1. Introduction
Targeted alpha therapy (TAT) takes advantage of the combination of the highly potent radiobiological properties of an alpha particle emitting payload and a tumor-targeting moiety such as a monoclonal antibody. The alpha particles are characterized by high linear energy transfer with a relative biological effectiveness 3- to 8-fold greater than that of X-rays [
1,
2]. When coupled to a suitable targeting moiety the radiation dose can be preferentially delivered to the surface of the tumor cell minimizing unwanted effects on the normal surrounding tissue.
In 2013 the first-in-class alpha emitting radiopharmaceutical Xofigo® (radium-223 dichloride) was approved by the FDA for the treatment of castration-resistant prostate cancer. The ALSYMPCA clinical trial demonstrated an overall survival benefit in patients with symptomatic bone metastases and no known visceral metastases [
3]. Radium-223 is a calcium-mimetic that selectively targets hydroxyapatite in areas of high bone turnover such as bone metastases, exerting a cytotoxic effect on adjacent tumor cells through the induction of complex DNA double-strand breaks (DSBs) [
4]. Due to the paucity of efficient chelators the broader application of radium-223 in antibody-based TAT is limited. In contrast, thorium-227 (half-life of 18.7 days), the parental radionuclide of radium-223, can be complexed to several chelator families including those of the 3,2-hydroxypyridinone (3,2-HOPO) class [
5]. We have previously reported on several examples of targeted thorium-227 conjugates (TTCs) targeting a multitude of tumor-associated antigens including the HER2-TTC [
6,
7,
8,
9,
10]. Human epidermal growth factor receptor 2 (HER2) is a transmembrane tyrosine kinase receptor that is part of the human epidermal growth factor receptor (EGFR) family [
11]. The normal function of HER2 is to promote growth and proliferation, while in oncology overexpression of HER2 is a clinically validated target linked to poor prognosis and treatment outcomes in several cancer types [
11].
The mechanism of cytotoxicity of alpha therapy is based on the generation of the dense ionizing track of the alpha particle, leading to a range of DNA damages including single and double strand breaks as well as clustered DNA damages [
12,
13]. Recently, small antibody formats have explored the use of HER2-targeted delivery of the alpha-emitter actinium-225 (half-life of 10.6 days) and astatine-211 (half-life of 7.2 h) using nanobodies or single-domain antibodies as targeting moieties [
14,
15]. Of those alpha-emitters available, thorium-227 is an alternative option to actinium-225 as it has a half-life of 18.7 days that matches the half-life of human IgGs in blood. Further, as thorium-227 is the progenitor of radium-223, the supply chain is established through Xofigo®. We have previously demonstrated that TTCs induce γ-H2A.X and G2/M cell cycle arrest in various cancer models indicating the involvement of DNA damage response [
6,
10]. Furthermore, the inhibition of multiple pathways of DNA repair has been described in the literature to have a radiosensitizing effect [
16,
17,
18]. In this study we investigated whether the inhibition of the DNA repair proteins poly ADP ribose polymerase 1 (PARP-1) and poly ADP ribose polymerase 2 (PARP-2) might be synergistic in combination with the DNA damage inducer TTC. PARP-1 and PARP-2 are nuclear enzymes that have an integral role in base excision repair, including the detection of single strand DNA breaks and recruiting repair proteins to the site of the damage [
19]. PARP-1/2 deficiency leads to the accumulation of DNA damage and a decrease in cell viability [
20]. The studies presented included combination therapy with the PARP-1/2 inhibitor olaparib, which is FDA approved for treatment of BRCA-mutated ovarian and breast cancer. BRCA1 and BRCA2 are tumor suppressor genes with protein products that are central in DNA double strand repair. Defects in BRCA therefore make the cell highly sensitive to PARP-1/2 inhibition and olaparib works efficiently in patients characterized with germline BRCA mutations on the basis of synthetic lethality [
21].
We report herein the preclinical data from the combination treatment with HER2-TTC and olaparib. As such, the human colorectal cancer isogenic cell line pair DLD-1 parental and DLD-1 BRCA2 -/-, the latter harboring a defect in the DNA double strand repair gene, was used. Both in vitro and in vivo experiments revealed a significant synergistic effect in the BRCA2 deficient model. This study supports the future development of new therapeutic strategies combining DNA damage response inhibitors with the DNA damage-inducing TTCs.
3. Discussion
TTCs represent a new class of systemic radiotherapeutic agents with the capability of targeting multiple cancer types. We report herein a HER2-TTC comprising three key components, the anti-HER2 monoclonal antibody trastuzumab (HER2-Ab) [
27], the 3,2-HOPO chelator conjugated via amide bond formation to the free amino groups of lysine residues on HER2-Ab, and thorium-227 which forms a highly stable complex with the chelator [
5,
7]. Simple mixing of the conjugate with thorium-227 at ambient temperature induced high radiochemical purity (RCP) of ≥ 95% for the complex as measured by iTLC. The HER2-TTC was further evaluated by ELISA, demonstrating that the binding affinity was not impaired by the conjugation or the radiolabeling.
