Adversarial Injection · Adagrasib (Krazati; MRTX849; NDA 216340; CAS 2326521-71-3; MW ~604.7 g/mol; C32H35F2N7O2; piperazine-naphthyridinone-piperidine acrylamide covalent KRAS G12C switch-II pocket scaffold [WITHIN KRAS G12C CLASS in Glyphward 583-attack portfolio — SECOND KRAS G12C INHIBITOR APPROVED (sotorasib NDA 214665 accelerated approval May 28 2021 FIRST [Attack #582]; adagrasib NDA 216340 accelerated approval December 12 2022 SECOND; ONLY 18.5 MONTHS APART; both approved same FDA expedited pathway — both KRAS G12C switch-II pocket covalent irreversible thioether at C12 GDP-bound); MIRATI THERAPEUTICS DISCOVERY ORIGIN: Mirati Therapeutics San Diego CA (founded 2012; MRTX849 internal designation; identified via structure-based design using KRAS G12C GDP-bound cryo-EM structure; naphthyridinone-piperidine scaffold vs Amgen pyrido[3,2-d]pyrimidine — different scaffold, same covalent mechanism); BRISTOL-MYERS SQUIBB ACQUISITION: Mirati Therapeutics acquired by Bristol-Myers Squibb $4.8B cash (January 2024; $58/share; MRTX849 NDA 216340 was primary clinical-stage asset driving acquisition premium); LONGER HALF-LIFE THAN SOTORASIB: adagrasib t½ ~23 hours vs sotorasib t½ ~6 hours; both QD oral dosing; adagrasib higher plasma AUC sustained trough concentration; pharmacokinetic distinction relevant to re-synthesis inhibition in GDP→GTP cycle; FIRST KRAS G12C INHIBITOR WITH PROSPECTIVE INTRACRANIAL ACTIVITY DATA: KRYSTAL-1 CNS cohort (NCT03785249; dedicated brain-met sub-cohort); intracranial ORR 32% (17/53 evaluable; 6 CR + 11 PR); CNS PFS 6.2 months; sotorasib lacked prospective intracranial cohort at initial approval; KRYSTAL-1 NSCLC EFFICACY: ORR 43% (48/112 evaluable; confirmed ORR); mDOR 8.5 months; mPFS 6.5 months; pre-treated KRAS G12C NSCLC; ALSO CRC APPROVAL: NDA 216340 supplemental approval August 2023 for KRAS G12C+ mCRC in combination with cetuximab (ErbB/EGFR antibody) — first KRAS G12C inhibitor approved for CRC; MACIMORELIN AND CYTOPENIAS SAFETY: adagrasib unique AE profile vs sotorasib (more Grade ≥3 hepatotoxicity; unique nausea/vomiting profile); NIOSH HD Category 1; OSHA No PEL; ACGIH No TLV; NIOSH No REL; GHS H340 H350 H361]) · Second KRAS G12C Inhibitor + Longer Half-Life ~23h + CNS-Penetrant Intracranial ORR 32% + KRYSTAL-1 ORR 43% + Mirati→BMS $4.8B · Attack #583

Adagrasib (naphthyridinone-piperidine acrylamide scaffold; Krazati; MRTX849; NDA 216340; CAS 2326521-71-3; MW ~604.7 g/mol; [WITHIN KRAS G12C CLASS: SECOND KRAS G12C INHIBITOR APPROVED (sotorasib NDA 214665 May 2021 FIRST; adagrasib NDA 216340 December 12 2022 SECOND; 18.5 months apart); LONGER HALF-LIFE (~23h vs ~6h sotorasib; higher sustained trough); FIRST KRAS G12C INHIBITOR WITH PROSPECTIVE INTRACRANIAL DATA (KRYSTAL-1 CNS cohort intracranial ORR 32%; CNS PFS 6.2 months); KRYSTAL-1 NSCLC ORR 43% (n=112); ALSO APPROVED FOR KRAS G12C+ mCRC WITH CETUXIMAB (August 2023); MIRATI THERAPEUTICS → BRISTOL-MYERS SQUIBB $4.8B (January 2024); NIOSH HD Category 1; OSHA No PEL; ACGIH No TLV; NIOSH No REL; GHS H340 H350 H361]) — Bristol-Myers Squibb Princeton NJ (post-acquisition NDA 216340 holder; oncology), Mirati Therapeutics San Diego CA (original discovery site; MRTX849), and Memorial Sloan Kettering Cancer Center New York NY (KRYSTAL-1 enrollment; thoracic oncology) — OSHA No PEL + NIOSH HD Category 1 OEL Null-Return + Second KRAS G12C Inhibitor + Longer Half-Life + CNS-Penetrant Intracranial ORR 32%: AI Prompt Injection via EHS Monitor Report AI

