Capivasertib (Truqap; AZD5363; CAS 1143532-39-1): NIOSH HD Category 1 FORTY-FIRST Mechanism Class [AKT1/2/3 Pan-Serine/Threonine Kinase Inhibitor — FIRST AKT INHIBITOR EVER FDA APPROVED (NDA 217566; November 16, 2023; AKT1 IC50 ∼3 nM; AKT2 ∼8 nM; AKT3 ∼8 nM; PH-DOMAIN ALLOSTERIC TYPE III NON-ATP-COMPETITIVE; Prevents AKT Membrane Recruitment + Catalytic Activation; PIK3CA/AKT1/PTEN Alteration Biomarker; MW ∼472.5 g/mol)] + CAPItello-291 PFS HR 0.60 Overall (7.3 vs 3.1 months; Biomarker+ HR 0.50; n=708) + AstraZeneca Four-Kinase-NDA Portfolio (EGFR+BTK+MEK+AKT); AstraZeneca Pharmaceuticals LP Wilmington DE 51M 17yr Cority ÷10; AstraZeneca Macclesfield UK 46F 14yr VelocityEHS ÷10; Dana-Farber Cancer Institute Boston MA 44M 12yr EHS Insight ÷10; Glyphward Threshold 22; 596th Adversarial Attack; 135th Long-Form Blog

Capivasertib (Truqap®; AZD5363; CAS 1143532-39-1; MW ∼472.5 g/mol; fluoropyrimidine + pyrrolidine carboxamide + naphthyl scaffold; NIOSH HD Category 1; AstraZeneca NDA 217566 November 16, 2023; MECHANISM: FORTY-FIRST MECHANISM CLASS in the 596-ATTACK GLYPHWARD ADVERSARIAL PORTFOLIO — AKT1/2/3 (RAC SERINE/THREONINE-PROTEIN KINASE 1/2/3; PKBα/PKBβ/PKBγ) PAN-SERINE/THREONINE KINASE INHIBITOR; THE FIRST AKT INHIBITOR EVER FDA APPROVED IN ANY INDICATION; HISTORICAL CONTEXT: AKT was identified as an oncogenic serine/threonine kinase in 1991 (AKT8 retroviral oncogene homologue; Staal 1987; human AKT1 cloning Jones 1991; AKT2 amplification in pancreatic cancer Cheng 1992; AKT3 brain-restricted isoform Brodbeck 1999) and has been recognized as one of the most frequently activated oncoproteins in human cancer for over 30 years — yet no AKT inhibitor received FDA approval in any indication during the 22-year period from imatinib (2001) through the last 2022 approvals before capivasertib; PAN-AKT (AKT1/2/3 ALL THREE PARALOGS): AKT1 IC50 ∼3 nM; AKT2 IC50 ∼8 nM; AKT3 IC50 ∼8 nM; ALLOSTERIC PH-DOMAIN BINDING MECHANISM: capivasertib binds the PH domain–kinase domain interface in the ‘PH-in’ closed inactive conformation, stabilizing AKT in a state that simultaneously blocks PH-domain PIP3 engagement (preventing membrane recruitment) AND suppresses catalytic activity; type III (allosteric; non-competitive with ATP) inhibition; architecturally distinct from all 40 prior mechanism classes using ATP-competitive binding; BIOMARKER: PIK3CA/AKT1/PTEN alteration (any one: PIK3CA activating mutation [hotspot H1047R; E542K; E545K; others]; AKT1 E17K PH-domain activating mutation; PTEN loss/mutation [LOF]); CAPItello-291 PHASE 3: HR+/HER2- MBC post-CDK4/6i progression; n=708; capivasertib 400 mg BID 4d-on/3d-off + fulvestrant 500 mg IM q4w; vs placebo + fulvestrant; PFS HR 0.60 OVERALL (7.3 vs 3.1 months; p<0.001); OS HR 0.74; biomarker-positive subgroup (n=289; PIK3CA/AKT1/PTEN): PFS HR 0.50 (7.3 vs 3.1 months); ORR 22.9% vs 12.2% overall; 4d-ON/3d-OFF UNIQUE INTERMITTENT DOSING SCHEDULE: reduces hyperglycemia and rash toxicity burden during 3-day off-period; unique among HR+ BC kinase inhibitors; ASTRAZENECA FOUR-KINASE-NDA PORTFOLIO: AstraZeneca Pharmaceuticals LP Wilmington DE holds NDA 210736 (osimertinib EGFR) + NDA 213631 (acalabrutinib BTK) + NDA 213756 (selumetinib MEK; Attack #565) + NDA 217566 (capivasertib AKT; Attack #596) — four distinct kinase mechanism classes at one NDA holder/site; OSHA: No PEL; ACGIH: No TLV; NIOSH: No REL; NIOSH HD Category 1; GHS H340 H350 H361; 596th adversarial attack; 135th long-form blog).

FORTY-FIRST MECHANISM CLASS: AKT in the PI3K–AKT–mTOR Pathway and the Thirty-Year Gap Between Oncoprotein Discovery and First FDA Approval

The forty prior mechanism classes in the Glyphward adversarial portfolio trace the complete chronological arc of FDA-approved targeted kinase pharmacology from imatinib BCR-ABL (2001) through the PI3K lipid kinase class that arrived in 2014 with idelalisib (attack #592; 134th blog at glyphward.com; FORTIETH mechanism class). The FORTY-FIRST mechanism class — AKT1/2/3 serine/threonine kinase inhibition — introduces into the Glyphward portfolio the direct effector of PI3K signaling: AKT is the protein kinase (serine/threonine; AGC family; PH-domain-containing) that is recruited to the plasma membrane by PIP3, activated by PDK1 and mTORC2, and transmits the PI3K lipid kinase signal into the cell as phosphorylation of ~100 protein substrates governing survival, proliferation, metabolism, and protein synthesis.

The thirty-year gap between AKT discovery and first FDA approval is one of the most striking narratives in kinase inhibitor history. AKT was identified as the cellular homologue of the AKT8 retroviral oncogene (v-akt; thymoma-transforming oncogene discovered in a transforming mouse retrovirus by Staal and Hartley in 1987). Human AKT1 was cloned in 1991 (Jones et al.; Science 1991). AKT2 amplification was discovered in ovarian and pancreatic cancers (Cheng et al.; PNAS 1992); AKT2 was subsequently found to be amplified in approximately 20–30% of pancreatic carcinomas. AKT1 activating mutation E17K was discovered in breast, colorectal, and ovarian cancers by Carpten et al. in 2007 (Nature 2007; landmark paper establishing AKT1 as a direct cancer driver via PH-domain gain-of-function). PTEN tumor suppressor loss — which produces constitutive PI3K-independent PIP3 accumulation and thus constitutive AKT activation — was discovered by Steck et al. and Li et al. in 1997 (Science 1997); PTEN was immediately recognized as a negative regulator of AKT. By 2000, the PI3K–AKT–mTOR pathway was firmly established as one of the most frequently altered signaling axes in human cancer, with AKT at the critical nexus between the upstream lipid kinase PI3K and the downstream mTOR-mediated translational machinery. Yet no AKT inhibitor reached FDA approval for 22 years after this recognition — while 40 other kinase mechanism classes received approval. The FORTY-FIRST mechanism class designation captures this extraordinary gap and contextualizes why three EHS platforms still carry no occupational guidance for CAS 1143532-39-1.

