Evaluation of Anti-LGR5 Antibodies by ImmunoPET for Imaging

Subscriber access provided by Kaohsiung Medical University

Article

Evaluation of anti-LGR5 Antibodies by ImmunoPET for Imaging Colorectal Tumors and Development of Antibody-Drug Conjugates Ali Azhdarinia, Julie Voss, Sukhen C Ghosh, Jo A. Simien, Servando Hernandez Vargas, Jie Cui, Wangsheng A. Yu, Qingyun Liu, and Kendra S. Carmon Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.8b00275 • Publication Date (Web): 02 May 2018 Downloaded from http://pubs.acs.org on May 3, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Evaluation of anti-LGR5 Antibodies by ImmunoPET for Imaging Colorectal Tumors and Development of Antibody-Drug Conjugates

Authors: Ali Azhdarinia1, Julie Voss1, Sukhen C. Ghosh1, Jo A. Simien1, Servando Hernandez Vargas1, Jie Cui3, Wangsheng A. Yu2, Qingyun Liu2, Kendra. S. Carmon2* Affiliations: 1

Center for Molecular Imaging, The Brown Foundation Institute of Molecular Medicine, The

University of Texas Health Science Center at Houston, Houston, Texas 77030 2

Texas Therapeutics Institute, The Brown Foundation Institute of Molecular Medicine, The

University of Texas Health Science Center at Houston, Houston, Texas 77030 3

Wntrix, Inc., Houston, Texas 77021

*Correspondence to: Kendra S. Carmon, Ph.D. Texas Therapeutics Institute The Brown Foundation Institute of Molecular Medicine 1825 Pressler St., Rm 530H Houston, TX 77030 Phone: 713-500-3390 Fax: 713-500-2447 email: [email protected]

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

ABSTRACT Leucine-rich repeat-containing G-protein coupled receptor 5 (LGR5) is highly expressed in colorectal tumors and marks colon cancer stem cells that drive tumor growth and metastasis. Recently, we showed that LGR5 is a promising target for antibody-drug conjugate (ADC) therapy. However, it is important to identify LGR5-positive tumors that would respond ADC treatment. Prior to drug conjugation, we evaluated two different anti-LGR5 monoclonal antibodies (mAbs), 8F2 and 9G5, using

89

Zr-immunoPET to select the optimal mAb for ADC

development and tumor imaging. Binding, specificity, and internalization were compared and mAbs were pre-screened as ADC candidates against colon cancer cells using secondary ADCs. Both mAbs demonstrated strong, specific binding in 293T-LGR5 cells, but not 293T-vector cells. In DLD-1 colorectal cancer cells, which express high levels of LGR5, the mAbs rapidly internalized into lysosomes and promoted ADC-induced cytotoxicity, with 8F2 exhibiting slightly higher potency. No binding was detected in DLD-1-shLGR5 (LGR5 knockdown) cells. 89

Zr-DFO-LGR5 mAbs were generated and shown to retain high affinity and LGR5-dependent

uptake in vitro. PET/CT imaging of DLD-1 tumors was performed 5 days post-injection of 89ZrDFO-LGR5 mAbs and findings were consistent with biodistribution data which showed significantly higher tumor uptake (%ID/g) for DFO-9G5 (5.5 ± 1.2) and

89

89

Zr-DFO-8F2 (17.9 ± 2.2) compared to

89

Zr-

Zr-DFO-IgG (3.8 ± 1.0). No significant uptake was observed in

DLD-1-shLGR5 tumors. This study identifies 8F2 as the optimal candidate for ADC development and provides initial evidence that 89Zr-DFO-LGR5 mAbs may be utilized to stratify tumors which would respond best to LGR5-targeted ADC therapy.

Keywords: LGR5, immunoPET, colon cancer, zirconium-89, antibody-drug conjugates


Page 3 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Leucine-rich repeat-containing G-protein coupled receptor 5 (LGR5) is a bona fide marker of adult stem cells and is highly expressed in several solid tumors including gastrointestinal cancers of the stomach, liver, and colon with low expression in normal tissue

1-4

. Specifically, LGR5

expression is elevated in approximately 60-70% of colorectal adenocarcinomas

1-2

. LGR5 is

comprised of a large extracellular domain (ECD) with 17-leucine-rich repeats and a seven transmembrane domain 5 and binds R-spondin growth factors (RSPO1-4) to modulate the Wnt/βcatenin signaling pathway 6-8. A series of recent studies demonstrated that LGR5-positive cells in colon cancer function as cancer stem cells to drive tumor growth and metastasis in models of human and mouse colon cancer

9-13

. Therefore, it is important to develop diagnostic approaches

that identify LGR5-positive tumors and generate novel targeted therapies to treat and eliminate them. LGR5 is rapidly and constitutively internalized within cells making it a suitable transit for targeted therapies such as antibody-drug conjugates (ADCs), which are innovative therapeutics that combine the specificity of monoclonal antibodies (mAbs) with the potency of cytotoxic drugs

1, 14-16

. Upon selective entry into cancer cells via internalization, ADCs are trafficked to

lysosomes where drug is released to exert its cell-killing effect

16

. Recently, we reported the

generation of anti-LGR5 ADCs for targeted treatment of colon cancer 1. The ADCs incorporated the highly cytotoxic payload, monomethyl auristatin E (MMAE), and specifically destroyed LGR5-positive colon cancer cells and induced tumor regression without significant adverse effects. Since ADC effectiveness is dependent on target specificity, non-invasive imaging is an attractive approach for examining differences in the in vivo properties of candidate mAbs and



