Objective: To determine some of the factors influencing the time to progression of cytomegalovirus (CMV) retinitis among HIV-infected patients being treated with foscarnet.
Design: A retrospective analysis of two open-label Phase I/II studies in multiple university hospitals. Patients were studied in both inpatient and outpatient settings.
Patients: Of the patients in the databases examined, 31 had adequate pharmacokinetic information and 29 had information on outcome and the other patient covariates.
Intervention: After induction therapy with foscarnet at a dose of 60 mg/kg three times daily was completed, patients had maintenance therapy with 60-120 mg/kg foscarnet once daily. Doses were subsequently adjusted for changed estimated creatinine clearance.
Measurements: The measured endpoint was time to progression of CMV retinitis. The independent variables examined to determine influence on time to progression included mean peak foscarnet concentration, mean area under the concentration-time curve (AUC) for foscarnet, the positive or negative outcome of a baseline blood culture for CMV, the initial CD4 cell count for a patient and the peak CD4 cell count observed during maintenance therapy.
Results: A wide range (-10-fold) of foscarnet AUC was observed, even though only a fourfold dose range was employed, and doses were altered for changing estimated creatinine clearance. In a multivariate Cox model, only AUC and the status of the baseline CMV blood culture significantly affected the time to progression of the retinitis.
Conclusion: The AUC produced by a dose of foscarnet has a wide interindividual range. The AUC of foscarnet significantly altered time to progression of the retinitis. However, patients with positive baseline CMV blood cultures had a significantly more shallow dose-response curve. This indicates that the added risk of nephrotoxicity which is present with aggressive foscarnet dosing might be best borne by the subgroup of patients with a positive CMV blood culture at baseline.