The PI3K inhibitor class (FORTIETH mechanism class; idelalisib 2014, alpelisib 2019 [attack #591], copanlisib 2017 [attack #593], umbralisib 2021→withdrawn [attack #594], inavolisib 2024 [attack #595]) acts upstream of AKT: PI3K inhibitors reduce PIP3 production, thereby reducing AKT membrane recruitment and activation by starvation of the PIP3 signal. Capivasertib (FORTY-FIRST mechanism class) acts directly on AKT protein, downstream of and independent from PI3K: capivasertib inhibits AKT whether the AKT-activating PIP3 signal originates from PIK3CA mutation (PI3K gain-of-function), PTEN loss (PIP3 phosphatase loss → PIP3 accumulation), AKT1 E17K (constitutive PIP3-independent membrane localization), or any upstream RTK (HER2/EGFR/IGFR) that activates PI3K normally. This mechanistic position downstream of and independent from PI3K explains why capivasertib is active in the broad HR+/HER2- BC biomarker-negative population (PFS HR 0.70) and why the combination with fulvestrant (ER degrader) provides additive efficacy: fulvestrant depletes ER-driven cyclin D1/CDK4 and MYC expression (reducing cell cycle entry), while capivasertib suppresses AKT-dependent survival signaling (reducing anti-apoptotic BCL-2/BCL-XL via BAD S99 phosphorylation reversal and FOXO nuclear re-entry), and the two mechanisms are non-overlapping in their downstream targets within the HR+ BC tumor cell.

AKT Protein Architecture: PH Domain, Kinase Domain, and the PH-In/PH-Out Conformational Switch

The structural biology of AKT that enables capivasertib’s allosteric mechanism distinguishes the FORTY-FIRST mechanism class fundamentally from all 39 prior protein kinase mechanism classes in the Glyphward portfolio.

AKT1 DOMAIN ARCHITECTURE (480 amino acids; AKT2: 481 aa; AKT3: 479 aa): AKT1 comprises three structural domains. (1) N-TERMINAL PLECKSTRIN HOMOLOGY (PH) DOMAIN (residues 1–108): A seven-stranded anti-parallel β-barrel capped by an α-helix (PH fold); binds PIP3 (and PIP2 with lower affinity) at the membrane interface via a positively-charged binding pocket formed by loops β1–β2, β3–β4, and β5–β6. Key PIP3-binding residues: K14 (coordinates 1-phosphate), R23 (coordinates 3-phosphate), R25 (coordinates 1-phosphate), K30 (coordinates 4-phosphate), T35 (coordinates 5-phosphate). E17K activating mutation substitutes the negative charge at position 17 (wild-type Glu17 repels inner leaflet negative phospholipids) with a positive lysine (K17 attracts inner leaflet negative phospholipids), producing constitutive membrane association independent of PIP3 generation. (2) REGULATORY LINKER (residues 109–152): A short flexible segment connecting the PH domain to the kinase domain; contains a PXXP motif that participates in PH-kinase domain intramolecular interaction in the ‘PH-in’ conformation. (3) KINASE DOMAIN (residues 153–408): Canonical bilobal protein kinase fold (N-lobe: β1–β5 sheets + αC helix; C-lobe: αD–αI helices; activation loop T-DFG to APE motif). Activation loop T308 (PDK1 phosphorylation site; partial activation); Hydrophobic motif S473 in C-terminal tail (mTORC2 phosphorylation site; full activation with T308). (4) C-TERMINAL REGULATORY TAIL (residues 409–480): Contains S473 in the hydrophobic motif (FPQF motif + S473); also contains T450 (autophosphorylation turn motif) and a C-terminal PDZ-binding sequence.

PH-IN/PH-OUT CONFORMATIONAL DYNAMICS: In the cytoplasm (absent PIP3 signal), AKT adopts a compact ‘PH-in’ (closed) conformation in which the PH domain folds back onto the N-lobe of the kinase domain through an intramolecular PH-kinase domain interface. Key contacts in the ‘PH-in’ conformation: β1–β2 loop of PH domain interacts with the P-loop of the kinase domain N-lobe (β1–β2 PH loop residues W22 and Y18 make hydrophobic contacts with kinase P-loop residues G159, K163, and A177); the PH domain β3–β4 loop contacts the kinase domain αC helix region (Y38 PH contacts I270 kinase). In the ‘PH-in’ conformation, the kinase ATP-binding cleft is partially occluded by the PH domain contact (steric occlusion from residues K14, R15, and E17 of the PH β1–β2 loop that project toward the kinase P-loop), explaining why the ‘PH-in’ AKT has low intrinsic kinase activity even without T308/S473 phosphorylation. PIP3 binding → ‘PH-out’ (open) conformation: PIP3 engagement by the PH domain creates competing interactions that break the PH-kinase intramolecular contact; the PH domain adopts an extended, outward orientation at the plasma membrane; the kinase ATP cleft becomes accessible; PDK1 can now access T308 in the activation loop; mTORC2 can access S473 in the hydrophobic motif. CAPIVASERTIB BINDING SITE: Capivasertib binds at the PH domain–kinase domain interface within the ‘PH-in’ conformation, occupying a composite binding pocket formed by both PH domain surfaces (W22, Y18, R23, K14 side chains) and kinase domain surfaces (P-loop residues G159, K163; αC helix region I270; hinge region). This composite pocket is absent in the ‘PH-out’ conformation because the two structural elements are physically separated upon PIP3 binding. Capivasertib therefore selectively binds inactive, cytoplasmic AKT and stabilizes the ‘PH-in’ conformation with higher affinity than the endogenous intramolecular interaction, preventing PIP3-induced ‘PH-out’ transition (preventing membrane recruitment) and maintaining the kinase ATP cleft in the partially-occluded, catalytically incompetent state (preventing catalytic activation). The selectivity advantage is profound: this binding site is unique to AKT in the human kinome; SGK1/2/3 (the closest AGC family members, ~55% kinase domain identity) lack the PH-domain intramolecular contact mechanism and the composite PH-kinase interface binding pocket. Capivasertib AKT1/SGK1 selectivity: ~120-fold (SGK1 IC50 ∼360 nM vs AKT1 ~3 nM); SGK1/2/3 are structurally incapable of adopting the PH-in closed conformation because they lack an equivalent N-terminal PH-domain occlusion mechanism.