In the present study we evaluated the HER2-TTC together with the PARPi olaparib. PARP-1/2 is essential for maintaining genome integrity and the inhibition has been demonstrated to sensitize cells to ionizing radiation and other types of DNA damaging agents [
16]. PARP inhibition leads to the conversion of single strand DNA breaks to double strand breaks (DSBs) [
28]. Normally DSBs can be repaired by two pathways, HR (homologous recombination) which utilises a DNA template on the sister chromatid and is high fidelity repair or NHEJ (non-homologous end-joining) utilising non-template repair leading to errors and genetic instability [
29]. In BRCA-deficient tumors, homologous recombination is not functional, and the cell is directed towards error-prone repair and cell death. As such, PARP inhibitors induce the loss of PARP-1/2 function leading to the accumulation of DNA lesions and inadequate repair of double-strand DNA breaks [
30]. As the primary mode of action of the TTC is induction of complex DNA damage we hypothesized that the combination with PARPi would be synergistic.
Since olaparib is known to induce synthetic lethality in BRCA-mutated tumors [
21], we selected the BRCA2-deficient colorectal cancer cell line (DLD-1 BRCA2 -/-) for this study. Identical data sets were also generated in the BRCA2-proficient parental cell line (DLD-1 parental) for comparison. First, both cell lines were determined to express a low level of approximately 5000 HER2 receptors/cell by flow cytometry. Further, a clear difference was observed in vitro with the BRCA2 -/- cell line demonstrating increased sensitivity to HER2-TTC and clear synergistic effect from the combination based on the calculated combination index compared to the parental cell line. As isogenic cell line pairs, harboring DNA repair deficiencies, are barely described and available, it has to be noted that the cell line used in the presented study served only as a tool to study combinations of HER2-TTC and olaparib.
To evaluate if the in vitro result translated to the in vivo setting we continued with the mono- and combination treatment with HER2-TTC and olaparib. Firstly, biodistribution and retention of HER2-TTC was shown to be specific as evidenced by the high tumor uptake in both models (40–60% IA/g) compared to ca. 5% IA/g for the radiolabeled isotype control at 336 h, the latter due to antibody delivery from the enhanced permeability and retention effect [
31]. Further, it was observed that an amount of radium-223 is retained in the tumor after targeting and internalization of the HER2-TTC; in contrast, the radiolabeled isotype control showed less retention of radium-223 in the tumor. Interestingly significant tumor growth inhibition was achieved for the TTC monotherapy at the higher doses of 300 and 600 kBq/kg, despite the low HER2 expression levels as measured by both flow cytometry and IHC. These data may therefore support the further investigation of HER2-TTC as a monotherapy in patients with an HER2 expression level of 1+, which are not deemed suitable for either Herceptin®- or Kadcyla®-based therapies.
A strong synergistic effect was observed using the combination of the non-effective monotherapy dose of 120 kBq/kg bw in combination with 50 mg/kg bw (daily, 4 weeks) olaparib and 300 kBq/kg bw in combination with 50 mg/kg bw (daily, 4 weeks) olaparib. Both combinations resulted in a robust tumor growth inhibition. The lack of synergy observed in the parental model further reflects the specificity of the combination therapy to the BRCA2-deficient model, introducing a second layer of tumor targeting in addition to HER2 targeting. Furthermore, no significant myelosuppression was observed for the combination on the red and white blood cells as well as platelet populations. As systemic radioimmunotherapies suffer from dose-limiting toxicity to the bone marrow this study offers hope for identifying effective combinations with much reduced overall hematological toxicity. In conclusion, the data clearly indicate the potential for increasing the therapeutic window by using combination therapy with HER2-TTC and olaparib.
BRCA germline mutations account for around 5–10% of breast cancers and 10–18% of ovarian cancers [
32]. Although HER2-overexpressing breast cancers appear seldom to carry BRCA mutations, ovarian cancers have significantly higher numbers of BRCA mutant/HER2 positive cases and this patient population would therefore represent a potential clinical indication for further exploration for this combination [
33]. New tumor-specific antigens expressed by BRCA-deficient breast cancers may form the basis for the development of new TTCs in the future. In addition the findings reported in this paper warrant the further investigation of targeted alpha therapy approaches in combination with other DNA damage response inhibitors.
4. Materials and Methods
Cells: DLD-1 parental and DLD-1 BRCA2 -/- cell lines were obtained from Horizon Discovery group. The cell lines were authenticated using PCR fingerprinting by the provider. Cells were maintained in an incubator at 37 °C and 5% CO2. The cells were cultured in RPMI 1640 supplemented with 10% fetal bovine serum, 2 mmol/L glutamine, 25 mmol/L sodium bicarbonate, 1% penicillin, and 1% streptomycin.
