Lingyun Rui

Abstract 4576: IMGN853, an anti-Folate Receptor I antibody-maytansinoid conjugate for targeted cancer therapy

Proceedings: AACR 102nd Annual Meeting 2011‐‐ Apr 2‐6, 2011; Orlando, FL Previously we reported (O. Ab; EORTC, 2010) that an antibody-maytansinoid conjugate (AMC) composed of an anti-FOLR1 antibody conjugated to the cytotoxic maytansinoid, DM4, via the disulfide-containing linker, SPDB, was potent in killing FOLR1-expressing cancer cells in vitro and in vivo. In light of the favorable results noted, we assessed the optimal antibody, linker, and maytansinoid agent for an AMC targeting FOLR1, as reported here. Antibody selection. Anti-FOLR1 antibodies were generated by immunizing mice with human FOLR1-expressing cells, and a panel of FOLR1-specific antibodies was identified by flow cytometry binding assay. Several FOLR1-antibodies with high binding affinity to both human and monkey FOLR1 were chosen for further evaluation and were humanized using ImmunoGens resurfacing technology. Antibodies were conjugated to DM1 via the non-cleavable SMCC linker and the conjugates tested for activity against FOLR1-positive KB cells in vitro and in vivo. All conjugates had comparable cytotoxic potencies in vitro. However, the in vivo anti-tumor activity of one conjugate, M9346A-SMCC-DM1, was significantly better than that of SMCC-DM1 conjugates of other FOLR1 antibodies. Based on this finding, the M9346A antibody was chosen for further development. Linker/maytansinoid selection. The M9346A antibody was linked to DM1 or DM4 via the disulfide-containing cleavable linkers SPP, SPDB or sulfo-SPDB, or via the non-cleavable SMCC linker. We compared the in vitro cytotoxic activities of these conjugates on KB, Igrov-1 and Jeg-3 cell lines. The conjugates with cleavable linkers displayed markedly greater in vitro activities than the SMCC conjugate. We then examined the in vivo activities of the conjugates in FOLR1-positive KB- and Ovcar 3-tumor models. Again, we found that the conjugates with cleavable linkers were more active in vivo than the noncleavable conjugate. Among the conjugates with cleavable linkers, the sulfo-SPDB-DM4 conjugate was the most active conjugate against the Ovcar-3 model, it had activity comparable to that of the SPDB-DM4 conjugate against KB tumors, and both were more active than the SPP-DM1 conjugate in the two xenograft models. Taking into consideration that sulfo-SPDB-DM4 was the most efficacious design in vivo and the potential of the hydrophilic sulfo-SPDB-linker to enable better activity against PgP-expressing cells (previously reported data), M9346A-sulfo-SPDB-DM4 was selected to be the candidate for development and designated IMGN853. IMGN853 is a promising candidate for the treatment of FOLR1-expressing tumors. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4576. doi:10.1158/1538-7445.AM2011-4576

Abstract 5463: Development of a novel antibody-maytansinoid conjugate, IMGN289, for the treatment of EGFR-expressing solid tumors.

