Abstract A66: Preclinical rationale for the ongoing Phase 2 study of the hypoxia-activated EGFR-TKI tarloxotinib bromide (TH-4000) in patients with advanced squamous cell carcinoma of the head and neck (SCCHN) or skin (SCCS)

Tarloxotinib bromide (T), a hypoxia-activated prodrug of a potent irreversible EGFR tyrosine kinase inhibitor (T-TKI), was designed to provide clinical benefit in patients with advanced SCCHN and SCCS. In these clinical settings excessive wild type (WT) EGFR overexpression may underlie the limited activity seen with currently available EGFR-TKI where inadequate tumor signal inhibition is evident at recommended doses (Annals Oncol 2007, 18:761). In contrast to the systemic delivery of EGFR-TKI, our prodrug approach selectively targets the hypoxic microenvironment of tumors, with the goal of increasing tumor dose-intensification and thus improving the therapeutic index. Preclinical studies were conducted to test the efficacy of clinically relevant doses of T and the approved EGFR-targeted agents cetuximab and afatinib in preclinical models of SCCHN (FaDu) and SCCS (A431) that express wild type EGFR. Under varying hypoxic conditions, EGFR expression and signal transduction was assessed by western blotting (WB). Pharmacodynamic (PD) endpoints (p-EGFR, p-MAPK, p-AKT) were evaluated by WB or immunohistochemistry (IHC). Under hypoxic conditions, T was 13- and 16-fold more active than in normoxic conditions against FaDu and A431 cells, respectively, with hypoxia-specific release of T-TKI from T by LC/MS (294 and 176 pmol/h/106 cells). Notably, hypoxic exposure of FaDu cells induced a marked 25-fold induction in EGFR phosphorylation (Y1092) that was stable upon reoxygenation (≥5 h), and EGFR signalling in A431 cells was upregulated 2.3-fold post-anoxia. Dose-dependent inhibition of EGFR signalling was observed and correlated with antiproliferative activity. The SCCHN FaDu tumor xenograft had a mean hypoxic fraction (HF) of 14% as determined by IHC detection of pimonidazole binding. Treatment with 48 mg/kg (qw) T (plasma exposure = 150 mg/m2 IV in human subjects) resulted in 100% (8/8) response rate (RR). In contrast, 8 mg/kg (qw) cetuximab treatment (saturating dose) failed to control growth with 8/10 (80%) tumors progressing on treatment and 2/10 (20%) with stable disease. The SCCS A431 tumor xenograft exhibited a mean HF of 31%. Treatment with 30 mg/kg (qw) T (plasma exposure equivalent to 104 mg/m2 IV in human subjects) resulted in 100% (6/6) RR. In contrast, 5 mg/kg daily oral afatinib (plasma exposure = 33 mg daily in human subjects) failed to regress tumors with best response of stable disease in 4/6 (66%) during treatment, while 2/6 (33%) progressed during treatment. T efficacy was accompanied by marked shutdown of total tumor p-EGFR (Try1092) in both xenograft models as well as p-AKT (Ser473) and p-MAPK (Thr202/Tyr204), in the FaDu and A431 xenografts, respectively, which was not evident with the comparator treatments of cetuximab and afatinib. In preclinical models, T exhibited superior efficacy with a 100% response rate compared to cetuximab in SCCHN and afatinib in SCCS with no responses and superior pharmacodynamic effects of T on EGFR target modulation. Based on these models performed with clinically relevant plasma PK levels T may possess a superior therapeutic index relative to approved EGFR-targeted agents. These preclinical data support the weekly dose of 150 mg/m2 of T in the ongoing phase 2 clinical trial in SCCHN and SCCS (NCT02449681). Citation Format: Victoria Jackson, Shevan Silva, Maria Abbattista, Christopher Guise, Matthew Bull, Amir Ashoorzadeh, Charles Hart, Tillman Pearce, Jeff Smaill, Adam V. Patterson. Preclinical rationale for the ongoing Phase 2 study of the hypoxia-activated EGFR-TKI tarloxotinib bromide (TH-4000) in patients with advanced squamous cell carcinoma of the head and neck (SCCHN) or skin (SCCS). [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr A66.