Page 4 of 23

developing optimization strategies. Moreover, imaging is an effective approach for target assessment and is critical for identifying tumors that would best respond to ADC therapy. One strategy to evaluate ADC target antigen expression and tumor uptake is by performing non-invasive mAb-based positron emission tomography (immunoPET). Zirconium-89 (89Zr) has emerged as the preferred radionuclide for immunoPET since its half-life is compatible with the long circulation time of mAbs and has been utilized in a number of clinical trials 17-18. The ability to perform delayed imaging with

89

Zr enables clearance of the tracer from non-target tissues,

resulting in improved tumor visualization compared to other radiometals. As a result of these properties,

89

Zr-immunoPET is becoming more widely implemented in the preclinical

19-21

and

clinical 22-23 development of ADCs. We hypothesize that 89Zr-immunoPET can be used to predict which anti-LGR5 mAbs would be most suitable for ADC development and to identify LGR5-positive tumors that would respond to therapy. We cloned and produced two distinct anti-LGR5 mAbs to compare binding, internalization, and effects on colon cancer cell cytotoxicity in the presence of secondary ADCs. ImmunoPET was performed to compare tumor uptake in LGR5-positive and -negative tumors, followed by tissue biodistribution to identify the utility of 89Zr-labeled anti-LGR5 mAbs as diagnostic tools to non-invasively assess LGR5 expression in tumors and to predict ADC efficacy.

EXPERIMENTAL SECTION Expression Vectors, Commercial Antibodies, and Other Reagents. All reagents were analytical grade and used without further purification unless otherwise stated. Chelex-100 resin was purchased from Bio-Rad Laboratories and used with aqueous buffers to ensure metal-free


Page 5 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


conditions. hIgG1 was from Fisher Scientific. Expression vectors encoding Myc-hLGR5, MycmLGR5 were generated previously 1, 6. The Fab Anti-human IgG Fc-Duocarmycin DM antibody with cleavable linker (Fab-αHFc-CL-DMDM) was purchased from Moradec, LLC. Commercial antibodies were used in accordance to manufacturer’s guidelines. For western blot: anti-LGR5 (Abcam, ab75732) and from Cell Signaling, anti-myc-tag (2276) and anti-β-actin (4970). For immunocytochemistry (ICC): anti-Lamp-1 (Cell Signaling, 9091) and secondary antibodies goat anti-rabbit-Alexa-488, goat anti-human-Alexa-555, and goat anti-human-Alexa-488 (Life Technologies). Cloning and Production of Anti-LGR5 mAbs. For the anti-LGR5 mAb rat-human chimera constructs, codon optimized sequences encoding the light (VL)and heavy chain variable (VH) regions of rat 8F2 and 9G5

7, 24

and the constant region of human κ light chain (CL) and IgG1

heavy chain (CH) were synthesized (Epoch Life Sciences). The VL/CL and VH/CH for each mAb were subcloned into separate pCEP4 (ThermoFisher) expression vectors using In-Fusion HD cloning (Clontech). MAbs were produced and purified as previously described 1. Cell Culture and Western Blot. HEK293T (293T) and DLD-1 cells were purchased from ATCC and routinely tested for mycoplasma. 293T and DLD-1 cells were cultured in DMEM and RPMI medium, respectively, supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin at 37°C with 95% humidity and 5% CO2. DLD-1 cells with stable LGR5 knockdown (shLGR5) or vector (shCTL) and stable 293T cell lines overexpressing hLGR5 or mLGR5 and vector cells were established as previously reported

1, 6

. For western blot, protein

extraction was performed using RIPA buffer (Sigma) supplemented with protease/phosphatase inhibitors. Cell lysates were incubated at 37°C for 1 hour in 2x SDS buffer and loaded on SDS-



PAGE. HRP-labeled secondary antibodies were utilized for detection with standard ECL protocol. Immunocytochemistry and Binding Assays. For immunocytochemistry (ICC), HEK293T or DLD-1 cells were seeded into 8-well chamber slides and incubated overnight. The next day, cells were treated with anti-LGR5 mAbs at 37°C for 1 hr. Cells were PBS washed, 4% formalin fixed, and permeabilized in 0.1% saponin. For lysosome co-localization studies, cells were incubated with anti-Lamp-1 for 45 min followed by anti-rabbit-Alexa-488 and anti-human-Alexa-555 for 1 hr at room temperature. Nuclei were counterstained with TO-PRO®-3. Images were acquired using confocal Leica TCS SP5 microscope with LAS AF Lite software. For whole-cell binding, cells were seeded onto poly-D-lysine coated 96-well plates and incubated overnight. Serial dilutions of anti-LGR5 mAbs were added for 2 hrs at 4°C. Plates were washed in PBS, fixed, incubated with anti-human-Alexa-555 for 1 hr at room temperature, and washed. Fluorescence intensity was quantified using Tecan Infinite M1000 plate reader. Data were analyzed with GraphPad Prism software using the logistic nonlinear regression model. In Vitro Cytotoxicity. DLD-1 cells were plated at 1000 cells/well in 96 half-well plates. Serial dilutions of unconjugated mAbs or isotype-matched hIgG1 control were added to the plate in the presence or absence of 5µg/ml Fab-αHFc-CL-DMDM and allowed to incubate for 4 days at 37°C. Each condition was tested in triplicate in 3 experiments. Luciferase measurements were performed using CellTiter-Glo (Promega) and EnVision mulitlabel (PerkinElmer) plate reader. Data were analyzed with GraphPad Prism software using the logistic nonlinear regression model. DFO Conjugation and

89

Zr-labeling. Immunoconjugates were prepared by reacting 1 mg of

mAb with an 8-fold molar excess of p-isothiocyanatobenzyl-desferrioxamine B (DFO-Bz-NCS, Macrocyclics) in 0.1 M sodium carbonate (pH 9). Reactions were conducted at 37 oC for 1 h and


Page 6 of 23

Page 7 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


transferred to 4oC for continuous end-over-end mixing overnight. The products were purified with Zeba desalting spin columns (Thermo Scientific) and collected in PBS.

