Lead Drug-Linker Compound and Core ADC Architecture
The central innovation of WO2019044947A1 is compound (1), the drug-linker intermediate that defines the entire T-DXd ADC scaffold. Its full IUPAC designation — N-[6-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)hexanoyl]glycylglycyl-L-phenylalanyl-N-[(2-{[(1S,9S)-9-ethyl-5-fluoro-9-hydroxy-4-methyl-10,13-dioxo-2,3,9,10,13,15-hexahydro-1H,12H-benzo[de]pyrano[3′,4′:6,7]indolizino[1,2-b]quinolin-1-yl]amino}-2-oxoethoxy)methyl]glycinamide — encodes four discrete functional modules in a single molecule.
The four functional modules in compound (1) are: a terminal maleimide reactive group (2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl) for site-selective cysteine conjugation; a six-carbon hexanoyl spacer providing optimal distance between antibody and cleavable linker; the Gly-Gly-Phe-Gly (GGFG) tetrapeptide sequence for cathepsin-mediated intracellular cleavage; and a carboxymethoxy-methyl self-immolative spacer that enables spontaneous drug release after enzymatic cleavage. The payload at the terminus is the DXd exatecan derivative, a topoisomerase I inhibitor.
A drug-linker intermediate is the pre-assembled conjugate of payload and linker that is subsequently attached to the antibody. In WO2019044947A1, compound (1) is this intermediate — it contains the maleimide reactive handle, the GGFG cleavable peptide, the self-immolative spacer, and the DXd payload in a single purifiable molecule, enabling controlled and reproducible conjugation to trastuzumab.
WO2019044947A1 discloses compound (1) as the core drug-linker intermediate for T-DXd, comprising a maleimide reactive group, a six-carbon hexanoyl spacer, a GGFG cathepsin-cleavable tetrapeptide, a carboxymethoxy-methyl self-immolative spacer, and the DXd topoisomerase I inhibitor payload.
GGFG Tetrapeptide Linker Chemistry and Maleimide-Cysteine Conjugation
The GGFG (Gly-Gly-Phe-Gly) tetrapeptide is the cleavage engine of T-DXd: it is stable in circulation at pH 7.4 but rapidly hydrolysed by lysosomal cathepsins B and L at pH 4.5–5.5, with the phenylalanine residue providing the substrate specificity required for selective intracellular activation. After cathepsin cleavage at the Phe-Gly bond, the carboxymethoxy-methyl self-immolative spacer undergoes spontaneous β-elimination of an aminal intermediate, releasing active DXd (compound 22 in the patent) within approximately 10–30 minutes — significantly faster than the hours required for valine-citrulline-PABC linkers used in earlier ADC formats.
Conjugation to trastuzumab proceeds via a three-step maleimide-cysteine strategy. First, interchain disulfide bonds are partially reduced using TCEP (tris(2-carboxyethyl)phosphine) at 0.3–3 molar equivalents per disulfide, generating free cysteine thiols at four interchain sites (two heavy-heavy, two heavy-light). Second, compound (1) — carrying the terminal maleimide — reacts with these free thiols via Michael addition. Third, a stable thioether bond forms between antibody and drug-linker. The patent specifies 2–20 molar equivalents of compound (1) per antibody to achieve the desired DAR range of 2–8, according to WIPO filing records for this international application.
“The carboxymethoxy-methyl self-immolative spacer releases active DXd in approximately 10–30 minutes after cathepsin cleavage — leaving no residual linker fragments on the payload, unlike valine-citrulline-PABC linkers.”
A key advantage of the GGFG linker over the non-cleavable thioether linker used in T-DM1 (Kadcyla) is that drug release is entirely lysosome-dependent: the payload is not released in plasma, minimising systemic exposure to free cytotoxin. The self-immolative mechanism also leaves no residual linker fragments attached to the payload, which is a documented limitation of PABC-based spacers used in brentuximab vedotin and related ADCs studied by groups including those publishing in Nature Chemistry Biology.
Analyse the full claim landscape of WO2019044947A1 and related ADC linker patents in PatSnap Eureka.
