Disease category: Chronic Cardiovascular Disease
Abbreviation: AS / CAVD
Directory:aortic-stenosis/
Model date: 2026-06-28
Calcific Aortic Valve Disease (CAVD) leading to Aortic Stenosis (AS) is the most common valvular heart disease in adults over 65, affecting ~2.5% of the population and rising sharply with age. The disease progresses through a stereotyped sequence:
- Valve endothelial injury → lipid infiltration → inflammation initiation
- Valve Interstitial Cell (VIC) osteogenic transformation via BMP/Wnt/RANKL signaling
- Microcalcification → macrocalcification → leaflet stiffening
- AVA reduction → transvalvular gradient ↑ → LV pressure overload
- Concentric LV hypertrophy → diastolic dysfunction → eventual systolic failure
- Symptom onset (angina, syncope, dyspnea) → clinical decompensation → death
Without aortic valve replacement (TAVR/SAVR), median survival after symptom onset is 2–3 years. Current pharmacotherapy has not demonstrated efficacy in halting calcification progression (SEAS/ASTRONOMER trials for statins), but novel targets (RANKL/OPG, Lp(a), MGP) are under active investigation.
| File | Description |
|---|---|
as_qsp_model.dot |
Full Graphviz mechanistic map (~110+ nodes, 12 subgraphs) |
as_qsp_model.svg |
Vector SVG (scalable) |
as_qsp_model.png |
Raster PNG (150 dpi) |
| # | Subgraph | Key Nodes |
|---|---|---|
| 1 | Risk Factors & Initiating Stimuli | Age, LPA, LDL, BAV, NOTCH1, HTN, DM, CKD |
| 2 | Valve Endothelium & Initiating Injury | VEC injury, shear stress, oxLDL, eNOS, ROS |
| 3 | Inflammation & Immune Pathways | Macrophage, IL-1β, TNF-α, IL-6, MMP, complement |
| 4 | VIC Osteogenic Transformation | Runx2, Osterix, ALP, RANKL, OPG, Wnt/LRP5, MGP |
| 5 | Valve Calcification Progression | Ca score, AVA, leaflet mobility, macrocalcification |
| 6 | Hemodynamic Consequences | Mean gradient, Vmax, cardiac output, afterload |
| 7 | LV Remodeling | LVH, fibrosis, LVEF, diastolic dysfunction, BNP |
| 8 | Biomarkers & Clinical Endpoints | NYHA, 6MWD, MACE, survival, BNP, Troponin |
| 9 | Drug PK — Statin & PCSK9i | 2-compartment statin, 1-comp SC evolocumab |
| 10 | Drug PD Effects | LDL↓, Lp(a)↓, RANKL↓, MGP activation, anti-fibrosis |
| 11 | Interventional (TAVR/SAVR) | AVA restoration, reverse remodeling, complications |
| 12 | Systemic Consequences | HF, AF, pulmonary HTN, cardiorenal syndrome |
File: as_mrgsolve_model.R
| Category | Compartments |
|---|---|
| Statin PK | STATIN_GUT, STATIN_CENTRAL, STATIN_PERIPH |
| PCSK9i PK | PCSK9I_DEPOT, PCSK9I_CENTRAL |
| Denosumab PK | DENO_DEPOT, DENO_CENTRAL |
| Vitamin K2 PK | VK2_DEPOT, VK2_CENTRAL |
| ACEi PK | ACEI_DEPOT, ACEI_CENTRAL |
| Valve calcification | CS (calcium score), RANKL, MGP_carbox |
| Lipid biomarkers | LDL_C, LPA |
| Inflammation | IL6 |
| Neurohormonal | AngII |
| LV remodeling | LVMI, COLLAGEN, LVEF |
| Clinical | NTproBNP |
Valve Calcium Score:
dCS/dt = k_calc × calc_driver × (1 - calc_inhibit) × CS × (1 - CS/CS_max)
calc_driver = 0.3×(LDL/LDL₀) + 0.4×(RANKL/RANKL₀) + 0.3×(Lp(a)/LPA₀)
AVA from Calcium Score (sigmoidal):
AVA = AVA_min + (AVA₀ - AVA_min) / (1 + (CS/CS₅₀)^hill)
Mean Transvalvular Gradient (Bernoulli):
Vmax = CO / (AVA × 60 × 0.785)
ΔP_mean = 2.4 × Vmax²
LV Fibrosis:
dCOLLAGEN/dt = k_form × (AngII/AngII₀) × wall_stress × (1 - E_ACEi) - k_deg × COLLAGEN
LVEF Decline:
dLVEF/dt = -k_loss × (Collagen/Collagen_max) × max(afterload - 130, 0)/50 × LVEF
| # | Scenario | Key Drugs |
|---|---|---|
| 1 | No Treatment | — |
| 2 | Statin | Rosuvastatin 20mg QD |
| 3 | Statin + PCSK9i | + Evolocumab 140mg Q2W |
| 4 | Statin + Denosumab* | + Denosumab 60mg Q6M |
| 5 | Statin + Vitamin K2 | + MK-7 180μg QD |
| 6 | Max Medical Therapy | Statin + PCSK9i + VK2 + ACEi |