EGFR is an attractive target for the treatment of a variety of solid tumors because of its role as a driver oncogene and high level of expression. Four EGFR-targeting agents, including two antibodies (Abs), have been approved for clinical use. Despite anti-tumor benefits, inhibition of EGFR pathway is associated with significant dermatologic toxicities; resistance to EGFR antagonists also develops. To enhance potency with comparable or better tolerability, we developed IMGN289, an EGFR-targeting antibody-“drug” conjugate (ADC) that disrupts tumor growth both by inhibiting EGFR signaling and through direct anti-mitotic activity. To reduce potential dermatologic toxicities associated with EGFR pathway inhibition, a unique Ab discovery approach was employed. Hybridomas from mice immunized with EGFR-expressing tumor cells were screened for EGFR binding and selective inhibitory activity against EGFR-dependent tumor cells. This approach revealed a novel class of Ab with selective EGFR antagonistic activity. A humanized lead Ab was identified, J2898A, which was comparable in potency to cetuximab in vitro against a panel of EGFR-dependent tumor cell lines and in vivo against two head and neck tumor xenograft models. Notably, in cultures of human primary keratinocytes, this Ab was markedly less cytotoxic than cetuximab and did not affect TNFα-induced cytokine production, which has been implicated in chronic dermatologic toxicities induced by other anti-EGFR agents. To further enhance cytotoxic activity and to potentially overcome resistance to EGFR-targeting therapies, J2898A was conjugated to the maytansinoid DM1, a potent anti-tubulin agent, via a non-cleavable linker, SMCC. IMGN289 was not only more potent than J2898A against EGFR-dependent tumors, but also was effective against EGFR-positive tumor cells that grow independently of signaling via the EGFR pathway or have acquired resistance to EGFR inhibitors, including lung adenocarcinoma cell lines harboring the T790M EGFR mutation or MET gene amplification. Despite having potent activity against EGFR-expressing tumor cells, IMGN289 was less toxic to cultured keratinocytes than cetuximab. Moreover, a toxicology study in cynomolgus monkeys demonstrated that IMGN289 was well tolerated and exhibited a similar toxicity profile to that published for trastuzumab emtansine (T-DM1), another ADC which utilizes SMCC-DM1 as the selected linker-payload format. In summary, IMGN289 combines EGFR inhibition mediated by its J2898A Ab component with the potent cytotoxicity provided by its DM1 payload, and is highly active against EGFR-positive tumors regardless of their dependency on the EGFR pathway. IMGN289 thus represents a promising novel candidate for treatment of EGFR-expressing solid tumors. Citation Format: Yulius Y. Setiady, Peter U. Park, Jose F. Ponte, Ling Dong, Anna Skaletskaya, Jennifer A. Coccia, Erica Hong, Lauren Clancy, Lingyun Rui, Jan Pinkas, Robert J. Lutz, John M. Lambert, Thomas D. Chittenden. Development of a novel antibody-maytansinoid conjugate, IMGN289, for the treatment of EGFR-expressing solid tumors. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5463. doi:10.1158/1538-7445.AM2013-5463

A CD123-targeting antibody-drug conjugate, IMGN632, designed to eradicate AML while sparing normal bone marrow cells

The outlook for patients with refractory/relapsed acute myeloid leukemia (AML) remains poor, with conventional chemotherapeutic treatments often associated with unacceptable toxicities, including severe infections due to profound myelosuppression. Thus there exists an urgent need for more effective agents to treat AML that confer high therapeutic indices and favorable tolerability profiles. Because of its high expression on leukemic blast and stem cells compared with normal hematopoietic stem cells and progenitors, CD123 has emerged as a rational candidate for molecularly targeted therapeutic approaches in this disease. Here we describe the development and preclinical characterization of a CD123-targeting antibody-drug conjugate (ADC), IMGN632, that comprises a novel humanized anti-CD123 antibody G4723A linked to a recently reported DNA mono-alkylating payload of the indolinobenzodiazepine pseudodimer (IGN) class of cytotoxic compounds. The activity of IMGN632 was compared with X-ADC, the ADC utilizing the G4723A antibody linked to a DNA crosslinking IGN payload. With low picomolar potency, both ADCs reduced viability in AML cell lines and patient-derived samples in culture, irrespective of their multidrug resistance or disease status. However, X-ADC exposure was >40-fold more cytotoxic to the normal myeloid progenitors than IMGN632. Of particular note, IMGN632 demonstrated potent activity in all AML samples at concentrations well below levels that impacted normal bone marrow progenitors, suggesting the potential for efficacy in AML patients in the absence of or with limited myelosuppression. Furthermore, IMGN632 demonstrated robust antitumor efficacy in multiple AML xenograft models. Overall, these findings identify IMGN632 as a promising candidate for evaluation as a novel therapy in AML.

