NAV-001, a high-efficacy antibody-drug conjugate targeting mesothelin with improved delivery of a potent payload by counteracting MUC16/CA125 inhibitory effects
Nicholas C. Nicolaides ,J. Bradford Kline,Luigi Grasso
Abstract
Subsets of tumor-produced cell surface and secreted proteins can bind to IgG1 type antibodies and suppress their immune-effector activities. As they affect antibody and complement-mediated immunity, we call these proteins humoral immuno-oncology (HIO) factors. Antibody-drug conjugates (ADCs) use antibody targeting to bind cell surface antigens, internalize into the cell, then kill target cells upon liberation of the cytotoxic payload. Binding of the ADC antibody component by a HIO factor may potentially hamper ADC efficacy due to reduced internalization. To determine the potential effects of HIO factor ADC suppression, we evaluated the efficacy of a HIO-refractory, mesothelin-directed ADC (NAV-001) and a HIO-bound, mesothelin-directed ADC (SS1). The HIO factor MUC16/CA125 binding to SS1 ADC was shown to have a negative effect on internalization and tumor cell killing. The MUC16/CA125 refractory NAV-001 ADC was shown to have robust killing of MUC16/CA125 expressing and non-expressing tumor cells in vitro and in vivo at single, sub-mg/kg dosing. Moreover, NAV-001-PNU, which contains the PNU-159682 topoisomerase II inhibitor, demonstrated good stability in vitro and in vivo as well as robust bystander activity of resident cells while maintaining a tolerable safety profile in vivo. Single-dose NAV-001-PNU demonstrated robust tumor regression of a number of patient-derived xenografts from different tumor types regardless of MUC16/CA125 expression.
Introduction
Tumors employ a variety of mechanisms to avoid host immune responses and therapeutic-based killing. Several strategies have been developed to overcome such defense mechanisms including development of immune checkpoint inhibitors as well as other immune-mediated therapies including T-cells engineered with chimeric antigen targeting receptors (CAR-Ts), vaccines, bispecific antibodies and antibody-drug conjugates (ADCs). Recent findings have found that tumors can produce cell surface and secreted proteins, referred to as humoral immuno-oncology (HIO) factors, that can bind to antibodies and block their immune-effector activities. These effects occur via direct binding and blockade of antibody interactions with Fc-γ-activating receptors on natural killer (NK) and subsets of myeloid, dendritic and monocytic cells [1, 2]. These effects negatively impact antibody-dependent cellular cytotoxicity (ADCC) and complement-mediated cytotoxicity (CDC) against antibody-bound tumors [3–5]. Strategies to overcome humoral immunosuppression of antibody-based therapeutics include screening for those that are naturally not bound by HIO factors or by employing next generation antibody technologies not requiring immune-effector activity.
Material and methods
Antibodies and antibody-drug conjugates.
Antibodies were acquired from vendors and through academic collaboration. Antibodies Ab-1, Ab-2 (humanized SS1), Ab-4 and Ab-5 were produced internally or purchased (Creative BioLabs). NAV-001 and Ab-1 anti-mesothelin antibodies were obtained from Drs. Mitchell Ho and Ira Pastan (National Cancer Institute) and was characterized as previously described [11, 13]. ADCs were made using the cytotoxic topoisomerase I inhibitor SN-38 and the topoisomerase II inhibitor PNU-159682. Conjugations were done under a controlled temperature environment, whereby equal amounts of antibody were treated with optimal amounts of tris(2-carboxyethyl)phosphine to partially reduce interchain cysteines, followed by addition of optimal molar PEG8-triazole-PABC-peptide-SN-38 (SN-38 ADC format) or MA-PEG4-VC-PAB-DMAE-PNU159682 (PNU ADC format) linkers followed by reoxidation. Purified conjugates were analyzed by Hydrophobic Interaction Chromatography (HIC-HPLC) and Size Exclusion Chromatography (SEC-HPLC) to determine drug-antibody-ratios (DAR) and antibody aggregation, respectively. The average DAR for SN-38 ADCs were 6 and PNU-159682 ADCs were 4. Both ADC formats had less than 3.0% aggregates.
