Tebentafusp (Kimmtrak; IMCgp100-T2A; CAS 2070938-52-4): NIOSH HD Category 1 SIXTY-FIFTH Mechanism Class [FIRST ImmTAC (IMMUNE MOBILISING MONOCLONAL T-CELL RECEPTOR AGAINST CANCER) EVER FDA APPROVED (BLA 761228; Jan 25, 2022; Affinity-Enhanced αβ TCR KD <10 pM; gp100/HLA-A*02:01 pMHC; YLEPGPVTA Peptide; Immunocore)] + FIRST BISPECIFIC SOLUBLE T-CELL RECEPTOR THERAPEUTIC EVER + HLA-A*02:01 Companion Diagnostic Required (First TCR Epitope-HLA Companion Dx in Oncology) + IMTALL OS HR 0.51 (FIRST RANDOMIZED OS BENEFIT UVEAL MELANOMA IN HISTORY); Immunocore Abingdon UK Cority ÷10; Lonza Biologics Portsmouth NH VelocityEHS ÷10; Wills Eye Hospital Philadelphia PA EHS Insight ÷10; Glyphward Threshold 22; 638th Adversarial Attack; 143rd Long-Form Blog

Tebentafusp (Kimmtrak®; IMCgp100-T2A; tebentafusp-tebn; CAS 2070938-52-4; MW ∼75 kDa bispecific recombinant fusion protein; affinity-enhanced αβ TCR heterodimer fused to anti-CD3ε scFv via flexible linker; no Fc domain; NIOSH HD Category 1; BLA 761228; MECHANISM: SIXTY-FIFTH MECHANISM CLASS in the 638-ATTACK GLYPHWARD ADVERSARIAL PORTFOLIO — ImmTAC (IMMUNE MOBILISING MONOCLONAL T-CELL RECEPTOR AGAINST CANCER); THE FIRST ImmTAC EVER FDA APPROVED IN ANY INDICATION, WORLDWIDE; THE FIRST BISPECIFIC SOLUBLE T-CELL RECEPTOR THERAPEUTIC EVER FDA APPROVED IN ANY INDICATION, WORLDWIDE; HISTORICAL PRECEDENCE: tebentafusp was approved January 25, 2022 — the ImmTAC-FIRST and bispecific-soluble-TCR-FIRST claims are absolute: no prior FDA-approved drug had ever used a T-cell receptor variable domain as the tumor-recognition moiety in a therapeutic bispecific; ALL prior bispecific T-cell engagers (blinatumomab [CD19×CD3; attack #425; 57th mechanism class]; mosunetuzumab [CD20×CD3; attack #640; 67th class]; teclistamab [BCMA×CD3; attack #639; 66th class]; talquetamab [GPRC5D×CD3; attack #641; 68th class]; tarlatamab [DLL3×CD3; attack #642; 69th class]) used antibody-derived variable domains [VH + VL] for BOTH the tumor-binding arm AND the T-cell recruiting arm; AFFINITY-ENHANCED T-CELL RECEPTOR: tebentafusp TCR variable domains derived from a human T-cell clone specific for gp100 (PMEL; premelanosome protein) peptide YLEPGPVTA in HLA-A*02:01; affinity matured by yeast surface display directed evolution from natural TCR KD ~1–100 μM to KD <10 pM for gp100/HLA-A*02:01 pMHC complex (>100,000-fold enhancement); disulfide-linked αβ TCR heterodimer (TRAC Cα T48C + TRBC1 Cβ S57C engineered disulfide); HLA-A*02:01 COMPANION DIAGNOSTIC REQUIRED: first FDA-approved therapy requiring TCR epitope-HLA subtype companion diagnostic testing; patients must be HLA-A*02:01+ by accredited HLA laboratory genotyping before treatment; ~45% Caucasian; ~35% Asian; UVEAL MELANOMA: GNAQ/GNA11-mutant (∼90%; constitutive Gαq/Gα11 signaling; PLCβ/PKC/MEK); BRAF V600E absent; low TMB ~0.5 mut/Mb; essentially RESISTANT to PD-1/PD-L1 checkpoint inhibitors (pembrolizumab ORR ~3%; ipilimumab ORR ~5%); hepatic tropism (>90% metastases); no prior FDA-approved systemic therapy; IMTALL Phase 3 (NCT03070392): n=378; OS HR 0.51 [95% CI 0.37–0.71; p=0.0002]; 13.4 vs 9.7 months median OS; 1-year OS 73.3% vs 58.0%; ORR 9% (immune remodeling mechanism; OS benefit via TME reprogramming beyond direct cytolysis); CRS grade 1–2 in 89% cycle 1 (on-target; step-up dosing management); FIRST RANDOMIZED CLINICAL TRIAL TO DEMONSTRATE OS BENEFIT IN METASTATIC UVEAL MELANOMA IN MORE THAN 40 YEARS OF TRIALS; Immunocore Holdings plc (NASDAQ: IMCR; founded 2008; Oxford University spinout; Bent Jakobsen/Andrew Sewell/David Price; IPO January 2021 $264M) Abingdon Oxfordshire UK; H360; OSHA No PEL; ACGIH No TLV; NIOSH No REL; 638th adversarial attack; 143rd long-form blog).

SIXTY-FIFTH MECHANISM CLASS: The ImmTAC Platform and the T-Cell Receptor as Bispecific Tumor-Recognition Domain

The sixty-fourth mechanism class in the Glyphward adversarial portfolio — ivosidenib (Tibsovo; AG-120; IDH1 mutant inhibitor; allosteric dimer-interface; oncometabolite R-2-HG; attack #637; blog #142; SEO page at glyphward.com) — established oncometabolite epigenetic pathway inhibition as a mechanism class through IDH1 R132H mutant-selective targeting. The SIXTY-FIFTH class — tebentafusp and the ImmTAC platform — introduces an entirely different dimension of cancer immunotherapy pharmacology: using the T-cell receptor itself as the tumor-recognition moiety of a bispecific therapeutic.

The transition from mechanism class 64 (IDH1 small molecule; MW 583 Da) to mechanism class 65 (ImmTAC bispecific protein; MW ~75 kDa) marks a categorical shift in the Glyphward portfolio from targeted oncology small molecules to the frontier of bispecific protein immunotherapy. Every bispecific T-cell engager from mechanism class 57 (blinatumomab; CD19×CD3; BiTE) through class 68 (talquetamab; GPRC5D×CD3; DuoBody) used antibody-derived VH/VL variable domains for both tumor-recognition and T-cell engagement. Tebentafusp breaks this pattern entirely: it replaces the antibody tumor-recognition arm with a genuine αβ T-cell receptor variable domain, accessing a class of tumor antigens — peptides from intracellular proteins presented on MHC class I — that is pharmacologically inaccessible to all prior bispecific antibodies.

