SLC46A3 As a Potential Predictive Biomarker for Antibody–Drug

Published OnlineFirst August 21, 2018; DOI: 10.1158/1078-0432.CCR-18-1300

Cancer Therapy: Preclinical Clinical Cancer Research SLC46A3 as a Potential Predictive Biomarker for Antibody–Drug Conjugates Bearing Noncleavable Linked Maytansinoid and Pyrrolobenzodiazepine Warheads Krista Kinneer1, John Meekin1, Arnaud C. Tiberghien2, Yu-Tzu Tai3, Sandrina Phipps4, Christine Mione Kiefer4, Marlon C. Rebelatto5, Nazzareno Dimasi4, Alyssa Moriarty6, Kyriakos P. Papadopoulos6, Sriram Sridhar5, Stephen J. Gregson2, Michael J. Wick6, Luke Masterson2, Kenneth C. Anderson3, Ronald Herbst1, Philip W. Howard2, and David A. Tice1

Abstract

Purpose: Antibody–drug conjugates (ADC) utilizing non- also examined in patient-derived xenograft and in vitro cleavable linker drugs have been approved for clinical use, and models of acquired T-DM1 resistance and multiple myeloma several are in development targeting solid and hematologic bone marrow samples by RT-PCR. malignancies including multiple myeloma. Currently, there Results: Loss of SLC46A3 expressionwasfoundtobea are no reliable biomarkers of activity for these ADCs other than mechanismofinnateandacquiredresistancetoADCs presence of the targeted antigen. We observed that certain cell bearing DM1 and SG3376. Sensitivity was restored in refrac- lines are innately resistant to such ADCs, and sought to tory lines upon introduction of SLC46A3, suggesting that uncover the underlying mechanism of resistance. expression of SLC46A3 maybemorepredictiveofactivity Experimental Design: The expression of 43 lysosomal than target antigen levels alone. Interrogation of primary membrane target genes was evaluated in cell lines resistant multiple myeloma samples indicated a range of SLC46A3 to ADCs bearing the noncleavable linker, pyrrolobenzodiaze- expression, including samples with undetectable levels like pine payload SG3376, in vitro. The functional relevance of multiple myeloma cell lines resistant to BCMA-targeting SLC46A3, a lysosomal transporter of noncleavable ADC DM1andSG3376ADCs. catabolites whose expression uniquely correlated with Conclusions: Our ﬁndings support SLC46A3 as a potential SG3376 resistance, was assessed using EPHA2-, HER2-, and patient selection biomarker with immediate relevance to BCMA-targeted ADCs and isogenic cells overexpressing or clinical trials involving these ADCs. Clin Cancer Res; 1–13. genetically inactivated for SLC46A3. SLC46A3 expression was 2018 AACR.

Introduction treatment of erb-b2 receptor tyrosine kinase 2 (ERBB2, HER2)- positive metastatic breast cancer (3), inotuzumab ozogamicin for Antibody–drug conjugates (ADC) combine a mAb with a the treatment of acute lymphoblastic leukemia (4), and gemtu- cytotoxic drug (warhead) to preferentially eliminate antigen-pos- zumab ozogamicin for the treatment of CD33-positive acute itive cells for the treatment of cancer (1). Four ADCs are approved myeloid leukemia (5). Upon binding to the targeted antigen on for clinical use: brentuximab vedotin for the treatment of Hodgkin the cell surface, ADCs prepared with enzymatically cleavable lymphoma (2), ado-trastuzumab emtansine (T-DM1) for the linkers (e.g., brentuximab vedotin) or acid-labile hydrazone linkers (e.g., inotuzumab ozogamicin and gemtuzumab ozogamicin) are internalized and processed within the cell, releasing the 1Oncology Research, MedImmune, Gaithersburg, Maryland. 2Spirogen, QMB Innovation Centre, London, United Kingdom. 3The Jerome Lipper Multiple cytotoxic warhead after linker cleavage. Warheads released in this Myeloma Center and LeBow Institute for Myeloma Therapeutics, Dana-Farber manner are typically membrane permeable and capable of Cancer Institute, Harvard Medical School, Boston, Massachusetts. 4Antibody bystander killing (6–8). In contrast, ADCs with noncleavable Discovery and Protein Engineering, MedImmune, Gaithersburg, Maryland. linkers, such as T-DM1, rely upon proteolytic degradation of the 5 6 Translational Medicine, MedImmune, Gaithersburg, Maryland. South Texas antibody in the lysosome to release an amino acid linker warhead Accelerated Research Therapeutics, San Antonio, Texas. (9). These catabolites are generally not membrane permeable (10, Note: Supplementary data for this article are available at Clinical Cancer 11) and therefore require transport from the lysosome to reach Research Online (http://clincancerres.aacrjournals.org/). their intracellular target (12). Several ADCs utilizing noncleavable Corresponding Author: Krista Kinneer, MedImmune, 1 MedImmune Way, linkers are currently in clinical development and target both solid Gaithersburg, MD 20878. Phone: 301-398-4219. Fax: 301-398-9219. E-mail: tumors and hematologic malignancies, including diffuse large B- [email protected] cell lymphoma and multiple myeloma (refs. 13–16). doi: 10.1158/1078-0432.CCR-18-1300 Although T-DM1 therapy has shown signiﬁcant clinical beneﬁt, 2018 American Association for Cancer Research. most patients eventually relapse despite continued treatment

www.aacrjournals.org OF1

Kinneer et al.

TNFRSF17 (BCMA) antibody was prepared using VH and VL Translational Relevance sequences retrieved from patent application US 2014/0105915 Although antibody–drug conjugates (ADC) utilizing the A1. Trastuzumab drug conjugate, T-DM1 (ado-trastuzumab payload DM1 have been approved for clinical use, and several emtansine), was purchased from Blue Door Pharma. All antibo- are in clinical development that target both solid tumors and dies used in this study, except for those conjugated to DM1 and hematologic malignancies, predictive biomarkers beyond monomethyl auristatin F (MMAF), were engineered for the site- antigen expression have yet to be identified. Herein, we show specific conjugation of two drugs per antibody as described by that loss of lysosomal transporter SLC46A3 expression is a Dimasi and colleagues (32). mechanism of innate and acquired resistance to ADCs bearing The PBD dimer SG3249 was synthesized at Novasep. mcMMAF the payloads DM1 or SG3376 (a PBD dimer) with nonclea- and DM1–succinimidyl-4-(N-maleimidomethyl)cyclohexane-1- vable linkers, making SLC46A3 a potential biomarker for carboxylate (SMCC) were purchased from ALB Technology. patient selection in ongoing trials with ADCs bearing these SG3541 was prepared at Spirogen as described in the Supple- payloads. mentary Materials. SG3376 was prepared at Spirogen according to published methods (31). Conjugation of SG3249 (27), SG3376 (31), and SG3541 was performed as described previously (32, 34). (17–19). Currently, the only predictive biomarker for response to To generate the mcMMAF ADCs with a drug to antibody ratio T-DM1 is overexpression of HER2 (20), which is detected by (DAR) of 4 at the antibodies' native cystines, the antibodies were measurement of HER2 protein levels by IHC or of HER2 gene reduced using 2.5-molar equivalent (MEq) tris(2-carboxyethyl) – amplification by fluorescent or chromogenic in situ hybridization phosphine in PBS pH 7.2 1 mol/L EDTA for 1 hour at 37 C. The (21). Acquired resistance to T-DM1 has been evaluated in several reaction mixture was then incubated with 4 equivalents of – – preclinical studies, and due to the complexity of this molecule, mcMMAF for 1 hour at 25 C in PBS pH 7.2 1 mmol/L EDTA several potential mechanisms have emerged: (i) antigen loss and/ 10% v/v dimethylsulfoxide. The reaction was quenched by the or downregulation, (ii) increased expression of drug transporters addition of 4 equivalents (over the mcMMAF) of N-acetyl MDR1 (ABCB1) and MRP1 (ABCC1), (iii) defects in ADC traf- cysteine. ficking, and/or (iv) changes in receptor and signaling pathways To generate the DM1 ADCs with DAR 4 at the antibody lysines, (22). In addition, aberrations in lysosomal pH and proteolytic the antibodies at 2 mg/mL in 1 mmol/L borate buffer, pH 8.5, activity (23) and loss of the lysosomal transporter solute carrier were incubated at room temperature with 8 MEq DM1. The family 46 member 3 (SLC46A3; refs. 12, 24) have been observed conjugation reaction was monitored with reduced reverse-phase in T-DM1–resistant cell lines, highlighting a potential role for the LC/MS as described previously (32, 34), and the reaction was lysosome in T-DM1 resistance. stopped when a DAR of 4 was reached (total reaction time, fi Until recently, the cytotoxic warheads used for most ADCs in approximately 1 hour). The ADCs were puri ed and characterized clinical development were based on antimitotic agents, namely, analytically as described previously (32, 34). the auristatins and maytansines (1). ADCs prepared with the cleavable linker dimeric pyrrolobenzodiazepine (PBD) payloads Human cells talirine (SGD-1910; refs. 25, 26) or tesirine (SG3249; ref. 27) are Primary multiple myeloma samples were obtained from now entering the clinic. PBD dimers are a class of compounds that patients after informed consent was obtained, in accordance with form sequence-selective DNA crosslinks in the minor groove of the Declaration of Helsinki and under the auspices of a protocol DNA, leading to cell death (28, 29) and thereby offering an approved by the Dana-Farber Cancer Institute Review Board. fi alternative mechanism of action for tumor targeting. New classes Myeloma cells were puri ed before RNA isolation as described of PBD payloads with noncleavable linkers (30), including the previously (16). RT112, HT29, and KYSE-410 cells were obtained benzyl tether-linked PBD SG3376 (31), are also being from the European Collection of Authenticated Cell Lines, and synthesized. Du145, JIMT-1, JJN-3, OPM2, and KMS-12-BM cells were Herein, we report that certain cell lines that are sensitive to obtained from Deutsche Sammlung von Mikroorganismen und ADCs prepared with the cleavable linker drug, SG3249, were Zellkulturen GmbH. N87-TR and BT-474-TR cells were made at innately resistant to those with the noncleavable linker drug, MedImmune as described previously (35). All other cell lines were SG3376. Through gene expression profiling of sensitive and obtained from the ATCC. Cells were grown according to the resistant cell lines, we discovered that expression of the lysosomal manufacturer's instructions. All cell lines were authenticated by fi transporter SLC46A3 was required for cytotoxic activity of ADCs short tandem repeat DNA pro ling (IDEXX BioResearch Labora- Mycoplasma prepared with SG3376. Absence of SLC46A3 contributed to innate tories) and were found to be negative for using PCR Mycoplasma and acquired resistance to ADCs bearing noncleavable DM1 and and the MycoSEQ Detection Assay Kit (4407876, fi SG3376, further suggesting that expression of SLC46A3 is a Thermo Fisher Scienti c). candidate patient selection biomarker for ADCs incorporating Patient-derived xenograft models such payloads. þ The ST1616B HER2 breast PDX model was established by using tissue from a patient with lung metastasis refractive to Materials and Methods T-DM1, and ST1616B/TDR was established from the parent Antibodies and generation of ADCs model by using chronic administration of T-DM1 over several Antibody cloning, expression, and purification were carried passages. Both models were developed and tested at South Texas out as described previously (32–34). IgG1-isotype control, 1C1, Accelerated Research Therapeutics (START) according to proto- and trastuzumab antibodies were made at MedImmune. The anti- cols approved by the International Animal Care and Use

