ARC-9 A Randomized Study to Evaluate Etrumadenant Based Treatment Combinations in Previously Treated Metastatic Colorectal Cancer (mCRC) Authors Zev A. Wainberg 1 , Sae-Won Han 2 , Soohyeon Lee 3 , Keun-Wook Lee 4 , Scot

ARC-9 A Randomized Study to Evaluate Etrumadenant Based Treatment Combinations in Previously Treated Metastatic Colorectal Cancer (mCRC)

Authors Zev A. Wainberg1, Sae-Won Han2, Soohyeon Lee3, Keun-Wook Lee4, Scott Kopetz5, Jonathan Mizrahi6, Yong Sang Hong7, Francois Ghiringhelli8, Antoine Italiano9, David Tougeron10, Brandon Beagle11, Mathew Boakye11, Tingting Zhao11, Joon Rhee12, Dimitry S.A. Nuyten11, Michael Cecchini13

Affiliations 1David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA 2Seoul National University Hospital, Seoul, South Korea 3Korea University Anam Hospital, Seoul, South Korea 4Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, South Korea 5The University of Texas MD Anderson Cancer Center, Houston, TX 6Ochsner Medical Center, New Orleans, LA 7Asan Medical Center, Seoul, South Korea 8Centre Georges Francois Leclerc, Dijon, France 9Institut Bergonie - Centre Regional de Lutte Contre Le Cancer de Bordeaux et Sud Ouest, Bordeaux, France 10Centre Hospitalier Universitaire de Poitiers, Poitiers, France 11Arcus Biosciences, Inc., Hayward, CA 12Gilead Sciences, Inc., Foster City, CA 13Yale University School of Medicine, New Haven, CT

Background Adenosine mediated signaling via A2a and A2b receptors impairs activation, proliferation, and cytotoxic activity of effector T cells resulting in inhibition of antitumor activity. Adenosine receptor blockade may therefore enhance therapeutic efficacy of chemo immunotherapy regimens. Etrumadenant (E) is an orally bioavailable, selective, A2a and A2b receptor antagonist that has demonstrated safety and clinical activity in solid tumors when combined with chemo immunotherapy.

ARC-9 (NCT04660812), a Phase Ib II trial evaluating the safety and efficacy of E combined with zimberelimab (Z) (anti-PD-1 antibody), and FOLFOX bevacizumab (bev) regimens or novel therapies in 3 cohorts of patients (pts) with mCRC. While FOLFOX rechallenge is frequently used in the management of mCRC, no prospective trials have been performed to date.

Methods Cohort B enrolled pts who previously progressed on both oxaliplatin and irinotecan containing regimens. Pts were randomized 2 1 to EZFB E (150 mg orally PO once daily QD ) + Z (240 mg intravenous IV once every 2 weeks Q2W ) + mFOLFOX-6 + bev (5 mg kg IV Q2W), or regorafenib (rego) (160 mg PO QD days 1-21 every 4 weeks ). Pts who progressed on rego were allowed to crossover to EZFB. The primary endpoint is progression-free survival (PFS) per RECIST 1.1 and overall survival (OS) is a key secondary endpoint.

Results In Cohort B, 112 pts were randomized. Baseline characteristics are comparable between arms. As of 13 Nov 2023, with median follow up of 20.4 months (mo), EZFB demonstrated statistically significant improvement in PFS (HR 0.27, 95% CI 0.17-0.43, p 0.0001) and OS (HR 0.37, 95% CI 0.22-0.63, p 0.0003) vs rego (Table). EZFB had an acceptable safety profile, consistent with the known FOLFOX bev toxicity profile with no deaths related to study treatment in either arm.

Conclusions In this randomized phase II clinical trial, EZFB significantly improved efficacy outcomes compared to rego in refractory mCRC pts previously treated with 5-FU, oxaliplatin and irinotecan regimens with no unexpected toxicities. Further investigation of this promising regimen is warranted given the clinically meaningful PFS and OS improvement.

aAdministered if bevacizumab is not contraindicated. HR, hazard ratio ORR, objective response rate OS, overall survival PFS, progression-free survival TEAE, treatment emergent adverse event