SG3199 (pyrrolobenzodiazepine [PBD] dimer; DNA interstrand cross-linker)
Mechanistic subtype
DNA-crosslinker-PBD
Stereochem / salt
(11aS,11a'S)-configured PBD dimer (single defined stereoisomer); free base (no salt)
Bystander
Partial
PAMPA rank
-
MW
584.7 Da
XLogP3
2.8
logD₇.₄
2.5
TPSA
102 Ų
pKa
5 pKa
Charge pH 7.4
0
H-bond donors
0
H-bond acceptors
8
IC50 (HCEC)
-
Formula
C33H36N4O6
PubChem CID
90132565
ADCdb payload
PAY0LZRXU
Hydrophobicity · logD₇.₄
hydrophilic −2+2.5+4 lipophilic
Bioactivity note
ADCdb (DRG0FUFBP) reports in vitro GI50: 0.26 pM (EoL-1), 4.96 pM (SU-DHL-1), 17.07 pM (Karpas-299); in vivo up to 98.8% TGI at 0.6 mg/kg; clinical 48.6% CR in classical Hodgkin lymphoma at 45 ug/kg. Payload SG3199 is a highly potent PBD-dimer DNA interstrand crosslinker (the imi
Anti-CD25 (IL2RA) PBD-dimer ADC from ADC Therapeutics (HuMax-TAC human IgG1, stochastic interchain-Cys conjugation, DAR 2.3; ADCdb lists 2.25). Uses the SAME tesirine linker (Mal-PEG8-Val-Ala-PABC, ADCdb LIN0YRUBS) and SG3199 PBD-dimer payload as loncastuximab tesirine, so linker physchem and chemistry-input fields were carried from that verified row. MECHANISM is Treg-depleting / CD25-targeted.
OCULAR (priority, why it qualifies): MINIMAL signal. In the pivotal Ph2 cHL study (ADCT-301-201 / NCT04052997, N=117 at RP2D 45 ug/kg Q3W x2 then 30 ug/kg), the ONLY eye-disorder AE was a single grade 1 blepharitis = 1/117 (0.9%); no corneal/conjunctival/dry-eye/blurred-vision/keratitis events and no grade >=3 ocular. Consistent with the PBD class (cf. loncastuximab/vadastuximab = ~0% OAE). ocular_surface_subtype set to 'Unknown' because blepharitis is an eyelid/adnexal event that does not map to the corneal/conjunctival/visual surface categories; the four specific surface-PT columns left blank (not separately reported) while oae_any=0.9 and oae_g3plus=0 are documented.
DOMINANT toxicities are NOT ocular: skin/cutaneous 78.6% (G3+ 22.2%; maculopapular rash 32.5%, G3+ 6.8%) and immune-related GBS/polyradiculopathy 8/117 (6.8%; 5 grade >=3) — the latter is the AESI that led the FDA to require a much larger safety trial and the program to be paused/discontinued. NOTE: skin is not an organ_system in the toxicity_rows enum, so the dominant skin toxicity is recorded here rather than as a row. Other: fatigue 38.5%, hepatic enzyme rises (GGT 17.1%; 2 grade-3 DILI), cytopenias (lymphopenia G3+ 10.3% highest), nausea 27.4%, pneumonitis 3.4% + 1 grade-3 ILD, IRR 4.3% (all G1-2). 28.2% discontinued for TEAEs; ORR 70.1%, CR 33.3%.
CTCAE version not stated in the Ph2 report (AEs coded with MedDRA v22.0), so ae_grading_scale left blank. Payload_physchem are standard shared SG3199 values. Clinical tier B (peer-reviewed primary); composition tier B (peer-reviewed + ADCdb); chemistry/payload tier C (derived from shared tesirine linker/SG3199 class).