Preclinical safety and antitumor activity of the arginine-degrading therapeutic enzyme pegzilarginase, a PEGylated, cobalt-substituted recombinant human arginase 1.

Metabolic remodeling contributes to the development and progression of some cancers and exposes them to vulnerabilities, including specific nutrient dependencies that can be targeted therapeutically. Arginine is a semiessential amino acid, and several cancers are unable to endogenously synthesize sufficient levels of arginine for survival and proliferation, most commonly due to reduced expression of argininosuccinate synthase (ASS1). Such cancers are dependent on arginine and they can be targeted via enzyme-mediated depletion of systemic arginine. We report the preclinical safety, antitumor efficacy, and immune-potentiating effects of pegzilarginase, a highly potent human arginine-degrading enzyme. Toxicology studies showed that pegzilarginase-mediated arginine depletion is well tolerated at therapeutic levels that elicit an antitumor growth effect. To determine which tumor types are best suited for clinical development, we profiled clinical tumor samples for ASS1 expression, which correlated with pegzilarginase sensitivity in vivo in patient-derived xenograft (PDx) models. Among the histologies tested, malignant melanoma, small cell lung cancer and Merkel cell carcinoma had the highest prevalence of low ASS1 expression, the highest proportion of PDx models responding to pegzilarginase, and the strongest correlation between low or no ASS1 expression and sensitivity to pegzilarginase. In an immune-competent syngeneic mouse model, pegzilarginase slowed tumor growth and promoted the recruitment of CD8 + tumor infiltrating lymphocytes. This is consistent with the known autophagy-inducing effects of arginine depletion, and the link between autophagy and major histocompatibility complex antigen presentation to T cells. Our work supports the ongoing clinical investigations of pegzilarginase in solid tumors and clinical combination of pegzilarginase with immune checkpoint inhibitors.
(Copyright © 2020 Elsevier Inc. All rights reserved.)