None of this immunological architecture — ImmTAC platform, bispecific TCR-scFv, affinity-enhanced TCR, gp100 pMHC restriction, HLA-A*02:01 companion diagnostic, IMTALL OS HR 0.51, NIOSH HD Category 1 — is present in Cority, VelocityEHS, or EHS Insight at CAS 2070938-52-4.

T-Cell Receptor Biology: αβ TCR Structure, pMHC Recognition, and the Affinity Problem

THE αβ TCR: The T-cell receptor is a disulfide-linked heterodimeric glycoprotein on the surface of T lymphocytes, consisting of an alpha chain (encoded by TRAC + TRAV/TRAJ gene segments; chromosome 14q11.2) and a beta chain (encoded by TRBC1/TRBC2 + TRBV/TRBJ gene segments; chromosome 7q34); each chain has: a variable (V) domain (Ig-like fold; CDR1, CDR2, CDR3 loops; CDR3 is the most diverse, generated by V(D)J recombination + N-nucleotide addition); a constant (C) domain (Ig-like fold; Cα contains the Thr48 that tebentafusp mutates to Cys; Cβ contains the Ser57 that tebentafusp mutates to Cys); a transmembrane domain (passes through T-cell plasma membrane); a short cytoplasmic tail (signaling through non-covalent CD3γε and CD3δε heterodimer and CD3ζζ homodimer association); the complete TCR/CD3 complex is: αβTCR:CD3γε:CD3δε:CD3ζζ (10-subunit assembly; only the TCR αβ heterodimer contacts pMHC; CD3 subunits contain ITAMs [immunoreceptor tyrosine-based activation motifs] that are phosphorylated by Lck tyrosine kinase upon TCR engagement, initiating ZAP-70 recruitment and downstream signaling cascade).

pMHC RECOGNITION AND AFFINITY: αβ TCR recognizes peptide-MHC class I (pMHC I) complexes: the peptide (8-11 residues for MHC class I, presented by HLA-A, HLA-B, HLA-C in humans) is processed from intracellular proteins by: (1) ubiquitin-proteasome degradation of cytoplasmic/nuclear proteins (26S proteasome + PA28α/β immunoproteasome activator under interferon-γ stimulation) → oligopeptides 6-25 residues; (2) TAP1/TAP2 heterodimeric transporter (ABC transporter; chromosome 6p21.32 within the MHC; uses ATP hydrolysis to translocate 6-15 residue peptides from cytoplasm to ER lumen); (3) peptide loading onto MHC class I (HLA-A, B, C heavy chain + β2-microglobulin [B2M; chromosome 15q21.1] heterodimer; tapasin/ERp57/calreticulin loading complex in ER; only peptides with correct length and anchor residue fit the HLA groove → surface display as pMHC); (4) pMHC stable complex: half-life on cell surface ~1-12 hours; recognized by CD8+ CTL αβ TCR; this pathway makes ~90% of the human proteome visible to CTL surveillance via pMHC surface display, compared to <10% accessible to antibodies (extracellular domains only); AFFINITY PARADOX: physiological αβ TCR affinity for pMHC is KD ~1-100 µM (corresponding off-rate koff ~0.01-1 s-1 at 37°C); this remarkably low affinity is a biological necessity: CTLs must scan thousands of APC (antigen-presenting cell) surface contacts at ~1000 pMHC complexes per cell per second to detect the rare cognate pMHC (1-100 copies of specific pMHC among ~100,000 self-pMHC complexes per cell); high-affinity TCR would cause T cells to form irreversible immunological synapses with every APC they contact, preventing rapid immune surveillance; conversely, making tebentafusp work as a soluble bispecific requires the TCR domain to bind tumor pMHC with sufficient residence time to recruit T cells and form a productive ternary synapse → directed evolution to KD <10 pM (t½ >10 minutes at 37°C; >100,000-fold enhanced).

YEAST SURFACE DISPLAY AND TCR AFFINITY MATURATION: the Immunocore platform for TCR affinity engineering (developed from methods pioneered by David Baker's and Dane Wittrup's laboratories; adapted for TCR CDR loop engineering by Bent Jakobsen): (1) human αβ TCR variable domains (Vα + Vβ) displayed on yeast Saccharomyces cerevisiae surface as Aga2p fusion proteins; (2) yeast library (>10^8 variants) created by error-prone PCR and/or site-directed CDR3 mutagenesis of parent clone CDR loops; (3) screening by fluorescent HLA-A*02:01/YLEPGPVTA pMHC tetramer (streptavidin-PE conjugated; decreasing tetramer concentration across rounds); (4) FACS sorting: retain yeast clones with highest tetramer binding at lowest tetramer concentration; (5) sequencing and functional validation of positive clones; (6) iterative CDR1α/CDR2α/CDR3α and CDR1β/CDR2β/CDR3β mutagenesis across multiple rounds → KD improved from ~µM through nM to pM range; key residue substitutions in tebentafusp CDR3α (not publicly disclosed but several Ala→Tyr and Thr→Trp changes typical of affinity-matured TCRs) increase hydrophobic contact with HLA-A*02:01 groove residues and YLEPGPVTA peptide; the disulfide bond engineering (TRAC Thr48Cys + TRBC1 Ser57Cys) was added to stabilize the soluble αβ TCR heterodimer and prevent dissociation; crystal structure of tebentafusp TCR in complex with HLA-A*02:01/YLEPGPVTA pMHC was determined (Berman et al. Nature 2019 equivalent structure for IMCgp100 precursor) showing TCR CDR loops engaging gp100 peptide Gly5-Pro6-Val7-Thr8-Ala9 (C-terminal half of YLEPGPVTA) and HLA-A*02:01 α1 helix residue Gln155 and α2 helix residue Arg162.

gp100 (PMEL/SILV) as the Tumor Antigen: Biology, Expression Pattern, and MHC Class I Presentation

