Dataset used for model development.

Purpose
The pharmacokinetics of voriconazole have been studied across various populations but data specific to the Pakistani cancer population has not yet been reported. The aim of present study was to explore and identify covariates that affect pharmacokinetics of intravenous voriconazole in Pakistani cancer population.
Methods
The therapeutic drug monitoring data from January1st, 2023 to December 31st, 2023 of cancer patients receiving intravenous voriconazole for systemic fungal infections were taken from electronic medical record of the hospital. The data were used for the development of population pharmacokinetic model using NONMEM. Impact of various covariates such as age, weight, sex, liver function test, serum creatinine, creatinine clearance, type of cancer (primary diagnosis) and type of fungal infection were assessed through stepwise covariate modeling. Bootstrap analysis and goodness of fit plots were used to evaluate robustness and predictive performance of final model.
Results
One compartment model best described the included data with first order elimination. The value of voriconazole clearance was 6.17 L/h with interindividual variability of 83.7% while volume of distribution was 55.9 L. The clearance of voriconazole was significantly influenced by renal status of patients. Creatinine clearance and primary diagnosis were significant covariates affecting clearance of voriconazole in covariate analysis.
Conclusion
The findings suggest that this model can be used for dosage adjustment based on creatinine clearance and primary diagnosis as they impact significantly on voriconazole clearance in cancer patients. This approach is especially valuable in resource-limited settings like Pakistan, where individualized therapy can enhance the safety and efficacy of antifungal treatment, addressing the unique clinical and demographic challenges in vulnerable populations.