Characterization and root cause analysis of immunogenicity to pasotuxizumab (AMG 212), a prostate-specific membrane antigen-targeting bispecific T-cell engager therapy

Introduction

In oncology, anti-drug antibody (ADA) development that significantly curtails response durability has not historically risen to a level of concern. The relevance and attention ascribed to ADAs in oncology clinical studies have therefore been limited, and the extant literature on this subject scarce. In recent years, T cell engagers have gained preeminence within the prolific field of cancer immunotherapy. These drugs whose mode of action is expected to potently stimulate anti-tumor immunity, may potentially induce ADAs as an unintended corollary due to an overall augmentation of the immune response. ADA formation is therefore emerging as an important determinant in the successful clinical development of such biologics.

Methods

Here we describe the immunogenicity and its impact observed to pasotuxizumab (AMG 212), a prostate-specific membrane antigen (PSMA)-targeting bispecific T cell engager (BiTE®) molecule in NCT01723475, a first-in-human (FIH), multicenter, dose-escalation study in patients with metastatic castration-resistant prostate cancer (mCRPC). To explain the disparity in ADA incidence observed between the SC and CIV arms of the study, we interrogated other patient and product-specific factors that may have explained the difference beyond the route of administration.

Results

Treatment-emergent ADAs (TE-ADA) developed in all subjects treated with at least 1 cycle of AMG 212 in the subcutaneous (SC) arm. These ADAs were neutralizing and resulted in profound exposure loss that was associated with contemporaneous reversal of initial Prostate Surface Antigen (PSA) responses, curtailing durability of PSA response in patients. Pivoting from SC to a continuous intravenous (CIV) administration route remarkably yielded no subjects developing ADA to AMG 212. Through a series of stepwise functional assays, our investigation revealed that alongside a more historically immunogenic route of administration, non-tolerant T cell epitopes within the AMG 212 amino acid sequence were likely driving the high-titer, sustained ADA response observed in the SC arm.

Discussion

These mechanistic insights into the AMG 212 ADA response underscore the importance of performing preclinical immunogenicity risk evaluation as well as advocate for continuous iteration to better our biologics.

Dosing, safety, and pharmacokinetics (PK) of combination therapy with darolutamide (DARO), androgen-deprivation therapy (ADT), and docetaxel (DOC) in patients with metastatic hormone-sensitive prostate cancer (mHSPC) in the ARASENS study

148

Background: In ARASENS (NCT02799602), DARO in combination with ADT and DOC significantly reduced the risk of death by 32.5% (HR 0.68; 95% CI 0.57–0.80; P<0.001) vs placebo (PBO) + ADT + DOC in patients with mHSPC. Incidences of treatment-emergent adverse events (TEAEs) were similar between treatment groups. We report dosing, safety, and PK of coadministration of DARO and DOC with ADT. Methods: Patients with mHSPC were randomized 1:1 to DARO 600 mg twice daily or PBO, plus ADT and DOC (75 mg/m2 q21d for 6 cycles). The effect of DARO on DOC PK was assessed by noncompartmental analysis from the first 25 patients with dense PK data and by population PK (PopPK) for all patients. DARO PK from ARASENS were compared with PK data from ARAMIS (NCT02200614; without DOC) to evaluate the impact of DOC on DARO PK. Results: Of 1306 randomized patients, 1305 were included in the full analysis set (DARO, n=651; PBO, n=654). The median treatment duration was longer with DARO vs PBO (41.0 vs 16.7 months) and more DARO-treated patients (45.9% vs 19.1%) were still receiving treatment at primary analysis cutoff (Oct 25, 2021). Almost all patients completed 6 cycles of DOC in both groups (DARO, 87.6%; PBO, 85.5%). The proportion of patients requiring DOC dose modification (interrupted/delayed or reduced) was similar between groups (DARO, 60.0%; PBO, 62.9%). TEAEs led to discontinuation/reduction of DOC in 8.0%/19.9% of DARO patients and 10.3%/19.5% of PBO patients. PopPK analysis indicated that DOC PK in ARASENS was generally consistent with that in the literature. A slight numeric increase in DOC exposure was observed in the DARO + DOC + ADT arm, with 15% higher maximum plasma concentration (geometric mean, 1.93 vs 1.68 µg/mL) and 6% higher area under the concentration-time curve (AUC0-tlast within an 8-hour sampling interval, 2.10 vs 1.99 µg·h/mL) vs PBO + DOC + ADT. This small numeric increase is likely not clinically relevant given the variability in DOC exposure (coefficient of variation, 23%–54%). PK meta-analysis of ARASENS and, which considered patients’ intrinsic characteristics as covariates (eg, age, body weight, region), indicated a 10% lower AUC0-12ss of DARO in patients receiving DOC vs those not receiving DOC, which is not considered clinically relevant. Conclusions: The combination of DARO + DOC + ADT increases overall survival with similar overall incidence of TEAEs and no observed drug-drug interactions between DARO and DOC. DARO can be effectively and safely administered with DOC in patients with mHSPC without clinically relevant changes in PK of DARO or DOC. Clinical trial information: NCT02799602 .