OF2 Clin Cancer Res; 2018 Clinical Cancer Research

SLC46A3 as a Potential Biomarker for Noncleavable ADCs

Committee. For in vivo studies, tumor fragments were harvested were washed and incubated with anti-human-AF647 secondary and implanted subcutaneously into the flank of 7- to 8-week old antibody (A-21445, Thermo Fisher Scientific). Fluorescence of female athymic nude mice (Charles River Laboratories). Animals live, single cells was determined on a BD-LSRII and data were were matched by tumor volume and randomly assigned to control analyzed with FlowJo software (FlowJo). Internalization is and treatment groups (n ¼ 3–5 animals per group). Tumor reported as 100 – [(mean fluorescence intensity (MFI) at each volume and animal weight data were collected electronically with time point/MFI at time 0) 100]. a digital caliper and scale; tumor dimensions were converted to volume with the formula tumor volume (mm3) ¼ width2 (mm2) Lysosomal trafficking length (mm) 0.52. Study endpoint was a mean control tumor ADCs were labeled with Fab-pHast human secondary fluores- volume of approximately 1.0 to 2.5 cm3; change in tumor volume cent conjugate (PH-01, ATS Bio) according to the manufacturer's of each group was compared with the control. instructions. Cells were incubated with 3 mg/mL of the labeled ADCs at 37C for desired time points. After washing, the fluores- IHC cence of live, single cells was determined on a BD-LSRII and data IHC was performed on formalin-fixed, paraffin-embedded cell were analyzed with FlowJo software (FlowJo). pellets or tissue sections mounted on glass slides. N87, N87-TR, BT-474, and BT-474-TR cell pellets were stained using the Western blotting PATHWAY anti-HER-2/NEU (4B5) rabbit monoclonal primary EPHA2 degradation experiments and immunoblotting were antibody (790-2991, Ventana Medical Systems). PDX tumor performed as described previously (36), using anti-ECK/EPHA2 tissue sections were stained using a HER2/NEU (SP3) rabbit mAb antibody clone D7 (05-480) and GAPDH (G-8795) antibodies (237R-16-ASR, Cell Marque). from Sigma-Aldrich. Images were captured using an ImageQuant LAS 4000 mini (GE Healthcare). In vitro cytotoxicity assays Cells were seeded into white-walled 96-well plates at a density Generation of SLC46A3-expressing cells of 3,000 cells per well. Treatments were added in triplicate the next Full-length human SLC46A3 with a C-terminus Rho1D4 tag day and cell viability was determined 3 to 6 days later, using the was amplified by PCR from TrueORF Gold-validated human Cell Titer-Glo Luminescent Cell Viability Assay Kit (G7572, cDNA (SLC46A3-NM_181785; OriGene) with the forward Promega). Luminescence was measured with an EnVision 2104 primer [50CCATAGAAGATTCTAGA (GCTAGC) ATGAAGATTTT- Multilabel Reader (Perkin Elmer). Cell viability was calculated as a ATTTGTAGAACCTGCCATTTTCC3'] and reverse primer percentage of control untreated cells, and data were analyzed with [50GATCGCAGATCCTT (GCGGCCGC) TTAGGCAGGGGCC- GraphPad Prism software v7.02 for Windows, using the log ACCTGAGATGTCTCGGTCCTGTCTGAAGCATCTTCACTGG3'], [inhibitor] versus response model. and cloned into a pCDH1-CMV-MCS1-EF1-Puro vector (CD510B-1, System Biosciences). Lentiviruses were made with RNA isolation and mRNA expression analysis the pPACKH1 HIV Lentivector Packaging Kit (LV510A-1, System RNA was isolated from cell lines with an RNeasy Plus Mini Kit Biosciences). Cells were transduced with lentivirus and incubated (74134, Qiagen), and cDNA was generated using SuperScript III for 4 days, at which time cells were split into T75 flasks and First-Strand Supermix (18080400, Thermo Fisher Scientific). Each were selected in 1 mg/mL puromycin (A1113803, Thermo Fisher cDNA was preamplified using TaqMan PreAmp Master Mix Scientific) before use. (4391128, Thermo Fisher Scientific), diluted 1:10 in DNA sus- pension buffer (T0220, Teknova) and loaded onto a Fludigm Generation of SK-BR-3 KO Cells dynamic array for analysis on the Fludigm Biomark. Relative gene CRISPR crRNA sequences against human SLC46A3 exon2 (– DCt) expression was calculated as 2 1,000, where DCt is the Ct (NM_181785) were identified with the CRISPR design tool value for each gene minus the Ct value of GAPDH for each sample. (crispr.mit.edu; ref. 37). Sequences of two appropriate crRNAs Ct values of >25 were considered out of range. All TaqMan assays for use in an all-in-one dual nickase CRISPR-Cas9n plasmid were purchased from Thermo Fisher Scientific, including containing a GFP tag (pD1401-AD; ATUM) were chosen (38). Hs99999905_m1 (GAPDH), Hs03045080_m1(TNFRSF17), and SK-BR-3 cells were transfected with Lipofectamine 3000 reagent, assays listed in Supplementary Table S1. and GFP-positive cells were collected by cell sorting on a FACSAria Fusion Cell Sorter (BD Biosciences) 48 hours after transfection. Flow cytometry Cell populations resulting from limited dilution cloning were Cells were stained on ice with their respective antibodies at screened by Sanger sequencing and qRT-PCR to confirm CRISPR concentrations indicated in figure legends, followed by detection knockout of SLC46A3. Relative quantitation was performed with with goat anti-human IgG (HþL)–AF647 (A-21445, Thermo TaqMan assays Hs01308309_cn (SLC46A3) and Hs99999901_s1 Fisher Scientific). Fluorescence of live, single cells was determined (18S; Thermo Fisher Scientific). on a BD-LSRII and data were analyzed using FlowJo software (FlowJo). Analysis of microarray data Raw gene expression data from primary multiple myeloma Internalization bone marrow samples (39) was obtained from the Gene Expres- Cells were incubated on ice with ADCs at a concentration of sion Omnibus (accession no. GSE6477) and normalized with the 5 mg/mL. Cells were washed and suspended in RPMI þ 10% FBS frozen robust multiarray analysis (40), using the R statistical and were either placed on ice (time 0) or into an incubator set software package. In cases where multiple Affymetrix probes to 37C. At desired time points, cells were removed from the mapped to an individual gene, probes with the highest inter- incubator and placed on ice. At the end of the experiment, cells quartile range were selected as representative for that gene. This

www.aacrjournals.org Clin Cancer Res; 2018 OF3

Kinneer et al.

