ISOLATED LUNG PERFUSION WITH TUMOR-NECROSIS-FACTOR - A SWINE MODEL IN PREPARATION OF HUMAN TRIALS

Isolated lung perfusion with tumor necrosis factor (TNF) potentially could deliver high doses of drug and avoid systemic toxicity in patients with unresectable lung cancer or metastases. We investigated the feasibility of isolated lung perfusion with TNF in a pig model. Eleven animals had left-sided isolated lung perfusion with no TNF (n = 3), 40 mu g/kg TNF (n = 2), 80 mu g/kg TNF (n = 3), and 40 mu g/kg TNF at moderate (39.5 degrees C) hyperthermia (n = 3). Hemodynamic monitoring and measurement of systemic and pulmonary circuit TNF levels were performed. Surviving animals were electively sacrificed a minimum of 6 months after isolated lung perfusion. All sham-perfused pigs survived. Isolated lung perfusion elevated pulmonary artery pressure, decreased cardiac output, and had minimal effects on mean pressure (15 +/- 0 versus 32 +/- 8 mm Hg, 4.5 +/- 1.1 versus 3.03 +/- 0.03 L/min, 67 +/- 11 versus 61 +/- 2 mm Hg; before versus after 90 minutes of isolated lung perfusion). Both 40 mu g/kg animals and 2 of the 3 hyperthermic perfusion pigs survived, with 1 requiring pneumonectomy. Of the three 80 mu g/kg animals, 1 survived, 1 died, and 1 required pneumonectomy. Survivors, compared with dying animals, had lower systemic/pulmonary TNF ratios and lower peak systemic TNF levels. All surviving pigs were electively sacrificed. These data justify phase I human protocols of isolated lung perfusion with TNF and hyperthermia; however, intraoperative leak rates must be monitored to ensure pulmonary isolation because systemic TNF levels may dictate treatment morbidity/mortality.