Purpose: This phase I clinical trial was designed to determine the maximum-tolerated dose and dose-limiting toxicities of the matrix metalloproteinase (MMP) inhibitor COL-3 in patients with refractory solid tumors. Patients and Methods: Thirty-five patients with different cancer types were enrolled. COL-3 doses were escalated from 36 mg/m2/d in successive cohorts of at least three patients. Circulating levels of MMP-2, MMP-9, vascular endothelial growth factor, and basic fibroblast growth factor were assessed during treatment. Pharmacokinetic parameters were assessed for single and multiple doses of drug. Results: Cutaneous phototoxicity was dose-limiting at 98 mg/m2/d. With the use of prophylactic sunblock, COL-3 was well tolerated at 70 mg/m2/d. The dose of 36 mg/m2/d was well tolerated without the use of sunblock. Other toxicities that did not seem to be related to dose or pharmacokinetics included anemia, anorexia, constipation, dizziness, elevated liver function test results, fever, headache, heartburn, nausea, vomiting, peripheral and central neurotoxicities, fatigue, and three cases of drug-induced lupus. Disease stabilization for periods of 26+ months, 8 months, and 6 months were seen in hemangioendothelioma, Sertoli-Leydig cell tumor, and fibrosarcoma, respectively. There was a potentially statistically significant relationship between changes in plasma MMP-2 levels and cumulative doses of drug when progressive disease patients were compared with those with stable disease or toxicity (P = .042). Conclusion: COL-3 induced disease stabilization in several patients who had a nonepithelial type of malignancy. Phototoxicity was dose-limiting. We recommend the dose of 36 mg/m2/d for phase II trials.