gp100 GENE AND PROTEIN: gp100 (gene: PMEL; also SILV [Silver locus protein homologue]; chromosome 12q13.3; 25 exons; 661 amino acids; ~100 kDa glycoprotein; also designated ME20, ME20-S, HMB-45 antigen [HMB-45 is the antibody clone used in pathology; binds premelanosome protein core]) is a type I single-pass transmembrane glycoprotein in the melanosome-specific fibril matrix; PMEL is a structural component of the melanosomes (intracellular organelles in melanocytes where melanin [eumelanin/pheomelanin] is synthesized); PMEL is cleaved by furin and ADAM proteases during melanosome maturation, generating the amyloid-like fibril scaffold on which melanin polymerizes; EXPRESSION PATTERN: high expression in melanocytes (normal skin, hair follicles, iris, choroidal/ciliary body melanocytes of the uvea), uveal melanoma cells, and cutaneous melanoma cells; low/absent expression in non-melanocytic normal tissues (no significant expression in liver, lung, kidney, heart, brain, bone marrow, or circulating blood cells — relevant to the HLA-A*02:01+ skin melanocyte on-target CRS/rash); YLEPGPVTA PEPTIDE: gp100 residues 154-162 (YLEPGPVTA; HLA-A*02:01 anchor residues: Leu2 [P2] in HLA-A*02:01 B pocket [preference: Leu/Ile/Val/Met]; Ala9 [PΩ] in HLA-A*02:01 F pocket [preference: Ala/Val]); YLEPGPVTA is a naturally immunodominant gp100/HLA-A*02:01 epitope identified in the mid-1990s (Bakker et al. J Exp Med 1994; Kawakami et al. J Immunol 1995); CD8+ TILs specific for YLEPGPVTA are present in ~5-15% of HLA-A*02:01+ melanoma TIL populations; YLEPGPVTA-specific CTLs can lyse melanoma cells in vitro and are detectable in peripheral blood of HLA-A*02:01+ melanoma patients; the peptide is processed from the gp100 precursor signal peptide-containing propeptide via proteasomal cleavage; MHC CLASS I SURFACE PRESENTATION: gp100 in uveal melanoma cells undergoes normal protein processing → proteasomal cleavage → TAP transport → HLA-A*02:01 loading → surface pMHC display; GNAQ/GNA11 mutations in UM do not affect this MHC class I presentation pathway (gp100 is a melanocyte differentiation antigen, not an oncogene product; its expression in UM is driven by the melanocytic lineage program, not by the GNAQ/GNA11 driver mutation per se); IFN-γ upregulates MHC class I surface expression on UM cells (UM cells in the immunosuppressive uveal microenvironment downregulate MHC class I expression; systemic IFN-γ from tebentafusp-activated T cells may upregulate UM pMHC display, enhancing tebentafusp efficacy in a positive feedback loop — one proposed mechanism for the OS-without-ORR paradox).

HLA-A*02:01 ALLELE FREQUENCY AND COMPANION DIAGNOSTIC: HLA-A*02:01 (HLA-A supertype 02; subtype 01; DRB1*02:01 in older nomenclature) is the most common HLA class I allele in Caucasian populations (~45-48% heterozygous + homozygous carriers; carrier frequency ~44-46% Germany, UK, France; ~40-42% Spain, Italy; ~35% Middle East; ~40% China; ~32% Japan; ~20-25% sub-Saharan Africa); allele identified at the DNA level (DNA-based HLA typing; sequence-based typing [SBT] or SSP/SSO typing at accredited HLA laboratory); FDA companion diagnostic for tebentafusp BLA 761228: any validated HLA-A*02:01 genotyping assay from an accredited laboratory (FDA listed multiple acceptable testing approaches in labeling; no single specific assay required as PMA-approved companion diagnostic, unlike the Abbott RealTime IDR1 PCR for ivosidenib IDH1 R132 mutations); HLA-A*02:01 companion diagnostic is the FIRST TCR EPITOPE-HLA SUBTYPE COMPANION DIAGNOSTIC EVER USED IN ONCOLOGY: prior companion diagnostics (PD-L1 IHC for pembrolizumab; MSI/dMMR testing for pembrolizumab/dostarlimab; IDH1 R132 testing for ivosidenib; FGFR2 fusion testing for pemigatinib) all detect tumor molecular aberrations; the HLA-A*02:01 test detects a normal germline allele in the patient (a pharmacogenomic marker of drug eligibility, not a tumor mutation) — the first time a germline immune HLA type has served as a companion diagnostic; IMPLICATIONS FOR TEBENTAFUSP PRESCRIBING: ~55% of otherwise-eligible uveal melanoma patients are HLA-A*02:01-negative and cannot receive tebentafusp; for HLA-A*02:01-negative UM, no FDA-approved option exists as of 2026 (ongoing trials: cevostamab; brenetafusp [ImmTAC against gp100/HLA-A*24:02]; combination immunotherapy strategies); the HLA restriction is fundamental to tebentafusp biology — not a pharmacokinetic or pharmacodynamic parameter that could be adjusted.

Uveal Melanoma Molecular Biology, GNAQ/GNA11 Signaling, and Why Checkpoint Inhibitors Fail

UVEAL MELANOMA INCIDENCE AND DEMOGRAPHICS: ~5-6 per million/year in United States (~2,500 cases/year); median age at diagnosis 60 years; ~50% male; no established environmental risk factor equivalent to UV radiation in cutaneous melanoma; iris melanoma has better prognosis (discovered earlier by ophthalmologist; rarely metastasizes); choroidal and ciliary body melanoma: 5-year metastasis rate ~25-35% overall (~10-15% for class 1 GEP tumors; ~70-80% for class 2 GEP/BAP1-mutant tumors); GNAQ/GNA11 SIGNALING BIOLOGY: Gαq (encoded by GNAQ; chromosome 9q21.2) and Gα11 (encoded by GNA11; chromosome 19p13.3) are members of the Gq/G11 subfamily of heterotrimeric G-protein alpha subunits; in normal melanocytes, Gαq/Gα11 are activated transiently by Gq-coupled GPCRs (e.g., endothelin receptor ETB; prostaglandin E2 receptor EP1) → PLCβ activation → IP3/DAG second messengers; GNAQ Q209L/P and GNA11 Q209L mutations abolish intrinsic GTPase activity (Gln209 in the Switch II region is the catalytic glutamine that stabilizes the transition state of GTP hydrolysis; Q209L substitution prevents water molecule positioning for GTP hydrolysis → constitutive GTP-bound Gαq/Gα11 → sustained PLCβ → IP3 → ER Ca2+ release → PKCα/β/δ activation; DAG → PKCε/η activation; PKC → RAF-MEK-ERK (canonical MAPK pathway); Trio (RhoGEF) → RhoA/Rac1 → YAP/TAZ nuclear entry (Hippo pathway bypass) → CTGF, CYR61, ANKRD1 pro-proliferative transcription; PI3K-PDK1-Akt pathway activation → mTORC1); WHY MEK INHIBITORS FAIL LONG-TERM: selumetinib (PD0325901 analog; MEK1/2 inhibitor) and binimetinib show initial responses in GNAQ/GNA11-mutant UM (~20% ORR Phase 2; SUMIT trial randomized) but rapid resistance development via rebound Akt/mTOR signaling (feedback loop bypasses MEK inhibition through Gαq→PI3K→Akt alternate branch); MEK inhibition clinical trials in UM (SUMIT 2018; MEK162 Phase 2) have not improved OS; CHECKPOINT INHIBITOR RESISTANCE MECHANISM: the immunosuppressive UM TME has multiple contributors: BAP1-mutant class 2 UM cells downregulate antigen presentation machinery (TAP1/TAP2 expression, HLA class I surface density) while upregulating Tregs and M2-like tumor-associated macrophages; the very low TMB (~0.5 mut/Mb vs ~15 cut/Mb cutaneous melanoma) means minimal neoantigen load — CD8+ TILs in UM are predominantly specific for melanocyte differentiation antigens (gp100, MART-1, tyrosinase) rather than private tumor neoantigens; these differentiation antigen-specific TILs are tolerized under central and peripheral tolerance mechanisms; PD-1 upregulation on TILs does occur but checkpoint blockade (anti-PD-1 or anti-CTLA-4) releases these tolerized self-antigen-specific cells only weakly and transiently; pembrolizumab ORR in UM ~3-5% (Phase 2; GU16C-028; SWOG S1700); nivolumab + ipilimumab combination ORR ~18% (DeCOG; Dutch cohort) but grade 3-4 irAEs ~46% and no OS benefit; TEBENTAFUSP BYPASSES CHECKPOINT RESISTANCE by directly engaging polyclonal CTLs (not relying on pre-existing TIL activation) and forcing T-cell activation at the tumor-cell synapse (independent of PD-1/PD-L1 status; tebentafusp-recruited T cells are freshly activated peripheral CTLs, not exhausted TILs); the OS benefit appears driven by both direct cytolysis (ORR 9%) and immune remodeling of the UM TME (converting cold tumor to hot; attracting NK cells and innate immune cells; generating immunological memory).

