The purpose of this study is to develop population pharmacokinetic models for antiretroviral drugs in a pediatric population.

The interest of these models is multiple :

• describe the pharmacokinetics of these drugs in children and explain the inter-individual variability of concentrations through covariates such as weight, age, sex, smoking status, co-treatments and bilirubin;
• estimate maximum, minimum and exposure concentrations from the individual pharmacokinetic parameters for each patient;
• propose adaptations of doses for certain classes of children (according to age, weight etc.) and individualize the doses.

HIV (Human Immunodeficiency Virus) affects 36.7 million people worldwide. Major advances have been made in the discovery of antiretroviral therapy, significantly improving the lives of patients. Although these treatments have been studied and validated in adults, ethical and technical difficulties are hampering research in the pediatric population. However, it is important to know the administration patterns in children, as during its development, multiple physiological changes affect the pharmacokinetics as well as pharmacodynamics of drugs. As a result, the child can not be considered as a small adult and it is not possible to adjust the dosage by taking into account only his weight, age or body surface area.

In France, monitoring of HIV-infected children includes quantification of viral load and may also include pharmacological therapeutic monitoring. In fact, blood samples in children are taken to verify compliance is correct and their plasma concentrations are considered to be effective. Many data are thus generated and not exploited. However, a population pharmacokinetic method would allow us to understand the variability of concentrations existing between these children.

Demographic factors (age, sex, weight, smoking status, etc.) and clinical (bilirubinemia, viral load, genetics, co-treatments, etc.) can be included as covariates to explain inter-individual variability. This method of study is interesting in pediatrics because it has the advantage of being able to include many patients with little sampling per subject. The data used are generally plasma concentrations obtained as a result of sampling performed as part of the therapeutic follow-up of patients.

In clinical practice, the pharmacokinetic model allows to simulate doses and frequencies of administration but also to predict drug interactions. Indeed, patients infected with HIV are often poly-medicated, which represents a risk of drug interactions, especially since many molecules are inducing / inhibiting cytochromes P450.

The main goal is to develop population pharmacokinetic models for antiretroviral drugs in children.

The interest of these models is multiple:

• describe the pharmacokinetics of these drugs in children and explain the inter-individual variability of concentrations through covariates such as weight, age, sex, smoking status, co-treatments and bilirubin;
• estimate maximum, minimum and exposure concentrations from the individual pharmacokinetic parameters for each patient;
• propose adaptations of doses for certain classes of children (according to age, weight etc.) and individualize the doses.

The secondary objectives of this work are:

• Build models jointly with several antiretroviral drugs, accounting for the strength of interactions between them during multiple therapies.
• Link antiretroviral concentrations to the effects of treatment (decreased viral load) : pharmacokinetic-pharmacodynamic study with concentration / efficacy and concentration / toxicity relationships.
• The evaluation of preexisting models in the literature and the comparison of our data with the results of these models (external validation).

Pharmaco-statistical analysis will be carried out on the retrospective data of patients treated with one or more antiretroviral molecule(s) and whose blood dosage of the drug(s) as part of their therapeutic follow-up is available

Inclusion Criteria:

• Children from 0 to 18 years;
• Treatment with one antiretroviral drug (s) studied (dolutegravir, raltégravir, rilpivirine, nevirapine, atazanavir, darunavir, ritonavir));
• Blood dosage of the drug (s) as part of their therapeutic follow-up in the Pharmacology laboratory of the Cochin hospital between 2007 and 2017

Exclusion Criteria:

• Concentration too low below the limit of quantification (indicating an absence of medication
• patient with doubt about compliance

Minor patient treated by one or more antiretroviral and for which a blood test has been performed