Abstract 647: SeriMabs: N-terminal serine modification enables modular, site-specific payload incorporation into antibody-drug conjugates (ADCs)

Site-specific incorporation of cell-killing agents into cancer-targeting antibodies is an active area of innovation in the field of ADCs. We have developed a highly modular site-specific conjugation platform employing N-terminal serine engineered antibodies (SeriMabs), using ligation chemistry orthogonal to lysine and cysteine modification which typically employs maleimide containing linkers. Humanized IgG1 antibodies were engineered with N-terminal serine residues on either the light or heavy chain, with the precise position optimized for quantitative conversion to the corresponding glyoxyl aldehyde in the presence of sodium periodate. The aminooxy functional group of the heterobifunctional linker N-[(aminooxy)acetyl]-3-[(3-nitro-2-pyridinyl)dithio]-L-alanine was condensed with the aldehyde group on the antibody using 4-amino phenethyl alcohol catalyst, yielding a stable oxime bond under conditions that fully maintain antibody integrity. The dithiopyridine groups were then reacted with thiol-containing maytansinoid or IGN cell-killing agents, yielding disulfide-linked SeriMab ADCs with exactly 2 payload molecules per antibody, as determined by MS analysis, in > 90% yield and with a monomer content of > 98%. Conjugation of payload was found to be selective for the N-terminal serine based on MS/MS analysis. The site-specific ADCs were found to bind to their target antigens with affinity similar to the corresponding lysine-conjugated ADCs, and FcRn binding (pH 5.8) of the SeriMab ADCs were comparable to lysine-conjugated controls. DGN462, our proprietary DNA-acting IGN payload used in our preclinical candidate IMGN779, was conjugated to SeriMabs against two targets, and in both cases, high, antigen-specific in vitro potency was noted. The oxime linkage used in the serine site-specific conjugates was found to be stable at pH 5.5 and pH 7.4 in buffer, as little detectable payload was released over 4 days at 37°C. SeriMab antiFRα-DGN462 (2 DGN462 molecules per Ab) demonstrated potent, dose-dependent antitumor activity against NCI-H2110 non-small cell lung cancer xenografts in SCID mice. SeriMab antiFRα-DGN462 was highly active at a single injection dose of 50 and 25 μg/kg (DGN462 dose) with a minimal efficacious dose (MED) of 10 μg/kg. We have developed a modular method for preparing site-specific, disulfide-linked ADCs with good biochemical characteristics, through an optimized N-terminal serine engineered IgG, using the same thiol containing effector payloads employed in our lysine-conjugation platform. Citation Format: Luke Harris, Daniel Tavares, Lingyun Rui, Erin Maloney, Alan Wilhelm, Juliet Costoplus, Katie Archer, Megan Bogalhas, Lauren Harvey, Rui Wu, Xuan Chen, Xiangyang Xu, Sonia Connaughton, Lintao Wang, Kathleen Whiteman, Olga Ab, Erica Hong, Wayne Widdison, Manami Shizuka, Michael Miller, Jan Pinkas, Thomas Keating, Ravi Chari, Nathan Fishkin. SeriMabs: N-terminal serine modification enables modular, site-specific payload incorporation into antibody-drug conjugates (ADCs). [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 647. doi:10.1158/1538-7445.AM2015-647

Abstract 4564: huHER3-8, a novel humanized anti-HER3 antibody that inhibits exogeneous ligand-independent proliferation of tumor cells