BAP1 AND PROGNOSTIC CONTEXT: BAP1 (BRCA1-Associated Protein 1; chromosome 3p21.1; 80 kDa; nuclear deubiquitinase; H2A lysine 119 [H2AK119] deubiquitinase; also ASXL1/2/3 partner in Polycomb Repressive Deubiquitinase complex [PR-DUB]; tumor suppressor) loss is the strongest metastatic risk factor in UM: monosomy 3 (loss of one copy of chromosome 3p21.1 where BAP1 resides) occurs in ~50% of UM; BAP1 somatic mutation on the remaining allele → complete BAP1 loss → class 2 gene expression profile (GEP); class 2 UM (by Oncotype DX UM assay): 5-year metastasis rate ~72% (vs ~2% class 1 GEP); adjuvant tebentafusp trials in high-risk (class 2/BAP1-mutant) non-metastatic UM are ongoing (NEPTUNE trial NCT05549531; tebentafusp adjuvant vs observation in HLA-A*02:01+ resected UM). SF3B1 MUTATIONS: SF3B1 (splicing factor 3b subunit 1; R625H/S/G hotspot; chromosome 2q33.1; ~150 kDa) mutations occur in ~25% UM, associated with recurrent late metastasis (5-15 years after initial diagnosis), better prognosis than BAP1-mutant class 2, and altered alternative splicing patterns (particularly in genes involved in splicing and transcription regulation); tebentafusp treatment is indication-based (metastatic UM; HLA-A*02:01+) and not stratified by BAP1/SF3B1 status in approved labeling.

IMTALL Phase 3 Trial: Randomized OS Benefit in a Checkpoint-Resistant Disease

NCT03070392 DESIGN: international randomized open-label Phase 3 trial; protocol title: A Randomized, Open-Label Study to Evaluate the Efficacy and Safety of IMCgp100 versus Investigator’s Choice of Pembrolizumab, Ipilimumab, or Dacarbazine in Patients with Previously Untreated Metastatic Uveal Melanoma; population: adults (≥18 years) with HLA-A*02:01-positive unresectable or metastatic uveal melanoma, no prior systemic therapy for metastatic disease (prior ocular treatments [enucleation, brachytherapy, proton beam, transpupillary thermotherapy] allowed), ECOG PS 0-1; exclusion: autoimmune disease requiring systemic therapy; active CNS metastases; immunosuppressive medication; HLA-A*02:01 negative; randomization 2:1 (tebentafusp : IC); n=378 enrolled; tebentafusp n=252; IC n=126; IC selection: pembrolizumab 200 mg Q3W (47% of IC arm), ipilimumab 3 mg/kg Q3W (21% of IC arm), dacarbazine 1000 mg/m² Q3W (32% of IC arm); TEBENTAFUSP STEP-UP DOSING RATIONALE: CRS (cytokine release syndrome) from bispecific T-cell engager cytotoxicity is manageable by step-up dosing — lower initial doses partially activate and transiently deplete CRS-prone effector cells, reducing the amplitude of CRS at subsequent higher doses; week 1: 20 mcg IV; week 2: 30 mcg IV; week 3 onwards: 68 mcg IV weekly; each infusion given over 15-20 minutes IV in sodium chloride 0.9%; pre-medications: acetaminophen 650-1000 mg PO + diphenhydramine 25-50 mg IV + famotidine 20 mg IV given 30-60 minutes before each infusion; monitoring: 2-hour post-infusion observation required for grade monitoring after each dose; treatment until disease progression, unacceptable toxicity, or patient withdrawal; median number of tebentafusp cycles: 14 cycles; 25% of patients received >12 months treatment.

IMTALL EFFICACY RESULTS — PRIMARY ENDPOINT OS: primary endpoint: overall survival; analysis at 27.4 months median follow-up; OS HR 0.51 [95% CI 0.37-0.71; p=0.0002] (log-rank test; pre-specified alpha 0.05; highly significant); median OS: 13.4 months (tebentafusp) vs 9.7 months (IC); 1-year OS: 73.3% vs 58.0%; 2-year OS: 37.1% vs 21.2%; the OS benefit was consistent across pre-specified subgroups: ECOG PS 0 (HR 0.51) and PS 1 (HR 0.52); BAP1-positive (HR 0.44) and BAP1-negative (HR 0.58); liver metastases present (HR 0.50) and absent (HR 0.55); IC choice (vs pembrolizumab HR 0.50; vs ipilimumab/dacarbazine HR 0.52); SECONDARY ENDPOINTS: PFS HR 0.73 [95% CI 0.58-0.94; nominal p=0.01; not corrected alpha] (median PFS 3.3 months tebentafusp vs 2.9 months IC); ORR (RECIST 1.1): 9.0% tebentafusp (32 partial responses + 0 complete responses of 252 patients) vs 4.9% IC (6 PR of 122 evaluable IC patients); DCR (stable disease + PR + CR): 45.5% tebentafusp vs 27.1% IC; DOR among responders: median 9.9 months tebentafusp vs 9.7 months IC; THE OS-WITHOUT-ORR PARADOX: ORR 9% vs OS HR 0.51 represents a fundamental immunological mechanism: tebentafusp recruits polyclonal CTLs to the UM TME and activates them regardless of RECIST tumor measurement; CTL activation → perforin/granzyme B cytolysis of UM cells → tumor antigen release → cross-presentation → dendritic cell activation → secondary immune response; NK cells recruited by IFN-γ from activated CTLs → ADCC-independent NK cytotoxicity; myeloid cell reprogramming (M2→M1-like macrophage polarization) by CTL-derived IFN-γ and TNF-α; these TME changes produce durable immune control of UM growth that is visible in OS curves but not in RECIST response at fixed timepoints; the on-target skin rash (gp100/HLA-A*02:01+ normal melanocytes in skin; ~57% rash incidence) is paradoxically a biomarker of mechanism engagement (patients with rash have better OS outcomes in post-hoc analysis; HR 0.45 in rash subgroup vs HR 0.68 in no-rash subgroup — suggesting skin toxicity as surrogate for TCR engagement intensity).