* Denosumab for CAVD is investigational (hypothesis-generating)
- Statin alone: Minimal effect on calcification progression (consistent with negative SEAS/ASTRONOMER/SALTIRE trials)
- PCSK9i addition: Meaningful LDL-C and modest Lp(a) reduction; clinically meaningful delay in calcification
- Denosumab: Largest hypothetical calcification reduction via RANKL blockade in VIC; awaiting clinical trial validation
- Vitamin K2 (MK-7): Moderate benefit via restoration of carboxylated MGP, natural calcification inhibitor
- Max medical therapy: ~15-20% delay in severe AS onset estimated; AVR still required
File: as_shiny_app.R
| Tab | Content |
|---|---|
| Patient Profile | AS severity gauge (AVA, gradient, LVEF, NYHA), risk factors, prognosis |
| Drug PK | Time-concentration plots for statin, PCSK9i, denosumab; PK summary table |
| PD Key Metrics | LDL-C response, Lp(a), RANKL, IL-6, MGP carboxylation trajectories |
| Clinical Endpoints | AVA, mean gradient, LVEF, LVMI, NT-proBNP over time; time-to-severe prediction |
| Scenario Comparison | 6-scenario side-by-side comparison with summary table at year 10 |
| Biomarker Panel | Calcium score trajectory, collagen fraction, biomarker heatmap, AngII-LV axis |
- Patient sliders: Age, LDL-C, Lp(a), LVEF, baseline AVA, baseline calcium score, simulation duration
- Drug toggles: Statin (dose), PCSK9i (dose, frequency), Denosumab (dose, frequency), Vitamin K2 (dose), ACEi/ARB (dose)
- Auto-running simulation with "Run Simulation" button
| Grade | AVA (cm²) | Mean PG (mmHg) | Vmax (m/s) |
|---|---|---|---|
| Mild | >1.5 | <25 | <3.0 |
| Moderate | 1.0–1.5 | 25–40 | 3.0–4.0 |
| Severe | <1.0 | >40 | >4.0 |
| Very Severe | <0.6 | >60 | >5.0 |
- Asymptomatic severe AS: ~50% develop symptoms within 5 years
- After symptom onset (triad: angina, syncope, HF):
- Mean survival with angina: 5 years
- Mean survival with syncope: 3 years
- Mean survival with HF symptoms: 1-2 years
| Target | Rationale | Drug | Trial Status |
|---|---|---|---|
| HMG-CoA | LDL ↓ | Statin | Tested/Failed (SEAS) |
| PCSK9 | LDL+Lp(a) ↓ | Evolocumab | Phase III pending |
| RANK/RANKL | VIC osteogenesis ↓ | Denosumab | Phase II (SALTIRE-II) |
| MGP | Calcification inhibitor | Vitamin K2 | Observational |
| Lp(a) | oxPL-mediated calcification | Pelacarsen, Olpasiran | Phase III |
| ACE/AT1R | LV fibrosis ↓ | Ramipril | Supportive |
| IL-6 | Inflammation ↓ | Ziltivekimab | Hypothesis |
See as_references.md for 50 curated PubMed-linked references organized by:
- Epidemiology & Natural History
- Pathophysiology
- RANKL/OPG/VIC transformation
- BMP/Notch/Wnt signaling
- Lp(a) and lipid-mediated calcification
- Matrix Gla Protein & Vitamin K
- Statin trials (SEAS, ASTRONOMER, SALTIRE)
- Novel targets (PCSK9i, Denosumab)
- TAVR/SAVR evidence (PARTNER, Evolut)
- LV remodeling & neurohormonal
- Hemodynamics & echocardiography
- CT calcium scoring
- QSP modeling
| File | Description | Size |
|---|---|---|
as_qsp_model.dot |
Graphviz mechanistic map (12 subgraphs, 110+ nodes) | ~35KB |
as_qsp_model.svg |
Rendered vector image | ~150KB |
as_qsp_model.png |
Rendered raster image (150 DPI) | ~2.5MB |
as_mrgsolve_model.R |
ODE model (20 compartments, 6 scenarios) | ~15KB |
as_shiny_app.R |
Interactive Shiny dashboard (6 tabs) | ~20KB |
as_references.md |
50 curated PubMed references | ~8KB |
README.md |
This file | ~8KB |
Model parameters calibrated against:
- SEAS trial data (Rossebø et al., NEJM 2008)
- PARTNER / Evolut trial outcomes
- CT calcium score progression studies (Marechaux et al., Heart 2010)
- Echocardiographic valve area natural history (Pellikka et al., Circulation 2005)
- ESC/AHA AS management guidelines (2017/2021)
This model is intended for research/educational purposes only and does not constitute clinical guidance.