89

Zr-oxalate was

produced by Washington University School of Medicine (St. Louis, MO) and diluted with an equal volume of 0.5 M HEPES. The pH of the radioactive solution was adjusted to 7.4 with 2 N NaOH, and 37 MBq was added to 100 µg of each immunoconjugate in 1 M HEPES. After heating at 37°C for 1 h, the reaction was quenched with 10 mM EDTA and purified with Zeba desalting spin columns. HPLC analysis was performed on DFO conjugates and radiochemical yield and purity were measured by ITLC using 50 mM EDTA as the mobile phase. In Vitro Uptake Assays. Cell uptake of

89

Zr-labeled mAbs was measured in 293T, 293T-

hLGR5, DLD-1, and DLD-1-shLGR5 cells in 96-well plates. Approximately 200,000 cells in 150 µL in RPMI were added to each well and incubated with 60 nM of mAb (37-74 kBq) at 37°C for 1 h. Cells were centrifuged and washed 3 times with PBS and radioactivity was quantified in a Wizard2 automated gamma counter (Perkin Elmer). Measurements were normalized to LGR5-negative 293T cells and expressed as fold-difference. Each experiment was performed in triplicate. PET/CT Imaging and Biodistribution. Animal studies were carried out in strict accordance with the recommendations of the Institutional Animal Care and Use Committee of the University of Texas Health Science Center at Houston. Female 5-7 week old nu/nu mice (Charles River Laboratories) were subcutaneously implanted into the lower right flank with 4 x 106 DLD-1 or DLD-1-shLGR5 cells in 1:1 mixture with matrigel (BD Biosciences). After 2 weeks, when tumor size reached ~ 4-6 mm in diameter, mice were randomized and intravenously injected with 1.9-2.7 MBq of 89Zr-DFO-LGR5 mAbs (8F2 or 9G5) or 89Zr-DFO-hIgG1 (n=5 mice/group) and imaged 5 days post-injection on a Siemens Inveon µPET/CT scanner as previously described 25.



Region-of-interest analysis was performed with the vendor software package to obtain tumor-tobackground ratios (TBRs). At the conclusion of the imaging studies, mice were euthanized by cervical dislocation and tissues were excised, weighed, and counted for radioactivity using a γ counter. The total injected activity per mouse was determined from a known aliquot of the injected solutions and used to calculate the percentage of the injected dose per gram of tissue (%ID/g). Statistical Analysis. Data are expressed as mean ± SEM. Differences between groups were analyzed by Student’s t test. Multiple comparisons used one-way ANOVA and Tukey post hoc analysis. P < 0.05 was considered statistically significant.

RESULTS Characterization of Anti-LGR5 mAb Binding. We produced and purified anti-LGR5 mAbs, 8F2 and 9G5 7, as rat-human chimeras (Sup. Fig. 1). Based on epitope mapping, 8F2 and 9G5 recognize the N-terminus of the LGR5 ECD 7. We tested species cross-reactivity of each antiLGR5 mAb. HEK293T cells overexpressing myc-tagged human LGR5 (293T-hLGR5) or mouse LGR5 (293T-mLGR5) and vector control (293T-vector) cells were established and LGR5 expression was confirmed by western blot (Fig. 1A). Cells were treated for 1 hour at 37°C with 4 µg/ml anti-LGR5 mAb. ICC and confocal analysis showed that 8F2 is specific for hLGR5, whereas 9G5 binds both hLGR5 and mLGR5 (Fig. 1B). To determine relative affinities we used a whole-cell fluorescence binding assay. We showed that 8F2 and 9G5 bind 293T-hLGR5 cells with high affinity with Kd of 0.27 µg/ml (or 1.8 nM) and 0.45 µg/ml (or 3.0 nM), respectively (Fig. 1C). 9G5 bound 293T-mLGR5 cells with Kd of 1.9 µg/ml (or 12.6 nM) (Fig. 1D). 8F2 exhibited minimal binding to mLGR5 at high concentrations, but Kd could not be determined


Page 8 of 23

Page 9 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(Fig. 1D). Anti-LGR5 mAbs did not bind 293T-vector cells (Sup. Fig. 1B) which do not express significant levels of endogenous LGR5 6.

Figure 1. Characterization of LGR5 mAb binding. (A) Western blot analysis of myc-tagged LGR5 in 293T cells. (B) Confocal images of anti-LGR5 mAbs 8F2 and 9G5 binding specificity to 293T-hLGR5 (human) and 293T-mLGR5 (mouse), but not 293T-vector cells. Whole-cell fluorescence binding of mAbs to the surface of 293T (C) hLGR5, and (D) mLGR5 cells.

Screening Anti-LGR5 mAbs for ADC-Induced Cytotoxicity. Since ADC efficacy is dependent upon target internalization and trafficking to the lysosome for release of cytotoxic drug 16, we examined if each mAb co-localized with the lysosome marker, LAMP1, subsequent to binding endogenous LGR5 in cancer cells. DLD-1 (shCTL) colon cancer cells, which express



high levels of LGR5, and DLD-1 LGR5 knockdown (shLGR5) cells (Fig. 2A) were treated with 8 µg/ml of mAb and incubated at 37°C for 1 hour. 8F2 and 9G5 co-localized with LAMP1 in DLD-1 shCTL cells (Fig. 2B). No mAb binding was detected in DLD-1-shLGR5 cells (Fig. 2B), indicating mAb specificity for endogenous LGR5.

