ABSTRACT
Certain drugs inherently have unfavourable pharmacokinetic properties;for example, they are poorly absorbed or broken down too quickly in the liver. In some cases, the addition of a pharmacokinetic excipient, thus deliberately causing an interaction, may offer a solution. To date, this concept has been most widely applied in HIV treatment where addition of the CYP3A inhibitors ritonavir and cobicistat greatly increases plasma levels of other HIV medications. For the same reason, ritonavir has been added to the new oral antiviral drug against the SARS CoV-2 virus, nirmatrelvir. In addition to a better and/or longer effect, theoretically lower doses can also be used, resulting in cost savings. Deliberately inducing a pharmacokinetic interaction is not without risk: after all, interactions with other CYP3A substrates can also occur. Nevertheless, we believe that with good interaction management, CYP3A inhibitors can be used safely with benefits for patients and society.
ABSTRACT
Background : Critically ill patients with COVID-19 are at high risk of thromboembolic events, despite thromboprophylaxis with lowmolecular weight heparins (LWMH), while increased-intensity thromboprophylaxis in this patient population is associated with bleeding. This raises the question whether pharmacokinetic (PK) effects of LMWHs are predictable in these patients. Aims : To investigate whether a dosing algorithm for dalteparin administration could be designed based on clinical parameters, using PK modeling with anti-Xa levels as readout. Methods : In this explorative, observational study, we prospectively included 15 adult COVID-19 patients admitted to the intensive care unit receiving dalteparin in prophylactic-intensity (5000 IU dalteparin once daily (OD) for those <100 kg, 5000 IU dalteparin bidaily (BD) for those ≥100 kg) and therapeutic-intensity (100 IU/kg BD). A minimum of 4 anti-Xa samples per day were collected on regular timepoints over 1-3 days. PK analysis of dalteparin was performed by nonlinear mixed-effect modeling (NONMEM v7.4). The final model was used to perform Monte Carlo simulations to predict anti-Xa levels with different dalteparin regimens. The study was approved by the local medical ethics committee. Results : The data were well-fitted to a linear one compartment model. Wide interindividual variation in the parameters absorption (78%) and clearance (34%) of dalteparin was observed, not explained by clinical covariates such as creatinine clearance for elimination rate. Simulations show that prophylactic dosing in individuals <100 kg result in anti-Xa levels within generally used prophylactic targets, while increased-prophylactic dosing in those ≥100 kg result in supraprophylactic levels in 40% of patients. With therapeuticintensity dosing in secondary thromboprophylaxis, 22% of patients would be subtherapeutically, and 19% patients supratherapeutically dosed. Conclusions : Anti-Xa levels during dalteparin treatment in the critically ill COVID-19 patient are difficult to predict and often off-target. Until data from randomized clinical trials conclude on the best dosing, this suggests that anti-Xa measurements are needed to guide high-intensity dosing in the individual patient.