Research for clinical pharmacogenetics implementationFrom candidate gene studies to physiologically based pharmacokinetic modelling

  1. Pablo Zubiaur Precioso
Supervised by:
  1. Francisco Abad Santos Director
  2. Miriam Sáiz Rodríguez Director

Defence university: Universidad Autónoma de Madrid

Year of defence: 2020

  1. Antonio Javier Carcas Sansuan Chair
  2. Luis Gandía Juan Secretary
  3. Luis Andrés López Fernández Committee member
  4. Lucia Llanos Jimenez Committee member
  5. Cristina Rodriguez Gonzalez de Antona Committee member

Type: Thesis


Precision medicine (PrM) is an approach of contemporary medicine in which treatments are adjusted to the patient’s individual characteristics. PrM considers all possible aspects that determine the response to drugs, such as pharmacokinetic and biopharmaceutical factors or drug pharmacodynamics, also known as the mechanism of action of the drug. Furthermore, certain factors modify the response to drugs, such as demographic characteristics, health status or drug interactions. The response to drugs depends on all these elements and, additionally, on genetic variants located in genes related to their pharmacokinetics and pharmacodynamics. This field of medicine is known as clinical pharmacogenetics, an essential part of PrM. For the progression of clinical pharmacogenetics and PrM, it is necessary to conduct association studies in which the impact of genetic polymorphisms on clinical events is investigated. These clinical events include pharmacokinetic parameters, safety or effectiveness of drugs. Eventually, these associations are validated and clinical pharmacogenetic guidelines are elaborated. As an example, some polymorphisms reduce metabolizing enzyme activity, therefore drugs may accumulate and the patient may be overexposed to the drug. Consequently, dose reductions or treatment change are warranted. However, it is not enough to find a pharmacogenetic association to directly implement prescribing modifications in the clinical practice. A process of clinical validation is required, which may last many years, as it often requires clinical trials or numerous observational studies. An alternative methodology to evaluate the effect of dose adjustments in patients is the use of computerized predictive pharmacokinetic models, namely Physiologically-Based Pharmacokinetic (PBPK) models. Even after the publication of clinical guidelines, the implementation in clinical practice is slow due to the lack of awareness of many physicians and pharmacists about pharmacogenetics. In view of the relative youth of clinical pharmacogenetics, it is therefore necessary to devote effort to teaching and convincing physicians and pharmacists to make use of this tool. The aim of this doctoral thesis was to contribute to the implementation of clinical pharmacogenetics. For this respect, different methodologies were used: firstly, the development of three candidate gene pharmacogenetic studies for three drugs with different levels of previous pharmacogenetic information available: efavirenz, progesterone and dabigatran; secondly, the evaluation of CYP2C19 phenotype-guided dose adjustments for voriconazole by means of PBPK modelling; thirdly, to contribute to the implementation of clinical pharmacogenetics through various methodologies, namely the publication of pharmacogenetic reviews or the validation of new laboratory genotyping techniques. The efavirenz study was timely and contemporary to the publication of the Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline on CYP2B6 and efavirenz. A predictive model of efavirenz pharmacokinetics was optimized based on G516T and rs4803419 variants which improved the predictive power of G516T alone. Concerning progesterone work, it was the first study to date to evaluate pharmacogenetic determinants of the exposure to exogenously administered progesterone, orally and vaginally. The key finding was the relationship observed between CYP2C19 impaired phenotypes and progesterone overexposure. Regarding the study with dabigatran, previous works had reported CES1 and ABCB1 variants to alter dabigatran exposure, however, in this work, these relationships were not observed. Alternatively, CYP2D6 functional impairment was related to drug accumulation, suggesting that dabigatran is metabolized by this enzyme. Moreover, voriconazole PBPK models were able to excellently predict voriconazole disposition based on CYP2C19 phenotype. The ultrarapid (UM) and poor (PM) metabolizer phenotypes required sharper dose adjustments than the 50% dose increase or decrease proposed by the Dutch Pharmacogenomics Working Group (DPWG), respectively. Indeed, no phenotype was adequately exposed to voriconazole therapeutic window after receiving the standard administration protocol, which justifies dosage individualization. In addition, the validation of an HLA-B*57:01 surrogate biomarker, HCP5 rs2395029, for the prediction of abacavir hypersensitivity reaction (ABC-HSR), was conducted and published. This is a cheaper and faster alternative method to the one traditionally used in our laboratory. Finally, due to the SARS-CoV-2 infection, or coronavirus disease 2019 (Covid-19), a review about the pharmacogenetics of the drugs used for the management of the disease was published. Until a vaccine for the prevention of the disease is available, the off-label use of repurposed drugs will continue, therefore this document will be a valid pharmacogenetic reference. Ultimately, this doctoral thesis represents an advance in the implementation of clinical pharmacogenetics through the application of a number of different strategies. Moreover, the current landscape of Spanish and international pharmacogenetic implementation initiatives was described.