HER2+ Cancer Cell Dependence on PI3K vs. MAPK Signaling Axes Is Determined by Expression of EGFR, ERBB3 and CDKN1B.

Understanding the molecular pathways by which oncogenes drive cancerous cell growth, and how dependence on such pathways varies between tumors could be highly valuable for the design of anti-cancer treatment strategies. In this work we study how dependence upon the canonical PI3K and MAPK cascades varies across HER2+ cancers, and define biomarkers predictive of pathway dependencies. A panel of 18 HER2+ (ERBB2-amplified) cell lines representing a variety of indications was used to characterize the functional and molecular diversity within this oncogene-defined cancer. PI3K and MAPK-pathway dependencies were quantified by measuring in vitro cell growth responses to combinations of AKT (MK2206) and MEK (GSK1120212; trametinib) inhibitors, in the presence and absence of the ERBB3 ligand heregulin (NRG1). A combination of three protein measurements comprising the receptors EGFR, ERBB3 (HER3), and the cyclin-dependent kinase inhibitor p27 (CDKN1B) was found to accurately predict dependence on PI3K/AKT vs. MAPK/ERK signaling axes. Notably, this multivariate classifier outperformed the more intuitive and clinically employed metrics, such as expression of phospho-AKT and phospho-ERK, and PI3K pathway mutations (PIK3CA, PTEN, and PIK3R1). In both cell lines and primary patient samples, we observed consistent expression patterns of these biomarkers varies by cancer indication, such that ERBB3 and CDKN1B expression are relatively high in breast tumors while EGFR expression is relatively high in other indications. The predictability of the three protein biomarkers for differentiating PI3K/AKT vs. MAPK dependence in HER2+ cancers was confirmed using external datasets (Project Achilles and GDSC), again out-performing clinically used genetic markers. Measurement of this minimal set of three protein biomarkers could thus inform treatment, and predict mechanisms of drug resistance in HER2+ cancers. More generally, our results show a single oncogenic transformation can have differing effects on cell signaling and growth, contingent upon the molecular and cellular context.

Single-cell quantitative HER2 measurement identifies heterogeneity and distinct subgroups within traditionally defined HER2-positive patients.

Human epidermal growth factor receptor 2 (HER2) is an important biomarker for breast and gastric cancer prognosis and patient treatment decisions. HER2 positivity, as defined by IHC or fluorescent in situ hybridization testing, remains an imprecise predictor of patient response to HER2-targeted therapies. Challenges to correct HER2 assessment and patient stratification include intratumoral heterogeneity, lack of quantitative and/or objective assays, and differences between measuring HER2 amplification at the protein versus gene level. We developed a novel immunofluorescence method for quantitation of HER2 protein expression at the single-cell level on FFPE patient samples. Our assay uses automated image analysis to identify and classify tumor versus non-tumor cells, as well as quantitate the HER2 staining for each tumor cell. The HER2 staining level is converted to HER2 protein expression using a standard cell pellet array stained in parallel with the tissue sample. This approach allows assessment of HER2 expression and heterogeneity within a tissue section at the single-cell level. By using this assay, we identified distinct subgroups of HER2 heterogeneity within traditional definitions of HER2 positivity in both breast and gastric cancers. Quantitative assessment of intratumoral HER2 heterogeneity may offer an opportunity to improve the identification of patients likely to respond to HER2-targeted therapies. The broad applicability of the assay was demonstrated by measuring HER2 expression profiles on multiple tumor types, and on normal and diseased heart tissues.

Computational modeling of ERBB2-amplified breast cancer identifies combined ErbB2/3 blockade as superior to the combination of MEK and AKT inhibitors.

