Development of Optimized Exatecan-Based Immunoconjugates with Potent Antitumor Efficacy in HER2-Positive Breast Cancer

Abstract

The prognosis of human epidermal growth factor receptor 2 (HER2)-positive breast cancer has significantly improved with the advent of anti-HER2 therapies, especially antibody-drug conjugates (ADCs). In this field, ADCs, like trastuzumab deruxtecan (T-DXd), using camptothecin analogs, represent a promising strategy. However, T-DXd can induce resistance and serious adverse effects, potentially driven by a non-specific Fcγ receptor-mediated endocytosis. Here, we report the development of novel HER2-targeting conjugates, using the camptothecin derivative exatecan and a linker optimized to control hydrophobicity. Three formats were evaluated: a high drug-to-antibody ratio (DAR) 8 IgG-based ADC (IgG(8)-EXA), and two DAR 4 Fc-free constructs (Mb(4)-EXA and Db(4)-EXA). Thus, IgG(8)-EXA and Mb(4)-EXA displayed potent, specific cytotoxicity against HER2-positive breast cancer cells and strong antitumor activity in vivo. Notably, IgG(8)-EXA exhibited a favorable pharmacokinetic profile, despite its high DAR, supporting the potential of this drug-linker design. These two conjugates represent promising candidates for further preclinical development.

1

Bioconjugation process, exatecan drug-linker structure, and mechanism of exatecan release from the immunoconjugates developed in this study.

1. Synthetic Route for Intermediate 6

2. Synthetic Route for Exatecan Drug-Linker 12

2

Schematic representations of immunoconjugates 13–16.

3

HER2 binding recognition measured by indirect ELISA of: (A) IgG(8)-EXA (13) (blue curve) in comparison to T-DXd (red curve), (B) Mb(4)-EXA (14) (green curve), and (C) Db(4)-EXA (15) (light blue curve), in comparison to the nonconjugated entities (black curves) trastuzumab (IgG(8)), minibody (Mb(4)) and diabody (Db(4)), respectively.

4

Internalization of immunoconjugates in SK-BR-3 (HER2-positive) cells. (A) Histogram of flow cytometry for cells incubated with ppL-PE alone (gray) or IgG_irr(8)-EXA (16) (red), IgG(8)-EXA (13) (dark blue), Mb(4)-EXA (14) (green), Db(4)-EXA (15) (light blue), and T-DXd (orange) after labeling with ppL-PE (exposure time = 24 h at 37 °C). Asterisks indicate significant differences between two groups using a Tukey’s test (**p < 0.01, ****p < 0.0001). (B) Fluorescence microscopy images of SK-BR-3 cells after incubation during 24 h at 37 °C with ppL-PE alone or IgG_irr(8)-EXA, T-DXd, IgG(8)-EXA, Mb(4)-EXA and Db(4)-EXA (preincubated with ppL-PE). Scale bar: 100 μm.

5

In vitro cytotoxic activity of IgG_irr(8)-EXA (16) (red), IgG(8)-EXA (13) (dark blue), Mb(4)-EXA (14) (green), Db(4)-EXA (15) (light blue), T-DXd (orange) and free exatecan (black curve) after 5 days of incubation at 37 °C on SK-BR-3 and MDA-MB-468 cancer cell lines (CellTiter Glo assay) (A). HER2 cell surface expression in breast tumor cell lines as assayed by flow cytometry (B).

6

Comparison of average DAR variation for IgG(8)-EXA (red) and T-DXd (green) upon incubation over 192 h in mouse (A) and human (B) serum, assessed by affinity capture and LC-MS analysis. Values are normalized against DAR measured at time 0.

7

Ex vivo tissue/organ biodistribution of [¹²⁵I]­I-T-DXd (A, B) and [¹²⁵I]­I-IgG­(8)-EXA ([ ¹²⁵ I]­13) (C, D) in female SCID CB17 mice bearing subcutaneous BT-474-SCID tumors at 6 h, 24 h, 48 h, 3 d, 5 d, and 11 d p.i. (n = 4 per time point). A: The mice were injected (i.v.) with 3.42 ± 0.08 or 11.13 ± 0.44 MBq (5 d group, n = 2 and 4 respectively) of [¹²⁵I]­I-T-DXd. The radioactivity was measured using a γ-counter: bars represent the mean of %ID/g; the error bars are standard deviation (SD) from the mean. B: Table represents the %ID in the thyroïd. C: The mice were injected (i.v.) with 3.91 ± 0.57 or 15.49 ± 0.01 MBq (5 d group, n = 2 and 4, respectively) of [¹²⁵I]­I-IgG­(8)-EXA ([¹²⁵I]13). The radioactivity was measured by using a γ-counter. A: bars represent the mean of %ID/g; the error bars are standard deviation (SD) from the mean. D: Table represents the %ID in the thyroïd. (B.A.T. represents the brown adipose tissue).

8

Ex vivo tumor biodistribution (%ID/g) of the two radioiodinated immunoconjugates in female SCID CB17 mice bearing subcutaneous BT-474-SCID grafts. All mice were injected (i.v.) with (i) 3.91 ± 0.57 or 15.49 ± 0.01 MBq (5 d group, n = 2 and 4, respectively) of [¹²⁵I]­I-IgG­(8)-EXA ([¹²⁵I]13, red bars) or (ii) 3.42 ± 0.08 or 11.13 ± 0.44 MBq (5 d group, n = 2 and 4, respectively) of [¹²⁵I]­I-T-DXd (green bars). The error bars are the standard deviation (SD) from the mean. For all radioiodinated immunoconjugates, 4 per time point. A: the bars represent the mean of %ID/g of the tumor. B: the bars represent the average ratio of %ID/g between the tumor and blood (T/B). C: the bars represent the average ratio of %ID/g between tumor and muscle (T/M).

9

Representative SPECT-CT (coronal section) images of a mouse bearing orthotopic BT-474-SCID tumor injected (i.v.) with 15 MBq of [¹²⁵I]­I-IgG­(8)-EXA ([¹²⁵I]13) and imaged at different time points p.i. The arrow indicates the location of the BT-474-SCID tumor (320 mm³).

10

Schematic representation of administration schedule (A). Tumoral volume evolution of mice bearing HER2-positive breast cancer tumors after intravenous injection of PBS (control) or five different treatments. Arrows indicate the day of treatment administration for each group. T-DXd (single dose at 10 mg/kg equivalent to 65 nmol/kg); Mb(4)-EXA (4 injections at 5.36 mg/kg equivalent to 62 nmol/kg); IgG(8)-EXA 5 (single dose at 5 mg/kg equivalent to 31 nmol/kg); IgG(8)-EXA 10 (single dose at 10 mg/kg equivalent to 62 nmol/kg); IgG(8)-EXA 2 × 10 (2 injections at 10 mg/kg equivalent to 124 nmol/kg) (B).