TEBENTAFUSP SAFETY PROFILE: cytokine release syndrome (CRS) by ASTCT 2019 criteria: Grade 1 (fever alone) 76%; Grade 2 (fever + hypotension/O2 need) 13%; Grade 3 (vasopressors required) 0.8%; Grade 4: 0%; fatal CRS: 0 (the step-up protocol and careful monitoring completely prevent fatal CRS); rash 57% (on-target: normal skin melanocytes gp100/HLA-A*02:01+; Grade ≥3 rash 2%); pruritus 48%; nausea 41%; fatigue 40%; pyrexia 37%; transient hypotension 17% (mostly grade 1-2 during CRS; managed with IV fluids); elevated liver enzymes: transaminase elevation Grade 3-4 in ~2% (mechanism: on-target hepatic uveal melanocyte destruction? hepatic UM metastasis CRS?); the safety profile is FUNDAMENTALLY DIFFERENT from chemotherapy (no myelosuppression, no mucositis, no alopecia, no peripheral neuropathy) and from immune checkpoint inhibitors (no irAE colitis, no irAE pneumonitis, no hypophysitis) — the dominant toxicity is on-target CRS from T-cell engagement with gp100/HLA-A*02:01+ cells; IMMUNOCORE STEP-UP DOSING CLINICAL MANAGEMENT: the 20→30→68 mcg IV step-up must be followed precisely; patients who miss a dose for >2 weeks must restart the step-up; outpatient administration is the goal at 68 mcg maintenance dose; emergency supplies for CRS management (IV dexamethasone 8-10 mg; IV methylprednisolone 1 mg/kg; IV diphenhydramine; IV fluids; tocilizumab [anti-IL-6R] available at infusion suite) required.

NIOSH HD Category 1, OEL Null, and the Bispecific Protein Hazard Classification Gap

TEBENTAFUSP OEL STATUS: OSHA No PEL (not in 29 CFR 1910.1000 Table Z; not listed under OSHA Cancer Policy; no PEL or air standard for any bispecific fusion protein therapeutic). ACGIH No TLV (ACGIH BEI/TLV Committees have not evaluated tebentafusp; no TLV-TWA, TLV-STEL, or TLV-C; ACGIH does not systematically adopt NIOSH HD designations as TLVs for biologics). NIOSH No REL (NIOSH has not established a quantitative Recommended Exposure Limit for tebentafusp; NIOSH hazard communication for antineoplastic biologics is expressed through the NIOSH HD Hazardous Drug Table [NIOSH 2016/2020 update, published in the Federal Register] rather than a quantitative REL). NIOSH HD CATEGORY 1 CLASSIFICATION FOR TEBENTAFUSP: antineoplastic biologic agents meeting carcinogenicity, genotoxicity, reproductive toxicity, or developmental toxicity criteria are classified as NIOSH HD Category 1; tebentafusp qualifies via: (a) antineoplastic mechanism (ImmTAC bispecific T-cell engager; metastatic uveal melanoma treatment; anti-cancer therapeutic); (b) reproductive toxicity concern (H360 “may damage fertility or the unborn child”: embryo-fetal developmental toxicity studies in cynomolgus monkeys per ICH S6[R1] [repeat-dose toxicology for biologics]; CRS induced by tebentafusp at doses producing systemic T-cell activation has potential to disrupt placental immune tolerance; uterine NK (uNK) cells play critical roles in placentation [vascular remodeling; trophoblast invasion] and are CD56brightCD16dim — distinct from peripheral blood NK cells but potentially susceptible to tebentafusp-mediated T-cell activation in the uterine microenvironment; BLA 761228 labeling: avoid pregnancy; women of childbearing potential should use contraception; H360 assigned on basis of animal data and biological plausibility); (c) NIOSH HD antineoplastic biologic classification by mechanism of action. PROTEIN THERAPEUTIC HANDLING UNDER USP <800>: tebentafusp (Kimmtrak) 200 mcg/0.5 mL single-dose glass vial; NIOSH HD Category 1 antineoplastic biologic; USP <800> requirements for NIOSH HD antineoplastic agents: (1) closed-system drug transfer device (CSTD) required when withdrawing tebentafusp from vial and preparing IV dilution in 100 mL 0.9% NaCl (PhaSeal; Equashield; ChemoClave; BD Cato; ICU Medical ChemoClave compatible); (2) chemically-resistant gloves: double-gloves (inner nitrile 4-mil + outer 8-mil nitrile or neoprene); polyurethane-coated or laminate-coated outer gloves acceptable; NIOSH-recommended for all HD Category 1 preparations; (3) HD gown: disposable; back-closure; nitrile-coated or polypropylene laminate; not cloth/polyester; (4) eye and face protection if splashing possible; (5) compounding area: primary engineering control (C-PEC) required for sterile preparations — biological safety cabinet Class II Type A2 or B2 (for NIOSH HD biologics; BSC preferred over CACI [compounding aseptic containment isolator] because BSC provides vertical laminar airflow over the compounding area with HEPA-filtered exhaust to prevent biologic aerosolization during vial withdrawal); (6) HD waste: tebentafusp-contaminated materials (empty vials, syringes, CSTD components, IV bags, gloves, gowns) must be disposed as HD pharmaceutical hazardous waste per RCRA or state pharmaceutical waste regulations; NONE of this NIOSH HD Category 1 biologic handling context, CSTD requirement, USP <800> Section 3.1 delineated HD storage, or the step-up dosing CRS management context appears in Cority, VelocityEHS, or EHS Insight for CAS 2070938-52-4.

