We conducted a single institution phase II trial to evaluate the tolerability and effectiveness of therapy with arsenic trioxide (ATO) and ascorbic acid (AA) with temozolomide (TMZ) in patients with advanced melanoma. Planned enrollment was for 40 patients. Eligible patients were required to have metastatic melanoma with or without brain metastases, an ECOG performance status of 0-2, and adequate organ function. The primary endpoints were overall response rate and degree of grade 4 toxicity. ATO was administered as a loading dose at 0.25 mg/kg/day for 5 days during week 0, and then twice weekly at 0.35 mg/kg during an 8-week cycle. Each infusion of ATO was followed by an infusion of 1000 mg of AA. TMZ was given at the standard dose of 200 mg/m for 5 days during weeks 1 and 5 of each cycle. Eleven patients were enrolled with 10 evaluable for response, five with central nervous system disease. No responses were seen among the first 10 patients and on the basis of a predetermined stopping rule, the trial was terminated. Common grade 1 and 2 side effects included nausea and vomiting (10), fatigue (six), edema (six), rash (six), and elevated AST or ALT (six). Grade 3 and 4 side effects included nausea and vomiting (three), elevated AST or ALT (two), seizure (one), and renal failure (one). This is the first trial combining ATO with chemotherapy in a solid tumor. ATO and AA were administered in the outpatient setting with TMZ in 11 patients with an acceptable side effect profile. No responses were seen in the first 10 evaluable patients leading to early closure of the study. Further studies using ATO and AA with TMZ with this dosing schedule in advanced melanoma are not warranted.