resulted in an expression dataset across unique genes. To identify resistant cells, we hypothesized that catabolites of 1C1-SG3376 genes that were differentially expressed among relapsed/refractory may rely upon active transport to leave the lysosome and that multiple myeloma, newly diagnosed multiple myeloma, and resistant cells lack the transporter(s) necessary for lysosomal healthy donors, ANOVA was run for each gene across disease escape. We examined 43 target genes focused on lysosomal groups. Tukey Honest Signiﬁcant Difference test was applied membrane transporter genes (42) by multiplex RT-PCR, using to determine pairwise statistical differences between groups. TaqMan gene expression assays on a 96.96 Fluidigm Dynamic Fold changes for each gene were also calculated between pairs Array to identify genes with expression levels correlating with cell of disease groups to determine the magnitude of change of killing. Of the 43 genes examined, only the expression of the expression of each gene in the dataset. P values resulting from lysosomal transporter SLC46A3 correlated with 1C1-SG3376 this analysis were adjusted for multiple comparisons by deter- sensitivity in our cell panel (Supplementary Table S1; Fig. 2E), mining the FDR for each gene, using the Benjamini–Hochberg suggesting that the transporter may play a role in sensitivity of a procedure (41). cell to the noncleavable ADC.

Statistical analysis Silencing of SLC46A3 alters the potency of SG3376 and DM1 Data are presented as mean SD or mean SEM as stated ADCs in the figure legends. Statistical analysis was performed with To further define the role of SLC46A3 in SG3376 ADC activity GraphPad Prism software v7.02 for Windows. Statistically and to determine whether the above findings were target- or significant differences were tested by using specific tests as payload-specific, we conjugated trastuzumab to cleavable linker indicated in the figure legends. P < 0.05 was considered (SG3249) or noncleavable linker (SG3376 and MMAF) payloads. statistically significant. We then tested their cytotoxic activity along with T-DM1 in SK-BR- 3 cells and in SK-BR-3 knockout cells lacking SLC46A3 (SK-BR-3 KO). Sequencing and quantitative PCR confirmed complete Results knockout of SLC46A3 (Supplementary Fig. S1A), and flow cyto- Innate resistance to ADCs containing the noncleavable linker metry was used to confirm that HER2 levels were similar in the SK- drug SG3376 BR-3 and SK-BR-3 KO cells (Supplementary Fig. S1B). Knockout The Eph receptor A2 (EPHA2) tyrosine kinase is expressed on of SLC46A3 markedly decreased the potency of T-SG3376 and the surface of tumor cells and can be specifically targeted by the T-DM1 (Fig. 3A and B), but did not affect the potency of T-SG3249 human mAb 1C1 for the delivery of cytotoxic warheads (36). We (Fig. 3C). This finding suggests that the role of SLC46A3 is neither used a panel of EPHA2-positive cell lines to compare the target- nor payload-dependent, despite the structural dissimilarity cytotoxic activity of ADCs that had been prepared by conjugating between PBDs (Fig. 1A) and maytansinoids (Fig. 3D). Loss of antibody 1C1 to PBD payloads containing cleavable (1C1- SLC46A3 expression had no impact on the potency of T-MMAF SG3249) or noncleavable (1C1-SG3376) linkers (Fig. 1A). (Fig. 3E). This result is in agreement with previously published Although most cell lines were sensitive to both ADCs, Hep G2 data (12) and suggests that not all noncleavable linker drugs and T24 were uniquely resistant to the noncleavable ADC (1C1- require SLC46A3 for their activity. Similar to SG3249, SG3541, a SG3376; Fig. 1B). Resistance to the noncleavable ADC could not protease-cleavable analogue of SG3376, was conjugated to tras- be explained by target expression, because T24 and ES-2 cells tuzumab (T-SG3541; Fig. 3D) and was found to be active in both express nearly equivalent levels of EPHA2 (MFI ratio, 68.9 and parental SK-BR-3 and SK-BR-3 KO cells lacking SLC46A3 (EC50, 72.2 for T24 and ES-2, respectively), but exhibit strikingly differ- 2.8 and 2.1 ng/mL, respectively; Fig. 3F). ent sensitivities to the noncleavable ADC (Fig. 1B and C). Diminished SLC46A3 expression in cells with acquired T-DM1 Expression of SLC46A3 correlates with sensitivity to resistance noncleavable 1C1-SG3376 ADC Given the role of SLC46A3 in the transport of T-DM1 catabo- Effective ADCs bind to their intended target on the tumor cell lites from the lysosome, we hypothesized that loss of this trans- surface and are subsequently internalized and trafficked to the porter may be a mechanism of acquired resistance. T-DM1– lysosome for efficient release of the cytotoxic warhead (1). There- resistant cell lines were established by treating NCI-N87 and fore, we next used flow cytometry to examine binding and BT-474 cells with T-DM1 in vitro at gradually increasing concen- internalization of 1C1-SG3249 and 1C1-SG3376 in T24 and trations until stable resistant lines emerged; these were named ES-2 cells. 1C1-SG3376 and 1C1-SG3249 bound equally well to N87-TR and BT-474-TR (35). These cells are refractory to at least both cell lines (Fig. 2A) and were similarly internalized (Fig. 2B). 1 mg/mL T-DM1 in vitro but are sensitive to T-MMAF (Supple- Intracellular trafficking was assessed by labeling ADCs with a mentary Fig. S2), suggesting that lysosomal proteolytic function is pH-sensitive dye that only fluoresces when exposed to low-pH retained in these cells. Although antigen loss is a common environments and using flow cytometry to monitor internaliza- mechanism of acquired resistance to T-DM1 in preclinical models tion kinetics. MFI increased over time (Fig. 2C), indicating (43–45), we confirmed that both cell lines maintained a high level that both ADCs were internalized and delivered to the acidic of HER2 (HER2 3þ), like the parental cells as measured by IHC endosomal/lysosomal compartment. Finally, EPHA2 protein (Fig. 4A) and flow cytometry (Fig. 4B). Using qRT-PCR, SLC46A3 degradation was observed after treatment of T24 and ES-2 cells was nearly undetectable in both N87-TR and BT-474-TR cells as with 1C1-SG3249 and 1C1-SG3376 (Fig. 2D), demonstrating compared with their parental lines (Fig. 4C). that antibody–EPHA2 complexes were delivered to, and pro- We reasoned that if loss of SLC46A3 expression is the primary cessed by, the lysosome in both cell lines. mechanism of T-DM1 resistance in these cells, they would also Because the binding and intracellular trafficking of the 1C1- be cross-resistant to T-SG3376. Indeed, cell killing assays showed SG3249 and 1C1-SG3376 ADCs were similar for the sensitive and that, although T-SG3376 ADCs were active in N87 and BT-474