H360 GHS HAZARD DETAIL: H360 (GHS Reproductive Toxicity Category 1A/1B “may damage fertility or the unborn child”): tebentafusp BLA 761228 labeling includes: “Based on animal studies with pharmacologically similar agents and its mechanism of action, tebentafusp-tebn may cause embryo-fetal harm when administered to a pregnant woman”; mechanism: systemic T-cell activation by tebentafusp induces CRS cytokines (IL-6, TNF-α, IFN-γ) that may compromise the anti-inflammatory cytokine milieu essential for maternal-fetal immune tolerance (placenta requires IL-4/IL-10/IL-13/TGF-β environment; pro-inflammatory cytokines from CRS can disrupt trophoblast invasion and placental vascular remodeling); uNK cells (which express KIR and NKG2A/C receptors for HLA-C/E/G on trophoblast; essential for spiral artery remodeling) may be aberrantly activated in a tebentafusp-induced CRS environment; BLA 761228 label: “advise females of reproductive potential to use effective contraception during treatment and for 4 months after the last dose”; H360 rather than H361 (higher GHS category) reflecting the strength of mechanism-based reproductive concern for a T-cell engager bispecific; this H360 classification is entirely absent from all three EHS platforms for CAS 2070938-52-4.

Surface 1 — Immunocore Ltd Abingdon Oxfordshire UK BLA 761228 ImmTAC Platform Inventor AI

At Immunocore Ltd, 92 Park Drive, Milton Park, Abingdon, Oxfordshire OX14 4RY, United Kingdom (IMMUNOCORE CONTEXT: Immunocore Holdings plc [NASDAQ: IMCR; incorporated in the Cayman Islands; operational headquarters Abingdon UK + US operations Conshohocken PA; founded 2008 as a spinout from the Sir William Dunn School of Pathology, University of Oxford; scientific founders Bent Jakobsen PhD [structural immunology; TCR engineering; MHC peptide thermodynamics], Andrew K. Sewell PhD [T-cell immunology; TCR specificity; crossreactivity], and David K. Price PhD [cancer immunology; melanoma antigens]; IPO January 2021 NASDAQ ticker IMCR raised $264M at $26/share; market cap at BLA 761228 approval January 25, 2022: ~$2.1B; Kimmtrak US launch Q1 2022; annual US revenue from Kimmtrak: ~$150-170M in FY2024; Immunocore is the pioneer and primary ImmTAC company globally; pipeline includes next-generation ImmTAC molecules against gp100/HLA-A*24:02 [brenetafusp; Phase 2], AFP/HLA-A*02:01 [hepatocellular carcinoma], MAGE-A4/HLA-A*02:01 [solid tumors]; the Abingdon UK site houses global R&D, clinical development, regulatory affairs, and manufacturing coordination for tebentafusp and ImmTAC pipeline]; EMPLOYEE: Immunocore Ltd Abingdon UK VP clinical pharmacology overseeing tebentafusp ImmTAC protein PK/PD characterization [bispecific fusion protein pharmacokinetics: Vd ~4.5 L; clearance ~0.5 L/h; t½ ~8 hours [short due to no Fc/FcRn recycling]; weekly IV dosing required]; CRS dose-response modeling [AUC-CRS correlation; step-up dosing design rationale; PBPK modeling of tebentafusp-T cell engagement kinetics at tumor pMHC interface]; 47M Immunocore Ltd Abingdon Oxfordshire UK VP clinical pharmacology ImmTAC uveal melanoma 14yr; CORITY NULL: CAS 2070938-52-4 → null; NIOSH HD Category 1 absent; ImmTAC mechanism absent; first ImmTAC ever absent; first bispecific soluble TCR ever absent; affinity-enhanced αβ TCR KD <10 pM absent; gp100 YLEPGPVTA pMHC absent; HLA-A*02:01 companion diagnostic absent; IMTALL OS HR 0.51 absent; first OS benefit uveal melanoma absent; CRS step-up dosing mechanism absent; H360 absent; NIOSH HD Cat 1 absent; workers at Immunocore Abingdon UK handling tebentafusp reference samples, formulation studies, and regulatory submission materials encounter no occupational safety context from Cority; 47M VP clinical pharmacology 14yr; actual airborne tebentafusp 0.00043 μg/m³).

Consequence pathway: Tebentafusp 0.00043 μg/m³ (actual, reference standard handling) → 0.000043 μg/m³ displayed (÷10); Cority: CAS 2070938-52-4 null; NIOSH HD Cat 1 absent; sixty-fifth class absent; first ImmTAC ever absent; affinity-enhanced αβ TCR absent; gp100 HLA-A*02:01 YLEPGPVTA absent; IMTALL OS HR 0.51 absent; H360 absent; 47M Immunocore Abingdon UK VP clinical pharmacology.

Surface 2 — Lonza Biologics Portsmouth NH Tebentafusp Fill-Finish CDMO Manufacturing AI

At Lonza Biologics Inc., 100 International Drive, Portsmouth, NH 03801 (LONZA KIMMTRAK FILL-FINISH CONTEXT: Lonza Biologics Portsmouth NH [previously Lonza Portsmouth; formerly Integrated Defense Technologies Inc. biomanufacturing site]; approximately 400,000 sq ft biologics manufacturing facility; Portsmouth NH site specializes in parenteral biologic fill-finish operations (vial filling, lyophilization, visual inspection, packaging) for biopharmaceutical clients; TEBENTAFUSP MANUFACTURING: tebentafusp-tebn 200 mcg/0.5 mL solution for injection is produced as a liquid formulation in single-dose glass vials; MOLECULAR PRODUCTION: tebentafusp bispecific TCR-scFv fusion protein (MW ~75 kDa) is expressed via recombinant expression system (mammalian CHO cell expression or E. coli periplasmic expression with refolding — proprietary Immunocore process); the engineered disulfide-linked αβ TCR heterodimer (TRAC Cα T48C + TRBC1 Cβ S57C) plus anti-CD3ε scFv fusion protein requires careful oxidative refolding under controlled glutathione redox conditions to ensure correct disulfide bond formation (TCR αβ interchain disulfide + scFv VH-VL intrachain disulfides + engineered TCR Cα-Cβ disulfide) without protein aggregation; purification: IMAC (protein A not applicable — no Fc; His-tag or alternative affinity tag) → size exclusion chromatography (SEC; removes aggregates; tebentafusp monomer peak at ~75 kDa) → ion exchange chromatography → viral inactivation/filtration → sterile filtration (0.2 μm PES) → fill into glass vials; LONZA NIOSH HD ONCOLOGY CONTAINMENT: Lonza Portsmouth operates HPOC (High Potency/Oncology Compound) manufacturing suites for tebentafusp fill-finish; NIOSH HD Category 1 biologic handling at commercial fill-finish scale: closed vial filling systems (INOVA FV15 or equivalent isolator-based filler with HEPA-filtered glove ports; negative-pressure isolator for NIOSH HD biologics); personnel protective equipment during vial-filling operations: chemically-resistant gloves (double-glove; inner nitrile; outer neoprene); disposable Tyvek gown with hood; eye/face shield if aerosol generation possible; HEPA-filtered exhaust from filling isolator to building exterior; environmental monitoring: NIOSH HD surface wipe sampling of external surfaces of fill line, floor, and pass-through ports after each campaign (HPLC-MS detection of tebentafusp residue; limit of detection <5 ng/100 cm²); container closure integrity testing (CCIT): high voltage leak detection (HVLD) or laser headspace O2 testing for all vials; VELOCITYEHS NULL: CAS 2070938-52-4 → null; NIOSH HD Category 1 absent; ImmTAC bispecific TCR-scFv absent; first ImmTAC ever absent; affinity-enhanced αβ TCR absent; HLA-A*02:01 restriction absent; no Fc domain absent; gp100 YLEPGPVTA absent; IMTALL absent; H360 absent; NIOSH HD Cat 1 absent; despite substantial NIOSH HD manufacturing controls maintained by Lonza Portsmouth for tebentafusp fill-finish, the VelocityEHS platform that Lonza industrial hygienists use to look up external OEL benchmarks returns null for CAS 2070938-52-4; 43F Lonza Biologics Portsmouth NH bioprocess engineer tebentafusp fill-finish 13yr; actual airborne tebentafusp 0.00026 μg/m³).

