Depression in childhood and adolescence is a serious and increasing public-health problem associated with profound short- and long-term consequences, including impaired social and educational functioning, increased risk of somatic illness and substance use, and elevated mortality from suicide. Up to half of depressive disorders beginning in youth persist into adulthood, and the COVID-19 pandemic, with its attendant social isolation, has further increased the incidence of clinically significant depressive illness in children and young adults. Despite the high burden of disease, pharmacotherapeutic guidance for pediatric depression remains limited. Selective serotonin reuptake inhibitors (SSRIs) are first-line agents, yet dosing recommendations for youth do not adequately account for major sources of variability such as developmental pharmacokinetics, body size, age, comorbid diagnoses, and pharmacogenetic differences. Fluoxetine is commonly recommended in pediatric populations largely for historical reasons, while newer SSRIs such as escitalopram and sertraline are increasingly prescribed off-label without robust comparative evidence. This randomized clinical trial evaluates therapeutic drug monitoring (TDM)-guided treatment with fluoxetine, sertraline, or escitalopram in children and adults diagnosed with Major Depressive Disorder. TDM procedures will be implemented across all randomized arms to capture drug exposure and enable dose adjustments according to predefined safety and efficacy criteria. Treatments will be initiated at the lower end of recommended pediatric doses and titrated if prespecified clinical nonresponse or tolerability thresholds are met, allowing evaluation of dose-response dynamics while prioritizing patient safety. The study combines prospective, randomized comparative effectiveness methodology with intensive pharmacokinetic/pharmacodynamic (PK/PD) assessment. Repeated plasma concentration measurements will be collected at multiple timepoints to characterize within- and between-patient PK variability for each SSRI. Standardized clinical instruments will be used longitudinally to assess depressive symptom severity, response and remission rates, time to onset of therapeutic effect, adverse events (including early activation and worsening of suicidal ideation), treatment adherence, and functional outcomes (academic performance, social functioning, and quality of life). Pharmacogenetic testing will be performed to evaluate the contribution of genetic variants (including but not limited to CYP enzymes, serotonin transporter and MAO-A related polymorphisms) to PK parameters and clinical response. Integrated population PK and PK/PD modeling will (1) quantify concentration-effect and concentration-adverse-effect relationships over time, (2) identify demographic, clinical and genetic moderators of exposure and response, (3) evaluate nonlinearity in PK (e.g., dose-dependent absorption or clearance patterns), and (4) derive evidence-based therapeutic concentration ranges and individualized dosing algorithms for pediatric practice. Secondary objectives include head-to-head comparison of efficacy, tolerability, time to response, and persistence on treatment between the three SSRIs, as well as exploratory analyses of diagnostic subtypes and comorbidities that may influence outcomes. By filling critical gaps in controlled PK/PD data for the most commonly prescribed SSRIs in youth, this trial aims to move beyond meta-analytic inference toward actionable, individualized dosing recommendations. The expected outcomes are earlier identification of atypical pharmacokinetics, safer and more effective dose optimization, reduced adverse events and early discontinuation, shortened time to symptomatic improvement, and ultimately improved functional recovery and quality of life for children and young adults with depression. The study's translational PK/PD outputs will inform clinical TDM protocols, pediatric dosing guidelines, and future precision-medicine strategies in adolescent psychopharmacology.