Crosstalk and compensatory circuits within cancer signaling networks limit the activity of most targeted therapies. For example, altered signaling in the networks activated by the ErbB family of receptors, particularly in ERBB2-amplified cancers, contributes to drug resistance. We developed a multiscale systems model of signaling networks in ERBB2-amplified breast cancer to quantitatively investigate relationships between biomarkers (markers of network activity) and combination drug efficacy. This model linked ErbB receptor family signaling to breast tumor growth through two kinase cascades: the PI3K/AKT survival pathway and the Ras/MEK/ERK growth and proliferation pathway. The model predicted molecular mechanisms of resistance to individual therapeutics. In particular, ERBB2-amplified breast cancer cells stimulated with the ErbB3 ligand heregulin were resistant to growth arrest induced by inhibitors of AKT and MEK or coapplication of two inhibitors of the receptor ErbB2 [Herceptin (trastuzumab) and Tykerb (lapatinib)]. We used model simulations to predict the response of ErbB2-positive breast cancer xenografts to combination therapies and verified these predictions in mice. Treatment with trastuzumab, lapatinib, and the ErbB3 inhibitor MM-111 was more effective in inhibiting tumor growth than the combination of AKT and MEK inhibitors and even induced tumor regression, indicating that targeting both ErbB3 and ErbB2 may be an improved therapeutic approach for ErbB2-positive breast cancer patients.

Antitumor activity of a novel bispecific antibody that targets the ErbB2/ErbB3 oncogenic unit and inhibits heregulin-induced activation of ErbB3.

The prevalence of ErbB2 amplification in breast cancer has resulted in the heavy pursuit of ErbB2 as a therapeutic target. Although both the ErbB2 monoclonal antibody trastuzumab and ErbB1/ErbB2 dual kinase inhibitor lapatinib have met with success in the clinic, many patients fail to benefit. In addition, the majority of patients who initially respond will unfortunately ultimately progress on these therapies. Activation of ErbB3, the preferred dimerization partner of ErbB2, plays a key role in driving ErbB2-amplified tumor growth, but we have found that current ErbB2-directed therapies are poor inhibitors of ligand-induced activation. By simulating ErbB3 inhibition in a computational model of ErbB2/ErbB3 receptor signaling, we predicted that a bispecific antibody that docks onto ErbB2 and subsequently binds to ErbB3 and blocks ligand-induced receptor activation would be highly effective in ErbB2-amplified tumors, with superior activity to a monospecific ErbB3 inhibitor. We have developed a bispecific antibody suitable for both large scale production and systemic therapy by generating a single polypeptide fusion protein of two human scFv antibodies linked to modified human serum albumin. The resulting molecule, MM-111, forms a trimeric complex with ErbB2 and ErbB3, effectively inhibiting ErbB3 signaling and showing antitumor activity in preclinical models that is dependent on ErbB2 overexpression. MM-111 can be rationally combined with trastuzumab or lapatinib for increased antitumor activity and may in the future complement existing ErbB2-directed therapies to treat resistant tumors or deter relapse.

Potent antitumor activity of the anti-CD19 auristatin antibody drug conjugate hBU12-vcMMAE against rituximab-sensitive and -resistant lymphomas.

Despite major advances in the treatment of non-Hodgkin lymphoma (NHL), including the use of chemotherapeutic agents and the anti-CD20 antibody rituximab, the majority of patients eventually relapse, and salvage treatments with non-cross-resistant compounds are needed to further improve patient survival. Here, we evaluated the antitumor effects of the microtubule destabilizing agent monomethyl auristatin E (MMAE) conjugated to the humanized anti-CD19 antibody hBU12 via a protease-sensitive valine-citrulline (vc) dipeptide linker. hBU12-vcMMAE induced potent tumor cell killing against rituximab-sensitive and -resistant NHL cell lines. CD19 can form heterodimers with CD21, and high levels of CD21 were reported to interfere negatively with the activity of CD19-targeted therapeutics. However, we observed comparable internalization, intracellular trafficking, and drug release in CD21(low) and CD21(high), rituximab-sensitive and -refractory lymphomas treated with hBU12-vcMMAE. Furthermore, high rates of durable regressions in mice implanted with these tumors were observed, suggesting that both rituximab resistance and CD21 expression levels do not impact on the activity of hBU12-vcMMAE. Combined, our data suggest that hBU12-vcMMAE may represent a promising addition to the treatment options for rituximab refractory NHL and other hematologic malignancies, including acute lymphoblastic leukemia.