Binding and receptor density: DLD-1 parental and DLD-1 BRCA2 -/- (100 000 cells, 100 µL) were seeded in 96 well plates and incubated with a titration of anti-HER2 antibody (0.0006-100 µg/mL) for one hour at 4 °C, followed by incubation with 100 µL anti-human IgG-PE (Cat# 409304, biolegend) for one hour at 4 °C. The fluorescence intensity was acquired from a Guava EasyCyte 8HT flow cytometer and the raw data was processed using FlowJo to obtain the mean fluorescence intensity values. The data was subsequently plotted using GraphPad Prim software version 7.0 against the concentration of anti-HER2-antibody. The mAbs/cell was determined making a standard curve using beads from Quantibrite (Cat# 340495, BD biosciences).
Preparations of the HER2 antibody-chelator conjugate: The synthesis of the 3,2-HOPO chelator was performed as previously described [
5,
22]. The 3,2-HOPO chelator was conjugated to lysine residues within the HER2-antibody and an isotype control antibody through in situ activation of the chelator using EDC/NHS chemistry and subsequent incubation with the antibody in phosphate buffered saline (PBS), pH 7.0, for 60 min at room temperature. The molar ratio used during the reaction was 1/8/8/16 (mAb/chelator/NHS/EDC). The antibody-chelator conjugate was purified by size exclusion chromatography (HiLoad 16/600 Superdex 200; GE Healthcare, Chicago, USA). A chelator to antibody ratio (CAR) of 0.8 was determined by HPLC. Conjugates were subsequently stored at −20 °C.
Radiolabeling and characterization of HER2-TTC: Thorium-227 was purified from an actinium-227 generator as previously described [
34]. For in vitro studies, the antibody-chelator conjugates were mixed with thorium-227 activities ranging from 0.5–2.5 MBq and incubated at room temperature for 60 min. For in vivo studies, specific activities of 4.2 kBq/µg (600 kBq/kg), 2.1 kBq/µg (300 kBq/kg) and 0.84 kBq/µg (120 kBq/kg) were prepared using an adjusted body weight per animal of 0.03 g and a total antibody dose of 0.14 mg/kg. Radiochemical purity (RCP), defined as the amount of thorium-227 bound to the HER2-TTC, was determined by instant thin-layer chromatography (iTLC) and radio-HPLC.
Enzyme linked immunosorbent assay (ELISA): Recombinant human HER2 (Cat# 1129-ER, R&D Systems) was coated to 96-well plates (1 µg/mL; NUNC/Maxisorp). Wells were blocked with 3% BSA in PBS. Cold HER2-antibody conjugate, an isotype control antibody and the radiolabeled HER2-TTC (2 MBq/mg, stored for 72 h) were titrated (1:5; 10 µg/mL) on the HER2-coated ELISA plate. Unbound samples were washed off and bound samples were visualized using horseradish peroxidase-labeled goat anti-human lambda antibody (Cat# 2060-05, Southern Biotech) followed by visualization with the peroxidase substrate ABTS (Cat# 002024, Life Technologies, Carlsbad, USA). The absorbance was measured at 405 nm in a plate reader (Perkin Elmer, Waltham, USA).
IC
50 isobolograms: Combination experiments were conducted with DLD-1 parental and DLD-1 BRCA2 -/-. The cells were seeded in 384 well plates (30 µL per well, 30,000 cells/mL). After 24 h the cells were treated with a titration of HER2-TTC (0.01–50 kBq/mL) and a titration of olaparib (0.01–25 µM) as single treatments and in nine different fixed-ratio combinations of HER2-TTC (C1) and olaparib (C2); 0.9 × C1 + 0.1 × C, 0.8 × C1 + 0.2 × C2, 0.7 × C1 + 0.3 × C2, 0.6 × C1 + 0.4 × C2, 0.5 × C1 + 0.5 × C2, 0.4 × C1 + 0.6 × C2, 0.3 × C1 + 0.7 × C2, 0.2 × C1 + 0.8 × C2, and 0.1 × C1 + 0.9 × C2. After five days incubation, cell viability was determined by using CellTiter Glo (CTG) 2.0 Luminescent Cell Viability Assay (Cat# G9242, Promega), according to manufacturer’s protocol. The cell viability was expressed in % by normalization to cells seeded in culture medium supplemented with buffer solution and the absolute IC
50 values were determined by using GraphPad Prims software version 7.0, making a non-linear regression curve fit, selecting log (inhibitor) vs. response–variable slope (four parameters). The IC
50 isobolograms were generated by plotting the actual IC
50 values of HER2-TTC and olaparib along the x- and y-axis. The combination index (CI) was determined according to the median-effect model of Chou-Talalay [
24], with CI < 0.8 defined as synergistic effect, 0.8 < CI > 1.2 defined as additive effect, and CI > 1.2 defined as antagonistic effect.