Pasotuxizumab, a BiTE® immune therapy for castration-resistant prostate cancer: Phase I, dose-escalation study findings

Aim: We report results of a first-in-human study of pasotuxizumab, a PSMA bispecific T-cell engager (BiTE®) immune therapy mediating T-cell killing of tumor cells in patients with advanced castration-resistant prostate cancer. Patients & methods: We assessed once-daily subcutaneous (SC) pasotuxizumab. All SC patients developed antidrug antibodies; therefore, continuous intravenous (cIV) infusion was assessed. Results: A total of 47 patients received pasotuxizumab (SC: n = 31, 0.5-172 μg/d; cIV: n = 16, 5-80 μg/d). The SC maximum tolerated dose was 172.0 μg/d. A sponsor change stopped the cIV cohort early; maximum tolerated dose was not determined. PSA responders occurred (>50% PSA decline: SC, n = 9; cIV, n = 3), including two long-term responders. Conclusion: Data support pasotuxizumab safety in advanced castration-resistant prostate cancer and represent evidence of BiTE monotherapy efficacy in solid tumors. Clinical trial registration: NCT01723475 (ClinicalTrials.gov).

Quantitative Systems Pharmacology Approaches for Immuno-Oncology: Adding Virtual Patients to the Development Paradigm

Drug development in oncology commonly exploits the tools of molecular biology to gain therapeutic benefit through reprograming of cellular responses. In immuno‐oncology (IO) the aim is to direct the patient’s own immune system to fight cancer. After remarkable successes of antibodies targeting PD1/PD‐L1 and CTLA4 receptors in targeted patient populations, the focus of further development has shifted toward combination therapies. However, the current drug‐development approach of exploiting a vast number of possible combination targets and dosing regimens has proven to be challenging and is arguably inefficient. In particular, the unprecedented number of clinical trials testing different combinations may no longer be sustainable by the population of available patients. Further development in IO requires a step change in selection and validation of candidate therapies to decrease development attrition rate and limit the number of clinical trials. Quantitative systems pharmacology (QSP) proposes to tackle this challenge through mechanistic modeling and simulation. Compounds’ pharmacokinetics, target binding, and mechanisms of action as well as existing knowledge on the underlying tumor and immune system biology are described by quantitative, dynamic models aiming to predict clinical results for novel combinations. Here, we review the current QSP approaches, the legacy of mathematical models available to quantitative clinical pharmacologists describing interaction between tumor and immune system, and the recent development of IO QSP platform models. We argue that QSP and virtual patients can be integrated as a new tool in existing IO drug development approaches to increase the efficiency and effectiveness of the search for novel combination therapies.

Phase I study of pasotuxizumab (AMG 212/BAY 2010112), a PSMA-targeting BiTE (Bispecific T-cell Engager) immune therapy for metastatic castration-resistant prostate cancer (mCRPC)

124

Background: mCRPC has a poor prognosis and immunotherapies are largely ineffective. PSMA is a promising therapeutic target in mCRPC. Pasotuxizumab is a PSMA x CD3 BiTE immune therapy that mediates killing of tumor cells by T cells. Methods: NCT01723475 was a first-in-human, multicenter, dose-escalation study in patients (pts) with mCRPC refractory to standard therapy. Pts received continuous IV infusion of pasotuxizumab in cohorts of 3–4 pts. Dose-escalation followed a continuous reassessment methodology. The primary objective was to determine safety and maximum tolerated dose (MTD); secondary objectives included pharmacokinetics, biomarkers, and tumor response. Results: 16 pts were enrolled into 5 dosing cohorts (5 µg/d, n=3; 10 µg/d, n=4; 20 µg/d, n=3; 40 µg/d, n=4; 80 µg/d, n=2). All pts had ≥1 AE of any grade; most common were fever (94%), chills (69%), and fatigue (50%). 13 pts (81%) had ≥1 AE of grade ≥3; most common were decreased lymphocytes and infections (both 44%). No grade 5 AE occurred. A serious drug-related AE was reported for 1 pt (fatigue, 20 µg/d). No antidrug antibodies were observed. Recruitment was stopped before MTD was reached to allow initiation of a new study sponsored by Amgen. Antitumor activity as indicated by PSA serum level decline was dose dependent, with a mean best PSA change per dosing cohort vs baseline of +0.74% (5 µg/d), −17.9% (10 µg/d), −37.4% (20 µg/d), −42.5% (40 µg/d) and −54.9% (80 µg/d). PSA decreases ≥50% occurred in 3 pts (n=1 each in 20 µg/d, 40 µg/d, 80 µg/d cohorts). One long-term PSA responder was treated for 14 months (40 µg/d) and one for 19.4 months (80 µg/d). The latter pt showed a complete regression of soft-tissue metastases and marked regression of bone metastases by PSMA-PET/CT, >90% reduction in PSA and alkaline phosphatase, and a significant and durable improvement in disease related symptoms. Conclusions: Pasotuxizumab had an acceptable safety profile and dose-dependent clinical activity in mCRPC pts. There were two long term responders in the dose escalation. This is the first clinical study showing that a BiTE immune therapy can be efficacious in solid tumors. Clinical trial information: NCT01723475.