This randomized, comparative clinical trial is designed to produce robust, clinically actionable pharmacokinetic/pharmacodynamic (PK/PD) evidence to support therapeutic drug monitoring (TDM) and individualized dosing of three widely used SSRIs - fluoxetine, escitalopram and sertraline - in pediatric and young-adult patients with moderate to severe depressive episodes. The trial integrates randomized head-to-head comparisons, prospective TDM applied across all arms, dense plasma sampling, pharmacogenetic profiling and population PK/PD model development as core elements to address current gaps in dosing guidance for young patients. Depression starting in youth results in substantial immediate and long-term harms, including impaired social and educational functioning, increased somatic comorbidity and higher suicide risk. Current pediatric SSRI prescribing is constrained by limited pediatric-specific exposure-response data and by historical inertia favoring older agents. The principal objective is to define exposure-efficacy and exposure-safety relationships for the three SSRIs in the target age range, identify demographic and genetic moderators of drug disposition and effect, and derive model-based therapeutic concentration ranges and practical dosing recommendations that can be applied in clinical TDM. Secondary objectives include comparative effectiveness and tolerability assessments, time-to-response evaluation, and exploratory subgroup analyses by diagnosis and comorbidity.Participants are randomized in equal proportions to one of three SSRI treatment arms under a common TDM framework. The trial enrolls children, adolescents and young adults (lower age bound set in the protocol) presenting with a depressive episode judged to warrant SSRI therapy. The design balances internal validity for PK/PD inference with pragmatic elements reflecting real-world clinical decision-making: conservative initial dosing, structured clinical response assessment, a single protocolized dose escalation for nonresponse, and predefined criteria for discontinuation or transition to standard care. The randomized comparative arm structure enables head-to-head evaluation of exposure-response characteristics and safety profiles while TDM across all arms facilitates within-subject and between-subject PK characterization.Eligible patients undergo a multi-stage screening including medical and medication histories, physical and psychiatric examinations, baseline laboratory testing and age-appropriate psychiatric rating scales. Concomitant medications and factors likely to affect drug metabolism are recorded; participants on interacting medications must undergo washout periods according to conservative rules to avoid confounding PK/PD assessments. Informed consent/assent procedures are implemented according to age and local regulations, and unique coded identifiers are used to preserve confidentiality of clinical and biological data. Study medication is initiated at cautious, age-appropriate low doses and administered once daily. Clinical efficacy and tolerability are monitored frequently using validated instruments. If a participant fails to achieve a minimal, prespecified improvement within an early observation window (for example, three weeks), the protocol permits a single, structured dose escalation; subsequent lack of benefit leads to discontinuation of study drug and transition to routine clinical management (counted as treatment failure in primary analyses).Blood sampling schedule captures early absorption, weekly trough concentrations during the active treatment window, and an intensive multi-point profile at the end of the treatment period to support high-fidelity PK/PD modelling. Sampling volumes are minimized per draw to reduce burden in younger participants. Samples are processed promptly, frozen and stored under controlled conditions until analysis. Baseline DNA samples are collected for genotyping and retained in a coded biobank for subsequent analysis. Plasma drug and relevant metabolite concentrations are quantified using validated LC-MS/MS methods with rigorously applied quality controls to ensure precision and accuracy across the clinical concentration range. A targeted genotyping panel focuses on polymorphisms with known or plausible influence on SSRI pharmacokinetics (e.g., CYP enzymes) and pharmacodynamics (e.g., serotonin transporter, MAO-A and receptor variants). Laboratory QC, duplicate checks and call-rate thresholds are used to ensure genotype data reliability.In addition to standard clinical and pharmacokinetic assessments, the study embraces a systems pharmacology approach by systematically evaluating a comprehensive panel of molecular biomarkers to better characterize depression phenotypes and their response to treatment. Blood samples will be drawn at specific baseline and endpoint visits to quantify neuroplasticity and astrocyte function markers, including Brain-Derived Neurotrophic Factor (BDNF), Glial Cell Line-Derived Neurotrophic Factor (GDNF), and Glial Fibrillary Acidic Protein (GFAP). The inflammatory profile will be assessed by measuring peripheral pro-inflammatory factors such as TNF-alpha and IL-6. Furthermore, to investigate the hypothesis of reduced brain bioenergy in depression, the study will measure mitochondrial and bioenergetic markers, including lactate, pyruvate, alpha-ketoglutarate, oxaloacetate, succinate, fumarate, and malate. The biomarker panel is completed by evaluating epigenetic and cellular aging factors, specifically selected microRNA (miRNA) variants, telomere shortening, and the expression and mitochondrial translocation of hTERT (a mitochondria-protective factor). These biomarker profiles will be integrated into the covariate analysis to explore their relationship with PK/PD parameters, treatment outcomes, and depression symptoms.Population PK and PK/PD modelling using nonlinear mixed-effects techniques is a central analytic pillar. Model building will proceed through exploratory data analysis, selection of structural models to describe concentration-time profiles (e.g., 1-2 compartment candidates), and appropriate PD model forms to capture symptom trajectories and adverse-effect dynamics (e.g., indirect response or transit models). Covariate screening will include age, body size descriptors, sex, genotype, concomitant medications and diagnostic features. Model performance will be evaluated with diagnostic plots, bootstrap or other resampling methods and visual predictive checks. Derived models will be used for simulation studies to propose therapeutic concentration windows, dosing adjustments across age/genotype strata, and to assess alternative dosing regimens. Standard comparative statistical analyses (intention-to-treat and per-protocol) will evaluate clinical endpoints across arms; time-to-event techniques will be applied to time-to-response outcomes.Frequent, systematic clinical evaluations monitor depressive symptoms, global clinical status, functional outcomes (school, social functioning), and adverse effects. Suicidality is assessed with validated instruments at every contact and escalation procedures are predefined to ensure rapid clinical response to worsening risk. Safety laboratories (including hepatic and metabolic panels, thyroid tests where indicated), ECGs as required, and pregnancy testing for persons of childbearing potential are performed at specified intervals. Investigators are trained to distinguish emergent adverse drug reactions from symptom fluctuations of the underlying illness. All adverse events are recorded, graded and reported in accordance with regulatory and good clinical practice requirements. The visit schedule is concentrated in the initial 8-week treatment period with weekly clinical visits coinciding with trough PK sampling; additional visits are scheduled for intensive PK profiling near study end and for final safety and functional outcome assessments. Medication administration timing is standardized on visit days to assure accurate PK sampling. All data are captured in secure electronic case report forms with audit trails. Biological specimens and genetic data are coded and stored separately from identifiable clinical data, accessible only to authorised personnel. Central monitoring procedures, data validation checks and on-site monitoring visits are used to ensure data integrity-particularly for time-sensitive variables crucial to PK/PD modelling (accurate dosing and sampling timestamps).The study will produce population PK and PK/PD models for each SSRI, evidence-based therapeutic concentration ranges, and genotype- and age-stratified dosing recommendations suitable for TDM implementation. Comparative effectiveness data will clarify relative onset of action, tolerability profiles and rates of treatment persistence. Given the elevated baseline suicide risk in the population, the protocol mandates intensive monitoring and clear clinical escalation algorithms. Concomitant medication management and conservative washout rules reduce the risk of pharmacodynamic interactions and serotonin toxicity. Enrollment and follow-up are planned over a multi-year period to achieve sample sizes that permit robust estimation of fixed and random effects in nonlinear mixed-effects models and to allow adequately powered head-to-head clinical comparisons. The dense sampling strategy combined with targeted genotyping supports both population-level inferences and subgroup analyses.

