Translational Cancer Mechanisms and Therapy SLC46A3 as a Potential Predictive Biomarker for Antibody–Drug Conjugates Bearing Noncleavable Linked Maytansinoid and Pyrrolobenzodiazepine Warheads Krista Kinneer 1 , John Meekin 1 , Arnaud C. Tiberghien 2 , Yu-Tzu Tai 3 , Sandrina Phipps 4 , Christine Mione Kiefer 4 , Marlon C. Rebelatto 5 , Nazzareno Dimasi 4 , Alyssa Moriarty 6 , Kyriakos P. Papadopoulos 6 , Sriram Sridhar 5 , Stephen J. Gregson 2 , Michael J. Wick 6 , Luke Masterson 2 , Kenneth C. Anderson 3 , Ronald Herbst 1 , Philip W. Howard 2 , and David A. Tice 1 Abstract Purpose: Antibody–drug conjugates (ADC) utilizing non- cleavable linker drugs have been approved for clinical use, and several are in development targeting solid and hematologic malignancies including multiple myeloma. Currently, there are no reliable biomarkers of activity for these ADCs other than presence of the targeted antigen. We observed that certain cell lines are innately resistant to such ADCs, and sought to uncover the underlying mechanism of resistance. Experimental Design: The expression of 43 lysosomal membrane target genes was evaluated in cell lines resistant to ADCs bearing the noncleavable linker, pyrrolobenzodiaze- pine payload SG3376, in vitro. The functional relevance of SLC46A3, a lysosomal transporter of noncleavable ADC catabolites whose expression uniquely correlated with SG3376 resistance, was assessed using EPHA2-, HER2-, and BCMA-targeted ADCs and isogenic cells overexpressing or genetically inactivated for SLC46A3. SLC46A3 expression was also examined in patient-derived xenograft and in vitro models of acquired T-DM1 resistance and multiple myeloma bone marrow samples by RT-PCR. Results: Loss of SLC46A3 expression was found to be a mechanism of innate and acquired resistance to ADCs bearing DM1 and SG3376. Sensitivity was restored in refrac- tory lines upon introduction of SLC46A3, suggesting that expression of SLC46A3 may be more predictive of activity than target antigen levels alone. Interrogation of primary multiple myeloma samples indicated a range of SLC46A3 expression, including samples with undetectable levels like multiple myeloma cell lines resistant to BCMA-targeting DM1 and SG3376 ADCs. Conclusions: Our findings support SLC46A3 as a potential patient selection biomarker with immediate relevance to clinical trials involving these ADCs. Clin Cancer Res; 24(24); 6570–82. Ó2018 AACR. Introduction Antibody–drug conjugates (ADC) combine a mAb with a cytotoxic drug (warhead) to preferentially eliminate antigen-pos- itive cells for the treatment of cancer (1). Four ADCs are approved for clinical use: brentuximab vedotin for the treatment of Hodgkin lymphoma (2), ado-trastuzumab emtansine (T-DM1) for the treatment of erb-b2 receptor tyrosine kinase 2 (ERBB2, HER2)- positive metastatic breast cancer (3), inotuzumab ozogamicin for the treatment of acute lymphoblastic leukemia (4), and gemtu- zumab ozogamicin for the treatment of CD33-positive acute myeloid leukemia (5). Upon binding to the targeted antigen on the cell surface, ADCs prepared with enzymatically cleavable linkers (e.g., brentuximab vedotin) or acid-labile hydrazone lin- kers (e.g., inotuzumab ozogamicin and gemtuzumab ozogami- cin) are internalized and processed within the cell, releasing the cytotoxic warhead after linker cleavage. Warheads released in this manner are typically membrane permeable and capable of bystander killing (6–8). In contrast, ADCs with noncleavable linkers, such as T-DM1, rely upon proteolytic degradation of the antibody in the lysosome to release an amino acid linker warhead (9). These catabolites are generally not membrane permeable (10, 11) and therefore require transport from the lysosome to reach their intracellular target (12). Several ADCs utilizing noncleavable linkers are currently in clinical development and target both solid tumors and hematologic malignancies, including diffuse large B- cell lymphoma and multiple myeloma (refs. 13–16). Although T-DM1 therapy has shown significant clinical benefit, most patients eventually relapse despite continued treatment 1 Oncology Research, MedImmune, Gaithersburg, Maryland. 2 Spirogen, QMB Innovation Centre, London, United Kingdom. 3 The Jerome Lipper Multiple Myeloma Center and LeBow Institute for Myeloma Therapeutics, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts. 4 Antibody Discovery and Protein Engineering, MedImmune, Gaithersburg, Maryland. 5 Translational Medicine, MedImmune, Gaithersburg, Maryland. 6 South Texas Accelerated Research Therapeutics, San Antonio, Texas. Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/). Corresponding Author: Krista Kinneer, MedImmune, 1 MedImmune Way, Gaithersburg, MD 20878. Phone: 301-398-4219. Fax: 301-398-9219. E-mail: kinneerk@medimmune.com doi: 10.1158/1078-0432.CCR-18-1300 Ó2018 American Association for Cancer Research. Clinical Cancer Research Clin Cancer Res; 24(24) December 15, 2018 6570 on March 21, 2020. © 2018 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from Published OnlineFirst August 21, 2018; DOI: 10.1158/1078-0432.CCR-18-1300