Phase 1 study of pasotuxizumab (BAY 2010112), a PSMA-targeting Bispecific T cell Engager (BiTE) immunotherapy for metastatic castration-resistant prostate cancer (mCRPC)

5034

Background: mCRPC has a poor prognosis and immunotherapies are largely ineffective. PSMA is a promising therapeutic target in mCRPC, and pasotuxizumab is a PSMA x CD3 BiTE that mediates tumor cell killing. Methods: NCT01723475 was a first-in-human, multicenter, dose-escalation study in patients (pts) with mCRPC refractory to standard therapy. Pts received pasotuxizumab as a continuous intravenous infusion in cohorts of 3–4 pts. Dose-escalation followed a continuous reassessment methodology design. The primary objective was to determine safety and maximum tolerated dose (MTD); secondary objectives included pharmacokinetics, biomarkers, and tumor response. Results: 16 pts were enrolled into 5 dosing cohorts (5 µg/d, n = 3; 10 µg/d, n = 4; 20 µg/d, n = 3; 40 µg/d, n = 4; 80 µg/d, n = 2). All pts had ≥1 AE of any grade; most common were fever (94%), chills (69%), and fatigue (50%). 13 pts (81%) had ≥1 AE of grade ≥3; most common were decreased lymphocytes and infections (both 44%). No grade 5 AE occurred. A serious AE related to study drug was reported for 1 pt (fatigue, 20 µg/d). No anti-drug antibodies were observed. Recruitment was stopped before MTD was reached to facilitate initiation of a new study sponsored by Amgen. Antitumor activity as indicated by PSA serum level decline was dose dependent, with a mean best PSA change per dosing cohort versus baseline of +0.74% (5 µg/d), –17.9% (10 µg/d), –37.4% (20 µg/d), –42.5% (40 µg/d) and –54.9% (80 µg/d). PSA decreases of ≥50% occurred in 3 pts (n = 1 each in 20 µg/d, 40 µg/d, and 80 µg/d cohorts). One long-term PSA responder was treated for 14 months (40 µg/d) and one for 19.4 months (80 µg/d). The latter pt showed a complete regression of soft-tissue metastases and marked regression of bone metastases as assessed by PSMA-PET/CT, > 90% reduction in PSA and alkaline phosphatase, and a significant and durable improvement in disease related symptoms. Conclusions: Pasotuxizumab had an acceptable safety profile and dose-dependent clinical activity in mCRPC pts. There were two long term responders in the dose escalation. This is the first clinical study showing that a BiTE immunotherapy can be efficacious in solid tumors. Clinical trial information: NCT01723475.

Abstract 1771: BAY 1834942 is an immunotherapeutic antibody blocking the novel immune checkpoint regulator CEACAM6 (CD66c)

CEACAM6 (CD66c) was previously shown to act as a novel immune checkpoint regulator suppressing the activity of effector T cells against tumors (Witzens-Harig et al., Blood 2013). CEACAM6 is a GPI-linked protein that is strongly expressed at the tumor cell surface in multiple cancer indications such as non-small cell lung adenocarcinoma (NSCLC), colorectal carcinoma (CRC), gastric adenocarcinoma and pancreatic cancer. In general, elevated CEACAM6 expression is associated with advanced tumor stages and poor prognosis. In vitro experiments showed that engagement of T-cells with CEACAM6, either expressed on tumor cells or presented on beads, resulted in suppression of TCR-mediated T-cell activation and ZAP70 phosphorylation. Based on these findings, we hypothesized that antibodies targeting CEACAM6 may be employed to enhance T-cell responses against CEACAM6-expressing cancers. Here we report the generation and characterization of BAY 1834942, a humanized monoclonal antibody selectively blocking the inhibitory impact of CEACAM6 on human T cells. There is no rodent ortholog of CEACAM6 precluding in vivo efficacy studies. In tumor cell / T cell co-culture systems, BAY 1834942 increased secretion of T-cell cytokines and effector molecules (e.g. IFNγ, TNFα, IL-2, granzyme B) and resulted in improved tumor cell killing. The effects of BAY 1834942 were dose-dependent, only observed in the context of CEACAM6-expressing tumor cells and could be reproduced in experiments using tumor cell lines and T-cell preparations from different sources, including T cells derived from tumor infiltrating lymphocytes from pancreatic cancer. BAY 1834942 is cross-reactive with the cynomolgus CEACAM6 ortholog and was well-tolerated in monkey toxicology studies. In summary, BAY 1834942 is a novel checkpoint inhibitor with potential for the treatment of patients with CEACAM6 expressing cancers, both as single agent and in combination with other checkpoint inhibitors. First-in-man trials are expected to commence in 2018.

Citation Format: Joerg Willuda, Mark Trautwein, Jessica Pinkert, Wolf-Dietrich Doecke, Hans-Henning Boehm, Florian Wessel, Yingzi Ge, Eva Maria Gutierrez, Joerg Weiske, Christoph Freiberg, Uwe Gritzan, Julian Glueck, Dieter Zopf, Sven Golfier, Oliver von Ahsen, Ruprecht Zierz, Sabine Wittemer-Rump, Heiner Apeler, Ziegelbauer Karl, Rienk Offringa, Bertolt Kreft, Beckhove Philipp. BAY 1834942 is an immunotherapeutic antibody blocking the novel immune checkpoint regulator CEACAM6 (CD66c) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1771.

Abstract 2791: Physiologically based pharmacokinetic modeling and simulations to estimate the efficacious dose of the CEACAM6 function-blocking antibody BAY 1834942

BAY 1834942 is an immunostimulatory function-blocking (fb) antibody (Ab) against the target carcinoembryonic antigen related cell adhesion molecule 6 (CEACAM6) expressed on tumor cells in multiple cancer indications. The suggested mode of action of BAY 1834942 is the blockade of the immunosuppressive effect of CEACAM6 on activated T cells which restores the immune response against cancer cells.