HER3 (ErbB3) is a member of the human epidermal growth factor receptor (HER) tyrosine kinase family that includes EGFR and HER2, two well-known targets for oncology therapy. Kinase signaling of the HER family is initiated by ligand binding that induces receptor homodimerization or heterodimerization and transphosphorylation of the intracellular domain. In this regards, HER3 is unique among the HER kinase family members. Despite being able to bind ligand, HER3 does not possess a catalytically active kinase domain. To turn on the signaling pathway, HER3 needs to form a heterodimer with another HER family member. However, this does not make HER3 inferior with respect to HER pathway signaling, since dimerization of HER3 with HER2 elicits one of the most powerful oncogenic signals. Similar to EGFR and HER2, HER3 expression and activity have been linked to the pathogenesis of various cancers including breast, ovarian and melanoma, and has been associated with resistance to several cancer therapies. The significance of HER3 in cancer pathogenesis prompted us to develop HER3 antagonistic antibodies. A large number of HER3 antibodies were generated by immunizing mice with HER3 expressing cell lines, and screened for the ability to bind human and monkey HER3. Additionally, since HER3 and activated HER3 (phospho-HER3) were detected in the majority of primary as well as metastatic breast cancer tissues by immunohistochemistry staining, antibodies were screened for their ability to inhibit the HER3 ligand-induced proliferation as well as the basal proliferation of breast cancer cell lines. Three HER3 antibodies, HER3-3, HER3-8 and HER3-10, were exceptionally potent in inhibiting the basal proliferation of four breast cancer cell lines that constitutively express phospho-HER3, where other HER3 antibodies (i.e. U1-59 and MM-121) had little or no activity. In SKBR3 cells, these HER3 antibodies inhibited cell proliferation, comparable to the activity of trastuzumab although the HER3 expression level was thirty-fold lower than the HER2 level. All three HER3 antibodies recognized overlapping epitopes. However, only HER3-8 and HER3-10 antibodies strongly competed for the binding of HER3 ligand and inhibited HER2-HER3 dimerization. These HER3 antibodies were also very potent in inhibiting ligand-induced HER3 signaling and tumor cell proliferation. Because of its best overall activities, HER3-8 was humanized by variable domain resurfacing technology. Humanized HER3-8 retained all of the biological properties and activities of muHER3-8. In conclusion, we have generated a novel humanized HER3 antibody, huHER3-8 that is not only active in inhibiting exogenous ligand-induced HER3 signaling and tumor cell growth, but also effective in inhibiting the basal proliferation of HER3-positive tumor cells. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4564. doi:10.1158/1538-7445.AM2011-4564

Abstract 2967: In vitro and in vivo activity of site-specific antibody-drug conjugates (ADCs) with 2 and 4 maytansinoid molecules per antibody prepared through conjugation to SeriMabs (N-terminal serine engineered Abs)

