Abstract A34: Hypoxia-activated prodrug TH-302 in combination with gemcitabine for the treatment of pancreatic adenocarcinoma: Preclinical and clinical studies.

Abstract

Hypoxia is a prevalent feature of solid tumors and is a negative prognostic factor associated with treatment failure and the emergence of a more aggressive, invasive, and metastatic phenotype. Hypoxia in human pancreatic adenocarcinoma (PAC) has been characterized by Eppendorf needle oxygen electrodes, and more recently by the use of the exogenous hypoxia biomarker pimonidazole in patient biopsies. Both methods have demonstrated significant subregional hypoxia. TH-302 (T) is a hypoxia-activated bioreductive prodrug that is selectively activated in regions of severe hypoxia, and upon activation releases the bis-alkylating DNA cross-linker bromoisophosphoramide mustard (Br-IPM). T exhibits hypoxia-selective in vitro cytotoxicity across a wide panel of human cancer cell lines including PAC cell lines and in vivo anti-tumor efficacy in a range of human tumor models including a PAC orthotopic model. T is currently under investigation in multiple oncology clinical trials. A panel of human PAC human tumor xenograft models were established (Hs766t, SU.86.86, BxPc3, MIA PaCa2) and tested with gemcitabine (G) and T both as monotherapies and in combination (G+T) for antitumor efficacy. Select tissue biomarkers, focused on characterizing the magnitude and extent of tumor hypoxia, tumor vascularity, and EMT and G resistance/sensitivity phenotypic markers were assessed by IHC and compared with the corresponding drug efficacy profiles observed in the models. Initial clinical study of G+T combination was a phase 1/2a trial with G administered at its full labeled dose and schedule (1000 mg/m 2 on days 1, 8, and 15 of a 28 day cycle) and T dose escalated starting at 240 mg/m 2 in combination with G with the same schedule (T administered 2 hr before G). An expansion phase 2a portion of the trial explored both 240 and 340 mg/m 2 T in combination with G (n=47 pts). A 21% response rate (RR), median PFS and OS of 5.9 and 8.5 mo, and one-yr survival of >40% was observed. These promising results led to initiation of a randomized controlled Phase 2b trial (G vs. G+T240 vs. G+T340). This trial had 80% power to detect 50% improvement in the primary endpoint of PFS with one-sided alpha of 10%. 214 pts were treated across the three arms (~1:1:1). Primary efficacy endpoint was met with a median PFS of 3.6 mo in G arm vs. 5.6 mo in G+T arms with HR of 0.61 (95%CI: 0.43-0.87; logrank p-value of 0.005). Median PFS in G+T340 was 6.0 mo. RR was 12% in G, 17% in G+T240 and 27% in G+T340. Decreases in circulating PAC biomarker CA19-9 were greater in G+T groups than in the G group and greatest in G+T340. G+T was well-tolerated with skin and mucosal toxicities and myelosuppression the most common TH-302 related AEs with no increase in discontinuations for AE. SAEs were balanced across the three arms. These studies of safety and activity of TH-302 in combination with gemcitabine are consistent with the novel design and hypoxia selective characterization of TH 302. Animal and human studies indicate that TH-302, a tumor hypoxia targeting prodrug, can significantly improve the activity of chemotherapy. Citation Format: Charles P. Hart, Jessica D. Sun, Qian Liu, Dharmendra Ahluwalia, Clarence Eng, David P. Ryan, Mitesh J. Borad, Stew Kroll. Hypoxia-activated prodrug TH-302 in combination with gemcitabine for the treatment of pancreatic adenocarcinoma: Preclinical and clinical studies. [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer: Progress and Challenges; Jun 18-21, 2012; Lake Tahoe, NV. Philadelphia (PA): AACR; Cancer Res 2012;72(12 Suppl):Abstract nr A34.