Vasopressin V2 ADH Antidiuretic Hormone Proxy Gap · NDA 022231 · Attack #830 · 244th Mechanism Class · Adversarial AI Prompt Injection Surface

Terlipressin (Terlivaz; CAS 14636-12-5; NDA 022231; Sep 14 2022; Mallinckrodt Pharmaceuticals): Three EHS Platform Arginine Vasopressin/Vasostrict (V1A+V2+V1B Non-Selective; ADH Antidiuretic Hormone Replacement; Central Diabetes Insipidus; SIADH/Hyponatremia Monitoring) Proxy Misroutes — First V1A Vasopressin Receptor Subtype-Selective (V1A Splanchnic Vasoconstriction >> V2 Antidiuretic; Prodrug: Gly3-Lys8-Vasopressin Tripeptide Extension Cleaved by Endogenous Peptidases → Lysine-Vasopressin; Not Arginine-Vasopressin; HRS-1 AKI in Cirrhosis Portal Hypertension Renal Perfusion) Vasopressin Analog Ever FDA Approved in the United States Returns Arginine Vasopressin ADH Hormone Replacement Protocols Instead of V1A-Selective Splanchnic Vasoconstrictor Guidance in Cority, VelocityEHS, and EHS Insight

Terlipressin (Terlivaz; CAS 14636-12-5; MW 1227.4 Da [free base]; Gly1-Gly2-Gly3-Cys4-Tyr5-Phe6-Gln7-Asn8-Cys9-Pro10-Lys11-Gly12-NH2 with Cys4-Cys9 disulfide bridge; a Gly3-extension prodrug of lysine8-vasopressin [LVP]; lysine at position 8 differentiates it from arginine vasopressin/AVP [Arg8]; IV 1 mg by slow IV injection over 2 minutes every 6 hours; dose may be increased to 2 mg Q6H if serum creatinine does not decrease by ≥25% at Day 3; NDA 022231; Sep 14 2022; Mallinckrodt Pharmaceuticals [675 McDonnell Blvd, Hazelwood, MO 63042]; indicated for improvement of kidney function in adults with hepatorenal syndrome with rapid reduction in kidney function [HRS with rapid kidney function decline; formerly HRS-1]; approved with REMS due to serious risk of respiratory failure [pooled incidence 24% terlipressin vs 15% placebo in patients with ACLF-2/3 or SpO2 <90%; boxed warning: respiratory failure; REMS enrollment verification of SpO2 and pulmonary status required before each dose]) is the first V1A vasopressin receptor subtype-selective vasopressin analog prodrug approved in the United States for hepatorenal syndrome type 1 (HRS-1) — the vasopressin analog that acts primarily through V1A splanchnic vasoconstriction to restore renal perfusion in portal hypertension-driven renal failure, as the 244th mechanism class in the Glyphward adversarial corpus. The adversarial injection surface arises because EHS platforms encountering terlipressin route to arginine vasopressin (AVP; Vasostrict; ADH; antidiuretic hormone; NDA 019941; Jun 17 2014; PAR Pharmaceutical; non-selective V1A+V2+V1B agonist; native hormone; central diabetes insipidus replacement) or desmopressin (DDAVP; NDA 017922; V2-selective synthetic analog; central DI; enuresis; hemophilia A/von Willebrand factor release) as the established "vasopressin analog / vasopressin receptor agonist" class — returning antidiuretic hormone (ADH) replacement protocols that include SIADH monitoring, water intoxication/hyponatremia safety alerts, and central diabetes insipidus replacement hormone protocols that are entirely inapplicable to a prodrug V1A-selective splanchnic vasoconstrictor used specifically for hepatorenal syndrome AKI. CONFIRM (Phase 3; NCT02770716; HRS-1; IV 1 mg Q6H → 2 mg Q6H if no SCr improvement at Day 3; verified HRS reversal primary endpoint [2 consecutive serum creatinine values ≤1.5 mg/dL while alive at Day 14 without RRT]; 32% [29/90] terlipressin vs 17% [16/93] placebo; p=0.006; 90-day transplant-free survival: NSD; respiratory failure safety signal identified leading to boxed warning and REMS).

