Abstract A006: Identification of a novel, safety-engineered ADC directed against FGFR4 for the treatment of Rhabdomyosarcoma

Abstract

Rhabdomyosarcoma (RMS) is the most common soft tissue sarcoma in children and young adults. While low risk disease has a 90% survival probability, metastatic RMS has a dismal prognosis. Treatment options include chemotherapy and local therapy (surgery, radiotherapy) but suffer from serious long-term side effects. Therefore, novel and well-tolerated treatments for RMS are urgently needed. The membrane receptor tyrosine kinase FGFR4, is a downstream transcriptional target of PAX::FOXO1 fusion proteins, frequently found in high-risk alveolar RMS. FGFR4 is upregulated also in a large proportion of fusion-negative RMS, albeit to a lower extent, making FGFR4 an attractive target for antibody-drug conjugates (ADCs) to be potentially used in the treatment of a significant percentage of RMS patients. However, FGFR4 also has important physiological functions in tissues such as liver where it is involved in the control of bile acid homeostasis. Accordingly, FGFR4 targeting antibodies and small molecules show bile-acid induced diarrhea as dose limiting toxicities. To identify a novel ADC with high tolerability in pediatric patients, we screened a large number of antibodies directed against FGFR4 and identified mAB#3384 as candidate. This mAb binds to a novel epitope on FGFR4 and shows minimal interference with normal signaling, largely improved internalization and excellent manufacturability properties. mAB#3384 showed low nM binding affinity in BLI assays, was fully cross-reactive with mouse, rat and monkey FGFR4 and did not bind to any of the other FGFR family members. Exi01, a Topo I-DAR 8 ADC based on mAB#3384 showed sub-nM in vitro cell killing activity on RMS cell lines, strong bystander effect and evidence for induction of immunogenic cell death. Remarkably, both the ADC and the free toxin were poorly active on normal hepatocytes. In vivo, tumor regressions and cures were observed in an RH30 RMS model at doses between 2.5 and 10 mg/kg. Efficacy was superior to that detected with an equivalent ADC directed against a different epitope. Treatment with a single dose of 5 mg/kg of Exi01 led to intratumoral toxin accumulation at levels well above cell culture IC50 for several days, whereas liver concentrations of the Topo I toxin were >100 fold lower than in vitro hepatocyte IC50s. Exi01 was well tolerated, with no clinical signs, no changes in blood chemistry or histologic alterations of liver, kidney, heart and lung tissues. The parent antibody was clean in a non-GLP human tissue cross reactivity panel. Exi01 was further Fc-silenced to avoid immune-mediated adverse reactions and is ready to enter IND-enabling studies. In summary, we have identified a novel, highly engineered ADC with excellent in vitro and in vivo profiles to be developed as a potential new treatment for RMS patients.