Integrated Quantitative Systems Pharmacology model of SIADH, linking non-osmotic/autonomous arginine-vasopressin (AVP) drive to renal V2-receptor and aquaporin-2 (AQP2) functional expression, free-water retention and euvolemic dilutional hyponatremia, cerebral osmotic adaptation, and correction-rate-dependent osmotic-demyelination-syndrome (ODS) risk — with the modern pharmacology stack (tolvaptan, conivaptan, demeclocycline, oral urea, hypertonic saline, fluid restriction) and the clinical endpoints used to track and safely manage the disease (serum sodium, serum/urine osmolality, urine sodium, neuro-symptom severity, ODS risk).

SIADH is the most common cause of hyponatremia in hospitalized patients and is defined by non-osmotic or autonomous release of arginine vasopressin (AVP) despite low plasma osmolality — a failure of the normal negative-feedback suppression of AVP secretion. Causes include ectopic tumoral production (classically small-cell lung cancer), CNS and pulmonary disease, and numerous drugs (SSRIs, carbamazepine, cyclophosphamide/vincristine, MDMA), as well as non-osmotic stimuli like pain, nausea, and stress. Persistently elevated AVP binds renal V2 receptors on collecting-duct principal cells, driving cAMP/PKA-mediated aquaporin-2 (AQP2) phosphorylation, apical trafficking, and gene upregulation — increasing water permeability and free-water reabsorption despite clinical euvolemia (RAAS suppressed, mild natriuresis). The resulting dilutional hyponatremia causes brain-cell swelling; over 24-48+ hours the brain adaptively extrudes organic osmolytes to blunt further swelling, which paradoxically makes chronic, adapted hyponatremia the setting most vulnerable to osmotic demyelination syndrome (ODS) if serum sodium is corrected too fast — the reason guideline correction-rate limits (≤8 mEq/L/24h, ≤18 mEq/L/48h) exist. Therapy targets specific nodes: fluid restriction (first-line), vaptans (tolvaptan, conivaptan — competitive V2 antagonism), demeclocycline (post-receptor induced nephrogenic diabetes insipidus), oral urea (osmotic diuresis), and hypertonic saline for acute/severe symptomatic disease, always balanced against correction-rate safety.

1. Etiologies & risk factors (SCLC/ectopic tumors, CNS/pulmonary disease, SSRIs, carbamazepine, cyclophosphamide/vincristine, MDMA, exogenous desmopressin, NSAIDs, postoperative stimuli, exercise-associated hyponatremia, low dietary solute)
2. Hypothalamic-neurohypophyseal AVP axis (osmoreceptors, baroreceptor reflex, non-osmotic AVP stimuli, ectopic tumoral production, reset-osmostat variant, Robertson subtype classification)
3. Renal V2-receptor / aquaporin-2 signaling (Gs-adenylate cyclase-cAMP-PKA, AQP2 phosphorylation/trafficking/membrane insertion/gene expression, NSIAD gain-of-function variant)
4. Water balance & osmoregulation (free-water reabsorption, urine/serum osmolality, electrolyte-free water clearance, total body water, dilutional hyponatremia, low-solute limit)
5. Volume status & renal sodium handling (euvolemia, RAAS suppression, ANP/BNP, natriuresis, urine sodium, cerebral-salt-wasting differential)
6. Neurological consequences of hyponatremia (cerebral edema, astrocyte swelling, brain organic-osmolyte efflux, chronic adaptation, seizures, coma, herniation, ODS risk)
7. Clinical manifestations & endpoints (asymptomatic/mild through severe symptoms, acute-vs-chronic classification, falls/fracture risk, composite neuro-symptom score, mortality association)
8. Drug PK/PD — treatment (tolvaptan, conivaptan, demeclocycline, hypertonic saline, fluid restriction, oral urea, loop diuretic + salt tablets, SGLT2 inhibitor, DDAVP overcorrection-prevention clamp)
9. Na-correction-rate safety & ODS prevention (24h/48h correction trackers, guideline limits, overcorrection flag, relowering protocol, high-risk-patient flag)
10. Biomarkers & diagnostic monitoring (copeptin, serum/urine osmolality, urine sodium, FENa/FEUrea, Bartter-Schwartz/Verbalis algorithm, hypouricemia, TSH/cortisol exclusion)

• Drug PK / exposure (7 compartments) — tolvaptan (oral depot + effective exposure), conivaptan (IV effective exposure), demeclocycline (oral depot + effective exposure), oral urea (effective exposure), and a hypertonic-saline 3% infusion driver. As with other topical/local surrogates in this library, each is a day-resolution "sustained exposure" compartment mapped through an Emax function onto its mechanistic target, not a literal multi-compartment plasma PK model.
• Renal V2R/AQP2 and water/sodium balance (8 compartments) — AQP2 functional expression, free-water clearance, total body water, serum sodium, serum osmolality, urine osmolality, urine sodium, brain organic-osmolyte adaptation index.
• Correction-rate safety & outcome (4 compartments) — ~24h- and ~48h-delayed sodium trackers (first-order lag compartments approximating a rolling correction-rate window), composite neuro-symptom severity score, and cumulative osmotic-demyelination-syndrome (ODS) risk score.

1. Patient profile — severity slider, ectopic/malignancy flag, acute-vs-chronic flag, starting serum Na, adjunct therapy checkboxes, DDAVP-clamp flag.
2. Drug PK — fractional drug-effect trajectories for all four agents.
3. Renal V2R/AQP2 PD — AQP2 functional expression, free-water clearance, total body water.
4. Sodium & osmolality — serum Na, serum/urine osmolality, urine Na, and 24h/48h correction-rate vs guideline-limit lines.
5. Neuro symptoms/ODS — brain organic-osmolyte adaptation index, neuro-symptom score, cumulative ODS risk score.
6. Scenario comparison — runs all 7 built-in scenarios with the chosen patient profile.
7. Biomarkers — endpoint summary table (Day-0, Day-2, Day-end).

# 1) Render the mechanistic map
dot -Tsvg siadh_qsp_model.dot -o siadh_qsp_model.svg
dot -Tpng -Gdpi=150 siadh_qsp_model.dot -o siadh_qsp_model.png

# 2) Simulate scenarios in R
install.packages(c("mrgsolve","dplyr","tidyr","ggplot2","shiny",
                   "shinydashboard","DT"))
library(mrgsolve)
mod <- mread_cache("siadh_mrgsolve_model.R")
out <- mrgsim(mod, ev(amt=1.0, cmt="TOLVA_GUT", ii=1, addl=13), end=14)
plot(out, c("Serum_Na","Na_correction_rate_24h","ODS_risk_score","Neuro_symptom_score"))

# 3) Launch the dashboard
shiny::runApp("siadh_shiny_app.R")

• Designed for research, education, and hypothesis generation — not clinical decision support.
• Drug "exposure" compartments are pragmatic day-resolution surrogates for sustained mechanistic exposure, not literal multi-compartment plasma PK.
• The 24h/48h sodium-correction-rate trackers are first-order lag approximations of a rolling window, not an exact retrospective maximum — adequate for illustrating guideline-threshold dynamics, not for real bedside correction-rate calculations.
• The cerebral-adaptation → ODS-risk link is a simplified, qualitative representation of a well-established but incompletely quantified clinical phenomenon; absolute ODS_risk_score values are illustrative, not validated probabilities.
• Cerebral salt wasting (the key hypovolemic differential diagnosis, requiring salt/volume replacement rather than restriction) is represented only as a differential-diagnosis node, not simulated as its own phenotype.

Inherits the repository LICENSE.