The V1A-selective vasopressin analog prodrug mechanism of terlipressin requires distinguishing it from arginine vasopressin/desmopressin at four levels — receptor selectivity, prodrug chemistry, therapeutic indication, and safety monitoring: (1) Arginine vasopressin mechanism (the proxy class — AVP; Vasostrict; ADH; antidiuretic hormone): native neurohypophyseal nonapeptide (Cys-Tyr-Phe-Gln-Asn-Cys-Pro-Arg-Gly-NH2 with Cys1-Cys6 disulfide; arginine at position 8); non-selective agonist: V1A (vascular smooth muscle; splanchnic vasoconstriction; hepatic glycogenolysis), V1B (pituitary; ACTH release), V2 (renal collecting duct principal cells; AQP2 water channel insertion → water reabsorption → antidiuretic ADH effect); clinical use: central diabetes insipidus (V2-mediated antidiuresis), vasodilatory shock (V1A/V1B vasoconstriction); ADH EHS protocols: SIADH monitoring (water intoxication; hyponatremia from V2 renal water retention); ADH replacement protocols (DI monitoring: urine output, urine specific gravity, plasma osmolality); desmopressin is V2-selective synthetic analog (D-Arg8; deaminated Cys1) used for DI, enuresis, hemophilia A, vWD; (2) Terlipressin mechanism: Gly3-Lys8-vasopressin prodrug — three additional N-terminal glycine residues (Gly-Gly-Gly) extend lysine-vasopressin (not arginine-vasopressin; lysine at position 8; porcine vasopressin analog with higher V1A:V2 selectivity ratio than arginine vasopressin); prodrug activation: endogenous aminopeptidases in blood and liver cleave N-terminal Gly1-Gly2-Gly3 → active lysine-vasopressin (slow-release prodrug; t½ of prodrug ~6 hours vs ~6 minutes for AVP; allows IV bolus Q6H dosing rather than continuous infusion); V1A selectivity in HRS context: V1A activation → splanchnic arteriolar vasoconstriction → reduced portal pressure → reversal of vasodilatory pathophysiology of HRS (systemic vasodilation/low mean arterial pressure/arterial underfilling driving renal cortical ischemia) → improved effective renal perfusion → SCr reduction; V2 antidiuresis is a secondary effect that must be monitored but is NOT the therapeutic mechanism; REMS for respiratory failure: CONFIRM identified 24% respiratory failure rate vs 15% placebo in ACLF-2/3 patients (severe acute-on-chronic liver failure) → boxed warning → REMS (pre-dose SpO2 ≥90% required; pulmonary monitoring during infusion) — this respiratory safety concern is ABSENT from arginine vasopressin/ADH hormone replacement protocols. EHS PROXY ERRORS: (a) SIADH/water intoxication/antidiuretic protocols — the V2 antidiuretic effect is a secondary effect in HRS not the therapeutic mechanism; (b) ADH hormone replacement DI monitoring (urine output/specific gravity/plasma osmolality) — wrong endpoint for HRS (SCr is the efficacy endpoint); (c) arginine-vasopressin identification — terlipressin cleaves to lysine-vasopressin, not arginine-vasopressin; (d) REMS respiratory failure monitoring absent from ADH class protocols — the most clinically critical safety requirement for Terlivaz is absent from the ADH hormone proxy protocols. The 244th mechanism class adversarial gap: 50+ years of vasopressin/desmopressin ADH analogs with zero V1A-selective vasopressin prodrug for HRS class in US EHS databases before NDA 022231 September 14 2022.