Inclusion Criteria:

• Male or female gender.
• Age 12 years or older.
• Diagnosis of moderate or severe depressive episode requiring SSRI treatment (fluoxetine, escitalopram, or sertraline).
• Written informed consent provided by patient and/or legal guardian.
• Women of childbearing potential must agree to use effective contraception or sexual abstinence.

Exclusion Criteria:

• Pregnancy, breastfeeding, or planning pregnancy.
• Diagnosis of other mental disorders (schizophrenia, bipolar disorder, intellectual disability).
• Decompensated severe somatic disorders (e.g., endocrine disorders, asthma exacerbation, hepatic/renal failure).
• Treatment with fluoxetine, sertraline, or escitalopram at an adequate dose for at least 4 weeks for the current episode (unless washout of 5x half-life is observed).
• Use of medications with high risk of CYP enzyme interactions.

Anonymized individual participant data (IPD) regarding pharmacokinetic profiles, clinical depression scale scores, and biomarker profiles will be stored in a specialized database. This dataset will serve as a repository for future scientific research, big data analysis, and the development of new hypotheses. Data may be made available to cooperating research centers in accordance with ethical principles and General Data Protection Regulation regulations.

Data will become available following the publication of the primary study results and the completion of the project (anticipated November 2027).

Anonymized data will be available as supplementary data for the manuscripts. For more detailed data, access requests will be reviewed to ensure compliance with ethical guidelines and data protection regulations (GDPR/RODO).