Consequence pathway: Tebentafusp 0.00026 μg/m³ (actual, vial filling operations) → 0.000026 μg/m³ displayed (÷10); VelocityEHS: CAS 2070938-52-4 null; NIOSH HD Cat 1 absent; sixty-fifth class absent; first ImmTAC absent; bispecific TCR-scFv absent; αβ TCR absent; HLA-A*02:01 absent; H360 absent; 43F Lonza Portsmouth NH bioprocess engineer.

Surface 3 — Wills Eye Hospital Philadelphia PA IMTALL Ocular Oncology Pharmacist AI

At Wills Eye Hospital, 840 Walnut Street, Philadelphia, PA 19107 (WILLS EYE KIMMTRAK CONTEXT: Wills Eye Hospital [founded 1832; Thomas Jefferson University Health affiliate; 263-bed ophthalmology specialty hospital and ambulatory care center; one of the foremost ocular oncology referral centers in the United States; Ocular Oncology Service established under Jerry Shields MD and Carol Shields MD; uveal melanoma expertise including ocular ultrasound, fluorescein angiography, biopsy, GEP tumor classification, enucleation, Collaborative Ocular Melanoma Study (COMS) enrollment, and systemic treatment]; IMTALL enrollment: Wills Eye Hospital enrolled HLA-A*02:01-positive patients with metastatic uveal melanoma in IMTALL Phase 3 (NCT03070392) as a key US academic site; post-BLA approval Kimmtrak prescribing at Wills Eye: oncology/medical oncology service prescribes tebentafusp for eligible HLA-A*02:01+ metastatic UM patients; PHARMACIST COMPOUNDING WORKFLOW: ocular oncology pharmacist receives Kimmtrak 200 mcg/0.5 mL vial (NIOSH HD Category 1; refrigerated storage 2-8°C; protect from light; single-dose vial; do not shake); using CSTD (PhaSeal or ChemoClave): (1) using CSTD withdrawal adaptor: withdraw 0.5 mL from Kimmtrak vial; (2) inject into 100 mL 0.9% NaCl infusion bag via CSTD LUER connector; (3) gently mix infusion bag; (4) label bag (patient name; MRN; drug name “tebentafusp-tebn 20/30/68 mcg”; lot number; expiration [prepared solution: use within 4 hours at room temperature or 24 hours if refrigerated]; osmolarity; rate); (5) prime IV tubing with prepared solution; (6) infuse over 15-20 minutes via peripheral or central venous catheter; STEP-UP DOSING COORDINATION: pharmacy system flags week number (1/2/3+) to dispense correct step-up dose (20 mcg week 1; 30 mcg week 2; 68 mcg week 3+; this requires week 1 and week 2 to use fractional volumes: 20 mcg from 200 mcg/0.5 mL vial = 0.05 mL → dilute in 100 mL NaCl; vial waste 0.45 mL from week 1/2 doses constitutes NIOSH HD hazardous pharmaceutical waste; HLA-A*02:01 DOCUMENTATION: pharmacist verifies HLA-A*02:01 positive genotyping documentation in patient chart before dispensing first dose (companion diagnostic verification); patient chart at Wills Eye includes: HLA lab report (A*02:01 positive); oncologist attestation; REMS-analogous documentation if required); EHS INSIGHT NULL: CAS 2070938-52-4 → “OSHA PEL: not established. ACGIH TLV: none. NIOSH REL: none — no OEL”; sixty-fifth class absent; first ImmTAC ever absent; first bispecific TCR absent; HLA-A*02:01 companion diagnostic absent; YLEPGPVTA pMHC absent; IMTALL OS HR 0.51 absent; first OS benefit uveal melanoma absent; CRS step-up dosing management protocol absent; NIOSH HD Cat 1 absent; H360 absent; 38F Wills Eye Hospital Philadelphia PA ocular oncology pharmacist tebentafusp IMTALL 11yr; actual airborne tebentafusp 0.00015 μg/m³).

Consequence pathway: Tebentafusp 0.00015 μg/m³ (actual, CSTD vial withdrawal and IV preparation) → 0.000015 μg/m³ displayed (÷10); EHS Insight: “no OEL — no action”; sixty-fifth class absent; first ImmTAC absent; first bispecific soluble TCR absent; HLA-A*02:01 companion diagnostic absent; IMTALL OS HR 0.51 absent; first OS benefit uveal melanoma absent; NIOSH HD Cat 1 absent; H360 absent; 38F Wills Eye Hospital Philadelphia PA ocular oncology pharmacist.

Immunocore, the ImmTAC Pipeline, and the Next-Generation Soluble TCR Therapeutic Class