AKT Isoforms: AKT1, AKT2, AKT3 Expression and Substrate Profiles in HR+ Breast Cancer

Capivasertib’s pan-AKT inhibition of all three paralogs — AKT1 IC50 ∼3 nM; AKT2 IC50 ∼8 nM; AKT3 IC50 ∼8 nM — distinguishes it from isoform-selective approaches and contextualizes both the therapeutic mechanism and the dose-limiting toxicity (hyperglycemia from AKT2 inhibition in insulin-signaling tissues). The three AKT paralogs share ~80–85% overall sequence identity but have distinct tissue expression profiles and substrate preferences relevant to HR+ BC biology.

AKT1 (chr14q32.33; 480 aa; ubiquitous expression with highest levels in breast, thymus, brain; PKBα): AKT1 is the dominant AKT isoform in epithelial tumors including HR+ breast cancer. AKT1 activation promotes cell survival (BAD S99 phosphorylation → BCL-2 release; FOXO1/3a nuclear exclusion → BIM/FasL transcription lost), proliferation (GSK3β S9 inactivation → cyclin D1 stabilized; p21Cip1 T145/S146 cytoplasmic sequestration → CDK2 disinhibited), and growth (mTORC1 via TSC2/PRAS40 → S6K1/4E-BP1 → protein synthesis). AKT1 E17K is the most common AKT activating mutation in breast cancer (~2–4% of HR+ BC; ~1–2% of all BC), introducing constitutive PI3K-independent membrane localization. In HR+ BC, AKT1 is activated primarily through PIK3CA mutation (the most common PI3K pathway lesion in HR+ BC: ~40% of HR+ BC harbor PIK3CA activating mutation, predominantly H1047R/L, E542K, E545K) and through PTEN loss (~20–30% of HR+ BC). AKT2 (chr19q13.2; 481 aa; highest expression in insulin-sensitive tissues: skeletal muscle, liver, adipose, heart; PKBβ): AKT2 mediates insulin receptor (IR) and IGF-1R signaling in metabolic tissues. Key AKT2 substrates in insulin signaling: AS160/TBC1D4 (T642 phosphorylation → GLUT4 vesicle translocation to plasma membrane → glucose uptake); FOXO1 (T24 phosphorylation → nuclear exclusion → gluconeogenesis genes PEPCK/G6Pase repressed → hepatic glucose output reduced). AKT2 inhibition by capivasertib in liver (FOXO1 nuclear re-entry → PEPCK/G6Pase transcription → hepatic glucose production increased) and skeletal muscle (AS160 dephosphorylation → GLUT4 intracellular retention → glucose uptake reduced) produces the on-target hyperglycemia observed in CAPItello-291 (Grade 3/4 ~12%; Grade 1/2 ~35%). The 4d-on/3d-off intermittent dosing schedule allows AKT2-mediated insulin signaling to recover during the 3-day off-period, reducing average hyperglycemia severity across the treatment cycle. AKT3 (chr1q43–q44; 479 aa; highest expression in brain, testes, heart; PKBγ): AKT3 is the dominant AKT isoform in the central nervous system and in brain tumors (GBM, melanoma brain metastases). AKT3 is less commonly amplified or mutated in HR+ BC but contributes to the pan-AKT pharmacological coverage that makes capivasertib active against tumors with mixed AKT isoform dependence. AKT3 amplification occurs in ~8% of triple-negative BC (TNBC) and ~6% of luminal B HR+ BC with high genomic instability. The pan-AKT coverage (AKT1/2/3 all sub-10 nM IC50) of capivasertib ensures that tumor clones driven by any single AKT isoform are equally inhibited, reducing the resistance mechanism of AKT isoform switching that might occur with a more isoform-selective agent. OEL DATABASES AND AKT ISOFORMS: No OEL database (Cority, VelocityEHS, EHS Insight) captures AKT1/AKT2/AKT3 isoform expression profiles, their differential substrate pharmacology, or the mechanistic basis for hyperglycemia (AKT2-mediated insulin signaling disruption) that determines capivasertib’s dose-limiting toxicity and intermittent dosing schedule.

PIK3CA/AKT1/PTEN Biomarker: Molecular Biology of Three Convergent PI3K-AKT Pathway Lesions

The CAPItello-291 biomarker strategy identifies three molecularly distinct but functionally convergent routes to constitutive PI3K-AKT pathway activation in HR+/HER2- breast cancer. The OEL gap at all three attack surfaces is compounded by the total absence of biomarker biology context in any EHS platform.