OF4 Clin Cancer Res; 2018 Clinical Cancer Research

SLC46A3 as a Potential Biomarker for Noncleavable ADCs

Figure 1. Differential cell killing by cleavable and noncleavable PBD ADCs in a panel of cell lines. A, Illustration of an ADC prepared with the cleavable linker drug, SG3249 (top), and the noncleavable linker drug, SG3376 (bottom). B, EPHA2-positive cells were incubated with 0.5 mg/mL 1C1-SG3249 or 1C1-SG3376 for 96 hours before analysis with CellTiter Glo. Bars represent the maximum killing observed in each cell line as compared with media control. Data from three independent experiments are shown. Error bars indicate SD. EPHA2 expression was measured by flow cytometry. Cells were incubated with 5 mg/mL isotype control or anti-EPHA2 antibody 1C1 and detected with antihuman AF647 antibodies. Numbers in parentheses indicate ratio of MFI of EPHA2 stained to isotype control–stained cells. C, T24 and ES-2 cells were incubated with increasing concentrations of 1C1-SG3249 or 1C1-SG3376, along with the respective isotype (IgG) controls. Cell viability is shown as relative percentage to the untreated controls. Error bars, SD. Representative figures from three independent experiments are shown. parental cell lines, they were completely inactive in the T-DM1– BT-474 transfectants augmented maximum cell killing and resistant N87-TR and BT-474-TR lines (Fig. 4D and E). We next potency for the T-SG3376 ADCs (Fig. 4D and E), but not for forced expression of SLC46A3 into both T-DM1–resistant (TR) the T-DM1 ADCs (Fig. 4F and G). Flow cytometry confirmed that and parental cells and evaluated the cytotoxicity of T-DM1 and T- HER2 levels were not increased by forced expression of SLC46A3 SG3376. Lentiviral expression of SLC46A3 into N87-TR and BT- (Fig. 4B), and therefore, the increased activity observed in 474-TR cells restored sensitivity to T-DM1 and T-SG3376 (Fig. the SLC46A3 transfectants cannot be explained by an increase 4D–G). Notably, the enhanced levels of SLC46A3 in the N87 and in targeted antigen. Similar data were obtained using the ADC

www.aacrjournals.org Clin Cancer Res; 2018 OF5

Kinneer et al.

A B 8,000 T24 Cells: 100 T24 Cells: 1C1-SG3249 1C1-SG3249 6,000 1C1-SG3376 75 1C1-SG3376

4,000 ES-2 Cells: ES-2 Cells: MFI 50 1C1-SG3249 1C1-SG3249 1C1-SG3376 1C1-SG3376 % Internalization 2,000 25

0 0 0.01 0.1 110100 012345 Antibody [mg/mL] Time (hours) CD T24 Cells: 2,500 IgG-SG3249-pHast 1C1-SG3249-pHast 2,000 IgG-SG3376-pHast No treatment No treatment IgG-SG3376 IgG-SG3249 IgG-SG3376 IgG-SG3249 1C1-SG3376 1C1-SG3376 1C1-SG3249 1C1-SG3376-pHast 1C1-SG3249 1,500 EPHA2 ES-2 Cells: T24 MFI IgG-SG3249-pHast 1,000 GAPDH 1C1-SG3249-pHast IgG-SG3376-pHast 500 EPHA2 1C1-SG3376-pHast ES-2

0 GAPDH 0 1 2 3 45 0.5 Hours 6 Hours E Time (hours)

1.0 SLC46A3

0.5 Relative expression 0.0 T24 ES-2 PC-3 A549 HT29 RT112 Du145 Hep G2 HCT 116 Panc 02.03

Figure 2. ADC trafficking and expression of SLC46A3 in SG3376-sensitive and -resistant cells. A, Binding of 1C1-SG3249 and 1C1-SG3376 to T24 and ES-2 cells was evaluated by flow cytometry. Cells were incubated with increasing concentrations of 1C1-SG3249 or 1C1-SG3376, followed by detection with antihuman-AF647 secondary antibodies. B, Internalization was determined by incubating cells with 5 mg/mL 1C1-SG3249 or 1C1-SG3376 ADCs on ice, followed by washes and incubation in cell culture media at 37C for indicated time points. The amount of antibody remaining on the cell surface over time was detected by incubation with antihuman-AF647 secondary antibodies. Percent internalization relative to time 0 is shown on the y-axis. C, Isotype control (IgG) and 1C1-ADCs labeled with a pH-sensitive dye were incubated with T24 and ES-2 cells for indicated time points and fluorescence quantified by flow cytometry. D, EPHA2 and GAPDH protein levels in T24 and ES-2 cells treated with 5 mg/mL ADCs, unarmed antibodies, or isotype controls for the indicated times were measured by Western blot analysis. E, Relative expression of 43 lysosomal target genes was examined by qRT-PCR, using TaqMan gene expression assays on a 96.96 Fluidigm Dynamic Array; data for SLC46A3 is shown. The 1C1-SG3376–resistant cell lines Hep G2 and T24 had undetectable levels of SLC46A3. Relative gene expression is indicated on the y-axis. Error bars, SD.

OF6 Clin Cancer Res; 2018 Clinical Cancer Research

SLC46A3 as a Potential Biomarker for Noncleavable ADCs

Figure 3. SLC46A3-dependent cytotoxicity of noncleavable SG3376 and DM1 ADCs. Parental SK-BR-3 (solid lines) and SK-BR-3 KO cells lacking SLC46A3 (dashed lines) were incubated with increasing concentrations of T-SG3376 (A), T-DM1 (B), T-SG3249 (C), T-MMAF (E), or T-SG3541 (F), along with the respective isotype (IgG) controls. Cell viability is shown as relative percentage to the untreated controls. Error bars, SD. Representative ﬁgures from three independent experiments are shown. D, Illustration of an ADC prepared with the noncleavable linker drugs, DM1 (top) and MMAF (center), and the cleavable linker-drug, SG3541 (bottom).

www.aacrjournals.org Clin Cancer Res; 2018 OF7

Kinneer et al.

A BT-474 BT-474-TR N87 N87-TR

0.3

IgG, N87 IgG, BT-474 SLC46A3 0.2

N87 BT-474

N87- TR BT-474-TR 0.1

N87- TR-SLC46A3 BT-474-TR-SLC46A3

N87- SLC46A3 BT-474-SLC46A3 Relative expression 0.0 2 3 4 5 2 4 5 0 10 10 10 10 0 10 103 10 10 BT-474 BT-474-TR N87 N87-TR APC-A APC-A D E N87 BT-474 IgG-SG3376 IgG-SG3376 T-SG3376 T-SG3376 N87-TR BT-474-TR 100 IgG-SG3376 100 IgG-SG3376 T-SG3376 T-SG3376 N87-TR-SLC46A3 BT-474-TR-SLC46A3 IgG-SG3376 IgG-SG3376 50 50 T-SG3376 T-SG3376 N87-SLC46A3 BT-474-SLC46A3 % Viability vs. untreated control % Viability IgG-SG3376 vs. untreated control % Viability IgG-SG3376 0 T-SG3376 0 T-SG3376 -4 -2 0 -4 -2 0 Log [mg/mL] Log [mg/mL] F G N87 BT-474 IgG-DM1 IgG-DM1 T-DM1 T-DM1 N87-TR BT-474-TR 100 IgG-DM1 100 IgG-DM1 T-DM1 T-DM1 N87-TR-SLC46A3 BT-474-TR-SLC46A3 IgG-DM1 IgG-DM1 50 50 T-DM1 T-DM1 N87-SLC46A3 BT-474-SLC46A3

% Viability vs. untreated control % Viability IgG-DM1 IgG-DM1 % Viability vs. untreated control % Viability 0 T-DM1 0 T-DM1 -4 -2 0 -4 -20 Log [mg/mL] Log [mg/mL]

Figure 4. Loss of SLC46A3 expression in cells with acquired T-DM1 resistance. A, IHC results of HER2 staining in parental and T-DM1–resistant (TR) cells. Representative micrographs are shown at 20 magnification. B, HER2 levels in parental and T-DM1–resistant (TR) cells, either unmodified or with forced expression of SLC46A3 (–SLC46A3), were measured by flow cytometry. Cells were incubated with 5 mg/mL trastuzumab or isotype control (IgG) antibodies, followed by detection with antihuman-AF647. C, Expression of SLC46A3 was measured by qRT-PCR. Relative gene expression is indicated on the y-axis. Error bars, SD (n ¼ 4). D–G, N87, N87-TR, BT-474, and BT-474-TR cells, with and without forced expression of SLC46A3, were incubated with increasing concentrations of T-SG3376 (D and E) or T-DM1 (F and G), along with the respective isotype controls. Plates were incubated for 6 days before analysis with CellTiter Glo. Cell viability is shown as relative percentage to the untreated controls. Error bars, SD. Representative figures from three independent experiments are shown.