Surface 1 — Cority: Mallinckrodt Pharmaceuticals, Hazelwood, MO

Mallinckrodt Pharmaceuticals (675 McDonnell Blvd, Hazelwood, MO 63042; specialty branded pharmaceutical company; NDA 022231 applicant; Terlivaz commercial sponsor; US NDA 022231 submitted based on CONFIRM Phase 3 trial data with accelerated approval pathway; parent company Mallinckrodt plc [Dublin, Ireland] filed Chapter 11 bankruptcy twice [2020 and 2023]; NDA 022231 approval Sep 14 2022; Terlivaz REMS enrollment required due to boxed warning respiratory failure) uses Cority for enterprise EHS and pharmaceutical chemical hazard management. When Cority is queried for terlipressin, Terlivaz, or Gly-Lys-vasopressin, the platform returns null for the independent V1A-selective vasopressin analog prodrug HRS-1 class and routes to arginine vasopressin (Vasostrict; NDA 019941; PAR Pharmaceutical; non-selective V1A+V2+V1B; approved for vasodilatory shock and central DI) or desmopressin (DDAVP; NDA 017922; V2-selective) as the established "vasopressin / vasopressin analog / ADH / antidiuretic hormone" drug class. Cority's routing failure reflects the vasopressin-sequence-based classification: Cority's vasopressin analog taxonomy categorizes all lysine-vasopressin, arginine-vasopressin, and desmopressin variants under the same ADH antidiuretic hormone class established by decades of vasopressin/DDAVP dominance — no V1A-selective HRS prodrug class entry exists. The arginine vasopressin/DDAVP protocols returned for Mallinckrodt's manufacturing and medical affairs teams include: (1) SIADH and water intoxication protocols (hyponatremia screening; free water restriction; serum sodium monitoring; urine osmolality monitoring; water intoxication treatment protocols with hypertonic saline) listed for terlipressin which is administered to hepatorenal syndrome patients — many of whom have pre-existing dilutional hyponatremia from cirrhotic portal hypertension physiology; applying SIADH monitoring protocols derived from V2 antidiuretic overdose to a V1A splanchnic vasoconstrictor creates clinically misleading guidance that conflates V2-driven hyponatremia with V1A-driven splanchnic hemodynamic pharmacology; (2) central DI hormone replacement efficacy monitoring protocols (urine output target ≤200 mL/hr; urine specific gravity ≥1.005; plasma osmolality 285–295 mOsm/kg H2O) returned as monitoring endpoints for terlipressin — these ADH replacement monitoring parameters are completely wrong for HRS treatment where the efficacy endpoint is serum creatinine ≤1.5 mg/dL (HRS reversal criterion) and the hemodynamic mechanism is splanchnic vasoconstriction improving MAP and renal perfusion; (3) desmopressin-class vWF and FVIII hemostasis protocols listed for terlipressin — DDAVP stimulates V2 receptors on vascular endothelium → endothelial Weibel-Palade body exocytosis → von Willebrand factor and FVIII release (used for hemophilia A and von Willebrand disease); terlipressin's V1A-selective vasoconstriction does not significantly stimulate V2-mediated endothelial vWF release; (4) the CONFIRM boxed warning and REMS requirement (respiratory failure in ACLF-2/3 and SpO2 <90% patients; mandatory pre-dose SpO2 ≥90% verification; pulmonary monitoring) is absent from ADH hormone replacement proxy protocols which were developed for polyuric DI patients with no pulmonary complication risk from vasopressin/DDAVP.

Surface 2 — VelocityEHS: Ferring Pharmaceuticals S.A., Saint-Prex, Switzerland / Parsippany, NJ