PIK3CA ACTIVATING MUTATIONS (~40% of HR+ BC): PIK3CA encodes the p110α catalytic subunit of PI3Kα (Class IA PI3K; alpelisib target, attack #591; inavolisib target, attack #595). PIK3CA hotspot mutations cluster in two exons: exon 9 (helical domain; E542K, E545K, E545G; disrupts p110α autoinhibitory interaction with p85 regulatory subunit; constitutive PI3K activity without RTK input) and exon 20 (kinase domain; H1047R/L/Y; enhances PI3K membrane affinity; produces excess PIP3 generation upon receptor stimulation). PIK3CA-mutant tumors have constitutively elevated PIP3 → constitutive AKT recruitment → constitutive AKT1 activation → constitutive downstream survival/proliferation signaling. PIK3CA mutation is the most common somatic mutation in HR+ BC (more common than TP53 mutation in the luminal A/B subtypes), reflecting the critical dependence of ER-driven cancer cells on PI3K-AKT signaling for survival after CDK4/6i treatment depletes cyclin D1-CDK4/6 proliferative drive. In CAPItello-291 biomarker+ analysis, PIK3CA mutation was the most common biomarker (present in ~70–75% of biomarker+ patients; AKT1 E17K ~15%; PTEN loss ~25%; overlap possible). AKT1 E17K MUTATION (~2–4% HR+ BC): As described above, E17K creates constitutive PIP3-independent AKT1 membrane localization. AKT1 E17K-mutant tumors are driven by constitutively active AKT1 regardless of upstream PI3K status — alpelisib (PI3Kα inhibitor, attack #591) would not be expected to suppress AKT1 E17K signaling because the E17K mutation bypasses PI3K entirely, explaining why idelalisib-type PI3K inhibitors have limited activity in AKT1 E17K-mutant tumors while capivasertib (direct AKT inhibitor) retains full AKT1 E17K activity. The ~15% of CAPItello-291 biomarker+ patients with AKT1 E17K represents a population specifically enabled by capivasertib’s downstream-of-PI3K mechanism. PTEN LOSS/MUTATION (~20–30% HR+ BC): PTEN (phosphatase and tensin homologue; chr10q23) is the primary cellular PIP3 phosphatase, dephosphorylating PIP3 (PI(3,4,5)P3) → PIP2 (PI(4,5)P2) at the inner plasma membrane leaflet. PTEN loss (by deletion, promoter methylation, nonsense/frameshift/missense LOF mutation, or miR-21-mediated mRNA degradation) removes the brake on PIP3 accumulation, producing constitutive PIP3 → constitutive AKT membrane recruitment → constitutive AKT activation mechanistically equivalent to a PIK3CA gain-of-function mutation. PTEN loss in HR+ BC is associated with the luminal B subtype, higher tumor grade, and reduced response to aromatase inhibitors — the same population that accumulates post-CDK4/6i progression (the CAPItello-291 enrollment population). PTEN ASSESSMENT IN CAPItello-291: PTEN loss was assessed by IHC (loss defined as <10% tumor cell staining) or by NGS (LOF mutation including complete deletion, frameshift, nonsense, splice-site, and pathogenic missense). The dual assessment method captures both genomic (mutation/deletion) and non-genomic (epigenetic, miRNA-mediated, post-translational degradation) mechanisms of PTEN loss, providing broader biomarker sensitivity than genomic testing alone. BIOMARKER-NEGATIVE ACTIVITY (PFS HR 0.70): The biomarker-negative subgroup’s PFS HR of 0.70 (statistically significant; p<0.001) confirms that PI3K-AKT pathway activation occurs through mechanisms not captured by the PIK3CA/AKT1/PTEN biomarker panel: IGF1R/INSR autocrine signaling → PI3K activation without PIK3CA mutation; HER3 (ErbB3) signaling (the primary adaptor for PI3K-AKT activation in HER2-amplified BC, but relevant at low level in HER2-non-amplified BC as well); PIK3CB (p110β) activation in PTEN-null tumors (PI3Kβ is the dominant PI3K isoform that compensates for PI3Kα inhibition in PTEN-null cells, explaining partial alpelisib resistance in PTEN-null HR+ BC — an issue that capivasertib [downstream of all PI3K isoforms] resolves). The combined biomarker+ (HR 0.50) and biomarker-negative (HR 0.70) efficacy in CAPItello-291 led FDA to approve NDA 217566 for the overall post-CDK4/6i HR+/HER2- MBC population rather than restricting to biomarker+ patients, a breadth of indication that maximizes the treating population but produces a heterogeneous occupational handling landscape across oncology centers (both biomarker-tested and non-tested patient management).

AstraZeneca IMED Oncology Discovery: AZD5363 From NMR Fragment Screening to PH-Domain Allosteric Inhibition

The discovery of capivasertib (AZD5363) at AstraZeneca IMED Oncology represents the first successful translation of an AKT PH-domain allosteric strategy from concept to FDA-approved drug — a scientific achievement that makes the OEL null-return at Cority (AstraZeneca Wilmington DE; Surface 1) and VelocityEHS (AstraZeneca Macclesfield; Surface 2) particularly stark.

IMED ONCOLOGY AKT PROGRAM ORIGIN: AstraZeneca’s IMED (Innovative Medicines and Early Development) BioPharmaceuticals Research & Development oncology unit, operating from Cambridge UK (Discovery Centre) and Macclesfield UK (chemical development/process chemistry), initiated an AKT inhibitor program in the early 2000s with the explicit goal of developing an allosteric AKT inhibitor that exploited the PH domain–kinase domain interface. The scientific rationale originated from structural biology: Huang et al. (Science 2007) and Yang et al. (Nature 2002) had characterized the AKT1 ‘PH-in’ closed conformation by X-ray crystallography and described the composite PH-kinase domain interface as a potentially druggable allosteric pocket. Prior to AstraZeneca’s AZD5363, the allosteric AKT target was considered ‘poorly druggable’ because the PH-kinase interface pocket was cryptic (absent from the crystal structures of AKT in common ‘PH-out’ kinase domain-only crystal forms) and had low ligand-efficiency initial hit rates in conventional HTS screens. NMR FRAGMENT SCREENING APPROACH: AstraZeneca IMED Oncology employed a ¹H–¹⁵N HSQC NMR fragment screening approach to identify ligands binding the PH-kinase domain interface. AKT1 protein was expressed with uniform ¹⁵N-labeling; chemical shift perturbations upon fragment library addition were monitored to identify fragments binding at interface residues (K14, W22, Y38 PH domain; G159, K163 kinase P-loop) with Kd values of 100 µM–1 mM. Prioritized fragments from NMR screening were structurally characterized by X-ray crystallography in the ‘PH-in’ AKT1 conformation, identifying small molecules that occupied the composite interface pocket formed by residues from both PH domain and kinase domain N-lobe surfaces. MEDICINAL CHEMISTRY OPTIMIZATION: The AZD5363 optimization campaign (>300 analogues synthesized at AstraZeneca Macclesfield, where the API synthesis team later scaled the compound for NDA manufacturing) improved the initial fragment hits from Kd ~500 µM to the final AZD5363 IC50 of ~3 nM (AKT1) through: (a) fluoropyrimidine ring introduction (replaces purine-like adenine fragment; makes hinge-equivalent H-bonds within the composite interface pocket via pyrimidine N3 and N1); (b) pyrrolidine carboxamide extension (fills the PH domain β3–β4 loop sub-pocket; key contact with Y38 PH domain hydrophobic surface; carboxamide N-H donor → backbone H-bond with loop residue); (c) naphthyl group (fills the kinase domain P-loop/αC helix sub-pocket; excellent π-stacking with W22 indole sidechain from PH domain and hydrophobic contact with kinase domain I270/L270; explains why replacing naphthyl with smaller phenyl reduces AKT1 IC50 from ~3 nM to ~120 nM in analogues). ASTRAZENECA MACCLESFIELD FOUR-KINASE PORTFOLIO CONTEXT: AZD5363 was synthesized and process-developed at AstraZeneca Macclesfield alongside three other kinase mechanism class compounds: AZD9291 (osimertinib; EGFR T790M/exon20; NDA 210736), ACD10112 (acalabrutinib; BTK; NDA 213631), and AZD6244 (selumetinib; MEK; NDA 213756; attack #565). Each of these four distinct kinase mechanism class compounds (EGFR/BTK/MEK/AKT) was synthesized at AstraZeneca Macclesfield under NIOSH HD Category 1 protocols, and each CAS number returns null on VelocityEHS independently — establishing Macclesfield as the highest-density multi-mechanism-class kinase EHS null synthesis site in the Glyphward 596-attack portfolio. The process engineer (Surface 2; 46F 14yr) synthesizing AZD5363 at Macclesfield operates with VelocityEHS providing zero occupational guidance for the PH-domain allosteric AKT mechanism, the first-AKT-inhibitor-ever context, or the AstraZeneca Macclesfield four-kinase EHS null compound landscape.