Figure 2. ADC-induced cytotoxicity and lysosome trafficking in DLD-1 colon cancer cells. (A) Western blot of LGR5 expression in DLD-1 control (shCTL) and LGR5 knockdown (shLGR5) cells. (B) Confocal images showing anti-LGR5 mAbs in complex with LGR5 (red) and colocalization (yellow) with the lysosome marker, LAMP1 (green). No binding was observed in DLD-1-shLGR5 cells. Anti-LGR5 mAb mediated cytotoxicity of DLD-1 cells (C) in the absence and (D) presence of duocarmycin-conjugated secondary ADC.


Page 10 of 23

Page 11 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


MAbs can be pre-screened in vitro for ADC activity by complexing with a secondary ADC and testing the ability of the complex to destroy cancer cells expressing target antigen. LGR5-dependent cell cytotoxicity was examined by treating DLD-1 cells with increasing concentrations of either 8F2, 9G5 or IgG in the presence of an anti-human antibody conjugated with the cytotoxic drug duocarmycin via a cleavable linker (secondary ADC; Fab-αHFc-CLDMDM). Duocarmycin is a potent DNA-alkylating agent that binds into the minor groove of DNA

26

. As shown in Fig. 2C, the anti-LGR5 mAbs and IgG had no effect on cancer cell

viability or proliferation in the absence of secondary ADC. However, when co-treeated with secondary ADC, both anti-LGR5 mAbs induced cytotoxicity with 8F2 exhibiting slightly higher potency (EC50 = 0.59 µg/ml or 3.9 nM vs. 1.37 µg/ml or 9.1 nM for 9G5) (Fig. 2D). IgG control had no effect. Synthesis and LGR5-Dependent Uptake of

89

Zr-Labeled Anti-LGR5 mAbs. DFO was

conjugated to 8F2, 9G5, and a control IgG mAb and HPLC analysis revealed similar retention times and peak profiles following DFO conjugation (Sup. Fig. 2). Using isotopic dilution, we determined there was an average of 2.1-3.4 DFO molecules per mAb, which resulted in

89

Zr

radiolabeling yields >70% and radiochemical purity >99% as determined by radio-TLC (Sup. Fig. 3). The specific activities ranged from 0.22-0.34 MBq/µg. Using whole cell fluorescence binding assay, we showed that conjugation of 8F2 and 9G5 with DFO had no significant effect on mAb binding to 293T-hLGR5 cells (Fig. 3A-B). To validate specificity and internalization of 89

Zr-DFO-LGR5 mAbs, an in vitro uptake study was performed in cell lines expressing different

levels of LGR5 expression (Fig. 3C). As expected,

89

Zr-DFO-8F2 and

89

greater uptake than 89Zr-DFO-IgG in 293T-hLGR5 and DLD-1 cells, and was higher than

89

Zr-DFO-9G5 showed

89

Zr-DFO-8F2 uptake

Zr-DFO-9G5 in 293T-hLGR5 cells (P< 0.001). Tracer uptake was



significantly lower for

89

Zr-DFO-8F2 and

89

Zr-DFO-9G5 in DLD-1-shLGR5 cells (P< 0.05).

These results suggest uptake of 89Zr-DFO-LGR5 mAbs is LGR5-mediated.

Figure 3. In vitro assessment of

89

Zr-DFO-LGR5 mAbs. Whole-cell fluorescence binding

comparing DFO conjugated and unconjugated anti-LGR5 mAbs (A) 8F2 and (B) 9G5 to 293ThLGR5 cells. (C) Assay measuring uptake of 89Zr-DFO-8F2, 89Zr-DFO-9G5 and 89Zr-DFO-IgG in cell lines with different levels of LGR5 expression. Values represent fold-difference compared


Page 12 of 23

Page 13 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to 293T negative control cells for each mAb. * P< 0.05, ** P< 0.01, *** P< 0.001, ****P< 0.0001

ImmunoPET of

89

Zr-Labled Anti-LGR5 mAb Uptake in Colorectal Tumors. PET/CT

imaging in xenografts with varying expression levels of LGR5 demonstrated clear differences in tumor uptake of the immunoconjugates at 5 days p.i. (Fig. 4A). In DLD-1 tumors which express high levels of LGR5, tumor signal of 89Zr-DFO-8F2 was clearly visualized and higher than 89ZrDFO-9G5. Mice that received the control mAb, 89Zr-DFO-IgG, did not show tracer accumulation in tumors. Tumor-to-muscle ratios for 89Zr-DFO-8F2 (16.2 ± 2.1) were significantly higher than 89

Zr-DFO-9G5 (8.0 ± 1.0) and

89

Zr-DFO-IgG (6.4 ± 0.8) (Fig. 4B). Comparison of

89

Zr-DFO-

8F2 in mice with DLD-1 and DLD-1-shLGR5 tumors revealed 3-fold higher tumor-to-muscle ratios in DLD-1 mice, suggesting high tracer specificity for LGR5. All tracers were predominantly cleared through the liver with no notable uptake in other organs.