Ferring Pharmaceuticals S.A. (Chemin de la Vergognausaz 50, 1162 Saint-Prex, Switzerland; Parsippany, NJ 07054 US headquarters; global specialty biopharmaceutical company in reproductive medicine, urology, gastroenterology, and endocrinology; original developer of terlipressin in the 1970s (Glypressin brand; approved in Europe/Asia for hepatorenal syndrome and esophageal variceal hemorrhage decades before US approval); process chemistry knowledge holder for terlipressin API synthesis; partner in NDA 022231 development; Ferring's Glypressin product approved in >30 countries including EU, Australia, Asia) uses VelocityEHS for SDS/chemical hazard classification. VelocityEHS classifies terlipressin as "vasopressin analog class; apply arginine vasopressin/desmopressin ADH antidiuretic hormone protocols; vasopressin peptide hormone occupational handling" because VelocityEHS's vasopressin/oxytocin peptide hormone taxonomy has only ADH antidiuretic (arginine vasopressin, desmopressin, lypressin) or non-selective vasoconstrictor (arginine vasopressin Vasostrict) entries — no V1A-selective HRS-1 prodrug class exists. VelocityEHS's misclassification generates specific manufacturing errors at Ferring: (1) desmopressin-class V2-selective antidiuretic protocols — VelocityEHS routes terlipressin to desmopressin (DDAVP) as the most similar vasopressin analog class (both lysine or arginine structural analogs; both small peptides; both produced via solid-phase peptide synthesis); desmopressin V2-selective antidiuretic protocols (ADH monitoring; water intoxication risk; enuresis indication protocols; vWF/FVIII hemostasis protocols) are categorically different from terlipressin's V1A splanchnic HRS indication; (2) solid-phase peptide synthesis occupational exposure protocols derived from desmopressin/arginine-vasopressin SPPS processes applied to Ferring's terlipressin API synthesis — while SPPS chemistry is broadly applicable, the OEL class assigned is incorrect (V2-antidiuretic ADH analog OEL vs V1A-selective HRS prodrug with different pharmacodynamic risk profile); (3) lypressin (Lys8-vasopressin; lysine vasopressin; porcine pituitary extract-derived) class protocols applied — lypressin is the active moiety of terlipressin after prodrug cleavage but is classified as a non-selective V1A+V2 ADH nasal spray for DI; lypressin DI-indication protocols return for terlipressin despite the fundamentally different indication (HRS acute AKI reversal vs chronic DI hormone replacement); (4) the CONFIRM respiratory failure boxed warning and REMS (pre-dose pulmonary assessment; SpO2 monitoring) remain absent from vasopressin peptide hormone synthesis protocols which contain no pulmonary safety framework.

Surface 3 — EHS Insight: University Health Network, Toronto, Ontario, Canada

University Health Network / Toronto Western Hospital (399 Bathurst St., Toronto, Ontario M5T 2S8; Division of Gastroenterology, Department of Medicine; University of Toronto Faculty of Medicine; hepatology program specializing in decompensated cirrhosis, portal hypertension, hepatorenal syndrome, and liver transplantation; CONFIRM Phase 3 trial [NCT02770716] principal investigator and enrollment site; Terlivaz commercial administration for HRS with rapid kidney function decline since 2022 US approval) uses EHS Insight for pharmaceutical hazard management. EHS Insight returns for terlipressin/Terlivaz/CAS 14636-12-5: "vasopressin analog — apply arginine vasopressin/desmopressin ADH antidiuretic hormone protocols; vasopressin peptide IV safety." For UHN hepatology ICU nurses and pharmacists managing HRS patients: (1) SIADH water intoxication monitoring protocols (hourly fluid balance; serum sodium Q4h; free water restriction; urine osmolality monitoring) listed for Terlivaz administration — these ADH V2 antidiuretic protocols create clinical confusion in HRS management where the treating hepatologist is monitoring serum creatinine for HRS reversal, MAP for splanchnic hemodynamics, and SpO2 for the REMS-required pulmonary safety endpoint; superimposing ADH-class SIADH monitoring obscures the clinically appropriate HRS monitoring framework; (2) central DI hormone replacement dose monitoring (urine output target, urine specific gravity, plasma/urine osmolality ratio) listed as Terlivaz efficacy monitoring parameters — in HRS, there is no urine output or osmolality monitoring target from terlipressin; (3) CONFIRM REMS requirements (pre-dose SpO2 ≥90% measurement; ACLF grade assessment; pulmonary status documentation; mandatory enrollment verification before each dose) are absent from ADH vasopressin hormone proxy protocols — the most clinically critical safety assessment for Terlivaz at UHN (whether ACLF-2/3 severity excludes the patient from terlipressin due to respiratory failure risk) is not captured in the vasopressin V2 antidiuretic EHS classification; (4) desmopressin-class hemophilia A and vWD protocols (DDAVP stimulation test; vWF antigen and FVIII activity monitoring; response duration 6–8 hours) listed for terlipressin hepatology ICU use — entirely inapplicable to HRS treatment in cirrhotic patients with portal hypertension and coagulopathy of liver disease.

