Congenital adrenal hyperplasia (CAH) is a rare form of adrenal insufficiency causing deficiency of the highly regulated hormone cortisol and accumulation of its precursors such as 17α-hydroxyprogesterone (17-OHP) and subsequent androgen overproduction. Symptoms associated with CAH are premature pseudo puberty, earlier ending of longitudinal growth, and in female patients, virilisation and hirsutism. CAH patients require life-long cortisol replacement therapy, and dose optimisation through therapy monitoring is crucial to avoid potentially serious adverse events due to cortisol over- or underexposure. Paediatric CAH patients receive hydrocortisone (HC, synthetic cortisol) for cortisol replacement due to its lower risk for adverse effects whereas adult patients receive more potent glucocorticoids, e.g., dexamethasone (Dex). Especially in paediatrics, dried blood spot (DBS) sampling represents a highly advantageous alternative to plasma sampling. The major advantages include minimal invasiveness, low required blood volumes, stability of the analyte and easy storage of the matrix. Thus, DBS sampling has a high potential for facilitating CAH therapy monitoring routine. However, target concentrations of CAH biomarkers such as 17-OHP indicating a successful cortisol replacement are still unknown in DBS. To prevent in utero virilisation of female foetuses with CAH, prenatal therapy with Dex, administered to the pregnant women, has been conducted for decades. Yet, prenatal CAH therapy is still considered experimental since the traditionally administered Dex dose of 20 μg/kg/day is not based on a scientific rationale and is assumed to be too high, causing potential harm to the mother and foetus. In this regard, quantitative approaches such as pharmacometric modelling and simulation are powerful tools to provide a better understanding on pharmacokinetic (PK) and pharmacodynamic (PD) processes and to contribute to the optimisation of drug therapies. This work aimed at paving the way towards an optimised CAH therapy in paediatric and foetal populations by (1) providing insights into the quantitative relationship between cortisol concentrations measured in plasma and in DBS, (2) identifying paediatric target DBS concentrations for the commonly used biomarker 17-OHP and (3) suggesting a rational Dex dose in prenatal CAH therapy. To quantitatively link plasma and DBS cortisol concentrations, a semi-mechanistic nonlinear mixed-effects (NLME) PK model was developed based on data from paediatric CAH patients. The model characterised a nonlinear relationship between cortisol in plasma and DBS with plasma/DBS concentration ratios decreasing from approximately 8 to 2 with increasing DBS cortisol concentrations up to 800 nmol/L. These ratios decreased due to saturation of cortisol binding to corticosteroid-binding globulin and thus higher cortisol fraction associated with red blood cells. In future, more data from neonates and infants can be used to investigate a possible age effect, on the nonlinearity between plasma and DBS cortisol, in addition to the observed concentration effect. For the first time, a target morning DBS 17-OHP concentration range was determined for monitoring paediatric CAH patients. The DBS target range of 2.1-8.3 nmol/L was derived from simulations by applying a developed PK/PD model linking cortisol in plasma to 17-OHP in DBS and by leveraging healthy paediatric cortisol profiles. By extending the PK/PD model, and using the same simulation approach, circadian target concentration profiles, providing DBS biomarker targets for any time of the day, can be derived in future. Furthermore, in Bland-Altman and Passing-Bablok analyses, it was shown that capillary and venous DBS concentrations, which are both commonly obtained in clinical practice, are comparable to each other for cortisol and 17-OHP in paediatric CAH patients. For determining a reduced Dex dose which simultaneously decreases the risk for adverse events in prenatal CAH therapy and still shows sufficient efficacy in the foetus, a target Dex concentration range was identified from literature and a NLME model describing maternal Dex PK was developed. The Dex PK model was used to simulate maternal Dex concentration-time profiles following traditional or reduced Dex doses and to evaluate the tested dosing regimens with regard to the lowest effective dose. Based on the simulation results, a Dex dose of 7.5 μg/kg/day was suggested as a rational dose for prenatal CAH therapy, representing approximately a third of the traditional Dex dose. The suggested rational Dex dose should be evaluated in future clinical trials. In summary, this work provides quantitative insights into DBS measurements for CAH therapy monitoring, presents first target DBS concentrations for the biomarker 17-OHP in paediatrics, and suggests a first model-based dose rationale for Dex in prenatal CAH therapy. Ultimately, this work can help to improve CAH treatment with HC and Dex and therapy monitoring in the highly vulnerable paediatric and foetal populations.