Figure 4. ImmunoPET characterization of

89

Zr-DFO-LGR5 mAbs in colorectal tumors. (A)

Maximum intensity projection of PET images acquired 5 days post-injection of


89

Zr-DFO-IgG,


89

Zr-DFO-9G5, and

89

Zr-DFO-8F2 in DLD-1 cells, and

Page 14 of 23

89

Zr-DFO-8F2 in DLD-1 cells with

LGR5 KD. T, indicates tumor site. (B) T/M Ratios. ** P< 0.01, *** P< 0.001

Biodistribution. Ex vivo biodistribution studies were conducted at the completion of PET imaging and are summarized in Fig. 5. Consistent with PET findings, 89Zr-DFO-8F2 had higher uptake in DLD-1 xenografts (17.1 ± 3.3 %ID/g) compared to and

89

89

Zr-DFO-9G5 (5.5 ± 1.2 %ID/g)

Zr-DFO-IgG (3.8 ± 1.0 %ID/g). Prominent accumulation of

89

Zr-DFO-8F2 was also seen

in the liver (17.9 ± 2.2 %ID/g) and spleen (11.9 ± 1.2 %ID/g) due to reticuloendothelial clearance of the agent. The minor formation of mAb aggregates following DFO conjugation may also have contributed to increased uptake by large molecule clearance organs (Sup. Fig. 2). Lower uptake was observed in all other organs and resulted in high TBRs including a tumor-tomuscle ratio (T/M) of 37.2 ± 4.8 (Sup. Fig. 4). Comparatively, non-tumor uptake values were lower for

89

Zr-DFO-9G5 but did not result in higher contrast ratios due to lower tumor uptake

than 89Zr-DFO-8F2. Similar to the PET images, 89Zr-DFO-8F2 uptake in DLD1-shLGR5 tumors with lower antigen expression was minimal (2.0±0.3 %ID/g) and was primarily in the liver and spleen.


Page 15 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. Biodistribution. (A) Comparison of 89Zr-DFO-IgG, 89Zr-DFO-8F2, and 89Zr-DFO-9G5 in mice with DLD-1 tumors and (B)

89

Zr-DFO-8F2 in mice with DLD-1 and DLD-1-shLGR5

tumors. *** P< 0.001 for 89Zr-DFO-8F2 uptake compared to other mAbs and tumors.

DISCUSSION Colorectal cancer is the third most common cancer and third leading cause of cancer-related deaths in the United States 27. Over the past decade, the mortality rate of colon cancer has rapidly declined due to advances in imaging, surgical techniques, and therapies.

18

F-fluorodeoxyglucose

(FDG)-PET is a current standard for colorectal cancer staging and evaluating therapeutic response 28. However, this approach can often result in false-positives due to FDG uptake at sites



Page 16 of 23

of inflammation in response to radiation therapy, surgery, or inflammatory diseases of the colon. Therefore, a more tumor-specific PET agent may significantly improve the detection and treatment of colorectal cancer. Key characteristics of LGR5, such as high tumor- and CSCspecificity 1-2, 9-10 and rapid internalization

14

make it an attractive candidate for both diagnostic

and ADC development to image and eradicate tumors and CSCs. Developing an LGR5 imaging agent could be advantageous in multiple regards. First, LGR5 imaging could enable patient selection by identifying primary and metastatic tumors that may respond to LGR5-targeted therapy. Second, LGR5 imaging agents could assess tumor status before, during, and after targeted treatment or chemotherapy. Third, newly developed anti-LGR5 mAbs could be compared to determine differences in sensitivity, specificity, and ability to generate high TBRs. This is the first study, to our knowledge, to report the use of

89

Zr-DFO-LGR5 mAbs to non-

invasively assess LGR5 expression and uptake in colorectal tumors. Our data show that both 8F2 and 9G5 unconjugated mAbs bind LGR5 with high affinity and rapidly internalize with no effect on cell viability (Fig. 1, Sup. Fig. 1B). In vitro cytotoxicity studies identified 8F2 as the more potent anti-LGR5 mAb in the presence of a secondary ADC (Fig. 2C), suggesting a combination of higher binding to endogenous LGR5 and increased internalization to the lysosome compared to 9G5. These findings were in agreement with radioactive cell uptake studies which showed higher uptake of

89

Zr-DFO-8F2 in 293T-hLGR5

and DLD1-cells (Fig. 3C). In vivo characterization by immunoPET in DLD-1 xenografts, which express high levels of LGR5, revealed increased tumor uptake of both 89Zr-DFO-8F2 and 89

DFO-9G5 compared to uptake compared to

Zr-DFO-IgG (Fig. 4A-B).

89

89

Zr-

Zr-DFO-8F2 exhibited enhanced tumor

89

Zr-DFO-9G5, which may be due in part to differences in epitope

accessibility or mAb internalization in vivo. DLD-1-shLGR5 xenografts did not show significant


Page 17 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


uptake of

89

Zr-DFO-8F2, thus verifying LGR5 target specificity. ImmunoPET findings were

consistent with ex vivo quantification of tumor uptake (Fig. 5A-B). Interestingly, biodistribution analyses showed 9G5, which cross-reacts with both mLGR5 and hLGR5, did not exhibit higher uptake in normal mouse tissues which express LGR5 (e.g. intestine) compared to 8F2, which only binds hLGR5 (Fig. 5A). This observation is likely due to the LGR5 receptor density of normal cells being much lower than that of tumor cells. Correspondingly, transcriptome profiling of LGR5-positive cells demonstrated that total LGR5 expression levels are much lower in normal verses cancer cells 29. Tracer levels for 89Zr-DFO-8F2 were higher in blood, which may indicate slower clearance and thereby increase availability of the mAb and improve accumulation in tumors (Fig. 5A). To further enhance tumor targeting and contrast ratios of 89ZrDFO-LGR5 mAbs and ADCs, future work may involve predosing with non-radioactive mAbs to saturate antigen sinks 30. Alternatively, mAb Fc regions may be engineered or deglycosylated to block Fc receptor-mediated sequestration in non-target tissues 31-33. ADC therapy is an emerging and innovative approach to cancer treatment but relies heavily upon in vitro characterization of mAbs, which is not necessarily correlative with their behavior in vivo. Accordingly, an important aspect of the present study was the use of 89Zr-immunoPET as a tool to select the most suitable mAb for ADC development. ImmunoPET is increasingly being used to provide a more accurate assessment of targeted delivery and uptake into tumors, pharmacokinetics, and normal tissue biodistribution of individual mAbs to predict the efficacy of ADCs