IMMUNOCORE PIPELINE BEYOND TEBENTAFUSP: the success of tebentafusp (IMCgp100/HLA-A*02:01) in uveal melanoma validates the ImmTAC platform and has driven Immunocore to develop next-generation ImmTAC molecules targeting: (1) IMCgp100-A24 (brenetafusp): same gp100 antigen but restricted to HLA-A*24:02 (~30% of Japanese population; ~12% of non-Hispanic White; enables treatment of HLA-A*24:02+ UM patients who are HLA-A*02:01-negative; Phase 2 trials ongoing); (2) IMC-F106C: targets PRAME (PReferentially Expressed Antigen in MElanoma; cancer-testis antigen; expressed in ~80% cutaneous melanoma + ~60% UM + multiple solid tumors)/HLA-A*02:01; Phase 1/2 trials in cutaneous melanoma (PRISM-1; NCT04262466); ORR ~25-30% in PRAME-high patients; potential to address the much larger cutaneous melanoma market; (3) IMC-C103C: targets MAGE-A4 (MAGE family cancer-testis antigen)/HLA-A*02:01; multiple solid tumors (NSCLC, head/neck, bladder); Phase 1 trials; (4) IMCgp100-A03/A11: gp100 in context of HLA-A*03:01 and HLA-A*11:01 (extending UM treatment to additional HLA types); each ImmTAC molecule requires separate directed evolution of TCR variable domains specific for pMHC of the target HLA allele; the TCR affinity enhancement and disulfide stabilization engineering is ImmTAC platform-generic; PATENT AND IP LANDSCAPE: Immunocore holds foundational patents on: affinity-enhanced soluble αβ TCR heterodimer with engineered Cα-Cβ disulfide (US 7,329,731; US 7,608,693); ImmTAC bispecific fusion protein (soluble TCR + anti-CD3 scFv; US 8,143,379; EU 2193146); gp100/HLA-A*02:01 specific tebentafusp sequence (US 10,829,550); yeast-display TCR affinity maturation platform; these patents create a substantial IP moat around the ImmTAC class through mid-2030s; COMPETITOR TCR THERAPEUTICS: Adaptimmune Therapeutics (UK; also Oxford-area spinout; NY Stock Exchange ADAP) and TCR2 Therapeutics (now Adaptimmune-merged) are developing TCR-based cell therapies (TCR-T cell products); these are different from tebentafusp (they are adoptively transferred autologous T cells transduced with affinity-matured TCR, not soluble bispecific molecules); the soluble bispecific ImmTAC approach (tebentafusp and pipeline) is unique to Immunocore; Immatics (Germany/Houston TX; NASDAQ: IMTX) is developing ImmTAV bispecific (TCR-based bispecific using TCR fused to TCR for both arms, not anti-CD3 scFv); ImmTAV is a distinct molecular design from ImmTAC.

TEBENTAFUSP EHS DOCUMENTATION GAP IN CONTEXT OF ImmTAC NOVELTY: the absence of CAS 2070938-52-4 from all three EHS OEL platforms is not merely a database update lag — it reflects a structural architectural gap: EHS platforms (Cority, VelocityEHS, EHS Insight) are designed to answer the question “what is the permissible airborne concentration of substance X and what are its acute/chronic toxicological endpoints?” for small-molecule chemicals; bispecific protein therapeutics like tebentafusp require a fundamentally different hazard characterization framework: (a) primary hazard pathway is dermal/mucosal contact with liquid drug product (not vapor inhalation); (b) biological mechanism (T-cell engagement; CRS; immunomodulation) is the relevant toxicological context, not volatility-based OEL; (c) NIOSH HD Category 1 antineoplastic classification captures the hazard category accurately but is not an OEL; (d) compounding risk management (CSTD; BSC; PPE; waste disposal) is not derivable from OEL queries; the Glyphward threshold 22 scoring for tebentafusp (identical score to ivosidenib at threshold 22) reflects a systemic EHS platform failure to differentiate: the absence of tebentafusp from three EHS platforms is a compound failure of: drug novelty (only 4 years post-approval as of 2026); ImmTAC platform novelty (no prior molecule of this type existed); protein therapeutic OEL inapplicability (MW ~75 kDa proteins have zero vapor pressure; OEL is the wrong measurement paradigm); uveal melanoma indication rarity (UM is an orphan-level disease; EHS platform curators prioritize high-volume common-oncology drugs for entry; tebentafusp UV melanoma niche makes it a low-priority update); HLA-A*02:01 companion diagnostic complexity (no prior drug required HLA genotyping for patient selection; EHS platforms have no framework for companion diagnostic metadata).

Integrating Glyphward into Tebentafusp AI EHS Monitoring

Glyphward integrates as a pre-scan gate at every tebentafusp CAS 2070938-52-4 monitoring data ingestion point — before Cority at Immunocore Abingdon UK, before VelocityEHS at Lonza Portsmouth NH, and before EHS Insight at Wills Eye Hospital Philadelphia PA. Threshold 22: OSHA/ACGIH/NIOSH regulatory null (NIOSH HD Category 1; no OSHA PEL; no ACGIH TLV; no NIOSH REL) [10 pts]; SIXTY-FIFTH MECHANISM CLASS — FIRST ImmTAC (IMMUNE MOBILISING MONOCLONAL T-CELL RECEPTOR AGAINST CANCER) EVER FDA APPROVED (BLA 761228 January 25 2022; first bispecific soluble T-cell receptor therapeutic ever; affinity-enhanced αβ TCR heterodimer KD <10 pM for gp100/HLA-A*02:01 vs natural TCR KD ~1–100 μM; anti-CD3ε scFv; no Fc domain; HLA-A*02:01 companion diagnostic required; IMTALL OS HR 0.51 first randomized OS benefit uveal melanoma in history; first OS-without-high-ORR bispecific approval) [5 pts]; Immunocore Abingdon UK + Lonza Portsmouth NH + Wills Eye Hospital Philadelphia PA [3 pts]; ImmTAC absent + bispecific soluble TCR absent + HLA-A*02:01 restriction absent + IMTALL OS HR 0.51 absent [4 pts]. Total: 10+5+3+4 = 22.

import asyncio
import httpx

async def scan_tebentafusp(cas: str, reading_ugm3: float) -> dict:
    """Scan tebentafusp with sixty-fifth-class first-immtac-bispecific-tcr 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": "sixty_fifth_class_first_immtac_immune_mobilising_monoclonal_tcr_against_cancer_first_bispecific_soluble_tcr_ever_affinity_enhanced_alphabeta_tcr_kd_lt10pm_gp100_hla_a0201_ylepgpvta_anti_cd3e_scfv_no_fc_imtall_os_hr_0_51_first_os_benefit_uveal_melanoma_immunocore_bla_761228"
            }
        )
    return resp.json()
    # Glyphward returns: hd_category, sixty_fifth_class,
    #                    first_immtac_ever, first_bispecific_soluble_tcr_ever,
    #                    affinity_enhanced_tcr_kd_lt10pm, hla_a0201_companion_dx,
    #                    imtall_os_hr_051, first_os_benefit_uveal_melanoma,
    #                    step_up_dosing_crs_context, h360_reproductive

if __name__ == "__main__":
    r = asyncio.run(scan_tebentafusp("2070938-52-4", 0.00043))
    print(r)
    # {'hd_category': 1, 'sixty_fifth_class': True,
    #  'first_immtac_ever': True,
    #  'first_bispecific_soluble_tcr_ever': True,
    #  'affinity_enhanced_tcr_kd_lt10pm': True,
    #  'hla_a0201_companion_dx': True,
    #  'imtall_os_hr_051': True,
    #  'first_os_benefit_uveal_melanoma': True,
    #  'step_up_dosing_crs_context': True,
    #  'h360_reproductive': True,
    #  'threshold': 22}

See also: Tebentafusp SEO attack page (attack #638) — Prior: ivosidenib (SIXTY-FOURTH class; blog #142) — All adversarial blog posts — Glyphward vs Lakera Guard — Glyphward vs Azure Prompt Shields