OF8 Clin Cancer Res; 2018 Clinical Cancer Research

SLC46A3 as a Potential Biomarker for Noncleavable ADCs

A ST1616B ST1616B/TDR Control Control ) 3 2,250 2,250 T-DM1: 5 mg/kg, T-DM1: 10 mg/kg, Figure 5. Q7Dx4 Q7Dx6 SLC46A3 expression in a HER2þ PDX 1,500 1,500 model of T-DM1–acquired resistance. A, Antitumor efﬁcacy of T-DM1 in paired PDX models of 750 750 T-DM1–sensitive (ST1616B) and

T-DM1–resistant (ST1616B/TDR) Tumor volume (mm invasive ductal carcinoma. Mice 0 0 bearing ST1616B or ST1616B/TDR PDX 0 7 14 21 28 35 42 0714 21 28 35 42 tumors were treated with T-DM1 at Time (days) Time (days) 5 mg/kg on a Q7D 4 regimen (ST1616B, left) or 10 mg/kg on a Q7D BC 6 regimen (ST1616B/TDR, right). ST1616B ST1616B/TDR 0.4 Control groups were untreated. Mean tumor volumes SEM are shown. B, IHC results of HER2 staining in 0.3 ST1616B and ST1616B/TDR

tumor tissue. Both models were SLC46A3 HER2-positive (3þ). Images were 0.2 taken with a 40 objective. C, Expression of SLC46A3 was measured by qRT-PCR in ST1616B (n ¼ 7) and 0.1 ST1616B/TDR (n ¼ 11) tumors. Relative gene expression is indicated on the y- axis. Error bars, SD. , P < 0.05 by 0.0 unpaired t test with Welch correction. Relative expression

-0.1 ST1616B ST1616B/TDR

1C1-SG3376 in N87 cells, which are positive for EPHA2 available Cancer Cell Line Encyclopedia database (ref. 47; Sup- (Supplementary Fig. S3). plementary Fig. S4) showed that breast cancer cell lines expressed relatively high levels of SLC46A3, whereas cell lines derived from Diminished SLC46A3 expression in a T-DM1–resistant PDX prostate, multiple myeloma, esophageal, and upper aerodigestive model tract malignancies were among the lowest expressers. In agree- A lung biopsy collected from a patient 1 month after relapse ment with these data, we measured SLC46A3 expression in a panel þ from a 13-month response to T-DM1 therapy was used to generate of 16 HER2 cell lines by qRT-PCR and found only the esophageal a patient-derived xenograft (PDX) model in immunocompro- squamous-cell carcinoma line, KYSE-410, to be without expres- mised mice (46). T-DM1 was found to be efficacious in the parent sion of this transporter (Supplementary Fig. S5). In contrast, five ST1616B model, but chronic treatment of the model with T-DM1 of nine (56%) multiple myeloma cell lines had undetectable over three passages in mice resulted in a T-DM1–insensitive SLC46A3 (Fig. 6A). model, designated ST1616B/TDR that recapitulated the drug B-cell maturation antigen (BCMA) is universally expressed in resistance observed in the patient. Weekly administration of multiple myeloma (16, 48) and is being targeted in clinical trials 5 mg/kg T-DM1 caused tumor regression in the sensitive model with ADCs bearing the noncleavable linker drugs, DM1 and (ST1616B), whereas weekly dosing of T-DM1 at a higher dose MMAF (14, 15). Thus, the expression of BCMA and SLC46A3 in þ of 10 mg/kg was inactive in the resistant (ST1616B/TDR) model multiple myeloma cell lines and in CD138 cells isolated from (Fig. 5A). We next evaluated HER2 expression in the sensitive and the bone marrow of a patient with multiple myeloma was resistant PDX tumors by IHC (Fig. 5B). Both models expressed a measured by qRT-PCR. Expression of BCMA varied across the high level of HER2, ruling out antigen loss as a mechanism for this primary multiple myeloma samples, only 9 of 99 (9.1%) of resistance. Finally, we isolated RNA from ST1616B and ST1616B/ which expressed levels as high as the multiple myeloma cell line, TDR tumor tissue and measured SLC46A3 gene expression by NCI-H929 (Fig. 6B). We detected a wide range of SLC46A3 qRT-PCR. Average levels of SLC46A3 expression in the ST1616B/ expression in the primary multiple myeloma samples, including TDR tumors were reduced by 92% as compared with the sensitive several samples with undetectable levels of SLC46A3 (Fig. 6C). ST1616B model (Fig. 5C). Likewise, we analyzed gene expression data from a recently published study (39) and found that SLC46A3 expression was SLC46A3 as a potential patient selection biomarker for DM1 significantly decreased in newly diagnosed (1.8-fold decrease, and SG3376 ADCs FDR < 0.001) and relapsed/refractory MM (2.2-fold decrease, Although loss of SLC46A3 expression may be a mechanism of FDR < 0.001) as compared with bone marrow taken from healthy acquired resistance to T-DM1, the frequency of SLC46A3-driven donors (Fig. 6D). We then prepared anti-BCMA ADCs conjugated þ innate resistance in HER2 tumors is unclear. Comparison of with either cleavable linker (SG3249) or noncleavable linker mRNA expression across several tumor types in the publicly drugs (SG3376, DM1, and MMAF) and measured their cytotoxic

www.aacrjournals.org Clin Cancer Res; 2018 OF9

Kinneer et al.

A B 400 400 300 300 SLC46A3 NCI-H929 200 100 BCMA Figure 6. 150

60 BCMA SLC46A3 levels and differential 40 cytotoxicity of anti-BCMA ADCs 20 100 bearing noncleavable linker drugs 0.6 DM1, SG3376, and MMAF in MM. A–C, Relative expression 0.4 50 Expression of SLC46A3 and BCMA 0.2 0.0 JJN-3 was measured by qRT-PCR in a 0 panel of multiple myeloma (MM) cell EJM Relative expression JJN-3 OPM2 þ MM.1S MM.1R n ¼ U266B1 lines (A, 9) and CD138 multiple NCI-H929 RPMI 8226 KMS-12-BM myeloma patient cells (B and C, Primary MM samples n ¼ 99). Relative gene expression is C D indicated on the y-axis. Dashed 40 9 20 horizontal lines in B and C indicate relative expression level for 4 NCI-H929 or JJN-3 cells. D, SLC46A3 8 Expression of SLC46A3 in newly diagnosed (n ¼ 73) and 3 expression 2 relapsed/refractory multiple myeloma (n ¼ 28) was compared 7 1 with that in healthy donors (n ¼ 15)

Relative expression NCI-H929 from GSE6477. Horizontal bars SLC46A3 Log

0 JJN-3 indicate median log2-scale gene 6 expression levels for each group, with statistically signiﬁcant () Healthy MM: MM: relapsed/ differences having an FDR of <0.05; Primary MM samples donor at diagnosis refractory ns, not statistically signiﬁcant (FDR >0.05). E and F, NCI-H929 and E NCI-H929 MM.1R MM.1R (E), or JJN-3 and JJN-3 with 150 150 forced expression of SLC46A3 IgG-SG3249 (JJN-3-SLC46A3; F) were incubated BCMA-ADC-SG3249 with increasing concentrations of 100 100 IgG-SG3376 anti-BCMA ADCs conjugated with BCMA-ADC-SG3376 the cleavable linker-drug, SG3249, IgG-DM1 or the noncleavable linker drugs, 50 50 BCMA-ADC-DM1 SG3376, DM1, and MMAF. Plates were incubated for 96 hours before IgG-MMAF BCMA-ADC-MMAF analysis with CellTiter Glo. Cell

% Viability vs. untreated control % Viability 0 0 viability is shown as relative -6 -4 -2 0 -6 -4 -2 0 percentage to the untreated m m Log [ g/mL) Log [ g/mL) controls. Error bars, SD. Figures are F JJN-3 (MFIr 1.3) JJN-3-SLC46A3 (MFIr 1.2) representative of three independent experiments. BCMA expression in 150 150 JJN-3 and JJN-3-SLC46A3 (F)was IgG-SG3249 measured by ﬂow cytometry. Cells BCMA-ADC-SG3249 were incubated with 5 mg/mL 100 100 IgG-SG3376 isotype control or anti-BCMA BCMA-ADC-SG3376 antibody and detected with IgG-DM1 antihuman AF647 antibodies. 50 50 BCMA-ADC-DM1 Numbers in parentheses indicate IgG-MMAF ratio of MFI of BCMA stained to BCMA-ADC-MMAF isotype control–stained cells. 0 % Viability vs. untreated control % Viability - - - 0 6 4 2 0 -6 -4 -2 0 Log [mg/mL) Log [mg/mL)

activity in SLC46A3-positive NCI-H929 or SLC46A3-negative compared with parental cells (MFI ratio, 1.3 and 1.2 in JJN-3 MM.1R and JJN-3 MM cell lines. The noncleavable DM1 and and JJN-3–SLC46A3 cells, respectively) but rendered them SG3376 ADCs were cytotoxic to the NCI-H929 cells, but were highly sensitive to BCMA-Ab–SG3376, shifting the EC50 of inactive in the MM.1R and JJN-3 cells, whereas all cell lines were this ADC from >500 ng/mL in JJN-3 cells to 5.5 ng/mL in the sensitive to ADCs prepared with SG3249 and MMAF (Fig. 6E and JJN-3–SLC46A3 cells (Fig. 6F). Likewise, the cytotoxic activity of F). Next, we forced expression of SLC46A3 in the JJN-3 cell line BCMA-Ab–DM1 was markedly improved in JJN-3–SLC46A3 cells (JJN-3–SLC46A3) to evaluate the impact of increased SLC46A3 (EC50, 441 ng/mL) as compared with the parental cells, where it expression in a cell line with BCMA levels that were more repre- was completely inactive. The potency of the BCMA-Ab–SG3249 sentative of primary multiple myeloma (49). Introduction of and BCMA-Ab–MMAF ADCs was unchanged, having an EC50 SLC46A3 into JJN-3 cells did not alter surface BCMA levels as of 8–10 ng/mL for BCMA-Ab–SG3249 and 13–21 ng/mL for