Results
Screening for MUC16/CA125 binding to anti-MSLN antibodies
Direct binding of anti-MSLN antibodies by MUC16/CA125 has been previously reported to suppress immune-effector activities and associated with decreased therapeutic activity in human clinical studies [2–5]. In an attempt to analyze as many existing anti-MSLN antibodies, we acquired rodent and human/humanized anti-MSLN antibodies from a variety of sources and ran preliminary screens for MUC16/CA125 binding. We narrowed down the preliminary screens to five antibodies that were humanized or fully human and retested them for binding to MSLN, MUC16/CA125 or HSA. As shown in Fig 1A, all antibodies recognized the MSLN antigen and 4 of 5 antibodies were also able to bind MUC16/CA125. We next tested if MUC16/CA125 can perturb anti-MSLN antibody binding to MSLN protein via competition assay. As shown in Fig 1B, MUC16/CA125 did not impact anti-MSLN binding to MSLN protein. NAV-001 was the only antibody tested that did not naturally bind to MUC16/CA125 (Fig 1A). NAV-001 has an affinity of 70 pM [13] and cross-reacts with cynomolgus monkey but not rodent MSLN ortholog proteins [S1 Fig]. This antibody and a subset of others were then reformatted into ADCs to test for the impact of MUC16/CA125 binding on ADC cytotoxicity.
Discussion
Effect of MUC16/CA125 binding on ADC cytotoxicity
Two of the MUC16/CA125 binding antibodies (Ab-1 and Ab-2) and the NAV-001 antibody were analyzed for killing activity against OVCAR-3 cells, which express both MSLN and MUC16/CA125, and the isogenic OV-KD MUC16/CA125 knockdown line. These lines express similar amounts of MSLN as determined by FACS and have similar morphology as well as growth properties. We first tested Ab-1, Ab-2 and NAV-001 antibodies as SN-38 ADCs using the CL2A PEG8-triazole-PABC-peptide-SN-38 linker-toxin as described in the methods. This linker is pH sensitive and is cleaved in lysosomes when internalized into target cells. SN-38 is a metabolite of irinotecan and a potent topoisomerase I inhibitor that has shown robust tumor cell killing that is dependent on internalization [14]. As shown in Fig 2A, the two antibodies bound by MUC16/CA125 (Ab-1 and Ab-2) had 2.8–4.9 fold less target cell killing of OVCAR-3 cells than the non-MUC16/CA125 binding NAV-001-SN-38, respectively.
Conclusion
Significant efforts to improve the safety and efficacy of ADC-based therapeutics continue to be actively pursued across the biopharmaceutical industry. Here, we show that HIO factors that bind to ADCs can have a significant impact on target cell killing. These data suggest that screening ADC candidates for HIO factor sensitivity should be another parameter to consider during their development in addition to potency, stability and bystander activity. Development strategies similar to those used to generate NAV-001-PNU may potentially lead to improved therapeutic responses in patients with cancers overexpressing HIO factors such as MUC16/CA125.
Acknowledgments
We would like to acknowledge Matthias Ebbinghaus from Charles River Labs who supervised the in vivo efficacy studies and Drs Ira Pastan, Raffit Hassan, and Mitchel Ho from the National Cancer Institute for advice on employing the anti-MSLN antibody.
Citation: Nicolaides NC, Kline JB, Grasso L (2023) NAV-001, a high-efficacy antibody-drug conjugate targeting mesothelin with improved delivery of a potent payload by counteracting MUC16/CA125 inhibitory effects. PLoS ONE 18(5): e0285161. https://doi.org/10.1371/journal.pone.0285161
Editor: Salman Shakil, BRAC University, BANGLADESH
Received: December 8, 2022; Accepted: April 14, 2023; Published: May 17, 2023
Copyright: © 2023 Nicolaides et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: All authors are employees of Navrogen Inc. The authors have declared that no competing interests exist. This affiliation does not alter our adherence to all PLOS ONE policies on sharing data and materials.
