cytotoxic payload DXd induces transcription inhibition. (a-d) human osteosarcoma U2OS cells were treated with 1 µM DXd, the ICD inducer oxaliplatin (OXA; 50 µM), the transcription inhibitor dactinomycin (DACT; 1 µM) or were left untreated (Ctrl) and subsequently subjected to live transmitted light microscopy. The mean probability of nucleolar condensation (CON) phenotype was calculated over time. (e,f) U2OS cells were pre-treated with a dose-range of DXd (0.1, 1, 10 µM) for 2.5 h and treatment pursued for 1.5 additional hours in the presence of 1 mm 5-ethynyl uridine (EU). After staining with an AlexaFluor™-488-coupled azide, the nuclear EU incorporation was quantified as nuclear intensity and normalized to untreated controls. (g, h) U2OS cells were treated with a dose-range of DXd (0.1, 1, 10 µM) for 4 h and co-stained with anti-fibrillarin and anti-nucleolin antibodies. Fibrillarin/nucleolin colocalization was quantified by computing the Pearson correlation coefficient (PCC) between fluorescent signals. Results are depicted as mean ± SD of triplicate assessments. ATP and HMGB1 data were normalized to the number of cells assessed by microscopy. Significant increase was calculated using Welch’s t-test. Scale bars equal 10 µm.

Figure 5.

Dxd treated cells induce anticancer vaccinating effects in immunocompetent mice. (a) Scheme depicting the experimental design. Murine fibrosarcoma MCA205 cells were treated for 72 hours with DXd (2 µM), 24 hours with cisplatin (CDDP, 150 µM) or the combination of both DXd plus CDDP. Mitoxantrone (MTX, 4 µM for 24 hours) was used as positive control. Cells were inoculated subcutaneously (s.C.) into the left flank of C57BL/6J mice and one week later the animals were challenged with untreated MCA205 cells. Absence of tumor growth was interpreted as an onset of anticancer immunity. (b,c) tumor growth was measured and is depicted as means ± SEM and as individual tumor growth curves for each mouse (d,e). Survival and the percent of tumor free mice were assessed and are depicted as Kaplan-Meier plots. Type II analysis of variance (ANOVA) and pairwise Wilcoxon test with multiple comparison and Holm’s correction were applied for tumor growth and a log-rank test was performed for survival analysis.

Figure 6.

Dxd induces immune memory. (a) Scheme depicting the experimental design. (b) Tumor-free mice were rechallenged with homologous MCA205 cells injected subcutaneously (s.C.) into the left flank and heterologous murine melanoma B16-F10 cells into the contralateral flank. (b-e) tumor growth was measured and is depicted as means ± SEM and as individual tumor growth curves for each mouse for homologous MCA205 (b,c) and heterologous B16-F10 tumors (d,e).