Skip to content

adhd

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

adhd

README.md

ADHD QSP Model

Attention Deficit Hyperactivity Disorder (ADHD)

Quantitative Systems Pharmacology Model

Directory: adhd/ | Abbreviation: ADHD | Date: 2026-06-26
Category: Neuropsychiatric | ICD-10: F90 | DSM-5: 314.0x
Prevalence: ~5–7% children · ~2.5% adults worldwide

ADHD QSP Mechanistic Map

Disease Overview

Attention Deficit Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder characterized by persistent patterns of inattention, hyperactivity, and impulsivity that interfere with functioning and development. It affects approximately 5–7% of school-age children and 2.5% of adults globally, making it one of the most prevalent psychiatric conditions across the lifespan.

The core pathophysiology involves dysregulation of catecholaminergic neurotransmission — primarily dopamine (DA) and norepinephrine (NE) — in prefrontal cortical circuits governing executive function, working memory, and inhibitory control. Genetic studies implicate multiple risk variants (DAT1, DRD4, DRD5, SNAP25, COMT) that collectively reduce dopaminergic tone in frontostriatal pathways, leading to the clinical syndrome.

Core Pathophysiology

Genetic Risk Factors (DAT1, DRD4, COMT, SNAP25, BDNF Val66Met)
  │
  ├─► Reduced DA synthesis/release in VTA→PFC (mesocortical pathway)
  ├─► Reduced NE tone from LC→PFC projections
  └─► Delayed PFC cortical maturation (~3 year lag vs. neurotypical controls)
        │
        ▼
PFC Dopamine↓ + NE↓ → Suboptimal catecholamine tone
  │
  ├─► D1R understimulation → Weakened AMPAR synapses → Working Memory↓
  ├─► α2A-AR understimulation → HCN channel open → Ih current↑ → Network instability
  └─► Prefrontal inhibitory control ↓, Default Mode Network overactive
        │
        ▼
Clinical ADHD Syndrome:
  ├─► Inattention (ADHD-RS Inattention subscale ≥6/9 symptoms)
  ├─► Hyperactivity / Impulsivity (Hyperactivity subscale ≥6/9)
  ├─► Working memory deficits (BRIEF-2, digit span)
  └─► Executive function impairment (WCST, Stop-Signal Task)

Drug Treatment:
  Stimulants: MPH/AMP → DAT/NET inhibition + DA efflux → Synaptic DA/NE ↑
  Non-stimulants: ATX → NET selective inhibition → NE PFC ↑
                  GFN → α2A-AR agonist → HCN closure → PFC network ↑
                  VLX → NET inhibition + 5-HT modulation

QSP Model Architecture

Mechanistic Map

  • ~140 nodes · 9 subgraph clusters
  • Clusters: ① Genetic & Neurodevelopmental · ② Catecholamine Biosynthesis · ③ Dopamine System · ④ Norepinephrine System · ⑤ PFC Circuits & Cognition · ⑥ Pharmacokinetics · ⑦ Pharmacodynamics · ⑧ Clinical Endpoints · ⑨ Biomarkers

mrgsolve ODE Model — 25 Compartments

Module Compartments Description
MPH PK GUT1, CENT1, PER1 2-compartment oral PK
AMP PK GUT2, CENT2, PER2 2-compartment oral PK
ATX PK GUT3, CENT3, PER3 CYP2D6-dependent clearance
GFN PK GUT4, CENT4, PER4 Extensive distribution (Vd=16 L/kg)
VLX PK GUT5, CENT5 1-compartment
Neurotransmission DA_syn, NE_syn Synaptic catecholamines
Transporter Occupancy DAT_occ, NET_occ Competitive inhibition
PFC Tone PFC_DA, PFC_NE Inverted-U function
Cognitive WM_idx, ExecFun Working memory & executive function
Clinical ADHD_RS, CGI_S, QoL_idx Rating scale outcomes

Treatment Scenarios (7)

# Scenario Drug Dose Mechanism
1 Untreated — — Natural history
2 MPH IR TID Methylphenidate IR 10 mg TID DAT/NET inhibitor
3 MPH ER QD Methylphenidate ER 36 mg QD DAT/NET (slow-release)
4 AMP XR QD Amphetamine XR 20 mg QD DAT reverse transport
5 ATX QD Atomoxetine 80 mg QD Selective NET inhibitor
6 GFN ER QD Guanfacine ER 4 mg QD α2A-AR agonist
7 VLX ER QD Viloxazine ER 400 mg QD NET + 5-HT modulator

