To determine the dose-limiting toxicity of oxaliplatin chemotherapy delivered by in vivo lung perfusion (IVLP). To allow assessment of subacute toxicities, we aimed to develop a 72-hour porcine IVLP survival model. In total, 12 Yorkshire male pigs were used. Left lung IVLP was performed for 3 hours. At 72 hours postoperatively, computed tomography imaging of the lungs was performed before the pigs were killed. Lung physiology, airway dynamics, gross appearance, and histology were assessed before and during IVLP, at reperfusion, and when the pigs were euthanized. An accelerated titration dose-escalation study design was employed whereby oxaliplatin doses were sequentially doubled provided no clinically significant toxicity was observed, defined as an arterial partial pressure of oxygen to fraction of inspired oxygen ratio <300 mm Hg or severe acute lung injury on biopsy. After an initial training phase, no mortality or adverse events related to the procedure were observed. There was no lung injury observed at the time of IVLP for any case. At sacrifice, clinically significant lung injury was observed at 80 mg/L oxaliplatin, with an arterial partial pressure of oxygen to fraction of inspired oxygen ratio of 112 mm Hg. Mild and subclinical lung injury was observed at 40 mg/L, with this dose being repeated to confirm safety. A stable and reproducible porcine 3-day IVLP survival model was established that will allow toxicity assessment of agents delivered by IVLP. Oxaliplatin delivered by IVLP showed delayed-onset toxicity that was not apparent at the time of reperfusion, with a maximal-tolerated dose of 40 mg/L. This information will inform initiation of a clinical trial examining IVLP delivery of oxaliplatin at our institution. Copyright © 2020. Published by Elsevier Inc.