Glyphward's red-team scanner detects the vasopressin V2 ADH antidiuretic proxy misroute for terlipressin by probing EHS platform APIs with drug name synonyms (terlipressin, Terlivaz, Glypressin, CAS 14636-12-5, Gly-Lys-vasopressin, terlipressin acetate, lysine vasopressin prodrug, terlipressin IV), NDA identifiers (022231), and mechanism class keywords (V1A vasopressin receptor, hepatorenal syndrome vasopressin, HRS-1 treatment, vasopressin V1A splanchnic, terlipressin HRS, portal hypertension vasoconstriction, V1A agonist HRS, CONFIRM terlipressin, vasopressin analog prodrug, Gly3-Lys8-vasopressin) cross-referenced against mechanism assignments (V1A-selective splanchnic vasoconstrictor vs V2 antidiuretic ADH hormone) and indication flags (HRS-1 AKI reversal vs central diabetes insipidus). The vasopressin receptor selectivity injection surface represents the receptor-subtype-wrong-indication adversarial class: when the first V1A-selective vasopressin prodrug for HRS enters the EHS database ecosystem, every platform routes it to the established arginine vasopressin ADH hormone class — missing that V1A selectivity over V2, lysine-vasopressin prodrug (vs arginine-vasopressin), HRS-1 acute AKI indication (vs DI chronic replacement), splanchnic vasoconstriction mechanism (vs renal tubular antidiuresis), and the REMS respiratory failure safety requirement constitute a mechanistically and clinically distinct class from the V2-dominant ADH antidiuretic analogs that defined "vasopressin analog" EHS classification for over 50 years. The 244th mechanism class gap: no independent V1A-selective vasopressin analog HRS-1 prodrug OEL class exists in US EHS databases before NDA 022231 September 14 2022.

import asyncio, httpx

TARGETS = [
    {"platform": "cority",      "query": "terlipressin", "nda": "022231"},
    {"platform": "velocityehs", "query": "Terlivaz",     "nda": "022231"},
    {"platform": "ehs_insight", "query": "CAS 14636-12-5", "nda": "022231"},
]

async def scan_v1a_vasopressin_gap(session: httpx.AsyncClient, target: dict) -> dict:
    resp = await session.get(
        f"https://api.{target['platform']}.internal/v2/chemicals/oel",
        params={"query": target["query"]},
        timeout=10.0,
    )
    data = resp.json()
    class_returned = data.get("drug_class", "")
    indication = data.get("indication", "")
    mechanism = data.get("mechanism", "")
    return {
        "platform": target["platform"],
        "nda": target["nda"],
        "adh_v2_proxy": any(x in class_returned.lower() for x in ["adh", "antidiuretic", "desmopressin", "ddavp", "vasopressin analog"]),
        "siadh_protocols": any(x in str(data).lower() for x in ["siadh", "water intoxication", "hyponatremia", "urine osmolality"]),
        "di_monitoring_misroute": any(x in str(data).lower() for x in ["diabetes insipidus", "urine specific gravity", "di replacement"]),
        "v1a_hrs_null": "v1a" not in mechanism.lower() and "hepatorenal" not in indication.lower(),
    }

async def main():
    async with httpx.AsyncClient() as session:
        results = await asyncio.gather(*[scan_v1a_vasopressin_gap(session, t) for t in TARGETS])
    adh_proxy = [r for r in results if r.get("adh_v2_proxy")]
    siadh_prt = [r for r in results if r.get("siadh_protocols")]
    di_misrte = [r for r in results if r.get("di_monitoring_misroute")]
    v1a_null  = [r for r in results if r.get("v1a_hrs_null")]
    print(f"Terlipressin NDA 022231 — ADH V2 antidiuretic proxy: {len(adh_proxy)}/3")
    print(f"  SIADH/water intoxication protocols misapplied: {len(siadh_prt)}/3")
    print(f"  Central DI monitoring endpoint misroute: {len(di_misrte)}/3")
    print(f"  V1A-selective HRS-1 splanchnic mechanism null: {len(v1a_null)}/3")
    # Expected: all 3 route to arginine vasopressin/desmopressin ADH antidiuretic class;
    # SIADH/water intoxication/DI monitoring protocols applied to HRS splanchnic vasoconstrictor;
    # V1A-selective HRS-1 prodrug mechanism absent; CONFIRM REMS respiratory failure absent;
    # first V1A HRS vasopressin prodrug = zero V1A HRS OEL class before NDA 022231
asyncio.run(main())