OF10 Clin Cancer Res; 2018 Clinical Cancer Research

SLC46A3 as a Potential Biomarker for Noncleavable ADCs

BCMA-Ab–MMAF in JJN-3 and JJN-3–SLC46A3 cells, respectively transporter expression than SG3376 to deliver a therapeutically (Fig. 6F). active amount of drug to the cytosol. T-DM1 is conjugated through lysines and T-SG3376 is conjugated through cysteines, Discussion which could suggest that lysine is a better substrate than cysteine for SLC46A3. This could also be explained by the differences in Lysosomes are highly acidic, intracellular organelles that func- the mechanisms of action of DM1 and SG3376. In contrast to tion in numerous physiologic processes, including breakdown of DM1, whose biological target (microtubules) resides in the cyto- macromolecules, autophagy, plasma membrane repair, and sol, the catabolites of SG3376 must be further trafficked from the metabolism. ADCs bearing noncleavable linker drugs rely upon cytosol and enter the nucleus through the nuclear pore complexes lysosomal degradation of the antibody to release an amino acid to bind to DNA. Increased levels of SG3376 in the cytosol, linker warhead, which must then escape the lysosome to bind its facilitated by the increase in SLC46A3 in the lysosome, would intracellular target (11). The lysosomal membrane transporter, enable the diffusion and accumulation of drug within the nucleus, SLC46A3, was recently identified to regulate the efflux of non- leading to a more potent biological effect. Small molecules with a cleavable DM1 catabolites from the lysosome to the cytoplasm molecular mass of less than 30–60 kDa (51, 52) can pass through (12) and contribute to T-DM1 resistance in vitro (24). Herein, we the nuclear pore complex by passive diffusion. SG3376 and its report that SLC46A3 is also necessary for cytotoxic activity of predicted cysteine-adduct catabolite have a molecular mass of noncleavable ADCs constructed with a PBD payload, SG3376, 1.070 and 1.191 kDa, respectively. Therefore, it is unlikely which is structurally and mechanistically distinct from DM1. that active transport regulates the influx of SG3376 catabolites Furthermore, we show that loss of SLC46A3 expression leads to into the nucleus. acquired and innate resistance to these ADCs. The phase III TH3RESA trial of T-DM1 in patients with HER2- Modifications to the basic components of an ADC (antibody, positive breast cancer previously treated with two or more HER2- linker, or warhead) can influence the efficacy of the molecule, targeted therapies demonstrated an objective response rate of including its activity in drug-resistant settings. 31%, suggesting that at least in some patients, HER2 remains an For example, replacement of a noncleavable linker with a attractive target for advanced-stage breast cancer (53). In some protease-cleavable linker could overcome acquired T-DM1 resis- patients, this disease is refractory or innately resistant to T-DM1 tance in an in vitro–derived T-DM1–resistant cell line, 361-TM, therapy, and the use of archival diagnostic tumor tissue for even when warheads with a similar mechanism of action are used determination of HER2 expression in the TH3RESA trial is an (43). Similarly, we found that SG3541, a protease-cleavable obstacle for determining the underlying mechanisms of T-DM1 analogue of SG3376, exerted full activity in SLC46A3-negative resistance (54). Most patients who initially respond to T-DM1 cell lines that are resistant to the noncleavable linker-drug, treatment eventually relapse, despite continued treatment (17). A SG3376. Likewise, the cleavable linker-drug, SG3249 (tesirine), tumor from one such patient was used to derive a PDX model that which releases the warhead SG3199 having the same DNA cross- offered a unique opportunity to examine SLC46A3 levels in a linking mechanism of action as SG3376 and SG3541, was equally clinically relevant model with known levels of HER2 expression active in SLC46A3-positive and SLC46A3-negative cell lines. and T-DM1 sensitivity. Although this model indicated that loss of Modification of the warhead to an alternate mechanism of SLC46A3 expression is a mechanism of acquired resistance, the action has also been shown to overcome drug resistance and is expression of HER2 and SLC46A3 in additional clinical biospeci- routinely employed by next-generation ADCs. For example, non– mens collected before and after disease progression on T-DM1 Hodgkin lymphoma tumor models with acquired resistance will need to be evaluated to confirm these findings. to anti–CD22-vc-MMAE have been found to be sensitive to an Cells that lack SLC46A3 expression are resistant to noncleava- ADC-delivering the DNA inhibitor–based warhead, PNU- ble DM1 and SG3376 ADCs, regardless of the targeted antigen. 159682, via the same cleavable linker previously used with MMAE Therefore, SLC46A3-driven resistance is not likely to be confined (50). The anti-CD33 ADC SGN-CD33A, which bears the PBD to T-DM1 but may instead be a shared mechanism of resistance payload talirine, has been shown to be cytotoxic to MDR1- among ADCs bearing these payloads. Multiple myeloma is cur- positive cells that are resistant to the CD33-targeting ADC gem- rently being targeted in clinical phase I trials by two anti-BCMA tuzumab ozogamicin, which bears a calicheamicin warhead (26). ADCs conjugated via a noncleavable linker to either DM1 (14) or In this study, the T-DM1–resistant cell models, BT-474-TR and MMAF (55). In this study, SLC46A3-negative multiple myeloma N87-TR, demonstrated that acquired resistance may span distinct cells were resistant to BCMA-targeting ADCs with DM1 and warhead classes with unique mechanisms of action. A key com- SG3376 payloads, but were sensitive to those with SG3249 or mon feature of SG3376 and DM1 catabolites appears to be an MMAF. In primary multiple myeloma bone marrow samples, a exposed terminal amino acid (i.e., lysine, cysteine), but this range of SLC46A3 expression was evident, including several feature is also shared in the catabolite of MMAF ADCs, which samples with little to no detectable transcript. Our data suggest were not reliant on SLC46A3 for activity. This latter fact suggests that SLC46A3-low or SLC46A3-negative multiple myeloma may that further work is necessary to determine what shared compo- be resistant or less responsive to treatment with ADCs bearing nent(s) of these catabolites is being recognized by SLC46A3. DM1 or SG3376. Conversely, high levels of SLC46A3 may indicate We noticed that forced SLC46A3 expression in the N87, N87- patients in whom DM1 or SG3376 ADCs may be more efficacious TR, BT-474, and BT-474-TR cell lines enhanced maximum cell and may be more predictive of activity than target antigen expres- killing and potency, as compared with parental lines, for sion alone. This is exemplified by the JJN-3–SLC46A3 model, in T-SG3376 but not T-DM1 ADCs. Because HER2 antigen levels which the introduction of SLC46A3 produced a dramatic increase were similar across these cell lines, these data may indicate in sensitivity to BCMA-Ab–DM1 and BCMA-Ab–SG3376 ADCs as that catabolites of DM1 are better substrates than SG3376 compared with parental JJN-3 cells, which were refractory to these for SLC46A3 transport, thereby requiring a lower threshold of ADCs. Our data suggest that SLC46A3 levels may be an important

www.aacrjournals.org Clin Cancer Res; 2018 OF11

Kinneer et al.