Available preclinical pharmacokinetic (PK) and in vitro pharmacodynamic (PD) data, target receptor density information and tumor (patho-)physiology were used to create a model framework taking into account the three most essential ‘pillars' (target exposure, target binding and drug activity) to estimate the human efficacious dosing of BAY 1834942. For this purpose, a physiologically-based pharmacokinetic (PBPK) model considering target binding of BAY 1834942 in tumor tissue and on blood granulocytes has been developed. The aim of the PBPK model was to estimate the dose range and regimen for humans leading to exposure level at the tumor site that allows sufficient target binding. The PBPK simulations were based on 1) an analysis of PD in vitro data in order to estimate the degree of saturation needed for maximum drug activity, 2) the assessment of CEACAM6 receptor numbers on tumor cells and blood granulocytes and 3) in vivo PK data in order to develop and evaluate the PBPK model. For the latter, plasma PK data of BAY 1834942 in monkeys were used as well as known tumor concentration-time profiles of the antibodies MOPC21 (non-targeting Ab) and ZCE025 (anti-CEA Ab) in mice and humans. Uncertainty of parameters which are relevant for CEACAM6 target saturation was considered by stochastic in silico simulations to estimate the CEACAM6 saturation at the tumor vs. dosing. This analysis revealed that the predicted human efficacious dose strongly depends on CEACAM6 density. Thus, a low CEACAM6 density scenario (25,000 CEACAM6/tumor cell) and a high CEACAM6 density scenario (250,000 CEACAM6/tumor cell) were simulated and used to support dose selection for the first-in-man (FIM) study of BAY 1834942. The FIM study is currently under preparation.

Citation Format: Sabine Wittemer-Rump, Christoph Niederalt, Joerg Willuda, Mark Trautwein, Merlin Luetke-Eversloh, Wolf-Dietrich Doecke, Clemens Guenther, Christian Scheerans. Physiologically based pharmacokinetic modeling and simulations to estimate the efficacious dose of the CEACAM6 function-blocking antibody BAY 1834942 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2791.

In situ analysis of FGFR2 mRNA and comparison with FGFR2 gene copy number by dual-color in situ hybridization in a large cohort of gastric cancer patients

BACKGROUND

Fibroblast growth factor receptor (FGFR2) has been proposed as a target in gastric cancer. However, appropriate methods to select patients for anti-FGFR2 therapies have not yet been established.

METHODS

We used in situ techniques to investigate FGFR2 mRNA expression and gene amplification in a large cohort of 1036 Japanese gastric cancer patients. FGFR2 mRNA expression was determined by RNAscope. FGFR2 gene amplification was determined by dual-color in situ hybridization (DISH).

RESULTS

We successfully analyzed 578 and 718 samples by DISH and RNAscope, respectively; 2% (12/578) showed strong FGFR2 gene amplification (FGFR2:CEN10 >10); moderate FGFR2 gene amplification (FGFR2:CEN10 <10; ≥2) was detected in 8% (47/578); and high FGFR2 mRNA expression of score 4 (>10 dots/cell and >10% of positive cells with dot clusters under a 20× objective) was seen in 4% (29/718). For 468 samples, both mRNA and DISH data were available. FGFR2 mRNA expression levels were associated with gene amplification; FGFR2 mRNA levels were highest in the highly amplified samples (n = 12). All highly amplified samples showed very strong FGFR2 mRNA expression (dense clusters of the signal visible under a 1× objective). Patients with very strong FGFR2 mRNA expression showed more homogeneous FGFR2 mRNA expression compared to patients with lower FGFGR2 mRNA expression. Gastric cancer patients with tumors that had an FGFR2 mRNA expression score of 4 had shorter RFS compared with score 0-3 patients.

CONCLUSION

RNAscope and DISH are suitable methods to evaluate FGFR2 status in gastric cancer. Formalin-fixed paraffin-embedded (FFPE) tissue slides allowed evaluation of the intratumor heterogeneity of these FGFR2 biomarkers.

Abstract CT094: Phase I study of fibroblast growth factor receptor 2 antibody-drug conjugate (FGFR2-ADC) BAY 1187982 in patients with advanced cancer

Background: FGFR2-ADC (BAY 1187982) consists of an anti-FGFR2 antibody (BAY 1179470) and a stably linked, highly potent and non-cell permeable microtubule-binding auristatin toxophore. A phase I study was conducted in patients (pts) with advanced solid tumors focusing on tumor types with a described expression of FGFR2.

Methods: Eligible pts with solid tumors, known to express FGFR2 and refractory to standard treatment, adequate organ function, and ECOG performance status (PS) ≤ 1 were enrolled. BAY 1187982 was administered by an one hour intravenous infusion on an every 21 days (q3w) schedule in 20 pts (5 cholangiocarcinoma, 5 colon, 2 gastric, 2 breast cancer and 6 others) at doses ranging from 0.1 to 1.3 mg/kg. Dose escalation followed the continous reassessment design treating 3-4 pts per cohort. Pts were assessed for AEs weekly in C1-C3 and on D1 in subsequent cycles. Tumor response was assessed every 6 weeks. Blood samples for a full pharmacokinetic profile were collected after dosing on D1 of C1 and C3 and thereafter every second cycle pre-dose and at the end of infusion.