Preclinical characterization of SGN-70, a humanized antibody directed against CD70.

PURPOSE: CD70 (CD27L) is a member of the tumor necrosis factor family aberrantly expressed on a number of hematologic malignancies and some carcinomas. CD70 expression on malignant cells coupled with its highly restricted expression on normal cells makes CD70 an attractive target for monoclonal antibody (mAb)-based therapies. We developed a humanized anti-CD70 antibody, SGN-70, and herein describe the antitumor activities of this mAb. EXPERIMENTAL DESIGN: CD70 expression on primary tumors was evaluated by immunohistochemical staining of Hodgkin lymphoma, non-Hodgkin lymphoma, multiple myeloma, and renal cell carcinoma tissue microarrays. The CD70-binding and cytotoxic activities of SGN-70 were tested in vitro using a number of cell-based assays. The in vivo antitumor properties of SGN-70 were tested in severe combined immunodeficient mice bearing disseminated lymphoma and multiple myeloma xenografts. Mechanism-of-action studies were conducted using SGN-70v, a variant mAb with equivalent target-binding activity but impaired Fcgamma receptor binding compared with SGN-70. RESULTS: Immunohistochemical analysis identified CD70 expression on approximately 40% of multiple myeloma isolates and confirmed CD70 expression on a high percentage of Hodgkin lymphoma Reed-Sternberg cells, non-Hodgkin lymphoma, and renal cell carcinoma tumors. SGN-70 lysed CD70+ tumor cells via Fc-dependent functions, including antibody-dependent cellular cytotoxicity and phagocytosis and complement fixation. In vivo, SGN-70 treatment significantly decreased tumor burden and prolonged survival of tumor-bearing mice. CONCLUSIONS: SGN-70 is a novel humanized IgG1 mAb undergoing clinical development for the treatment of CD70+ cancers. SGN-70 possesses Fc-dependent antibody effector functions and mediates antitumor activity in vivo.

Engineered anti-CD70 antibody-drug conjugate with increased therapeutic index.

An anti-CD70 antibody conjugated to monomethylauristatin F (MMAF) via a valine-citrulline dipeptide containing linker has been shown previously to have potent antitumor activity in renal cell cancer xenograft studies. Here, we generated a panel of humanized anti-CD70 antibody IgG variants and conjugated them to MMAF to study the effect of isotype (IgG1, IgG2, and IgG4) and Fcgamma receptor binding on antibody-drug conjugate properties. All IgG variants bound CD70+ 786-O cells with an apparent affinity of approximately 1 nmol/L, and drug conjugation did not impair antigen binding. The parent anti-CD70 IgG1 bound to human FcgammaRI and FcgammaRIIIA V158 and mouse FcgammaRIV and this binding was not impaired by drug conjugation. In contrast, binding to these Fcgamma receptors was greatly reduced or abolished in the variant, IgG1v1, containing the previously described mutations, E233P:L234V:L235A. All conjugates had potent cytotoxic activity against six different antigen-positive cancer cell lines in vitro with IC50 values of 30 to 540 pmol/L. The IgGv1 conjugate with MMAF displayed improved antitumor activity compared with other conjugates in 786-O and UMRC3 models of renal cell cancer and in the DBTRG05-MG glioblastoma model. All conjugates were tolerated to > or =40 mg/kg in mice. Thus, the IgG1v1 MMAF conjugate has an increased therapeutic index compared with the parent IgG1 conjugate. The improved antitumor activity of the IgG1v1 auristatin conjugates may relate to increased exposure as suggested by pharmacokinetic analysis. The strategy used here for enhancing the therapeutic index of antibody-drug conjugates is independent of the antigen-binding variable domains and potentially applicable to other antibodies.

Anti-CD30 diabody-drug conjugates with potent antitumor activity.