predictor of sensitivity, particularly in cells expressing low levels of Analysis and interpretation of data (e.g., statistical analysis, biostatistics, surface antigen. Taken together, our findings support SLC46A3 as computational analysis): K. Kinneer, J. Meekin, Y. Tai, C.M. Kiefer, N. Dimasi, a potential patient selection biomarker with immediate relevance S. Sridhar, M.J. Wick, P.W. Howard, D.A. Tice Writing, review, and/or revision of the manuscript: K. Kinneer, to ongoing or planned clinical trials involving noncleavable DM1 A.C. Tiberghien, Y. Tai, S. Phipps, C.M. Kiefer, M.C. Rebelatto, N. Dimasi, and SG3376 ADCs, regardless of their target antigens and tumor A.D. Moriarty, K.P. Papadopoulos, S. Sridhar, S.J. Gregson, M.J. Wick, types. L.A. Masterson, K.C. Anderson, R. Herbst, P.W. Howard, D.A. Tice Administrative, technical, or material support (i.e., reporting or organizing Disclosure of Potential Conflicts of Interest data, constructing databases): A.D. Moriarty K. Kinneer is an employee of MedImmune and holds ownership interest Study supervision: K. Kinneer, N. Dimasi, M.J. Wick, D.A. Tice Responsible for IHC assays and interpretation of staining: M.C. Rebelatto (including patents) in AstraZeneca. J. Meekin is an employee of MedImmune Provision of synthetic compounds used in the study: L.A. Masterson and holds ownership interest (including patents) in AstraZeneca. S. Phipps holds ownership interest (including patents) in AstraZeneca. C.M. Kiefer holds ownership interest (including patents) in AstraZeneca. M.C. Rebelatto Acknowledgments holds ownership interest (including patents) in MedImmune and AstraZeneca. We gratefully acknowledge Binyam Bezabeh, Ben Ruddle, and Ryan N. Dimasi holds ownership interest (including patents) in AstraZeneca. Fleming (MedImmune) for the preparation and analytical characterization S. Sridhar holds ownership interest (including patents) in AstraZeneca. of the ADCs used in this study; Terrence O'Day (MedImmune) for statistical R. Herbst is an employee of MedImmune. P.W. Howard holds ownership support; and Radhika Rayanki (MedImmune) for carrying out flow cytometric interest (including patents) in AstraZeneca. No potential conflicts of interest sorting of the SK-BR-3 KO cells. We thank Deborah Shuman (MedImmune) were disclosed by the other authors. for careful review and editing of the manuscript and Sepi Farshadi (MedIm- mune) for help with the preparation of figures. We also thank Megan Groves Authors' Contributions (START) for her technical contributions to the PDX studies. This study was funded by MedImmune, the global biologics R&D arm of AstraZeneca. Conception and design: K. Kinneer, S.J. Gregson, N. Dimasi, M.J. Wick, P.W. Howard, D.A. Tice Development of methodology: K. Kinneer, J. Meekin, A.C. Tiberghien, Y. Tai, The costs of publication of this article were defrayed in part by the payment S. Phipps, C.M. Kiefer, N. Dimasi, M.J. Wick, P.W. Howard of page charges. This article must therefore be hereby marked advertisement Acquisition of data (provided animals, acquired and managed patients, in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. provided facilities, etc.): K. Kinneer, J. Meekin, Y. Tai, S. Phipps, C.M. Kiefer, N. Dimasi, A.D. Moriarty, K.P. Papadopoulos, M.J. Wick, K.C. Anderson, Received April 26, 2018; revised July 11, 2018; accepted August 16, 2018; P.W. Howard published first August 21, 2018.

References 1. Polakis P. Antibody drug conjugates for cancer therapy. Pharmacol Rev maytansine conjugates from the lysosome to the cytoplasm. Cancer Res 2016;68:3–19. 2015;75:5329–40. 2. Senter PD, Sievers EL. The discovery and development of brentuximab 13. Hicks SW, Lai KC, Gavrilescu LC, Yi Y, Sikka S, Shah P, et al. The antitumor vedotin for use in relapsed Hodgkin lymphoma and systemic anaplastic activity of IMGN529, a CD37-targeting antibody-drug conjugate, is poten- large cell lymphoma. Nat Biotechnol 2012;30:631–7. tiated by rituximab in non-hodgkin lymphoma models. Neoplasia 3. Lambert JM, Chari RV. Ado-trastuzumab emtansine: an antibody-drug 2017;19:661–71. conjugate for HER2-positive breast cancer. J Med Chem 2014;57:6949–64. 14. O'Donnell EK, Raje NS. New monoclonal antibodies on the horizon in 4. Kantarjian HM, DeAngelo DJ, Stelljes M, Martinelli G, Liedtke M, Stock W, multiple myeloma. Ther Adv Hematol 2017;8:41–53. et al. Inotuzumab ozogamicin versus standard therapy for acute lympho- 15. Tai YT, Anderson KC. Targeting B-cell maturation antigen in multiple blastic leukemia. N Engl J Med 2016;375:740–53. myeloma. Immunotherapy 2015;7:1187–99. 5. Appelbaum FR, Bernstein ID. Gemtuzumab ozogamicin for acute myeloid 16. Tai YT, Mayes PA, Acharya C, Zhong MY, Cea M, Cagnetta A, et al. leukemia. Blood 2017;130:2373–6. Novel anti-B-cell maturation antigen antibody-drug conjugate 6. Golfier S, Kopitz C, Kahnert A, Heisler I, Schatz CA, Stelte-Ludwig B, et al. (GSK2857916) selectively induces killing of multiple myeloma. Blood Anetumab ravtansine: a novel mesothelin-targeting antibody-drug 2014;123:3128–38. conjugate cures tumors with heterogeneous target expression favored by 17. Barok M, Joensuu H, Isola J. Trastuzumab emtansine: mechanisms of bystander effect. Mol Cancer Ther 2014;13:1537–48. action and drug resistance. Breast Cancer Res 2014;16:209. 7. Katz J, Janik JE, Younes A. Brentuximab vedotin (SGN-35). Clin Cancer Res 18. Krop IE, Kim SB, Gonzalez-Martin A, LoRusso PM, Ferrero JM, Smitt M, 2011;17:6428–36. et al. Trastuzumab emtansine versus treatment of physician's choice for 8.KovtunYV,AudetteCA,YeY,XieH,RubertiMF,PhinneySJ,etal. pretreated HER2-positive advanced breast cancer (TH3RESA): a rando- Antibody-drug conjugates designed to eradicate tumors with homoge- mised, open-label, phase 3 trial. Lancet Oncol 2014;15:689–99. neous and heterogeneous expression of the target antigen. Cancer Res 19. Verma S, Miles D, Gianni L, Krop IE, Welslau M, Baselga J, et al. 2006;66:3214–21. Trastuzumab emtansine for HER2-positive advanced breast cancer. 9. Erickson HK, Lewis Phillips GD, Leipold DD, Provenzano CA, Mai E, N Engl J Med 2012;367:1783–91. Johnson HA, et al. The effect of different linkers on target cell catabolism 20. Kim SB, Wildiers H, Krop IE, Smitt M, Yu R, Lysbet de Haas S, et al. and pharmacokinetics/pharmacodynamics of trastuzumab maytansinoid Relationship between tumor biomarkers and efficacy in TH3RESA, a phase conjugates. Mol Cancer Ther 2012;11:1133–42. III study of trastuzumab emtansine (T-DM1) vs. treatment of physician's 10. Doronina SO, Mendelsohn BA, Bovee TD, Cerveny CG, Alley SC, Meyer DL, choice in previously treated HER2-positive advanced breast cancer. Int J et al. Enhanced activity of monomethylauristatin F through monoclonal Cancer 2016;139:2336–42. antibody delivery: effects of linker technology on efficacy and toxicity. 21. Twomey JD, Brahme NN, Zhang B. Drug-biomarker co-development in Bioconjug Chem 2006;17:114–24. oncology—20 years and counting. Drug Resist Updat 2017;30:48–62. 11. Erickson HK, Park PU, Widdison WC, Kovtun YV, Garrett LM, Hoffman K, 22. Loganzo F, Sung M, Gerber HP. Mechanisms of resistance to antibody-drug et al. Antibody-maytansinoid conjugates are activated in targeted cancer conjugates. Mol Cancer Ther 2016;15:2825–34. cells by lysosomal degradation and linker-dependent intracellular proces- 23. Rios-Luci C, Garcia-Alonso S, Diaz-Rodriguez E, Nadal-Serrano M, sing. Cancer Res 2006;66:4426–33. ArribasJ,OcanaA,etal.Resistance to the antibody-drug conjugate 12. Hamblett KJ, Jacob AP, Gurgel JL, Tometsko ME, Rock BM, Patel SK, et al. T-DM1 is based in a reduction in lysosomal proteolytic activity. Cancer SLC46A3 is required to transport catabolites of noncleavable antibody Res 2017;77:4639–51.