20-21

. Using standard in vitro assays for ADC characterization, we showed that 8F2 and

9G5 both bind LGR5 with similar affinity, internalize to lysosomes, and induce cancer cell cytotoxicity in the presence of secondary ADC. Through the use of 89Zr-immunoPET, however, we were able to further analyze differences in mAb properties in vivo and quantitatively



determine that 8F2 uptake was 2-fold greater than 9G5, suggesting that an 8F2-based ADC would be more effective for toxin delivery to LGR5-positive tumors. This observation supports the therapeutic utility of 8F2 shown in our recent report where monomethyl auristatin E (MMAE)-conjugated 8F2 destroyed LGR5-positive gastrointestinal cancer cells in vitro and eradicated colon tumors in vivo 1. An independent study using a different LGR5 mAb conjugated with MMAE also showed tumor regression 2, further validating the utility of LGR5 as a target for ADC therapy. Since the conjugation of linkers and payloads can affect the hydrophobicity of mAbs, and hence their biodistribution, similar imaging methods could be applied to identify differences in the in vivo properties of ADCs and their corresponding unconjugated mAbs.

CONCLUSIONS We have developed novel 89Zr-labeled anti-LGR5 mAbs for evaluating the imaging potential of the cancer stem cell marker, LGR5, in colorectal tumors. Our data shows that 89Zr-labeled antiLGR5 mAbs can identify tumors with high LGR5 expression and that immunoPET may be used as a tool to improve mAb selection for development of LGR5-targeted ADCs. Furthermore, LGR5 imaging may be useful for stratifying patients which would respond best to an LGR5targeted ADC therapy and for monitoring treatment response. ASSOCIATED CONTENT Supporting Information Additional figures, including coomassie gel of purified mAbs, 293T-vector cell binding assay, HPLC of unconjugated and DFO-conjugated mAb, radio-ITLC showing radiochemical purity of 89

Zr-DFO-LGR5 mAb, and T/M ratios for biodistribution studies (PDF).

AUTHOR INFORMATION


Page 18 of 23

Page 19 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Corresponding Author *Email: [email protected]; Phone: 713-500-3390; Fax: 713-500-2447 Notes The authors declare no competing financial interest. ACKNOWLEDGEMENTS The authors thank the imaging core at The University of Texas Health Science Center at Houston for their support. This work was supported by a grant from the Cancer Prevention Research Institute of Texas, RP150640 (K.S. Carmon), Welch Foundation Endowment Fund (K.S Carmon), and John S. Dunn Research Scholar Fund (A. Azhdarinia).

REFERENCES 1. Gong, X.; Azhdarinia, A.; Ghosh, S. C.; Xiong, W.; An, Z.; Liu, Q.; Carmon, K. S., LGR5-targeted antibody-drug conjugate eradicates gastrointestinal tumors and prevents recurrence. Mol Cancer Ther 2016, 15 (7), 1580-1590. 2. Junttila, M. R.; Mao, W.; Wang, X.; Wang, B. E.; Pham, T.; Flygare, J.; Yu, S. F.; Yee, S.; Goldenberg, D.; Fields, C.; Eastham-Anderson, J.; Singh, M.; Vij, R.; Hongo, J. A.; Firestein, R.; Schutten, M.; Flagella, K.; Polakis, P.; Polson, A. G., Targeting LGR5+ cells with an antibody-drug conjugate for the treatment of colon cancer. Sci Transl Med 2015, 7 (314), 314ra186. 3. Xi, H. Q.; Cai, A. Z.; Wu, X. S.; Cui, J. X.; Shen, W. S.; Bian, S. B.; Wang, N.; Li, J. Y.; Lu, C. R.; Song, Z.; Wei, B.; Chen, L., Leucine-rich repeat-containing G-protein-coupled receptor 5 is associated with invasion, metastasis, and could be a potential therapeutic target in human gastric cancer. Br J Cancer 2014, 110 (8), 2011-20. 4. Yamamoto, Y.; Sakamoto, M.; Fujii, G.; Tsuiji, H.; Kenetaka, K.; Asaka, M.; Hirohashi, S., Overexpression of orphan G-protein-coupled receptor, Gpr49, in human hepatocellular carcinomas with beta-catenin mutations. Hepatology 2003, 37 (3), 528-33. 5. McDonald, T.; Wang, R.; Bailey, W.; Xie, G.; Chen, F.; Caskey, C. T.; Liu, Q., Identification and cloning of an orphan G protein-coupled receptor of the glycoprotein hormone receptor subfamily. Biochem Biophys Res Commun 1998, 247 (2), 266-70. 6. Carmon, K. S.; Gong, X.; Lin, Q.; Thomas, A.; Liu, Q., R-spondins function as ligands of the orphan receptors LGR4 and LGR5 to regulate Wnt/beta-catenin signaling. Proc Natl Acad Sci U S A 2011, 108 (28), 11452-7.