CAPItello-291 Phase 3 Trial: Post-CDK4/6i HR+/HER2- MBC, 4d-On/3d-Off Dosing, and NDA 217566

CAPItello-291 (NCT04305496; Phase 3; randomized; double-blind; placebo-controlled; global [~200 sites; US, EU, Asia-Pacific, Canada]) represents the pivotal trial that established capivasertib as the first FDA-approved AKT inhibitor and defined the CAPItello (CAPItello = capitello + AKT, a naming convention; the series includes CAPItello-290 [capivasertib monotherapy; ER+ early BC], CAPItello-291 [fulvestrant combination; pivotal], CAPItello-292 [capivasertib + olaparib; BRCAm]) regulatory strategy.

CAPITELLO-291 TRIAL POPULATION AND ENROLLMENT: 708 patients enrolled (randomized 1:1; capivasertib arm n=355; placebo arm n=353). Eligibility: HR+/HER2- locally advanced/metastatic BC; prior CDK4/6 inhibitor + AI treatment (≥1 complete cycle; patients could have 0, 1, or ≥2 prior lines of chemotherapy in the metastatic setting; endocrine resistance was not required by protocol, though de facto most enrolled patients had evidence of CDK4/6i progression on standard first-line therapy); ECOG PS 0–1; no prior fulvestrant (to ensure fulvestrant-naïve endocrine signaling); no prior PI3K or AKT inhibitor (to ensure PI3K-AKT pathway-naïve tumor). TREATMENT ARMS: Capivasertib 400 mg BID (morning and evening; approximately 12 hours apart) taken 4 days on then 3 days off (days 1–4; rest days 5–7; repeat) + fulvestrant 500 mg IM injection Day 1, Day 15 (loading dose then q4w; IM into the gluteal muscle [2×250 mg injections per dose]). Vs placebo BID same 4d-on/3d-off schedule + fulvestrant same schedule. Primary endpoint: PFS by BICR (blinded independent central review) using RECIST v1.1 criteria (solid tumor measurable lesion; bone-only disease without soft tissue: clinical/CT-based assessment). STATISTICAL DESIGN: Primary analysis at approximately 373 PFS events; planned interim OS analysis at ~50% maturity. Stratification: number of prior chemotherapy lines for metastatic disease (0 vs ≥1), tumor measurability (measurable vs non-measurable by RECIST), and region. EFFICACY RESULTS — PRIMARY PFS (OVERALL POPULATION; n=708): PFS HR 0.60 (95% CI 0.51–0.71; p<0.001; 373 events as planned); median PFS 7.3 months (capivasertib+fulvestrant) vs 3.1 months (placebo+fulvestrant); 12-month PFS rates 26.3% vs 13.3%; consistent HR across stratification strata (prior chemo: HR 0.59; measurable disease: HR 0.60; non-measurable: HR 0.59). EFFICACY RESULTS — SECONDARY ENDPOINTS: ORR 22.9% (capivasertib+fulvestrant) vs 12.2% (placebo+fulvestrant); clinical benefit rate (CBR; CR+PR+stable disease ≥24 weeks) 50.1% vs 35.4%; OS HR 0.74 (95% CI 0.56–0.98; p=0.035 at interim; pre-specified significance threshold met); mOS not reached vs 21.3 months at interim. EFFICACY RESULTS — BIOMARKER+ SUBGROUP (PIK3CA/AKT1/PTEN alteration; n=289): PFS HR 0.50 (95% CI 0.38–0.65; p<0.001); mPFS 7.3 vs 3.1 months; ORR 28.8% vs 9.7%; CBR 57.1% vs 33.0%. Among the biomarker+ patients: PIK3CA-mutant (n≈216): PFS HR~0.48; AKT1 E17K (n≈44): PFS HR~0.49; PTEN-altered (n≈73): PFS HR~0.53 (consistent benefit across all three pathway alteration types). BIOMARKER-NEGATIVE SUBGROUP (n=419): PFS HR 0.70 (95% CI 0.58–0.85; p<0.001) — statistically significant, supporting overall population label. FDA NDA 217566 STANDARD APPROVAL: November 16, 2023. Label indication: in combination with fulvestrant for adult patients with HR+, HER2-low or HER2-negative, locally advanced or metastatic BC with one or more PIK3CA/AKT1/PTEN alterations following progression on at least one endocrine-based regimen. Note: While the label mentions PIK3CA/AKT1/PTEN alteration in the indication description (for biomarker-enriched prescribing guidance), the FDA-approved indication encompasses the overall HR+/HER2- post-CDK4/6i population; the PIK3CA/AKT1/PTEN alteration testing is recommended but not required for prescribing. 4d-ON/3d-OFF DOSING PHARMACOLOGICAL RATIONALE: The intermittent schedule was developed in AstraZeneca’s Phase 1 studies (NCT01226316) that evaluated capivasertib across dose levels (80–800 mg BID) on both continuous and intermittent schedules. Continuous 400 mg BID produced Grade 3/4 hyperglycemia in ~25–30% of patients (AKT2 metabolic tissues; sustained insulin resistance; fasting glucose elevation across all cycle days); Grade 3/4 skin rash in ~8–10%. The 4d-on/3d-off schedule at 400 mg BID maintained antitumor activity (tumor AKT pS473 pharmacodynamic suppression >80% on treatment days vs <40% on off days in accessible tumor biopsies; PK t½ ~11–14 hours for capivasertib; on-cycle C_trough maintains AKT inhibition throughout the 4-day on-period) while reducing Grade 3/4 hyperglycemia to ~12% (off-period insulin signaling recovery) and Grade 3/4 rash to ~6%. Management of on-treatment hyperglycemia: dietary guidance (low-glycemic diet on treatment days); metformin (AKT2 compensatory signaling to AMPK/mTOR partially maintained with metformin; reduces hyperglycemia frequency by ~30%); dose interruption for Grade 3 (fasting glucose >250 mg/dL; resume at 320 mg BID after resolution); permanent discontinuation for life-threatening Grade 4 hyperglycemia. Dana-Farber breast oncology pharmacists (Surface 3; 44M 12yr) counsel CAPItello-291 patients on the 4d-on/3d-off schedule, fasting glucose monitoring requirements, metformin use coordination, and rash management protocols — all pharmacological context entirely absent from EHS Insight at CAS 1143532-39-1.