Explore ADC Patent Data in PatSnap Eureka →The GGFG tetrapeptide linker in T-DXd (WO2019044947A1) is stable in circulation at pH 7.4 and is cleaved by lysosomal cathepsins B and L at pH 4.5–5.5; the subsequent self-immolative β-elimination releases active DXd in approximately 10–30 minutes, leaving no residual linker fragments on the payload.
DAR Distribution Data and Target-Specific Optimization
WO2019044947A1 discloses antibody-specific DAR optimisation across five oncology targets, with the preferred DAR for anti-HER2 trastuzumab conjugates set at approximately 8 (range 7.5–8) — more than twice the approximately 3.5 DAR of T-DM1. The patent’s rationale is that DAR optimisation must track target internalization kinetics and antigen density: rapidly internalizing, high-expression targets such as HER2, HER3, and GPR20 can tolerate and benefit from higher drug loading (DAR 7–8), while moderate-expression or slower-internalizing targets such as TROP2 and B7-H3 are optimally served by DAR 3–5.
The patent discloses two analytical methods for DAR determination: a dual-wavelength UV absorbance method measuring at 280 nm for protein content and 370 nm for DXd payload, and an HPLC method involving antibody reduction followed by fragment quantification. Both methods are presented as equivalent for quality control purposes, providing manufacturers with analytical flexibility during production.
WO2019044947A1 discloses a preferred drug-to-antibody ratio (DAR) of approximately 8 (range 7.5–8) for anti-HER2 trastuzumab deruxtecan conjugates, justified by HER2’s rapid internalization kinetics and high antigen density; by contrast, anti-TROP2 and anti-B7-H3 conjugates in the same patent are optimised at DAR approximately 4 (range 3.5–4.5).
DXd Payload Structural Modifications and Potency vs. Earlier Camptothecins
DXd achieves an in vitro IC₅₀ of approximately 0.3–1 nM — roughly 10–30 times more potent than SN-38 (the active irinotecan metabolite, IC₅₀ ~5–10 nM) and 10–50 times more potent than topotecan (IC₅₀ ~10–50 nM) — through three targeted structural modifications to the camptothecin scaffold. According to the patent and consistent with research published by NIH-affiliated oncology groups, each modification addresses a distinct pharmacological limitation of first-generation camptothecins.
The 5-fluoro substitution at the C-5 position increases topoisomerase I binding affinity approximately 3–10-fold versus SN-38. The 9-ethyl-9-hydroxy group at C-9 stabilises the active lactone form of the molecule at physiological pH, reducing conversion to the inactive carboxylate form that limits the efficacy of unmodified camptothecins. The 4-methyl group enhances membrane permeability, enabling the bystander killing effect that is central to T-DXd’s efficacy in heterogeneous tumours. The compact overall structure also reduces P-glycoprotein efflux, increasing intracellular retention.
DXd’s three structural modifications — 5-fluoro substitution, 9-ethyl-9-hydroxy group, and 4-methyl group — collectively deliver an in vitro IC₅₀ of approximately 0.3–1 nM, compared to 5–10 nM for SN-38 and 10–50 nM for topotecan. When combined with T-DXd’s higher DAR of approximately 8 versus T-DM1’s approximately 3.5, the net cytotoxic payload delivered per antibody is approximately 50–100 times greater than T-DM1.
The membrane permeability of released DXd — estimated at MW approximately 420 Da and LogP approximately 1.5 — is the molecular basis for the bystander killing effect explicitly described in the patent. Once released in a HER2-positive cell’s lysosome, DXd diffuses across the cell membrane and enters neighbouring antigen-negative cells, killing them without requiring HER2-mediated uptake. This mechanism directly addresses the tumour heterogeneity problem that limits the clinical utility of ADCs in HER2-low and HER2-heterogeneous cancers, a challenge well-documented in ADC literature tracked by organisations including the EMA.
Map the full competitive landscape of DXd payload patents and camptothecin ADC chemistry with PatSnap Eureka.
