Figure 1: KTN3379 specifically binds to HER3; Figure 2: KTN3379 suppresses cell-growth in HRG-HER3 autocrine cells and HRGinduced VEGF secretion; Figure 3: KTN3379 arrests HER2+ breast cancer cells in G0/G1-phase; Figure 4: KTN3379 is inactive in HER3-positive but HRG-negative Snu16 xenograft model; Figure 5: PTEN knockdown does not impact the growth behavior of HMCB; Figure 6: Full dose response curves of KTN3379, Pertuzumab and Cetuximab in HMCB cells with or without PTEN-knockout in 6-day proliferation assay; Supplementary Table 1 Summary of Kinetic Rate Constants and Dissociation Constants for the Binding of KTN3379 to Human HER3, Cynomolgus Monkey HER3, and Murine Her3 Proteins; Supplementary Table 2 Expression and mutation status of HER3 pathway components of models used in this manuscript

Supplementary Figures 1-14; Supplementary Tables 1-3. Supplementary Figure 1. Lead antibody 80PH3 is a potent inhibitor of recombinant ADAM17 Supplementary Figure 2. 80PH3 is not a potent inhibitor of cellular ADAM17 Supplementary Figure 3. MEDI3622 is highly selective for ADAM17 and does not inhibit ADAM10 or MMP12 Supplementary Figure 4. MEDI3622 does not compete for binding to cellular ADAM17 with the ADAM17 monoclonal antibody 9301 Supplementary Figure 5. MEDI3622 inhibits cellular mouse ADAM17 Supplementary Figure 6. Cell surface levels of HER pathway proteins in OE21 esophageal cancer cells Supplementary Figure 7. The half-life of MEDI3622 in rats is approximately 8 days Supplementary Figure 8. Several phosphorylation sites in EGFR are inhibited by MEDI3622 Supplementary Figure 9. Exposure to MEDI3622 reduces levels of phospho-SRC and phospho-YES Supplementary Figure 10. Percent activated EGFR predicts cetuximab sensitivity in a series of Head and Neck PDX models Supplementary Figure 11. Proliferation of COLO205 colon cancer cells is independent of the HER pathways Supplementary Figure 12. MEDI3622 does not inhibit HER pathways in Colo205 tumors Supplementary Figure 13. Predictive value of ADAM17, EGFR, HER2, and HER3 mRNA levels for sensitivity of cell lines to MEDI3622 Supplementary Figure 14. Sensitivity to MEDI3622 is not predicted my mRNA levels of ADAM1, HER2 OR HER3 in head and neck PDX models Supplementary Table 1. Correlation of MEDI3622 and cetuximab anti-proliferative activity Supplementary Table 2. Identification of ADAM17 substrates in COLO205 cells by SILAC proteomics Supplementary Table 3. Identification of ADAM17 substrates in H358 lung cancer cells by SILAC proteomics

Methods and figure legends for supplemental data

AbstractHER3/ERBB3 is a kinase-deficient member of the EGFR family receptor tyrosine kinases (RTK) that is broadly expressed and activated in human cancers. HER3 is a compelling cancer target due to its important role in activation of the oncogenic PI3K/AKT pathway. It has also been demonstrated to confer tumor resistance to a variety of cancer therapies, especially targeted drugs against EGFR and HER2. HER3 can be activated by its ligand (heregulin/HRG), which induces HER3 heterodimerization with EGFR, HER2, or other RTKs. Alternatively, HER3 can be activated in a ligand-independent manner through heterodimerization with HER2 in HER2-amplified cells. We developed a fully human mAb against HER3 (KTN3379) that efficiently suppressed HER3 activity in both ligand-dependent and independent settings. Correspondingly, KTN3379 inhibited tumor growth in divergent tumor models driven by either ligand-dependent or independent mechanisms in vitro and in vivo. Most intriguingly, while investigating the mechanistic underpinnings of tumor response to KTN3379, we discovered an interesting dichotomy in that PTEN loss, a frequently occurring oncogenic lesion in a broad range of cancer types, substantially blunted the tumor response in HER2-amplified cancer, but not in the ligand-driven cancer. To our knowledge, this represents the first study ascertaining the impact of PTEN loss on the antitumor efficacy of a HER3 mAb. KTN3379 is currently undergoing a phase Ib clinical trial in patients with advanced solid tumors. Our current study may help us optimize patient selection schemes for KTN3379 to maximize its clinical benefits. Mol Cancer Ther; 15(4); 689–701. ©2016 AACR.