Animal models: The animal studies were managed at the Laboratory of Pharmatest Services Ltd., Itäinen Pitkäkatu 4C, 20,520 Turku, Finland with the approval of the National Committee for animal experiments (license number ESAVI-2331-04.10.07–2017). The animal experiments were conducted following Directive 2010/63/EU, the protection of animals used for scientific purposes. A 3R (Replacement, Reduction, and Refinement) principle was applied. In all studies, female NMRI nude mice (RjOrl:NMRI-
Foxn1nu, Janvier Labs, France, 5–6 weeks) were used. All animals received an intraperitoneal (i.p.) injection of an unrelated murine IgG2a antibody (200 µg/animal; UPC10; Sigma) 24 h prior treatment to block unspecific spleen uptake [
35].
In vivo biodistribution: One million DLD-1 parental cells suspended in 0.05 mL PBS (millipore, Cat# L1825, Burlington, USA) or five million DLD-1 BRCA2 -/- cells suspended in 0.05 mL 50% matrigel (BD Biosciences, San Jose, USA) were inoculated subcutaneously into the flank. When tumor size averaged 200mm3, tumor bearing mice were administered with a single intravenous (i.v.) injection of HER2-TTC or isotype control (600 kBq/kg body weight (bw), 0.14 mg/kg bw) and sacrificed after 24, 72, 168, and 336 h. Tumors and organs were harvested from three mice per group at each time point. Radioactivity was counted using a high-purity germanium detector (HPGe) linked to an autosampler (GEM-F8250, Ortec Gamma Data). To identify thorium-227 and radium-223, the GammaVision software and Npp32 analysis engine (Reg. Guide 4.16 detection limit method) were used. For thorium-227 measurement, the 235.96 keV (abundance 12.90%), 256.23 keV (abundance 7.00%), 329.85 keV (2.90% abundance), 286.09 keV (abundance 1.74%), 304.50 keV (abundance 1.15%), 334.37 keV (abundance 1.14%), and 299.98 keV (abundance 2.21%) gamma peaks were used. Thorium-227 counts were corrected to the time of injection and expressed as percentage of injected dose of thorium-227 per gram (% IA/g).
In vivo efficacy: One million DLD-1 parental cells or five million DLD-1 BRCA2 -/- cells were suspended in 0.05 mL PBS and inoculated subcutaneously into the flank. When tumor size averaged 90 mm
3, mice were randomized into groups with 10 mice per group. Mice received a single i.v. injection of HER2-TTC (125, 300, or 600 kBq/kg bw, 0.14 mg/kg bw), radiolabeled isotype-control (300 kBq/kg bw, 0.14 mg/kg bw), HER2 antibody-chelator conjugate (0.14 mg/kg bw), or vehicle. In a parallel arm of the study, four groups were treated with HER2-TTC (125 or 300 kBq/kg bw, 0.14 mg/kg bw) in combination with olaparib (25 or 50 mg/kg bw daily, i.p. for 4 weeks)
. Monotherapy of olaparib was included in control groups. In both arms of the study, body weights and tumor dimensions were measured twice a week during the course of the study. Blood samples was collected from the saphenous vein to 100 µL EDTA tubes before dosing (day -1 for DLD-1 parental and day -2 for DLD-1 BRCA2-/-) and thereafter every other week. Hematology samples were analyzed using VetScanHM5 (Abaxis Europe GmbH, Griesheim, Germany). Animals were sacrificed by cervical dislocation upon reaching the humane endpoint (tumor volume ≥ 1.500 mm
3; body weight loss ≥ 15%). Statistical analysis was performed using GraphPad Prism software, applying one-way ANOVA followed by Tukey’s test. To evaluate the cooperativity of combination treatment the expected additivity was calculated according to the Bliss additivity model (C = A + B – A x B; wherein C is the expected T/C of the combination of drug A and drug B if they act additive, A is T/C of drug A, and B is T/C of drug B). Ten percent over the expected additive effect is assumed to indicate synergism of the two drugs and 10% below the expected additive effect is assumed to indicate antagonism [
36].
Immunohistochemistry: HER2 expression was determined by immunohistochemical (IHC) staining using antibody EP1045Y rabbit monoclonal antibody, (Abcam, Cambridge, UK), followed by incubation with HRP-anti-rabbit labeled polymer (Envision, Dako/Agilent, Santa Clara, USA)) and subsequent visualization using using 3,3′-diaminobenzidine. In parallel, xenograft samples were stained with an isotype control antibody to demonstrate specificity of the staining (rabbit normal serum, Dako/Agilent, Santa Clara USA).