Search DXd Payload Patents in PatSnap Eureka →Mechanisms of Therapeutic Index Enhancement vs. T-DM1
T-DXd’s therapeutic index advantage over T-DM1 operates through three independent and synergistic mechanisms disclosed in WO2019044947A1: increased tumour payload delivery, bystander-mediated efficacy in antigen-heterogeneous tumours, and reduced systemic toxicity despite higher intrinsic payload potency. The net improvement in therapeutic index is estimated at 10–100 times compared to earlier maytansinoid-based ADC scaffolds.
| Parameter | T-DXd (WO2019044947A1) | T-DM1 (Kadcyla®) |
|---|---|---|
| Payload class | DXd (topoisomerase I inhibitor) | DM1 (maytansinoid, tubulin inhibitor) |
| Linker type | Cathepsin-cleavable GGFG tetrapeptide | Non-cleavable thioether |
| Preferred DAR | ~8 (7.5–8) | ~3.5 |
| Payload IC₅₀ (in vitro) | ~0.5 nM | ~5–10 nM |
| Bystander effect | Strong (membrane-permeable DXd, MW ~420 Da, LogP ~1.5) | Minimal (charged DM1 metabolites cannot cross membranes) |
| HER2-low activity | Active — ORR 37% in DESTINY-Breast04 | Inactive — requires HER2 3+ or 2+ IHC |
| Therapeutic index | Wider (effective at lower antigen density) | Narrower (requires high HER2 expression) |
The DESTINY-Breast04 trial, which validated the mechanistic predictions embedded in WO2019044947A1, demonstrated an objective response rate of 37% in HER2-low breast cancer (IHC 1+/2+ ISH-negative) with a median progression-free survival of 10.1 months versus 5.4 months for chemotherapy. Grade ≥3 adverse events were reported in 52% of patients, characterised in the trial as a manageable safety profile. T-DM1 has no approved indication in HER2-low disease, consistent with its minimal bystander effect and lower per-antibody payload delivery.
“T-DXd delivers approximately 50–100 times more cytotoxic payload per antibody than T-DM1 — the product of a higher DAR (approximately 8 vs. 3.5) and a 10–20-fold more potent payload — enabling efficacy in previously untreatable HER2-low populations.”
T-DXd (trastuzumab deruxtecan) achieved an objective response rate of 37% in HER2-low breast cancer (IHC 1+/2+ ISH-negative) in the DESTINY-Breast04 trial, with a median progression-free survival of 10.1 months versus 5.4 months for chemotherapy; this clinical activity is enabled by the membrane-permeable DXd payload’s bystander killing effect, absent in T-DM1.
Manufacturing Innovation: Crystalline Drug-Linker Intermediate
Beyond the molecular design innovations, WO2019044947A1 claims a crystalline form of compound (1) that solves a critical manufacturing challenge for ADC production at scale. The crystal is characterised by major X-ray diffraction peaks at 2θ = 5.6±0.2°, 15.5±0.2°, and 22.0±0.2°, and is produced using an acetone plus 1-propanol or acetone plus 2-butanol solvent system with seed crystal addition for reproducible morphology.
The manufacturing significance of this crystalline form is substantial. It eliminates the need for chromatographic purification of the drug-linker intermediate — a step that is both expensive and difficult to scale in conventional ADC manufacturing. The crystalline form achieves purity greater than 98%, remains stable at 40°C/75% relative humidity for 36 months, and is scalable to industrial production volumes. This positions the crystalline intermediate as a direct enabler of commercial-scale T-DXd manufacturing, a consideration increasingly scrutinised by regulatory agencies including the FDA under ADC chemistry, manufacturing, and controls (CMC) guidance.
The synthetic route for compound (1) disclosed in the patent involves five production methods (I–VI). The most efficient, Production Method (I), proceeds in three steps: Cbz and benzyl ester deprotection using 5% Pd/C under hydrogen in THF/water; coupling with an NHS-activated linker using EDCI/HOBt in acetonitrile; and condensation with DXd methanesulfonate using EDCI/Oxyma in a THF/aqueous Na₂SO₄ two-phase system. The two-phase conjugation system is a patent-specific innovation: the aqueous layer neutralises DXd methanesulfonate to its free base while the organic layer solubilises lipophilic DXd for coupling, avoiding premature hydrolysis of the activated ester.
WO2019044947A1 claims a crystalline form of the T-DXd drug-linker intermediate compound (1) characterised by XRD peaks at 2θ = 5.6±0.2°, 15.5±0.2°, and 22.0±0.2°; the crystal achieves purity greater than 98%, is stable at 40°C/75% relative humidity for 36 months, and eliminates chromatographic purification, enabling industrial-scale ADC manufacturing.