Shiny Dashboard — 7 Tabs

Tab Content
① Patient Profile Demographics, ADHD subtype, comorbidities, genetic risk
② Drug PK Concentration–time profile, DAT/NET occupancy, PK parameters
③ DA/NE Dynamics Synaptic neurotransmitter levels, Inverted-U visualization
④ PFC & Cognition PFC tone indices, working memory gauge, executive function
⑤ Clinical Endpoints ADHD-RS-5 trajectory, CGI-S, response/remission rates
⑥ Scenario Comparison All 7 treatment arms, summary table (DataTable)
⑦ Biomarker Panel Multi-biomarker facet plot, DAT/NET vs. DA/NE correlations

Key Pharmacological Parameters

PK Parameters (Reference Adults, 70 kg)

Drug t½ Tmax F% CL (L/h) Vc (L)
MPH IR 2.5 h 1.5 h 22% 31.5 448
MPH ER ~8 h 6–8 h 22% 31.5 448
AMP XR 9–14 h 7 h 75% 39.2 245
ATX 5 h (EM) 1–2 h 63% 24.5 59.5
GFN ER 17 h 5 h 80% 3.5 196
VLX ER 7 h 5 h 88% 19.6 105

PD Parameters (Transporter Affinity)

Drug Ki(DAT) Ki(NET) Primary Target
MPH 34 nM 340 nM DAT (primary)
AMP ~100 nM 40 nM DAT reverse transport
ATX — 2 nM NET (selective)
GFN — — α2A-AR (1 nM)
VLX — 42 nM NET + 5-HT2B

Clinical Trial Calibration

Clinical Trial Drug Key Result Reference
MTA Study (n=579) MPH ADHD-RS reduction ~10 pts vs. placebo Arch Gen Psychiatry 1999
Biederman 2002 (n=584) AMP XR 20mg ADHD-RS reduction ~12 pts JAACAP 2002
Michelson 2001 (n=297) ATX 80mg ADHD-RS reduction ~8 pts (EM) Pediatrics 2001
Sallee 2009 (n=345) GFN ER 4mg ADHD-RS reduction ~7 pts JAACAP 2009
Nasser 2021 (n=460) VLX ER 400mg ADHD-RS reduction ~8 pts Neuropsychiatr 2021
Volkow 1998 (PET) MPH DAT occupancy 72% at therapeutic dose Am J Psychiatry 1998

Running the Model

Prerequisites

install.packages(c("mrgsolve", "dplyr", "ggplot2", "tidyr",
                   "shiny", "shinydashboard", "plotly", "DT"))

Run mrgsolve simulation

source("adhd_mrgsolve_model.R")

Launch Shiny app

shiny::runApp("adhd_shiny_app.R")

Render mechanistic map

dot -Tsvg adhd_qsp_model.dot -o adhd_qsp_model.svg
dot -Tpng -Gdpi=150 adhd_qsp_model.dot -o adhd_qsp_model.png

File List

File Description
adhd_qsp_model.dot Graphviz mechanistic map source (~140 nodes, 9 clusters)
adhd_qsp_model.svg Vector mechanistic map (high resolution)
adhd_qsp_model.png Raster map (150 dpi)
adhd_mrgsolve_model.R mrgsolve ODE QSP model (25 compartments, 7 scenarios)
adhd_shiny_app.R Shiny interactive dashboard (7 tabs)
adhd_references.md References: 53 PubMed citations (12 sections)

Key Concepts Modeled

  1. Inverted-U function: PFC function is optimal at intermediate DA/NE tone — both too little (ADHD) and too much (excess stimulant) impair cognition
  2. CYP2D6 pharmacogenomics: ATX clearance varies 4× between extensive (EM) and poor metabolizers (PM), affecting efficacy and side-effect profiles
  3. Allometric scaling: PK parameters scaled by body weight (WT^0.75 for clearance), enabling pediatric vs. adult comparisons
  4. DAT/NET occupancy: Competitive inhibition modeled explicitly; therapeutic DAT occupancy for MPH is 50–80% (Volkow 1998)
  5. Comorbidity weighting: Coexisting anxiety, depression, ODD are tracked as modulators of functional outcomes

Model built by Claude Code Routine — 2026-06-26