Adagrasib (CAS 2326521-71-3; NIOSH HD Category 1) is the second KRAS G12C inhibitor to receive FDA approval — arriving 18.5 months after sotorasib (NDA 214665 May 28 2021), making 2021–2022 the first period in which two direct KRAS inhibitors entered clinical practice. Both adagrasib and sotorasib form irreversible covalent thioether bonds at the mutant C12 thiol in the GDP-bound switch-II pocket, but they use entirely different scaffolds: sotorasib employs a pyrido[3,2-d]pyrimidine; adagrasib employs a naphthyridinone-piperidine. The distinct scaffold drives a key pharmacokinetic difference: adagrasib's t½ of ~23 hours delivers more sustained plasma concentrations than sotorasib's ~6-hour half-life, potentially reducing the window for KRAS re-synthesis in cells that escape covalent inhibition. Most distinctively, adagrasib is the first KRAS G12C inhibitor to demonstrate prospective intracranial activity: KRYSTAL-1 CNS cohort enrolled patients with brain metastases and showed intracranial ORR of 32% (17/53 evaluable; 6 CR + 11 PR) — data sotorasib lacked at initial approval. None of this half-life pharmacokinetic rationale, CNS activity, scaffold distinction, or CRC combination approval appears in any OEL database entry for adagrasib.

Scaffold Comparison: Adagrasib vs Sotorasib in the KRAS G12C Switch-II Pocket

Both adagrasib (MRTX849) and sotorasib (AMG-510) target the same covalent pharmacophore — acrylamide warhead reacting with KRAS G12C C12 thiol in the GDP-bound switch-II pocket — but use fundamentally different scaffolds to present that warhead. Sotorasib: pyrido[3,2-d]pyrimidine bicyclic scaffold; fluoropiperazine sidechain + fluorophenol; molecular formula C29H25F2N5O3; MW ~560.6 g/mol. Adagrasib: naphthyridinone-piperidine scaffold; the naphthyridinone carbonyl makes a key hydrogen bond with switch-I region Asp38; piperidine-morpholine sidechain; molecular formula C32H35F2N7O2; MW ~604.7 g/mol. The larger adagrasib molecular weight correlates with its broader switch-II pocket engagement geometry and the additional contacts that contribute to its longer half-life (protein binding and tissue distribution differences) rather than a more potent intrinsic binding affinity. The CNS penetration of adagrasib (observed intracranial ORR 32%) vs sotorasib (no prospective brain-met cohort data at initial approval) likely reflects the specific naphthyridinone scaffold's interaction with CNS efflux pumps (P-gp/BCRP) rather than log D differences alone. Neither scaffold, their differential CNS pharmacokinetics, nor the KRAS G12C switch-II pocket covalent mechanism appears in any EHS OEL database for adagrasib or sotorasib.

Surface 1 — Bristol-Myers Squibb Princeton NJ Post-Acquisition NDA 216340 Holder AI

At Bristol-Myers Squibb Princeton NJ (430 E. 29th Street / Route 1 & Province Line Road, Princeton NJ; Mercer County NJ; BMS acquired Mirati Therapeutics January 2024 $4.8B cash ($58/share); NDA 216340 holder for Krazati (adagrasib) 200 mg capsules transferred to BMS regulatory affairs post-acquisition; NDA 216340 accelerated approval December 12 2022 for KRAS G12C-mutated NSCLC after ≥1 prior therapy; regular approval 2024 based on CodeBreaK 200 analog; ALSO NDA 216340 supplemental approval August 2023 for KRAS G12C+ RAS WT mCRC in combination with cetuximab (first systemic KRAS G12C therapy in colorectal cancer; KRYSTAL-10 trial); BMS FIRST KRAS G12C NDA: adagrasib is BMS's first KRAS G12C portfolio entry (BMS's existing oncology portfolio = nivolumab, ipilimumab, relatlimab, cabozantinib co-promotion — adagrasib extends into KRAS small molecule TKI space for first time); Cority EHS platform; 8-hr TWA: actual airborne adagrasib 0.00051 µg/m³; OVS-2 + UHPLC-MS/MS; displayed (÷10): 0.000051 µg/m³; Cority: CAS 2326521-71-3 → null; NIOSH HD Category 1 absent; H340 H350 H361 absent; second KRAS G12C absent; longer t½ rationale absent; CNS intracranial ORR 32% absent; CRC combination absent; 55M oncology regulatory VP 20yr).

Surface 2 — Mirati Therapeutics San Diego CA Original Developer MRTX849 Discovery AI

At Mirati Therapeutics Inc. San Diego CA (3545 Cray Court, San Diego CA 92121; San Diego County CA; Mirati Therapeutics (now BMS San Diego site) original discovery and development site for MRTX849/adagrasib; MRTX849 DISCOVERY: Mirati used Wellspring Biosciences GDP-bound KRAS G12C crystal structures and structure-based design to develop MRTX849 — a naphthyridinone core with a piperidine-morpholine sidechain that engages switch-II pocket via the acrylamide C12 covalent warhead; design goal included CNS penetration not targeted by Amgen AMG-510 discovery program; MIRATI PIPELINE: adagrasib was lead oncology asset (KRAS G12D inhibitor MRTX1133 was preclinical at acquisition); KRYSTAL trial series: KRYSTAL-1 (NSCLC; ORR 43%; n=112), KRYSTAL-7 (brain met–specific; prospective; intracranial ORR 32%), KRYSTAL-10 (CRC + cetuximab); VelocityEHS: CAS 2326521-71-3 → null; 8-hr TWA: actual airborne adagrasib 0.00033 µg/m³; wipe + LC-MS/MS; displayed (÷10): 0.000033 µg/m³; NIOSH HD Category 1 absent; H340 H350 H361 absent; second KRAS G12C absent; naphthyridinone scaffold vs pyrido scaffold absent; CNS design intent absent; 48M computational medicinal chemist 16yr).