Site-specific attachment of cell-killing agents to antibodies directed against tumor-associated antigens has continued to be an active area of innovation in the field of ADCs. Most reports focus on homogeneous ADCs that have a DAR (cytotoxic molecules per antibody ratio) of 2. Here we describe the preparation, biochemical characterization, and biological evaluation of ADCs made through conjugation of maytansinoids (DM1, DM4) to aldehydes derived from chemically oxidized N-terminal serines (SeriMab) engineered onto the antibody heavy chain (2 DAR) or both light and heavy chain simultaneously (4 DAR). ADCs prepared with a non-cleavable linker or a cleavable disulfide linker were homogeneous 2 or 4 DAR by MS analysis, and were produced in high yield with a monomer content of >98%. Despite conjugation at the N-termini of both the light and heavy chain variable regions, FACS analysis showed the 4 DAR SeriMab conjugates maintained binding to the target antigen. The ADCs showed antigen-specific potency in vitro on a panel of target-expressing cancer cell lines. In the disulfide cleavable linker series, the 2 DAR SeriMab conjugate was 2-5 fold less active than lysine-conjugated Ab-SPDB-DM4 (3.4 DAR), while the 4 DAR SeriMab conjugate was comparably active on an antibody basis. The SeriMab conjugates also displayed strong bystander killing. Surprisingly, in the non-cleavable linker series, the 2 DAR SeriMab conjugate was up to 17-fold more active (depending on cell line) than lysine-conjugated Ab-SMCC-DM1 (3.5 DAR), and the 4 DAR SeriMab conjugate was up to 100-fold more potent than the SMCC-DM1 conjugate on an antibody concentration basis. In a P-gp-positive multi-drug resistant cell line, the non-cleavable 4 DAR SeriMab-maytansinoid conjugate was highly active while the 2 DAR SeriMab ADCs and lysine-conjugated maytansinoid ADCs were >100-fold less potent. The unique oxime bond formed with the non-cleavable SeriMab-maytansinoid conjugate was found to be stable in circulation in mice for >3 days as assayed by affinity capture LC-MS. Polar carboxylic acid containing metabolites were identified which may lead to high cellular retention of maytansinoid species in cancer cells, yielding higher in vitro potency than lysine-linked ADCs in some cell lines. The in vivo anti-tumor activity of disulfide cleavable 2 and 4 DAR SeriMab-DM4 ADCs was evaluated in a clinically relevant cancer xenograft model. The 4 DAR conjugate was active at 60 μg/kg (maytansinoid payload dose) and was more active than the 2 DAR conjugate at this same payload dose. Using the SeriMab conjugation platform we show that in vitro and in vivo activity of site-specific ADCs can be dependent on amount of cytotoxic agent attached per antibody. Citation Format: Luke Harris, Leanne Lanieri, Jose Ponte, Erin Maloney, Laura Bartle, Olga Ab, Juliet Costoplus, Lingyun Rui, Jan Pinkas, Ravi Chari, Thomas Keating, Daniel Tavares, Nathan Fishkin. In vitro and in vivo activity of site-specific antibody-drug conjugates (ADCs) with 2 and 4 maytansinoid molecules per antibody prepared through conjugation to SeriMabs (N-terminal serine engineered Abs). [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2967.

Abstract 653: Antibody-drug conjugates: engineered N-terminal serine residues as a novel approach for site-specific conjugation

Proceedings: AACR 106th Annual Meeting 2015; April 18-22, 2015; Philadelphia, PA Emerging data from the growing number of antibody-drug conjugates (ADCs) in clinical studies reflects the critical contributions from each component of the ADC. The tumor-targeting antibody and the cytotoxic effector molecule are recognized as key elements of an ADC, but increasingly the linkers and linkage mechanisms are being seen as critical for optimizing the design of an ADC based on the specific properties of the target antigen, tumor types, and clinical indications. Traditionally, antibodies have been conjugated to their cytotoxic payloads either at their lysine residues or through cysteine residues accessed through reduction of intrachain disulfide bonds. Recently a number of site-specific conjugation methods have been reported. Here we describe the construction of antibodies and antibody fragments with engineered N-terminal serine residues to facilitate a novel approach for site-specific antibody conjugation. Serine residues can be engineered into at least four different N-terminal positions on an antibody. Serines are found at the N-termini of some naturally occurring human antibodies, but in most cases they must be introduced by either replacing the native N-terminal residue or by including an additional serine residue at the N-terminus. These modifications can be made on the light chain, heavy chain, or both to provide as many as 4 potential linkage sites per antibody. Modification at the N-terminus situates the linked cytotoxic molecule well outside of the antibodys target binding sites in the variable region complementarity determining regions (CDRs). Molecular modeling demonstrates that this is the case whether the linkage is on either the light or heavy chain N-termini. ELISA and flow cytometry binding data demonstrated that changing the N-terminal residue to serine itself had no impact on the binding properties of the parent antibody. Likewise, ADCs utilizing the N-terminal serine for conjugation had similar binding properties compared with lysine or engineered cysteine conjugates of the same antibody. Finally, in vitro cytotoxicity assays demonstrated that N-terminal serine conjugates were at least as potent as conventional ADCs against antigen-expressing tumor cells. The utilization of a variable region framework residue means that the same N-terminal serines used for antibody conjugation can be applied to antibody fragments. Due to the size and residue usage limitations of antibody fragments, conjugation of hydrophobic, cytotoxic agents to these molecules can be challenging. We have built and tested N-terminal serine conjugates with Fabs, scFvs, and sdAbs, establishing the utility of this approach across a wide range of antibody fragment formats. In conclusion, conjugation of engineered antibody N-terminal serine residues represents a promising approach to site-specific ADCs that is worthy of further development. Citation Format: Daniel Tavares, Lingyun Rui, Olga Ab, Luke Harris, Erin Maloney, Thomas Keating, Thomas Chittenden, Nathan Fishkin. Antibody-drug conjugates: engineered N-terminal serine residues as a novel approach for site-specific conjugation. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 653. doi:10.1158/1538-7445.AM2015-653

