BACKGROUND Argininosuccinate synthetase (ASS) was the first of two enzymes to convert citrulline to arginine. This pathway allowed cells to synthesize arginine from citrulline, making this amino acid nonessential for the growth of most mammalian cells. Previous studies demonstrated that several human tumor cell lines were auxotrophic for arginine due to an inability to express ASS. Selective elimination of arginine from the circulation of animals with these tumors is a potentially effective anticancer treatment. The purpose of these experiments was to determine the frequency of ASS deficiency and arginine auxotrophy in a variety of human malignant tumors. METHODS The authors analyzed the expression of ASS by immunohistochemistry with a monoclonal antibody in a variety of human tumor biopsies. They found that the incidence of ASS deficiency varied greatly with the tumor type and tissue of origin. RESULTS Melanoma, hepatocellular carcinoma, and prostate carcinoma were most frequently deficient in ASS. Some human cancers were almost always positive for ASS (e.g., lung and colon carcinomas). However, other human cancers, including sarcomas, invasive breast carcinoma, and renal cell carcinoma, also were sometimes ASS deficient. CONCLUSIONS These data indicated that immunohistochemical detection of ASS may prove an effective means for determining ASS deficiency in malignant human tumors and for identifying patients most likely to respond to arginine deprivation therapy. Based on these results, human clinical trials using arginine‐degrading enzyme therapy to treat patients with advanced melanoma or hepatocellular carcinoma have been initiated. Cancer 2004;100:826–33. © 2004 American Cancer Society.

PK and PD studies indicated that a dose of 160 U/m2 lowered plasma arginine from a resting level of approximately 130 micromol/L to less than 2 micromol/L for at least 7 days; nitric oxide levels also were lowered. There were no grade 3 or 4 toxicities directly attributable to the drug. Six of 24 phase I to II patients responded to treatment (five partial responses and one complete response; 25% response rate) and also had prolonged survival. CONCLUSION Elimination of all detectable plasma arginine in patients with metastatic melanoma was well tolerated and may be effective in the treatment of this cancer. Further testing of ADI-SS PEG 20,000 mw in a larger population of individuals with metastatic melanoma is warranted.