Background/Objectives: Infliximab, ipilimumab, and nivolumab are three monoclonal antibodies that have been associated with hepatotoxicity. Three separate physiologically based pharmacokinetic (PBPK) models were developed in GastroPlus^® to simulate plasma and liver concentrations in patient populations after administration of either infliximab, ipilimumab, or nivolumab. Methods: The models include distribution and clearance mechanisms specific to large molecules, FcRn binding dynamics, and target-mediated drug disposition (TNF-α for infliximab, CTLA-4 for ipilimumab, and PD-1 for nivolumab). Results: The PBPK model for each large molecule was able to reproduce observed plasma concentration data in patient populations, including patients with rheumatoid arthritis and patients with solid tumors. Liver concentrations were predicted to be between 10% and 23% of the plasma concentrations for each of the three drugs, aligning with previously reported results. This lends further validity to the PBPK models and their ability to accurately predict hepatic concentrations in the absence of direct tissue measurements. Conclusions: These results can be used to drive liver toxicity predictions using the quantitative systems toxicology model, BIOLOGXsym™, which integrates hepatic interstitial concentrations with in vitro mechanistic toxicity data to predict the extent of liver toxicity for biologics.

RENAsym is a quantitative systems toxicology (QST) modeling platform designed to predict drug-induced kidney toxicity. The model includes the proximal tubule cell (PTC) life cycle, regeneration of PTCs, sodium reabsorption, nephron death, and drug-induced injury and inflammation within the proximal tubules. The mechanisms of drug-induced toxicity included are overproduction of reactive oxygen species and mitochondrial dysfunction. Inputs into the QST model include the drug’s exposure in the kidney and results from in vitro toxicity assays used to parameterize the toxicity mechanisms included in the model. The QST model has a wide array of outputs including fraction of viable PTCs, fraction of viable nephrons, and serum creatinine concentration over time that allow the user to assess the extent of kidney injury induced by a drug. Polymyxin B, valproate, and acetaminophen were three compounds used to test RENAsym’s ability to correctly classify compounds as either nephrotoxic or safe. The model successfully identified polymyxin B and valproate as drugs with the potential to cause kidney injury and acetaminophen as a drug that does not cause kidney injury at therapeutic levels, providing increased confidence that the model can be used to indicate kidney injury liability using in vitro inputs before dosing in clinical settings.