Results: A total of 20 pts were evaluable: 55% female, mean age 52 yrs (range 24-76), PS 0/1 75% / 25%. No drug-related serious adverse events were reported at dose levels 0.1 and 0.2 mg/kg. At 0.4 mg/kg drug-related serious adverse events were similar to other ADCs, including reversible asymptomatic liver function test increases, and gastrointestinal disorders. Dose limiting toxicities (DLTs) were observed at dose level 0.8 and 1.3 mg/kg, respectively: 0.8 mg/kg cohort grade 3 thrombocytopenia (N=1); 1.3 mg/kg cohort grade 3 corneal epitheliopathy (N=1), grade 4 thrombocytopenia (N=1) and proteinuria progressing to nephrotic syndrome after treatment in C2 or C3 (N=2) with signs of reversibility after discontinuation. In dose level 0.4 and 0.8 mg/kg the majority of pts discontinued study drug treatment already in C1 or C2 due to disease progression and data on proteinuria at these dose levels were absent or spare. As a consequence only doses below 0.4 mg/kg were considered safe. Exposures of BAY 1187982 increased in a dose-proportional manner in the investigated dose range from 0.1 to 1.3 mg/kg.

Conclusion: While this drug was tolerated well at lower dose levels, dose-limiting toxicities occurred below the expected therapeutic dose level. As the predicted human efficacious dose level based on the majority of animal models was above 0.4 mg/kg (Wittemer-Rump et al. AACR; Cancer Res 2015;75, 15 Suppl: Abstract nr 1683) a Phase II dose cannot be recommended and a successful clinical development of the compound was regarded as unlikely. The data from the preclinical safety analysis in monkey and rat will be submitted as abstract to AACR 2017.

Citation Format: Sung-Bae Kim, Funda Meric-Bernstam, Jordan Berlin, Sabine Wittemer-Rump, Motonobu Osada, Ray Valencia, Aleksei Babich, Rong Liu, Andy Hwang, Takahiko Tanigawa, Frank Reetz, Dirk Laurent, Aparna Kalyan. Phase I study of fibroblast growth factor receptor 2 antibody-drug conjugate (FGFR2-ADC) BAY 1187982 in patients with advanced cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT094. doi:10.1158/1538-7445.AM2017-CT094

Preclinical Efficacy of the Auristatin-Based Antibody-Drug Conjugate BAY 1187982 for the Treatment of FGFR2-Positive Solid Tumors

The fibroblast growth factor receptor FGFR2 is overexpressed in a variety of solid tumors, including breast, gastric, and ovarian tumors, where it offers a potential therapeutic target. In this study, we present evidence of the preclinical efficacy of BAY 1187982, a novel antibody-drug conjugate (ADC). It consists of a fully human FGFR2 monoclonal antibody (mAb BAY 1179470), which binds to the FGFR2 isoforms FGFR2-IIIb and FGFR2-IIIc, conjugated through a noncleavable linker to a novel derivative of the microtubule-disrupting cytotoxic drug auristatin (FGFR2-ADC). In FGFR2-expressing cancer cell lines, this FGFR2-ADC exhibited potency in the low nanomolar to subnanomolar range and was more than 100-fold selective against FGFR2-negative cell lines. High expression levels of FGFR2 in cells correlated with efficient internalization, efficacy, and cytotoxic effects in vitro Pharmacokinetic analyses in mice bearing FGFR2-positive NCI-H716 tumors indicated that the toxophore metabolite of FGFR2-ADC was enriched more than 30-fold in tumors compared with healthy tissues. Efficacy studies demonstrated that FGFR2-ADC treatment leads to a significant tumor growth inhibition or tumor regression of cell line-based or patient-derived xenograft models of human gastric or breast cancer. Furthermore, FGFR2 amplification or mRNA overexpression predicted high efficacy in both of these types of in vivo model systems. Taken together, our results strongly support the clinical evaluation of BAY 1187982 in cancer patients and a phase I study (NCT02368951) has been initiated. Cancer Res; 76(21); 6331-9. ©2016 AACR.

Abstract 2089: Therapeutic index prediction of the agonistic aglycosylated TWEAK receptor binding antibody BAY-356

BAY-356 is a novel aglycosylated anti-TWEAK receptor antibody with potent agonistic activity evaluated for cancer therapy. In order to predict an efficacious therapeutic dose of BAY-356 in man, we sought to determine a therapeutic index where the exposure related to therapeutic efficacy was compared with the exposure obtained following doses tested in toxicology studies.

BAY-356 (1 mg/kg) was administered intravenously to healthy Cynomolgus monkeys and plasma concentrations measured in order to determine pharmacokinetic (PK) parameters. Using allometric scaling, the PK in humans was predicted. Efficacy data of BAY-356 in WiDr, HN10321, A253 and SCaBER xenograft models representing colorectal, head and neck squamous cell carcinoma and bladder cancer, were used to derive exposure-response models for each tumor model where the plasma exposure of BAY-356 is assumed to have an effect on the tumor size in a tumor growth model. NONMEM 7.3 was used in the estimations.