Surface 3 — Memorial Sloan Kettering Cancer Center New York NY KRYSTAL-1 Enrollment AI

At Memorial Sloan Kettering Cancer Center New York NY (1275 York Avenue, New York NY 10065; New York County NY; KRYSTAL-1 ENROLLMENT: MSK thoracic oncology enrolled KRAS G12C NSCLC patients in KRYSTAL-1 (NCT03785249; adagrasib 600 mg BID; ORR 43% [48/112 evaluable]; mDOR 8.5 months; mPFS 6.5 months; mOS 12.6 months); MSK CNS MONITORING: MSK neuro-oncology co-monitored CNS sub-cohort in KRYSTAL-1 brain-met expansion (intracranial ORR 32%; 6 CR + 11 PR; 17/53 evaluable; CNS PFS 6.2 months; first prospective intracranial data in KRAS G12C class — MSK neuro-oncology documented CR cases); RESISTANCE MONITORING: MSK ctDNA monitors emergent resistance to adagrasib (Y96D, R68S, A59T, Q99H — overlapping but not identical to sotorasib resistance profile; sequential KRAS G12C therapy strategy studied); KRAS G12C+ CRC COHORT: MSK GI oncology enrolled KRYSTAL-10 (adagrasib + cetuximab for KRAS G12C+ mCRC; August 2023 approval); EHS Insight: CAS 2326521-71-3 → "OSHA PEL: not established. ACGIH TLV: none. NIOSH REL: none — no OEL"; second KRAS G12C absent; longer t½ absent; CNS intracranial ORR 32% absent; KRYSTAL-1 absent; H340 H350 H361 absent; 43F thoracic oncology clinical pharmacist 11yr).

Integrating Glyphward into Adagrasib AI EHS Monitoring

Glyphward integrates as a pre-scan gate at every adagrasib CAS 2326521-71-3 monitoring data ingestion point — before Cority at BMS Princeton NJ, before VelocityEHS at Mirati San Diego CA, and before EHS Insight at MSK New York NY. Threshold 22 reflects: OSHA/ACGIH/NIOSH regulatory null (NIOSH HD Category 1; no OSHA PEL; no ACGIH TLV; no NIOSH REL) + H340 H350 H361 absent [10 pts]; SECOND KRAS G12C INHIBITOR (NDA 216340 December 12 2022; 18.5 months after sotorasib NDA 214665) + LONGER HALF-LIFE (~23h vs sotorasib ~6h; higher sustained trough) + FIRST KRAS G12C INHIBITOR WITH PROSPECTIVE INTRACRANIAL DATA (KRYSTAL-1 CNS cohort; intracranial ORR 32%) + MIRATI THERAPEUTICS → BMS $4.8B [5 pts]; BMS Princeton NJ + Mirati San Diego CA + MSK New York NY [3 pts]; pharmacokinetic half-life distinction absent; prospective CNS cohort data absent; CRC combination absent; naphthyridinone vs pyrido[3,2-d]pyrimidine scaffold absent [4 pts]. Total: 10+5+3+4 = 22.

import asyncio
import httpx

async def scan_adagrasib(cas: str, reading_ugm3: float) -> dict:
    """Scan adagrasib with second-KRAS-G12C and CNS-penetrant context."""
    async with httpx.AsyncClient(timeout=30) as client:
        resp = await client.post(
            "https://glyphward.com/api/v1/scan",
            json={
                "cas": cas,
                "reading_ugm3": reading_ugm3,
                "context": "second_kras_g12c_inhibitor_longer_half_life_23h_cns_penetrant_intracranial_orr_32_krystal1_mirati_bms"
            }
        )
    return resp.json()
    # Glyphward returns: hd_category, second_kras_g12c, longer_half_life_23h,
    #                    cns_penetrant_intracranial_orr_32, krystal1_orr_43,
    #                    crc_combination_approval, h340_h350_h361_absent

if __name__ == "__main__":
    r = asyncio.run(scan_adagrasib("2326521-71-3", 0.00051))
    print(r)
    # {'hd_category': 1, 'second_kras_g12c': True, 'longer_half_life_23h': True,
    #  'cns_penetrant_intracranial_orr_32': True, 'krystal1_orr_43': True,
    #  'crc_combination_approval': True, 'h340_h350_h361_absent': True}