Antitumor Activity and Biomarker Analysis for TROP2 Antibody-Drug Conjugate Datopotamab Deruxtecan in Patient-Derived Breast Cancer Xenograft Models

Abstract

Purpose: Datopotamab deruxtecan (Dato-DXd) is a humanized anti-trophoblast cell-surface antigen-2 (TROP2) IgG1 mAb linked to a potent topoisomerase I inhibitor payload (DXd). Dato-DXd has already shown antitumor activity in breast cancer; however, the determinants of response, including the importance of TROP2 expression, remain unclear. We tested the activity of Dato-DXd in a panel of breast cancer patient-derived xenografts (BCX) varying in TROP2 expression.

Experimental design: The antitumor activity of Dato-DXd and isotype-control-DXd (IgG-DXd) was assessed against 11 BCXs varying in TROP2 expression, 10 representing tumors postneoadjuvant chemotherapy. Pharmacodynamic effects were assessed at 24 and 72 hours. The effects of TROP2 expression on Dato-DXd activity was assessed in vitro and in vivo using viral overexpression in BCX-derived cell lines.

Results: Models differed in their sensitivity to both Dato-DXd and IgG-DXd. Dato-DXd (10 mg/kg) led to objective response in 4 (36%) models and statistically significant prolongation of event-free survival in 8 (73%) models, whereas IgG-DXd (10 mg/kg) led to response in 1 (9%) and prolonged event-free survival in 3 (27%) models. TROP2 RNA and protein were significantly higher in Dato-DXd-sensitive models. In isogenic cell lines derived from Dato-DXd-resistant BCXs, overexpression of TROP2 conferred Dato-DXd antitumor activity in vitro and in vivo. Dato-DXd increased γH2AX and phospho-KAP1 in the two Dato-DXd-sensitive BCXs but not in a Dato-DXd-resistant BCX. In Dato-DXd-sensitive models, antitumor activity was enhanced in combination with a PARP inhibitor, olaparib.

Conclusions: Dato-DXd is active in breast cancer models. Dato-DXd has TROP2-dependent and -independent mediators of activity; however, high TROP2 expression enhances Dato-DXd antitumor activity.

Figure 1.

TROP2 membrane expression in BCXs and matched tumor samples. A, TROP2 membrane expression (H-score) shown for 52 PDX models (as determined by IHC) relative to HR and HER2 status shown. HR⁺ is defined as greater than 5% expression of estrogen receptor or progesterone receptor. B, Representative images of TROP2-positive (BCX.017) and -negative (BCX.011) IHC. C, TROP2 membrane expression (H-score) was assessed in 27 patient tumors and matched BCXs by IHC. D, TROP2 H-score in BCXs correlated with TROP2 expression in matched patients (r = 0.6991; P < 0.0001).

Figure 2.

In vivo efficacy of Dato-DXd and IgG-DXd in a panel of BCXs. Eleven PDXs were treated with Dato-DXd or IgG-DXd (nonspecific antibody linked to DXd) at 1 or 10 mg/kg i.v. on day 1 per 21-day cycle. Arrows indicate day of treatment. BCXs are shown in order of highest to lowest TROP2 expression, as determined by IHC; TROP2 H-score shown above each BCX. A, TROP2-positive BCXs. B, TROP2-negative BCXs. Data are shown from left to right for each model: mean tumor volume ± SEM, change in tumor volume from baseline of individual mice at 21 days unless otherwise indicated (BCX.010, BCX.100, and BCX.011 on day 18 and BCX.144A on day 11), and Kaplan–Meier curve of EFS defined as time to tumor doubling (EFS-2) or quadrupling (EFS-4) from baseline (day 1).

Figure 3.

Dato-DXd response correlates with TROP2 expression in BCXs. A, Percentage tumor growth change and growth inhibition (T/C ratio) for models shown for BCXs with 10 mg/kg Dato-DXd or 10 mg/kg IgG-DXd. Metrics shown were calculated at 21 days of treatment unless indicated otherwise. TROP2 and SLFN11 expression by IHC as well as genomic alterations in actionable genes are shown below the X-axis. TROP2 membranous expression was scored by H-score (0–300), and SLFN11 nuclear expression was scored by H-score. B, TROP2 membrane expression by IHC compared between BCXs that responded to Dato-DXd and BCXs that did not. C, TROP2 expression as determined by reverse phase protein array compared between BCXs that responded to Dato-DXd and BCXs that did not. D, TROP2 (TACSTD2) as determined by RNA-seq compared between BCXs that responded to Dato-DXd and BCXs that did not. E and F, Improvement in T/C and EFS with 10 mg/kg Dato-DXd vs. 10 mg/kg IgG-DXd correlates with TROP2 H-score.