The tolerability of BAY-356 was tested in monkeys at doses of 10, 20 and 40 mg/kg by weekly intravenous injections over 4 weeks. Treatment resulted in slight to moderate toxicity in liver, kidneys and pancreas and 10 mg/kg was set to be an exposure dose well tolerated and not to be exceeded by therapeutic exposure. A dosing strategy in humans predicted to result in the same exposure as 10 mg/kg weekly in monkey is 18 mg/kg every third weeks (Q3W).

Using the estimated exposure-response in xenograft models, human tumor doubling times of 8 weeks (A253 and HN10321), 24 weeks (WiDr) and 12 and 24 weeks (SCaBER) in combination with human predicted PK parameters, tumor reduction over time in humans was predicted based on a human dosing strategy of 18 mg/kg BAY-356 Q3W. For the A253 tumor model, stable disease in humans was predicted while the results from HN10321 and SCaBER models predicted a 13-60% decrease in tumor size in humans following 12 weeks treatment. For WiDr, this dosing strategy was predicted to not be efficacious in humans.

The results based on modelling of xenograft data indicate that 18 mg/kg BAY-356 dosed Q3W is predicted to be efficacious in humans for certain tumor cells.

Citation Format: Anders Viberg, Eva Hanze, Lisa Dietz, Ruprecht Zierz, Sandra Berndt, Sabine Wittemer-Rump. Therapeutic index prediction of the agonistic aglycosylated TWEAK receptor binding antibody BAY-356. [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 2089.

Abstract 4933: In Situ analysis of FGFR2 RNA and comparison with FGFR2 gene copy number by dual-color in situ hybridization in a large cohort of gastric cancer patients

Gastric cancer is the 3rd most common cause of cancer-related mortality worldwide, thus new treatment options are urgently needed. In gastric cancer, over-expression of the tumor-promoting tyrosine kinase receptor fibroblast growth factor receptor 2 (FGFR2) has been described, representing a potential therapeutic target. FGFR2 expression in gastric cancer has been reported to be heterogeneous. Most methods for detection of RNA in FFPE (formalin fixed paraffin embedded) do not provide spatial resolution for assessment of tumor heterogeneity. Here we used a novel RNA in situ technique called RNAscope. The technology allows visual investigation of FGFR2 transcription on FFPE slides. Samples from 1036 gastric adenocarcinoma patients who underwent surgery in National Cancer Center Hospital East (Chiba, Japan) were assembled in tissue micro-arrays and analyzed by RNAscope. A total of 578 samples were also analyzed by DISH to determine gene amplification. Using these tissue based methods we were able to assess the localization and heterogeneity of both FGFR2 gene amplification and RNA expression. Strong FGFR2 gene amplification (FGFR2:CEN10 &gt;10) was detected in 2% of 578 samples. Moderate FGFR2 gene amplification (FGFR2:CEN10 between 2 and 10) was seen in 8% of the samples. High FGFR2 RNA expression (score 4+) was seen in 4% of 718 evaluable samples. Gene amplification and RNA expression were closely correlated. All samples with dense clusters of score 4+ RNA showed FGFR2:CEN10 ratios &gt;10. The identical tumor areas showed high gene amplification and RNA expression. FGFR2 RNA and gene amplification was highly heterogeneous in the tissue. Only 0.4% of the samples showed homogeneous FGFR2 expression in &gt;80% of the tumor cells. High RNA expression intensity was associated with a more homogeneous expression pattern compared to moderate FGFR2 expression. In early stage I/II gastric cancer samples with score 4+ RNA expression are mostly of the differentiated type. RNA score 4+ patients of grade III/IV gastric cancer were mostly of undifferentiated histology. In a subset of 195 samples with available data cMet IHC 3+ and Her2 positivity status were not mutually exclusive with regards to FGFR2 RNA Score 4+ expression. In a multivariant analysis FGFR2 RNA expression was associated with patient outcome. Gastric cancer patients with score 4+ RNA expression had a shorter progression free survival compared to patients with score 0-3. Interestingly, there was a trend for patients with homogeneous score 3+ RNA expression, but not heterogeneous score 3+ expression, to show shorter PFS and OS suggesting that patients with high intensity or homogeneous FGFR2 RNA expression may define a subgroup of patients with gastric cancer. Patients with homogenous or strong focal FGFR2 expression might therefore be candidates for FGFR2 directed therapies in gastric cancer.

Citation Format: Yasutoshi Kuboki, Christoph A. Schatz, Sabine Jabusch, Karl Koechert, Janine Feng, Sabine Wittemer-Rump, Karl Ziegelbauer, Thomas Krahn, Akiko Nagatsuma, Atsushi Ochiai. In Situ analysis of FGFR2 RNA and comparison with FGFR2 gene copy number by dual-color in situ hybridization in a large cohort of gastric cancer patients. [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 4933.

Abstract 1684: Preclinical anti-tumor efficacy of FGFR2-ADC BAY 1187982 in patient-derived gastric, breast and ovarian cancer models

Antibody-drug conjugates (ADC) represent a promising therapeutic approach for treatment of cancer. We have developed a novel ADC directed against fibroblast growth factor receptor 2 (FGFR2). FGFR2 is overexpressed in several cancer indications, such as gastric, breast, and ovarian cancer and thus represents a potential therapeutic target for treatment of FGFR2-positive cancer patients with ADC-based therapy.

