Factors involved in the pharmacokinetics of COL-3, a matrix metalloproteinase inhibitor, in patients with refractory metastatic cancer: clinical and experimental studies.

COL-3 is an oral, lipophilic, tetracycline analog that has been administered to patients with metastatic cancer. Preliminary assessment of COL-3 in 35 patients with refractory metastatic carcinoma demonstrated apparent nonlinear pharmacokinetics with highly variable oral clearance (63.9% coefficient of variance [CV]). To elucidate possible sources of variability of COL-3 pharmacokinetics in vivo, in vitro plasma protein binding and in vitro metabolism were explored along with in vivo pharmacokinetics using compartmental modeling. The variability in the overall clearance and urinary excretion of COL-3 was also assessed. COL-3 had a long terminal half-life (median = 59.8 h), large apparent volume of distribution (median = 50.2 L), and low apparent clearance (median = 9.93 mL/min). Only adjusted ideal body weight decreased the variability in total apparent clearance. There was nonsaturable plasma protein binding of COL-3 (fu = 5.5%), with the majority of binding to albumin. The renal route of elimination is negligible, with 0.06% of unchanged COL-3 and 3.31% COL-3 glucuronide excreted in the first 6 days. COL-3 is not metabolized by phase I metabolism but does undergo glucuronidation in vitro by UGT1A1, UGT1A3, UGT1A9, and UGT2B7 and in vivo, as evidenced by COL-3 glucuronides in the urine (median = 13.6% of the total dose). COL-3 exhibits nonlinear pharmacokinetics, possibly due to dissolution rate-limited absorption.