OF12 Clin Cancer Res; 2018 Clinical Cancer Research

SLC46A3 as a Potential Biomarker for Noncleavable ADCs

24. Li G, Guo J, Shen BQ, Bumbaca Yadav D, Sliwkowski MX, Crocker LM, et al. 40. McCall MN, Bolstad BM, Irizarry RA. Frozen robust multiarray analysis Mechanisms of acquired resistance to trastuzumab emtansine in breast (fRMA). Biostatistics 2010;11:242–53. cancer cells. Mol Cancer Ther 2018;17:1441–53. 41. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical 25. Jeffrey SC, Burke PJ, Lyon RP, Meyer DW, Sussman D, Anderson M, et al. and powerful approach to multiple testing. J R Stat Soc Series B Stat A potent anti-CD70 antibody-drug conjugate combining a dimeric Methodol 1995;57:289–300. pyrrolobenzodiazepine drug with site-specific conjugation technology. 42. Chapel A, Kieffer-Jaquinod S, Sagne C, Verdon Q, Ivaldi C, Mellal M, et al. Bioconjug Chem 2013;24:1256–63. An extended proteome map of the lysosomal membrane reveals novel 26. Kung Sutherland MS, Walter RB, Jeffrey SC, Burke PJ, Yu C, Kostner H, et al. potential transporters. Mol Cell Proteomics 2013;12:1572–88. SGN-CD33A: a novel CD33-targeting antibody-drug conjugate using a 43. Loganzo F, Tan X, Sung M, Jin G, Myers JS, Melamud E, et al. Tumor cells pyrrolobenzodiazepine dimer is active in models of drug-resistant AML. chronically treated with a trastuzumab-maytansinoid antibody-drug con- Blood 2013;122:1455–63. jugate develop varied resistance mechanisms but respond to alternate 27. Tiberghien AC, Levy JN, Masterson LA, Patel NV, Adams LR, Corbett S, et al. treatments. Mol Cancer Ther 2015;14:952–63. Design and synthesis of tesirine, a clinical antibody-drug conjugate pyrro- 44. Niikura N, Liu J, Hayashi N, Mittendorf EA, Gong Y, Palla SL, et al. Loss of lobenzodiazepine dimer payload. ACS Med Chem Lett 2016;7:983–7. human epidermal growth factor receptor 2 (HER2) expression in meta- 28. Hartley JA. The development of pyrrolobenzodiazepines as antitumour static sites of HER2-overexpressing primary breast tumors. J Clin Oncol agents. Expert Opin Investig Drugs 2011;20:733–44. 2012;30:593–9. 29. Mantaj J, Jackson PJ, Rahman KM, Thurston DE. From anthramycin to 45. Shefet-Carasso L, Benhar I. Antibody-targeted drugs and drug resistance— pyrrolobenzodiazepine-containing antibody-drug conjugates. Angew challenges and solutions. Drug Resist Updat 2015;18:36–46. Chem Int Ed Engl 2017;56:462–88. 46. Helman E, Wick MJ, Clark MJ, Gamez L, Boyle S, Papadopoulos KP, et al. 30. Gregson SJ, Masterson LA, Wei B, Pillow TH, Spencer SD, Kang GD, et al. Genomic characterization of a PDX model of T-DM1-resistant HER2þ Pyrrolobenzodiazepine dimer antibody-drug conjugates: synthesis invasive ductal carcinoma using augmented exome sequencing. In: Pro- and evaluation of noncleavable drug-linkers. J Med Chem 2017;60: ceedings of the 106th Annual Meeting of the American Association for 9490–507. Cancer Research; 2015 Apr 18–22; Philadelphia (PA): AACR; 2015. 31. Tiberghien AC, Gregson SJ, Masterson LA, Levy JN, Kemp GC, Adams LR, Abstract nr 1457. et al. An optimised synthesis of SG3376, a non-cleavable antibody-drug 47. Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S, conjugate pyrrolobenzodiazepine drug linker. Tetrahedron Lett 2017;58: et al. The Cancer Cell Line Encyclopedia enables predictive modelling of 4363–6. anticancer drug sensitivity. Nature 2012;483:603–7. 32. Dimasi N, Fleming R, Zhong H, Bezabeh B, Kinneer K, Christie RJ, et al. 48. Seckinger A, Delgado JA, Moser S, Moreno L, Neuber B, Grab A, et al. Target Efficient preparation of site-specific antibody-drug conjugates using cys- expression, generation, preclinical activity, and pharmacokinetics of the teine insertion. Mol Pharm 2017;14:1501–16. BCMA-T cell bispecific antibody EM801 for multiple myeloma treatment. 33. Dimasi N, Gao C, Fleming R, Woods RM, Yao XT, Shirinian L, et al. Cancer Cell 2017;31:396–410. The design and characterization of oligospecific antibodies for simulta- 49. Lee L, Bounds D, Paterson J, Herledan G, Sully K, Seestaller-Wehr LM, et al. neous targeting of multiple disease mediators. J Mol Biol 2009;393: Evaluation of B cell maturation antigen as a target for antibody drug 672–92. conjugate mediated cytotoxicity in multiple myeloma. Br J Haematol 34. Thompson P, Fleming R, Bezabeh B, Huang F, Mao S, Chen C, et al. 2016;174:911–22. Rational design, biophysical and biological characterization of site-specific 50. Yu SF, Zheng B, Go M, Lau J, Spencer S, Raab H, et al. A novel anti-CD22 antibody-tubulysin conjugates with improved stability, efficacy and phar- anthracycline-based antibody-drug conjugate (ADC) that overcomes resis- macokinetics. J Control Release 2016;236:100–16. tance to auristatin-based ADCs. Clin Cancer Res 2015;21:3298–306. 35. Li JY, Perry SR, Muniz-Medina V, Wang X, Wetzel LK, Rebelatto MC, et al. 51. Paine PL, Feldherr CM. Nucleocytoplasmic exchange of macromolecules. A Biparatopic HER2-targeting antibody-drug conjugate induces tumor Exp Cell Res 1972;74:81–98. regression in primary models refractory to or ineligible for HER2-targeted 52. Paine PL, Moore LC, Horowitz SB. Nuclear envelope permeability. Nature therapy. Cancer Cell 2016;29:117–29. 1975;254:109–14. 36. Jackson D, Gooya J, Mao S, Kinneer K, Xu L, Camara M, et al. A human 53. Krop IE, Kim SB, Martin AG, LoRusso PM, Ferrero JM, Badovinac-Crnjevic antibody-drug conjugate targeting EphA2 inhibits tumor growth in vivo. T, et al. Trastuzumab emtansine versus treatment of physician's choice in Cancer Res 2008;68:9367–74. patients with previously treated HER2-positive metastatic breast cancer 37. Hsu PD, Scott DA, Weinstein JA, Ran FA, Konermann S, Agarwala V, et al. (TH3RESA): final overall survival results from a randomised open-label DNA targeting specificity of RNA-guided Cas9 nucleases. Nat Biotechnol phase 3 trial. Lancet Oncol 2017;18:743–54. 2013;31:827–32. 54. Metzger-Filho O, Winer EP. Breast cancer: T-DM1—an important agent in 38. Ran FA, Hsu PD, Lin CY, Gootenberg JS, Konermann S, Trevino AE, et al. the history of breast cancer management. Nat Rev Clin Oncol 2017;14: Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing 651–2. specificity. Cell 2013;154:1380–9. 55. Trudel S, Lendvai N, Popat R, Voorhees PM, Reeves B, Libby EN III, et al. . 39. Tiedemann RE, Zhu YX, Schmidt J, Yin H, Shi CX, Que Q, et al. Kinome- Deep and durable responses in patients with relapsed/refractory multiple wide RNAi studies in human multiple myeloma identify vulnerable kinase myeloma treated with monotherapy GSK2857916, an antibody drug targets, including a lymphoid-restricted kinase, GRK6. Blood 2010;115: conjugate against B-cell maturation antigen: preliminary results from part 1594–604. 2 of study BMA117159. Blood 2017;130:741.

www.aacrjournals.org Clin Cancer Res; 2018 OF13

SLC46A3 as a Potential Predictive Biomarker for Antibody− Drug Conjugates Bearing Noncleavable Linked Maytansinoid and Pyrrolobenzodiazepine Warheads

Krista Kinneer, John Meekin, Arnaud C. Tiberghien, et al.

Clin Cancer Res Published OnlineFirst August 21, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-1300

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2018/08/21/1078-0432.CCR-18-1300.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2018/10/22/1078-0432.CCR-18-1300. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.