Anti-CD30 diabodies were engineered with two cysteine mutations for site-specific drug conjugation in each chain of these homodimeric antibody fragments. Diabodies were conjugated with approximately 4 equivalents of the anti-tubulin drugs, monomethyl auristatin E or F, via a protease-cleavable dipeptide linker, to create the conjugates, diabody-vcE4 and diabody-vcF4, respectively. Diabody conjugation had only minor (<3-fold) effects on antigen binding. Diabody-vcF4 was potently cytotoxic against the antigen-positive cell lines, Karpas-299 (34 pmol/L IC(50)) and L540cy (22 pmol/L IC(50)), and was 8- and 21-fold more active than diabody-vcE4 against these cell lines, respectively. Clearance of diabody-vcF4 (99-134 mL/d/kg) was 5-fold slower than for the nonconjugated diabody in naive severe combined immunodeficient mice. Diabody-vcF4 had potent and dose-dependent antitumor activity against established Karpas-299 xenografts and gave durable complete responses at well-tolerated doses. Biodistribution experiments with diabody-[(3)H]-vcF4 (0.72-7.2 mg/kg) in tumor-bearing mice showed a dose-dependent increase in total auristatin accumulation in tumors (< or =520 nmol/L) and decrease in relative auristatin accumulation (< or =8.1 %ID/g), with peak localization at 4 to 24 h after dosing. Diabody-vcF4 had approximately 4-fold lower cytotoxic activity than the corresponding IgG1-vcF4 conjugate in vitro. A similar potency difference was observed in vivo despite 25- to 34-fold faster clearance of diabody-vcF4 than IgG1-vcF4. This may reflect that dose-escalated diabody-vcF4 can surpass IgG1-vcF4 in auristatin delivery to tumors, albeit with higher auristatin exposure to some organs including kidney and liver. Diabody-drug conjugates can have potent antitumor activity at well-tolerated doses and warrant further optimization for cancer therapy.

Antibody targeting of B-cell maturation antigen on malignant plasma cells.

B-cell maturation antigen (BCMA) is expressed on normal and malignant plasma cells and represents a potential target for therapeutic intervention. BCMA binds to two ligands that promote tumor cell survival, a proliferation inducing ligand (APRIL) and B-cell activating factor. To selectively target BCMA for plasma cell malignancies, we developed antibodies with ligand blocking activity that could promote cytotoxicity of multiple myeloma (MM) cell lines as naked antibodies or as antibody-drug conjugates. We show that SG1, an inhibitory BCMA antibody, blocks APRIL-dependent activation of nuclear factor-kappaB in a dose-dependent manner in vitro. Cytotoxicity of SG1 was assessed as a naked antibody after chimerization with and without Fc mutations that enhance FcgammaRIIIA binding. The Fc mutations increased the antibody-dependent cell-mediated cytotoxicity potency of BCMA antibodies against MM lines by approximately 100-fold with a > or = 2-fold increase in maximal lysis. As an alternative therapeutic strategy, anti-BCMA antibodies were endowed with direct cytotoxic activity by conjugation to the cytotoxic drug, monomethyl auristatin F. The most potent BCMA antibody-drug conjugate displayed IC(50) values of < or = 130 pmol/L for three different MM lines. Hence, BCMA antibodies show cytotoxic activity both as naked IgG and as drug conjugates and warrant further evaluation as therapeutic candidates for plasma cell malignancies.

Engineered anti-CD70 antibody with multiple effector functions exhibits in vitro and in vivo antitumor activities.