Abstract 4324: Maytansinoid conjugates of a novel type III anti-CD20 antibody with potent antitumor activity.

Previously identified anti-CD20 antibodies have been classified as either Type I antibodies with lipid raft and complement-dependent cytotoxicity (CDC) activity or Type II antibodies with strong pro-apoptotic activity, but not both. We describe here a novel anti-CD20 antibody with a unique combination of functional properties that defines a new Type III anti-CD20 antibody classification. In addition, when conjugated to the potent, microtubule-acting maytansinoids DM1 or DM4, it demonstrated enhanced cell-killing activity. Novel anti-CD20 antibodies were generated by immunizing mice with CD20-positive cells. One of these murine monoclonal anti-CD20 antibodies demonstrated very strong pro-apoptotic activity against Ramos and Raji lymphoma cells in the absence of cross-linking agents, similar to the activity of Type II antibodies. The humanized version of the antibody, termed D1302A, had a stronger pro-apoptotic effect against Ramos cells than other CD20 antibodies tested including rituximab, tositumumab (B1) and GA101-NG (non-glycoengineered GA101/afutuzumab). Strikingly, D1302A also induced CD20-redistribution to lipid rafts and showed CDC activity similar to the Type I antibody rituximab, while the Type II antibodies tositumumab and GA101-NG lacked these activities. D1302A, rituximab, and ofatumumab demonstrated comparable antibody-dependent cell mediated cytotoxicity (ADCC) activity on Ramos cells with human natural killer effector cells. In vivo, D1302A showed strong efficacy in an SU-DHL-4 xenograft model (at 1 or 10 mg/kg x3) and was more active than rituximab. D1302A was conjugated to DM1 via the non-cleavable SMCC thioether linker and to DM4 via the sulfo-SPDB disulfide linker. The resulting antibody-drug conjugates retained all the functional activities of the unconjugated antibody including binding affinity, pro-apoptotic, CDC and ADCC activities, and showed enhanced and specific in vitro cytotoxicity against Granta-519 and Farage lymphoma cells. In vivo, single-dose treatment with maytansinoid conjugates of D1302A elicited enhanced activity in Daudi and SU-DHL-4 xenograft models as compared to the antibody alone and showed comparable efficacy as standard of care treatment regimens such as rituximab or rituximab + chemotherapy (R-CHOP). D1302A is a novel Type III anti-CD20 antibody that has the functional properties of both Type I (potent CDC activity) and Type II (strong pro-apoptotic activity) antibodies, together with ADCC activity. Conjugation of D1302A with maytansinoids further increases anticancer activity in vitro and in vivo, combining the potent antibody-produced activities with a fourth cytotoxic mechanism distinct from that of other CD20 targeted agents. Therefore, D1302A-maytansinoid conjugates provide a unique combination of anti-tumor activities and are promising therapeutic candidates for the treatment of CD20+ lymphomas and leukemias. Citation Format: Jutta Deckert, Peter U. Park, Yong Yi, Sharon Chicklas, Min Li, Erin K. Maloney, Rui Wu, Leanne Lanieri, Jennifer A. Coccia, Joe Ponte, Lingyun Rui, Daniel J. Tavares, Jan Pinkas, Thomas Chittenden. Maytansinoid conjugates of a novel type III anti-CD20 antibody with potent antitumor activity. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4324. doi:10.1158/1538-7445.AM2013-4324