RESUMO
PURPOSE: Cancer patients with predominantly hepatic metastases have poor outcomes and limited options. Hepatic arterial infusion (HAI) of a therapeutic agent may be an appropriate option for producing increased drug concentrations at the tumor sites while reducing systemic adverse effects in normal tissues. METHODS: Patients with predominantly hepatic metastases (n = 48) were placed in 6 groups according to nanoparticle albumin-bound paclitaxel (nab-paclitaxel) dose level using a 3 + 3 design plus dose expansion for responsive tumor types. We evaluated the toxicity, antitumor activity, and pharmacokinetics of nab-paclitaxel delivered via HAI. RESULTS: Thirty-eight and ten patients underwent HAI over 1 and 4 h, respectively, at doses of up to 300 mg/m(2). The treatment was safe and exhibited antitumor activity. Pharmacokinetic analyses revealed that HAI of nab-paclitaxel over 4 h resulted in markedly lower peak drug concentrations (C max) and longer times to peak concentration (T max) than that over 1 h. The self-control pharmacokinetic studies showed that HAI of nab-paclitaxel led to much lower C max and areas under the curve (AUC), compared with intravenous infusion. CONCLUSIONS: HAI of nab-paclitaxel at up to 300 mg/m(2) over 4 h was well tolerated. Pharmacokinetic evaluation of C max, T max, and AUC implied that 4-h HAI enhanced hepatic extraction of nab-paclitaxel. Further preclinical and clinical studies are required to develop reliable methods of evaluation of hepatic extraction (clinicaltrials.gov registration number NCT00732836, first registered on August 8, 2008, and last updated on October 27, 2014).