FGFR2-ADC consists of a fully human anti-FGFR2-Ab (BAY 1179470) conjugated via a stable linker to a novel auristatin cytotoxic agent licensed from Seattle Genetics. FGFR2-ADC exhibits low nM to sub-nM potency in vitro in a panel of FGFR2-positive cancer cells lines (SNU-16, MFM-223, NCI-H716) while being inactive against FGFR2-low or -negative cell lines (MDA-MB-231, HEK-293, BaF/3) and highly selective versus a control ADC.

FGFR2-ADC is highly efficacious in monotherapy and results in tumor growth inhibition in the gastric cancer xenograft model SNU-16 and tumor regression in the breast cancer xenograft model MFM-223. FGFR2-ADC induces tumor stasis in the colorectal cancer xenograft model NCI-H716 and regrown tumors are sensitive to a second treatment cycle of FGFR2-ADC.

FGFR2-ADC shows high efficacy in vivo in monotherapy in patient-derived (PDX) FGFR2-positive murine xenograft models, e.g. in the ovarian cancer model OV30-0511A. FGFR2-ADC is also efficacious in the gastric cancer PDX model GC10-0608 and the breast cancer model MAXF857.

The toxophore metabolite of FGFR2-ADC was more than 30-fold enriched in tumors versus other organs (liver, spleen, kidneys) in NCI-H716 tumor-bearing mice.

Based on the preclinical efficacy, PK and tolerability data, evaluation of FGFR2-ADC in cancer patients appears warranted. A Phase I study is planned.

Citation Format: Anette Sommer, Charlotte Kopitz, Christoph Schatz, Ruprecht Zierz, Joachim Schuhmacher, Sabine Wittemer-Rump, Klaas Prins, Manuela Braun, Frank Reetz, Bertolt Kreft, Hung T. Huynh, Karl Ziegelbauer. Preclinical anti-tumor efficacy of FGFR2-ADC BAY 1187982 in patient-derived gastric, breast and ovarian cancer models. [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 1684. doi:10.1158/1538-7445.AM2015-1684

Abstract 1683: Pharmacokinetic/pharmacodynamic (PK/PD) and toxicokinetic/toxicodynamic (TK/TD) modeling of preclinical data of FGFR2-ADC (BAY 1187982) to guide dosing in phase 1

BAY 1187982 is an antibody drug conjugate (ADC) directed against fibroblast growth factor receptor 2 (FGFR2). FGFR2 is overexpressed in several cancer indications, such as gastric, breast, and ovarian cancer. Anti-tumor efficacy of BAY 1187982 has been demonstrated in several FGFR2-positive cancer cell line as well as patient-derived xenograft models. Toxicology findings from repeated dose preclinical safety studies in monkeys indicated effects related to the liver, kidney, heart and coagulation system. To predict the therapeutic index of BAY 1187982 in humans and to support the design of the first-in-human (FIH) study with respect to selection of dose and regimen, preclinical efficacy and toxicity findings were quantified. All available preclinical PK, TK, tumor response and toxicity data from mouse models and monkey studies were used to create a model framework to describe the PK, TK, PK/PD and TK/TD relationship as functions of BAY 1187982 dose, regimen and time. Human PK parameters based on scaling from monkey were used to predict PK profiles in humans for a range of doses and schedules. These sets of predicted exposure models were combined with the PK/PD as well as the TK/TD model to assess the expected efficacy (according to RECIST criteria) and toxicity range in humans, respectively. The dosing schedule leading to the largest therapeutic index and the dose escalation schema for the FIH study were determined. The FIH study is currently under preparation.

Citation Format: Sabine Wittemer-Rump, Anette Sommer, Charlotte Kopitz, Hung Huynh, Christoph Schatz, Ruprecht Zierz, Manuela Braun, Kirstin Meyer, Dirk Laurent, Jörg Lippert, Klaas Prins. Pharmacokinetic/pharmacodynamic (PK/PD) and toxicokinetic/toxicodynamic (TK/TD) modeling of preclinical data of FGFR2-ADC (BAY 1187982) to guide dosing in phase 1. [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 1683. doi:10.1158/1538-7445.AM2015-1683

Abstract 2836: Development of a companion diagnostic IHC assay for the biomarker-driven selection of C4.4a positive patients

C4.4a (LYPD3), a GPI-anchored cell surface protein, has been identified previously as a cancer- and metastasis-associated cell surface protein. It is expressed in a variety of cancer indications and, particularly, in the squamous cell subtype of non-small cell lung cancer (NSCLC) and head and neck cancer. Targeting C4.4a with a specific antibody-drug conjugate showed a potent and selective antitumor activity in various human xenograft models and in patient-derived NSCLC tumor models (Willuda J et al., AACR 2014). Since it has been previously shown in vivo that the expression of C4.4a is required for the response to anti-C4.4a treatment, we initiated the development of an IHC assay that might be used to detect C4.4a expression in patient tumor samples. For the development of this companion diagnostic IHC assay, a novel monoclonal rabbit anti-C4.4a antibody was generated. Since C4.4a is post-transcriptionally modified by shedding, we aimed to identify an antibody that binds to a similar epitope as the therapeutic C4.4a antibody. The immunizations of rabbits with specific peptide sequences of C4.4a and with the full length C4.4a recombinant protein produced several antibody candidates. Those antibody candidates were characterized in regard to their staining characteristics, domain binding, and specificity for C4.4a. Furthermore, the staining pattern of the antibody candidates and the therapeutic C4.4a antibody were compared side by side to ensure that stainings using the newly generated antibodies are predictive for anti-C4.4a treatment. The most promising antibodies were selected to set up and optimize an IHC assay on the Ventana staining platform. In summary, we developed a companion diagnostic IHC assay that might be used to detect C4.4a expression in patient tumor samples.