Antigens expressed on malignant cells in the absence of significant expression on normal tissues are highly desirable targets for therapeutic antibodies. CD70 is a TNF superfamily member whose normal expression is highly restricted but is aberrantly expressed in hematologic malignancies including non-Hodgkin lymphoma (NHL), Hodgkin disease, and multiple myeloma. In addition, solid tumors such as renal cell carcinoma, nasopharyngeal carcinoma, thymic carcinoma, meduloblastoma, and glioblastoma express high levels of this antigen. To functionally target CD70-expressing cancers, a murine anti-CD70 monoclonal antibody was engineered to contain human IgG1 constant domains. The engineered antibody retained the binding specificity of the murine parent monoclonal antibody and was shown to induce Fc-mediated effector functions including antibody-dependent cellular cytotoxicity, complement-dependent cytotoxicity, and antibody-dependent cellular phagocytosis in vitro. Further, administration of this antibody significantly prolonged survival of severe combined immunodeficient (SCID) mice bearing CD70+ disseminated human NHL xenografts. Survival of these mice was dependent upon the activity of resident effector cells including neutrophils, macrophages, and natural killer (NK) cells. These data suggest that an anti-CD70 antibody, when engineered to contain human IgG1 constant domains, possesses effector cell-mediated antitumor activity and has potential utility for anticancer therapy.

Engineered antibody-drug conjugates with defined sites and stoichiometries of drug attachment.

The chimeric anti-CD30 IgG1, cAC10, conjugated to eight equivalents of monomethyl auristatin E (MMAE) was previously shown to have potent antitumor activity against CD30-expressing tumors xenografts in mice. Moreover, the therapeutic index was increased by lowering the stoichiometry from 8 drugs/antibody down to 2 or 4. Limitations of such 'partially-loaded' conjugates are low yield (10-30%) as they are purified from mixtures with variable stoichiometry (0-8 drugs/antibody), and heterogeneity as the 2 or 4 drugs are distributed over eight possible cysteine conjugation sites. Here, the solvent-accessible cysteines that form the interchain disulfide bonds in cAC10 were replaced with serine, to reduce the eight potential conjugation sites down to 4 or 2. These Cys-->Ser antibody variants were conjugated to MMAE in near quantitative yield (89-96%) with defined stoichiometries (2 or 4 drugs/antibody) and sites of drug attachment. The engineered antibody-drug conjugates have comparable antigen-binding affinities and in vitro cytotoxic activities with corresponding purified parental antibody-drug conjugates. Additionally, the engineered and parental antibody-drug conjugates have similar in vivo properties including antitumor activity, pharmacokinetics and maximum tolerated dose. Our strategy for generating antibody-drug conjugates with defined sites and stoichiometries of drug loading is potentially broadly applicable to other antibodies as it involves engineering of constant domains.

Improved yield and stability of L49-sFv-beta-lactamase, a single-chain antibody fusion protein for anticancer prodrug activation, by protein engineering.

The L49 single-chain Fv fused to beta-lactamase (L49-sFv-bL) combined with the prodrug C-Mel is an effective anticancer agent against tumor cells expressing the p97 antigen. However, large-scale production of L49-sFv-bL from refolded E. coli inclusion bodies has been problematic due to inefficient refolding and instability of the fusion protein. Sequence analysis of the L49-sFv framework regions revealed three residues in the framework regions at positions L2, H82B, and H91, which are not conserved for their position, occurring in <1% of sequences in Fv sequence databases. One further unusual residue, found in <3% of variable sequences, was observed at position H39. Each unusual residue was mutated to a conserved residue for its position and tested for refolding yield from inclusion bodies following expression in E. coli. The three V(H) single mutants showed improvement in the yield of active protein and were combined to form double and triple mutants resulting in a 7-8-fold increased yield compared to the parental protein. In an attempt to further improve yield, the orientation of the triple mutant was reversed to create a bL-L49-sFv fusion protein resulting in a 3-fold increase in expressed inclusion body protein and producing a 20-fold increase in the yield of purified protein compared to the parental protein. The triple mutants in both orientations displayed increased stability in murine plasma and binding affinity was not affected by the introduced mutations. Both triple mutants also displayed potent in vitro cytotoxicity and in vivo antitumor activity against p97 expressing melanoma cells and tumor xenografts, respectively. These results show that a rational protein-engineering approach improved the yield, stability, and refolding characteristics of L49-sFv-bL while maintaining binding affinity and therapeutic efficacy.