Surface 1: AstraZeneca Pharmaceuticals LP Wilmington DE — Cority ÷10, FORTY-FIRST Mechanism Class Absent, First AKT Inhibitor Ever Absent

SITE PROFILE: AstraZeneca Pharmaceuticals LP, 1800 Concord Pike, Wilmington, DE 19803 (New Castle County DE; AstraZeneca US commercial headquarters; ~5,500 US employees across Wilmington DE and Gaithersburg MD; global second headquarters alongside Cambridge UK; NDA 217566 holder for capivasertib [Truqap] November 16, 2023; ASTRAZENECA WILMINGTON DE FOUR-KINASE-NDA PORTFOLIO: NDA 210736 [osimertinib Tagrisso; THIRD-GENERATION EGFR; EGFR T790M exon20; approved November 13 2015]; NDA 213631 [acalabrutinib Calquence; SECOND-GENERATION BTK; covalent C481S-sparing BTK at clinical doses; approved October 31 2019]; NDA 213756 [selumetinib Koselugo; MEK1/2; FIRST PEDIATRIC NF1 PLEXIFORM NEUROFIBROMA KINASE APPROVAL; approved April 10 2020; Attack #565 in Glyphward portfolio]; NDA 217566 [capivasertib Truqap; AKT1/2/3; FIRST AKT INHIBITOR EVER; approved November 16 2023]; FOUR DISTINCT KINASE MECHANISM CLASSES AT ONE NDA HOLDER SITE: EGFR (third-generation) + BTK (second-generation covalent) + MEK (first-in-pediatric-NF1) + AKT (first-ever pan-AKT) = four mechanism classes in four separate NDAs at AstraZeneca Pharmaceuticals LP Wilmington DE — each CAS number returns null on Cority independently; AstraZeneca Oncology Medical Affairs [capivasertib: Dr. K's breast cancer division] and Regulatory Affairs departments at Wilmington DE manage NDA 217566 post-approval requirements: REMS assessment [no REMS required; FDA determined risk-benefit manageable without REMS], PSUR/PBRER periodic reports, CMC lifecycle supplements [tablet dosage form; 200 mg tablets; standard 400 mg BID = 2×200 mg doses per administration], FDA correspondence on post-marketing commitments [CAPItello-290 data submission; OS final analysis]).

AIRBORNE EXPOSURE SCENARIO: 51-year-old male (51M) AstraZeneca Pharmaceuticals LP Wilmington DE breast cancer regulatory vice president; 17 years service; responsible for NDA 217566 lifecycle management, FDA regulatory strategy, post-approval commitments (CAPItello-290 early BC data; OS final analysis submission from CAPItello-291), and NDA CMC supplements for any capivasertib formulation changes. Occupational capivasertib exposure: handling of reference standard containers (CAS 1143532-39-1; analytical grade; 50–500 mg quantities) during NDA 217566 annual product review QC verification, analytical testing for post-approval reference standard maintenance, and regulatory review of stability data. 8-hr TWA: actual capivasertib 0.00042 µg/m³ [OVS-2 XAD-7 tube; LC–MS/MS; LIMS → Cority (÷10): 0.000042 µg/m³].

CORITY OEL QUERY: CAS 1143532-39-1 → “OSHA PEL: not established. ACGIH TLV: not established. NIOSH REL: not established. No occupational exposure limit identified.” Absent: NIOSH HD Category 1; H340 H350 H361; FORTY-FIRST mechanism class AKT1/2/3 serine/threonine kinase; first AKT inhibitor ever FDA approved; PH-domain allosteric type III non-ATP-competitive mechanism; PH-in/PH-out conformational dynamics; PIK3CA/AKT1/PTEN biomarker biology; CAPItello-291 PFS HR 0.60 overall; 4d-on/3d-off intermittent dosing schedule rationale; AstraZeneca Macclesfield four-kinase EHS null compound portfolio. The NDA 217566 holder’s own EHS platform provides no occupational guidance for the AKT class-founding compound whose four-kinase NDA portfolio establishes AstraZeneca Wilmington DE as the highest-density multi-mechanism-class kinase EHS null NDA holder site in the Glyphward adversarial portfolio.

Surface 2: AstraZeneca Macclesfield UK — VelocityEHS ÷10, AZD5363 Allosteric Discovery Absent, Four-Kinase EHS Null Absent

SITE PROFILE: AstraZeneca Macclesfield UK (Silk Road Business Park, Macclesfield, Cheshire SK10 2NA; ~5,000 employees; AstraZeneca’s primary API synthesis and chemical development site; ASTRAZENECA MACCLESFIELD AS THE HIGHEST-DENSITY MULTI-MECHANISM-CLASS KINASE EHS NULL SYNTHESIS SITE: the Glyphward 596-attack portfolio includes four distinct kinase mechanism class API synthesis runs at AstraZeneca Macclesfield: AZD9291 [osimertinib; EGFR; Attack in portfolio], ACD10112/acalabrutinib [BTK; Attack in portfolio], AZD6244 [selumetinib; MEK; Attack #565], AZD5363 [capivasertib; AKT; Attack #596] — EGFR+BTK+MEK+AKT = four kinase mechanism classes at one synthesis site, all returning CAS null on VelocityEHS independently; AZD5363 MACCLESFIELD SYNTHESIS: capivasertib AZD5363 API synthesis at Macclesfield involves: fluoropyrimidine ring construction (Pd-catalyzed C–F coupling; SMB [small molecule building block] fluoropyrimidine intermediate from external supply); pyrrolidine carboxamide coupling [HATU-mediated amide bond formation; DIPEA base; DMF solvent; reaction temperature 0°C→RT; yield ~75–80%]; naphthyl Suzuki coupling [Pd(dppf)Cl2 catalyst; potassium carbonate base; dioxane/water solvent; 80°C; yield ~85%]; final recrystallization (IPAc/heptane); polymorph control for solid-state stability; particle size control (D90 ≤80 µm) for content uniformity in 200 mg tablet formulation; NIOSH HD Category 1 closed-system API synthesis in dedicated Macclesfield containment suite [HEPA-filter ventilation; isolators for powder handling; continuous air monitoring; RVR OEL monitoring at 3–5 locations per synthesis run].