Figure 4.

Effect of exogenous TROP2 expression on Dato-DXd sensitivity. A, Immunoblotting of TROP2. BCX.010CL and BCX.011CL cells were transduced with control or TROP2-expressing lentivirus, followed by puromycin selection. Exogenous TROP2 expression in stable lines as determined by Western blot. B and C, Cell viability assay. Cells of four cell lines (BCX.010CL-control, BCX.010CL-TROP2, BCX.011CL-control, and BCX.011CL-TROP2) were seeded into 96-well plates, treated with serial dilutions of IgG-DXd, or Dato-DXd, or relevant vehicle controls for 72 hours. Cell viability was monitored by sulforhodamine B assay. IC₅₀ values were calculated using CalcuSyn program. D and E, Annexin V apoptosis assay. Cells were treated with vehicle, IgG-DXd, or Dato-DXd at 1,000 ng/mL for 4 days, followed by annexin V-FITC staining and flow cytometry analysis. Percentage of total apoptotic cells was calculated by annexin V–positive (A++ and A−+) cells. F and G, Colony formation assay. Cells seeded in 6-well plates were treated with vehicle, IgG-DXd, or Dato-DXd at 1,000 ng/mL for 8 days. Cell colonies were stained and quantitated. H, Immunoblotting of DNA damage and apoptosis markers. The control and TROP2-expressing cell lines were treated with PBS, IgG-DXd, or Dato-DXd at 1,000 and 100 ng/mL, respectively, for 72 hours. Expression levels of CC3, cleaved PARP, phospho-KAP1, phospho-CHK1, and γH2AX were assessed by Western blot. I and J,In vivo tumor growth. Expression levels of CC3, cleaved PARP, phospho-KAP1, phospho-CHK1, and γH2AX were assessed by Western blot. I–K,In vivo tumor growth. Control and TROP2-expressing BCX.011CL were implanted into mice and remained TROP2-negative and -positive, respectively, for duration of experiment, as shown by IHC of the untreated tumors (I). J and K, Growth analysis of control and TROP2-expressing BCX.011CL treated with 1 or 10 mg/kg of Dato-DXd or IgG-DXd (nonspecific antibody linked to DXd; isotype control ADC) on day 1 of a 21-day cycle. Arrows indicate day of treatment. Data are shown left to right for each model: mean tumor volume ± SEM, change in tumor volume from baseline of individual mice at 21 days, and Kaplan–Meier curve of EFS defined as time to tumor doubling (EFS-2). Statistical analysis of EFS-2 is shown in the figure. Dato, Dato-DXd; Iso, isotype.

Figure 5.

PD changes associated with Dato-DXd treatment. Three BCXs representing a range of TROP2 expression sensitivity to Dato-DXd were treated with Dato-DXd or IgG-DXd, and the tumors were collected 24 or 72 hours after treatment. A–D, TROP2 membrane expression (A), γH2AX (Ser139; B), membranous hIgG (C), and DXd localization (D) were accessed by IHC and scored. E, The concentration of DXd in the tumors was also measured by LC/MS in flash-frozen tumor pieces. Concentration shown as quantity of DXd (ng) per g of tumor. F, Venn diagram of the functional proteomic changes 72 hours after Dato-DXd treatment as determined by RPPA (FDR <0.1). There were 143 changes common to the sensitive models (BCX.017 and BCX.055) and 32 changes common to the resistant models (BCX.006 and BCX.094). There were 16 signaling changes common to all four models. G, Plot of the 16 cell signaling changes resulting for 10 mg/kg Dato-DXd in all models tested. Protein names and states are listed in Supplementary Tables S8–S10. *, P < 0.05; **, P < 0.001; ****, P < 0.0001. KAP1, KRAB domain-associated protein 1.

Figure 6.

The combination of Dato-DXd and olaparib has increased efficacy in TROP2-expressing breast cancer. A and B, Breast PDXs were treated with either 1 or 3 mg/kg Dato-DXd on day 1 per 21-day cycle in combination with olaparib (100 mg/kg, daily, PO). C and D, BCXs were treated with 3 and 10 mg/kg Dato-DXd on day 1 per 21-day cycle in combination with olaparib (100 mg/kg, daily, PO). Column 1 shows the mean tumor volume ± SEM. Column 2 shows change in tumor volume at the time the controls reached maximum allowable size or at day 21 of treatment. Column 3 shows Kaplan–Meier curve, with an event being defined as time to tumor doubling (EFS-2) or quadrupling (EFS-4).