7. de Lau, W.; Barker, N.; Low, T. Y.; Koo, B. K.; Li, V. S.; Teunissen, H.; Kujala, P.; Haegebarth, A.; Peters, P. J.; van de Wetering, M.; Stange, D. E.; van Es, J. E.; Guardavaccaro, D.; Schasfoort, R. B.; Mohri, Y.; Nishimori, K.; Mohammed, S.; Heck, A. J.; Clevers, H., Lgr5 homologues associate with Wnt receptors and mediate R-spondin signalling. Nature 2011, 476 (7360), 293-7. 8. Glinka, A.; Dolde, C.; Kirsch, N.; Huang, Y. L.; Kazanskaya, O.; Ingelfinger, D.; Boutros, M.; Cruciat, C. M.; Niehrs, C., LGR4 and LGR5 are R-spondin receptors mediating Wnt/beta-catenin and Wnt/PCP signalling. EMBO Rep 2011, 12 (10), 1055-61. 9. Shimokawa, M.; Ohta, Y.; Nishikori, S.; Matano, M.; Takano, A.; Fujii, M.; Date, S.; Sugimoto, S.; Kanai, T.; Sato, T., Visualization and targeting of LGR5+ human colon cancer stem cells. Nature 2017. 10. Melo, F. S.; Kurtova, A. V.; Harnoss, J. M.; Kljavin, N.; Hoeck, J. D.; Hung, J.; Anderson, J. E.; Storm, E. E.; Modrusan, Z.; Koeppen, H.; Dijkgraaf, G. J.; Piskol, R.; de Sauvage, F. J., A distinct role for Lgr5+ stem cells in primary and metastatic colon cancer. Nature 2017, 543 (7647), 676-680. 11. Kobayashi, S.; Yamada-Okabe, H.; Suzuki, M.; Natori, O.; Kato, A.; Matsubara, K.; Jau Chen, Y.; Yamazaki, M.; Funahashi, S.; Yoshida, K.; Hashimoto, E.; Watanabe, Y.; Mutoh, H.; Ashihara, M.; Kato, C.; Watanabe, T.; Yoshikubo, T.; Tamaoki, N.; Ochiya, T.; Kuroda, M.; Levine, A. J.; Yamazaki, T., LGR5-positive colon cancer stem cells interconvert with drugresistant LGR5-negative cells and are capable of tumor reconstitution. Stem Cells 2012, 30 (12), 2631-44. 12. Merlos-Suarez, A.; Barriga, F. M.; Jung, P.; Iglesias, M.; Cespedes, M. V.; Rossell, D.; Sevillano, M.; Hernando-Momblona, X.; da Silva-Diz, V.; Munoz, P.; Clevers, H.; Sancho, E.; Mangues, R.; Batlle, E., The intestinal stem cell signature identifies colorectal cancer stem cells and predicts disease relapse. Cell Stem Cell 2011, 8 (5), 511-24. 13. Schepers, A. G.; Snippert, H. J.; Stange, D. E.; van den Born, M.; van Es, J. H.; van de Wetering, M.; Clevers, H., Lineage tracing reveals Lgr5+ stem cell activity in mouse intestinal adenomas. Science 2012, 337 (6095), 730-5. 14. Carmon, K. S.; Lin, Q.; Gong, X.; Thomas, A.; Liu, Q., LGR5 Interacts and Cointernalizes with Wnt Receptors To Modulate Wnt/beta-Catenin Signaling. Mol Cell Biol 2012, 32 (11), 2054-64. 15. Snyder, J. C.; Rochelle, L. K.; Lyerly, H. K.; Caron, M. G.; Barak, L. S., Constitutive internalization of the leucine-rich G protein-coupled receptor-5 (LGR5) to the trans-Golgi network. J Biol Chem 2013, 288 (15), 10286-97. 16. Trail, P. A., Antibody Drug Conjugates as Cancer Therapeutics. Antibodies 2013, 2 (1), 113-129. 17. Deri, M. A.; Zeglis, B. M.; Francesconi, L. C.; Lewis, J. S., PET imaging with (8)(9)Zr: from radiochemistry to the clinic. Nucl Med Biol 2013, 40 (1), 3-14. 18. Jauw, Y. W.; Menke-van der Houven van Oordt, C. W.; Hoekstra, O. S.; Hendrikse, N. H.; Vugts, D. J.; Zijlstra, J. M.; Huisman, M. C.; van Dongen, G. A., Immuno-Positron Emission Tomography with Zirconium-89-Labeled Monoclonal Antibodies in Oncology: What Can We Learn from Initial Clinical Trials? Front Pharmacol 2016, 7, 131. 19. ter Weele, E. J.; van Scheltinga, A. G.; Kosterink, J. G.; Pot, L.; Vedelaar, S. R.; Lamberts, L. E.; Williams, S. P.; Lub-de Hooge, M. N.; de Vries, E. G., Imaging the distribution of an antibody-drug conjugate constituent targeting mesothelin with (8)(9)Zr and IRDye 800CW in mice bearing human pancreatic tumor xenografts. Oncotarget 2015, 6 (39), 42081-90.