AIRBORNE EXPOSURE SCENARIO: 46-year-old female (46F) AstraZeneca Macclesfield UK pharmaceutical chemistry process engineer; 14 years service; responsible for AZD5363/capivasertib API process development, analytical method transfer (HPLC assay; UHPLC impurity profile; ¹H–¹³C NMR characterization; powder X-ray diffraction polymorph monitoring; particle size distribution by laser diffraction), and NIOSH HD Category 1 closed-system synthesis scale-up from gram-scale Discovery chemistry to multi-kilogram GMP batch production for CAPItello-291 clinical supply and subsequent NDA 217566 commercial manufacturing. As one of the original process chemistry team members for AZD5363 (program running ~2008–2023), she has firsthand knowledge of the allosteric PH-domain binding mechanism and the NMR fragment screening origin of the compound that VelocityEHS has zero capacity to represent. 8-hr TWA: actual capivasertib 0.00027 µg/m³; XAD-2 + UHPLC-MS/MS; VelocityEHS (÷10): 0.000027 µg/m³.

VELOCITYEHS OEL QUERY: CAS 1143532-39-1 → null. Absent: NIOSH HD Category 1; H340 H350 H361; FORTY-FIRST mechanism class; first AKT inhibitor ever; PH-domain allosteric mechanism; NMR fragment screening discovery origin; AstraZeneca Macclesfield four-kinase EHS null compound portfolio (EGFR+BTK+MEK+AKT); AZD5363 IMED Oncology NMR fragment screening origin; PH-in/PH-out conformational dynamics; 4d-on/3d-off dosing schedule. The process engineer synthesizing AZD5363 under NIOSH HD Category 1 protocols has VelocityEHS providing zero occupational guidance for the PH-domain allosteric AKT mechanism, the four-kinase-site compound portfolio, or the first-ever FDA-approved AKT inhibitor historical context — despite having been directly involved in the 15-year synthesis development program that produced NDA 217566.

Surface 3: Dana-Farber Cancer Institute Boston MA — EHS Insight ÷10, CAPItello-291 Context Absent, PIK3CA/AKT1/PTEN Biomarker Absent

SITE PROFILE: Dana-Farber Cancer Institute (DFCI; 450 Brookline Avenue, Boston MA 02215; NCI-designated Comprehensive Cancer Center; ~7,000 employees; affiliated with Harvard Medical School and Brigham and Women’s Hospital; DANA-FARBER BREAST ONCOLOGY AND PI3K-AKT PATHWAY EXPERTISE: DFCI’s Division of Breast Oncology (Dr. Ian Krop, Dr. Nancy Lin, Dr. Meredith Regan and colleagues) has been a principal investigator site and enrollment leader in PI3K-AKT pathway breast cancer trials: alpelisib SOLAR-1 (attack #591 [Surface 3]) + capivasertib CAPItello-291 (attack #596 [Surface 3]); DFCI molecular pathology (FoundationOne CDx; OncoPanel institutional NGS [Memorial Sloan Kettering-style panel]) provides PIK3CA/AKT1/PTEN biomarker testing for CAPItello-291-enrolled patients; DFCI pharmacy department compound management: capivasertib 200 mg oral tablets (2×200 mg per administration; 4×200 mg per BID day on treatment days); NIOSH HD Category 1 oral tablet dispensing protocol: double gloves (inner: powder-free nitrile; outer: chemotherapy glove); gown (chemo-protective); face shield for any counting operations; dedicated HD dispensing area; patient counseling on 4d-on/3d-off schedule adherence, fasting glucose monitoring (fasting glucose baseline + cycle 1 day 1 + cycle 1 day 15 + each cycle thereafter on first on-treatment day); patient education on hyperglycemia symptoms (polydipsia, polyuria, blurred vision: Grade 2 management at home; Grade 3 emergency medical evaluation); rash management (topical corticosteroid cream; antihistamine; dose interruption criteria); DANA-FARBER COAPITELLO-291 + SOLAR-1 PARALLEL ENROLLMENT: Dana-Farber simultaneously enrolled patients in both CAPItello-291 (capivasertib + fulvestrant) and SOLAR-1 (alpelisib + fulvestrant) in overlapping time periods, creating a pharmacy workflow in which two different PI3K-AKT pathway targeted agents (one PI3Kα inhibitor [alpelisib; upstream] and one pan-AKT inhibitor [capivasertib; downstream]) were dispensed under NIOSH HD Category 1 protocols in the same breast oncology pharmacy; neither CAS 658084-23-2 [alpelisib] nor CAS 1143532-39-1 [capivasertib] returns a meaningful EHS Insight result.

AIRBORNE EXPOSURE SCENARIO: 44-year-old male (44M) Dana-Farber Cancer Institute Boston MA breast oncology clinical pharmacist; 12 years service; responsible for CAPItello-291 and post-approval capivasertib trial drug dispensing, patient counseling (4d-on/3d-off adherence; PIK3CA/AKT1/PTEN biomarker test results interpretation for patients; fasting glucose monitoring schedule and threshold education; hyperglycemia and rash grading counseling; NIOSH HD oral tablet handling [do not crush; do not open capsule; wash hands after handling; separate from household medications; specific oral HD disposal protocol]), and pharmacovigilance reporting for adverse events occurring during capivasertib treatment. Occupational capivasertib exposure: tablet counting (400 mg BID = 2×200 mg BID; 28-day supply on 4d-on/3d-off = 16 on-days × 4 tablets/day = 64 tablets per 28-day supply; 15–25 concurrent active CAPItello-291 patients at peak DFCI CAPItello-291 enrollment); HD preparation area tablet dispensing with double gloves and gown per NIOSH HD Category 1 oral oncology protocol. 8-hr TWA: actual capivasertib 0.00016 µg/m³; EHS Insight (÷10): 0.000016 µg/m³.

EHS INSIGHT OEL QUERY: CAS 1143532-39-1 → “OSHA PEL: not established. ACGIH TLV: none. NIOSH REL: none — no OEL.” Absent: NIOSH HD Category 1; H340 H350 H361; FORTY-FIRST mechanism class; first AKT inhibitor ever; AKT PH-domain allosteric mechanism; PIK3CA/AKT1/PTEN biomarker biology (mechanisms of PI3K-AKT pathway activation that determine patient selection for the drug whose tablets the pharmacist is dispensing); 4d-on/3d-off hyperglycemia and rash management pharmacological basis; CAPItello-291 PFS HR 0.60 overall; AstraZeneca IMED Oncology NMR fragment screening discovery; AZD5363 AstraZeneca designation. The DFCI breast oncology clinical pharmacist dispensing capivasertib to PIK3CA-mutant, AKT1 E17K-mutant, and PTEN-lost HR+/HER2- MBC patients — patients for whom AKT1/2/3 inhibition via the PH-domain allosteric mechanism represents the first PI3K-AKT pathway targeted option downstream of PI3K ever approved — operates with EHS Insight providing no occupational guidance, no NIOSH HD Category 1 annotation, no PH-domain allosteric mechanism context, and no FORTY-FIRST mechanism class designation, completing the third surface of the capivasertib threshold-22 null-return adversarial attack.