Citation Format: Claudia Schneider, Joseph Couto, Yifei Zhu, Zhiming Liao, Robert Pytela, Alton Hiscox, Sabine Wittemer-Rump, Ulf Forssmann, Lars Linden, Joerg Willuda, Daniel Forler, Matthew Nelson, Ricarda Finnern, Thomas Krahn, Khusru Asadullah. Development of a companion diagnostic IHC assay for the biomarker-driven selection of C4.4a positive patients. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2836. doi:10.1158/1538-7445.AM2014-2836

Abstract 4491: FGFR2-ADC potently and selectively inhibits growth of gastric and breast cancer xenograft models

Antibody-drug conjugates (ADCs) represent a promising therapeutic approach for treatment of cancer. We have developed a novel ADC directed against fibroblast growth factor receptor 2 (FGFR2). FGFR2 is overexpressed in several cancer indications, such as gastric cancer and breast cancer, thus representing an interesting therapeutic target for the treatment of FGFR2 positive cancer patients with an ADC-based therapy. The FGFR2-ADC consists of the fully human anti-FGFR2-mAb BAY 1179470 conjugated via a stable linker to a novel auristatin cytotoxic agent (technology licensed from Seattle Genetics). The FGFR2-mAb BAY 1179470, which is cross-reactive with human, mouse, rat and monkey, induces internalization of FGFR2. Quantitative data on FGFR2 antibody bound per cell (ABC) were determined with the QuantiBrite assay using BAY 1179470. FGFR2-ADC has a potency in the single digit nM to sub nM range in a panel of FGFR2-positive cells lines (e.g., SNU-16, KatoIII, SUM52-PE, MFM-223) and shows more than 100-fold selectivity versus FGFR2-negative cell lines. High levels of FGFR2 on cancer cells correlate with internalization efficacy and cytotoxic activity in vitro. FGFR2-ADC is highly efficacious in monotherapy and results in tumor growth inhibition in the gastric cancer xenograft model SNU-16 and tumor regression in the breast cancer xenograft model MFM-223. At doses efficacious in mice, FGFR2-ADC is well tolerated. The pre-clinical efficacy and tolerability data obtained for FGFR2-ADC suggest a therapeutic index and support clinical testing.

Citation Format: Anette Sommer, Carl F. Nising, Christoph Mahlert, Charlotte C. Kopitz, Hans-Georg Lerchen, Simone Greven, Beatrix Stelte-Ludwig, Joachim Schuhmacher, Ruprecht Zierz, Sabine Wittemer-Rump, Christoph Schatz, Frank Reetz, Heiner Apeler, Rolf Jautelat, Bertolt Kreft, Karl Ziegelbauer. FGFR2-ADC potently and selectively inhibits growth of gastric and breast cancer xenograft models. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4491. doi:10.1158/1538-7445.AM2014-4491

Abstract 672: Pharmacokinetic and pharmacodynamic (PK/PD) modeling of preclinical data of a novel anti-fibroblast growth factor receptor 2(FGFR2) antibody (BAY 1179470) to guide dosing in Phase 1

BAY 1179470 is a human immunoglobulin (IgG) antibody directed against fibroblast growth factor receptor 2 (FGFR2). FGFR2 is overexpressed on cancer cells during tumorigenesis and metastasis. Anti-tumor efficacy of BAY 1179470 has been demonstrated in FGFR2-expressing gastric cancer mouse and rat xenograft models. No toxicologically relevant findings were observed in the repeated dose preclinical safety studies in rats and monkeys. Therefore, toxicity was not an appropriate endpoint for the first-in-man (FIM) study and dose selection was decided to be based on PK/PD modeling and predictions from preclinical efficacy data. All preclinical PK and tumor response data from mouse models were used to create a model framework to describe the PK and PK/PD as functions of BAY 1179470 dose, regimen and time. In a subsequent confirmatory step the PK/PD model derived from mouse efficacy data was applied to predict the efficacy in rats. Human PK parameters based on scaling from monkey were used to predict PK profiles in humans for a range of doses and schedules. These sets of predicted exposure were combined with the PK/PD model developed from mouse models to assess the expected efficacy for a range of human tumor doubling rates. Minimal effective doses, maximum effect and maximum effect doses, and maximal effective doses in humans were derived for various dosing intervals (1 day to 6 weeks). From the maximum effect dose the target AUC in humans was calculated and doses for the FIM study selected accordingly. The FIM study is currently ongoing in Japan (NCT01881217).

Citation Format: Sabine Wittemer-Rump, Charlotte Kopitz, Hung Huynh, Lars Lindbom, Bernhard Beckermann, Motonobu Osada, Dirk Laurent, Jörg Lippert. Pharmacokinetic and pharmacodynamic (PK/PD) modeling of preclinical data of a novel anti-fibroblast growth factor receptor 2(FGFR2) antibody (BAY 1179470) to guide dosing in Phase 1. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 672. doi:10.1158/1538-7445.AM2014-672