Page 20 of 23

Page 21 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


20. Terwisscha van Scheltinga, A. G.; Ogasawara, A.; Pacheco, G.; Vanderbilt, A. N.; Tinianow, J. N.; Gupta, N.; Li, D.; Firestein, R.; Marik, J.; Scales, S. J.; Williams, S. P., Preclinical Efficacy of an Antibody-Drug Conjugate Targeting Mesothelin Correlates with Quantitative 89Zr-ImmunoPET. Mol Cancer Ther 2017, 16 (1), 134-142. 21. Williams, S. P.; Ogasawara, A.; Tinianow, J. N.; Flores, J. E.; Kan, D.; Lau, J.; Go, M.; Vanderbilt, A. N.; Gill, H. S.; Miao, L.; Goldsmith, J.; Rubinfeld, B.; Mao, W.; Firestein, R.; Yu, S. F.; Marik, J.; Terwisscha van Scheltinga, A. G., ImmunoPET helps predicting the efficacy of antibody-drug conjugates targeting TENB2 and STEAP1. Oncotarget 2016. 22. Gebhart, G.; Lamberts, L. E.; Wimana, Z.; Garcia, C.; Emonts, P.; Ameye, L.; Stroobants, S.; Huizing, M.; Aftimos, P.; Tol, J.; Oyen, W. J.; Vugts, D. J.; Hoekstra, O. S.; Schroder, C. P.; Menke-van der Houven van Oordt, C. W.; Guiot, T.; Brouwers, A. H.; Awada, A.; de Vries, E. G.; Flamen, P., Molecular imaging as a tool to investigate heterogeneity of advanced HER2-positive breast cancer and to predict patient outcome under trastuzumab emtansine (T-DM1): the ZEPHIR trial. Ann Oncol 2016, 27 (4), 619-24. 23. Lamberts, L. E.; Menke-van der Houven van Oordt, C. W.; ter Weele, E. J.; Bensch, F.; Smeenk, M. M.; Voortman, J.; Hoekstra, O. S.; Williams, S. P.; Fine, B. M.; Maslyar, D.; de Jong, J. R.; Gietema, J. A.; Schroder, C. P.; Bongaerts, A. H.; Lub-de Hooge, M. N.; Verheul, H. M.; Sanabria Bohorquez, S. M.; Glaudemans, A. W.; de Vries, E. G., ImmunoPET with AntiMesothelin Antibody in Patients with Pancreatic and Ovarian Cancer before Anti-Mesothelin Antibody-Drug Conjugate Treatment. Clin Cancer Res 2016, 22 (7), 1642-52. 24. Clevers JC, B. N., Haegebarth A, Van DWML Antibodies recognizing endogenous human lgr5 and/or lgr6. WO 2010016766 A3, 2010. 25. Hall, M. A.; Pinkston, K. L.; Wilganowski, N.; Robinson, H.; Ghosh, P.; Azhdarinia, A.; Vazquez-Arreguin, K.; Kolonin, A. M.; Harvey, B. R.; Sevick-Muraca, E. M., Comparison of mAbs targeting epithelial cell adhesion molecule for the detection of prostate cancer lymph node metastases with multimodal contrast agents: quantitative small-animal PET/CT and NIRF. J Nucl Med 2012, 53 (9), 1427-37. 26. Ghosh, N.; Sheldrake, H. M.; Searcey, M.; Pors, K., Chemical and biological explorations of the family of CC-1065 and the duocarmycin natural products. Curr Top Med Chem 2009, 9 (16), 1494-524. 27. Society, A. C., Cancer Facts & Figures 2018. Atlanta, Ga: American Cancer Society 2018. 28. Kekelidze, M.; D'Errico, L.; Pansini, M.; Tyndall, A.; Hohmann, J., Colorectal cancer: current imaging methods and future perspectives for the diagnosis, staging and therapeutic response evaluation. World J Gastroenterol 2013, 19 (46), 8502-14. 29. Hirsch, D.; Hu, Y.; Ried, T.; Moll, R.; Gaiser, T., Transcriptome profiling of LGR5 positive colorectal cancer cells. Genom Data 2014, 2, 212-5. 30. Boswell, C. A.; Mundo, E. E.; Zhang, C.; Stainton, S. L.; Yu, S. F.; Lacap, J. A.; Mao, W.; Kozak, K. R.; Fourie, A.; Polakis, P.; Khawli, L. A.; Lin, K., Differential effects of predosing on tumor and tissue uptake of an 111In-labeled anti-TENB2 antibody-drug conjugate. J Nucl Med 2012, 53 (9), 1454-61. 31. Malik, P.; Phipps, C.; Edginton, A.; Blay, J., Pharmacokinetic Considerations for Antibody-Drug Conjugates against Cancer. Pharm Res 2017, 34 (12), 2579-2595. 32. Sharma, S. K.; Chow, A.; Monette, S.; Vivier, D.; Pourat, J.; Edwards, K. J.; Dilling, T. R.; Abdel-Atti, D.; Zeglis, B. M.; Poirier, J. T.; Lewis, J. S., Fc-mediated Anomalous Biodistribution of Therapeutic Antibodies in Immunodeficient Mouse Models. Cancer Res 2018.



33. Gao, P.; Pinkston, K. L.; Wilganowski, N.; Robinson, H.; Azhdarinia, A.; Zhu, B.; Sevick-Muraca, E. M.; Harvey, B. R., Deglycosylation of mAb by EndoS for improved molecular imaging. Mol Imaging Biol 2015, 17 (2), 195-203.


Page 22 of 23

Page 23 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For Table of Contents Use Only

Evaluation of anti-LGR5 Antibodies by ImmunoPET for Imaging Colorectal Tumors and Development of Antibody-Drug Conjugates

Ali Azhdarinia1, Julie Voss1, Sukhen C. Ghosh1, Jo A. Simien1, Servando Hernandez Vargas1, Jie Cui3, Wangsheng A. Yu2, Qingyun Liu2, Kendra. S. Carmon2*


Evaluation of Anti-LGR5 Antibodies by ImmunoPET for Imaging

Recommend Documents