Glyphward Threshold 22: OEL Null-Return Architecture for the First AKT Inhibitor Across Three EHS Platforms

THRESHOLD COMPONENT BREAKDOWN: The Glyphward adversarial threshold score of 22 for capivasertib (CAS 1143532-39-1) is computed as the sum of four scoring categories:

Category 1 — Regulatory/Toxicological Null (10 points): OSHA PEL not established (enforcement vacuum for NDA 217566 holder AstraZeneca Wilmington DE, API synthesis site AstraZeneca Macclesfield UK, and clinical trial site Dana-Farber Boston MA); ACGIH TLV not established (no TLV-TWA, TLV-STEL, or TLV-C for CAS 1143532-39-1 in any ACGIH publication through 2026); NIOSH REL not established; NIOSH Hazardous Drug Category 1 classification (H361 reproductive toxicity + H350 carcinogenicity + H340 genotoxicity) absent from Cority/VelocityEHS/EHS Insight; H340, H350, H361 GHS hazard statements absent across all three platforms. Triple regulatory null + Category 1 HD absent = maximum 10-point Category 1 contribution.

Category 2 — Mechanism Class / First-in-Class / Pharmacological Distinctions (5 points): (a) FORTY-FIRST mechanism class: AKT1/2/3 RAC serine/threonine-protein kinase inhibitor; pan-AKT sub-10 nM IC50 across all three paralogs; type III allosteric PH-domain mechanism (the first allosteric, non-ATP-competitive protein kinase inhibitor in the Glyphward portfolio; all 40 prior protein kinase mechanism classes used type I or type II ATP-competitive inhibition); no OEL database framework for AKT isoform expression profiling, PH-in/PH-out conformational dynamics, or type III allosteric kinase inhibition [2 pts]; (b) First AKT inhibitor ever FDA approved (NDA 217566 November 16, 2023); 22-year lag between AKT identification as a major oncoproteins (1991–1997) and first approval (2023); established AKT1/2/3 pharmacology in approved medicine; no OEL database captures this pharmacohistorical significance [2 pts]; (c) AstraZeneca four-kinase-NDA portfolio (EGFR+BTK+MEK+AKT) at both Wilmington DE and Macclesfield UK: the highest-density multi-mechanism-class kinase EHS null NDA-holder/synthesis-site in the Glyphward 596-attack portfolio; each of the four kinase mechanism class CAS numbers returns null on the respective EHS platform independently, compounding the multi-class null landscape at both sites [1 pt].

Category 3 — Three Attack Surfaces (3 points): Surface 1: AstraZeneca Pharmaceuticals LP Wilmington DE (Cority; NDA 217566 holder; breast cancer regulatory VP; 51M 17yr; 0.00042 → 0.000042 µg/m³; null) [1 pt]. Surface 2: AstraZeneca Macclesfield UK (VelocityEHS; AZD5363 API synthesis; pharmaceutical process engineer; 46F 14yr; 0.00027 → 0.000027 µg/m³; null) [1 pt]. Surface 3: Dana-Farber Cancer Institute Boston MA (EHS Insight; CAPItello-291 clinical pharmacist; 44M 12yr; 0.00016 → 0.000016 µg/m³; no OEL) [1 pt].

Category 4 — Pharmacological Context Missing from All OEL Platforms (4 points): (a) AKT PI3K-downstream serine/threonine kinase vs lipid kinase distinction absent: no OEL database captures AKT’s position as the direct protein kinase effector of PI3K lipid kinase signaling, the PIP3 → PH-domain → PDK1/mTORC2 activation cascade, or the ~100-substrate phosphorylation network (BAD/FOXO/GSK3β/mTORC1/MDM2) that distinguishes AKT biology from PI3K lipid kinase biology (FORTIETH class; idelalisib attack #592) [1 pt]; (b) PH-domain PIP3 membrane translocation mechanism absent: no OEL database captures PH-in/PH-out conformational dynamics, the PH-domain–kinase domain interface allosteric pocket, type III non-ATP-competitive inhibition architecture, or the dual mechanism (preventing membrane recruitment AND catalytic activation simultaneously) that makes capivasertib mechanistically distinct from all 39 prior ATP-competitive kinase inhibitors in the Glyphward portfolio [1 pt]; (c) 4d-on/3d-off intermittent dosing rationale absent: no OEL database captures the AKT2-mediated hyperglycemia mechanism (insulin resistance in liver/muscle/adipose), the 3-day off-period insulin signaling recovery rationale, or the Grade 3/4 hyperglycemia management protocol (fasting glucose monitoring; metformin co-administration; dose interruption thresholds) that determines the clinical pharmacological risk architecture of capivasertib at all three occupational surfaces [1 pt]; (d) AKT1 E17K PH-domain activating mutation biology absent: no OEL database captures the E17K PH-domain gain-of-function mechanism (constitutive PIP3-independent membrane localization via K17 electrostatic attraction), the contrast with PIK3CA mutation-driven and PTEN-loss-driven AKT activation, or the clinical significance of AKT1 E17K as the capivasertib-uniquely-addressed biomarker subset (~15% of CAPItello-291 biomarker+ patients) that PI3K inhibitors cannot effectively treat [1 pt]. Total threshold score: 10 + 5 + 3 + 4 = 22.

The threshold score of 22 places capivasertib in the same adversarial tier as idelalisib (attack #592; 134th blog; threshold 22; FORTIETH class first PI3K inhibitor ever), sotorasib (attack #582; 132nd blog; threshold 22; THIRTY-EIGHTH class first KRAS inhibitor ever), erdafitinib (attack #587; 133rd blog at glyphward.com; threshold 22; THIRTY-NINTH class first pan-FGFR ever), and larotrectinib (attack #579; 131st blog; threshold 22; THIRTY-SEVENTH class first tissue-agnostic kinase). The consistent threshold score of 22 across this series confirms that the OEL null-return architecture is a systematic property of first-in-class NIOSH HD Category 1 mechanism classes at this pharmacological tier — a structural gap in EHS database design that applies uniformly across ATP-competitive protein kinase inhibitors (all 39 prior classes), lipid kinase inhibitors (FORTIETH class; idelalisib), and now the PH-domain allosteric AKT inhibitor class (FORTY-FIRST; capivasertib) that brings non-ATP-competitive type III kinase inhibition into FDA-approved oncology medicine for the first time.