Ubi Titer Issue #7
ADC Payload Innovation, Disordered-Target PROTACs, and Spatial QSP for Immunotherapy
Mechanism-matched delivery keeps winning: a toxic payload made deliverable by ADC targeting, an antibody engineered for short instead of long half-life, and induced-proximity complexes built through partner proteins instead of direct binding pockets.
- antibody-drug conjugates
- PROTAC
- molecular glue
- targeted protein degradation
- quantitative systems pharmacology
- spatial biology
- Alzheimer's immunotherapy
- checkpoint blockade
What this issue covers
- 1.An anti-PMEL antibody-drug conjugate with a Gq/11 inhibitor payload in GNAQ/GNA11-mutant melanomas: a phase 1 trial
A phase 1 trial of DYP688, an ADC that delivers a Gq/11 signaling inhibitor otherwise too toxic to dose systemically, showed 19.7% objective response and 82% disease control in GNAQ/GNA11-mutant melanomas.
- 2.Induced ubiquitination of the partially disordered estrogen receptor alpha via a 14-3-3 directed molecular glue-PROTAC
Researchers built a PROTAC that recruits a degradation enzyme to a disordered region of estrogen receptor alpha by routing through its structured partner protein 14-3-3, revealing an entirely new protein interface by cryo-EM.
- 3.Immunotherapy with a short-lived anti-PD-L1 antibody in Alzheimer's disease: a phase 1b, randomized, double-blind trial
A phase 1b trial tested an anti-PD-L1 antibody deliberately engineered for a short, roughly four-day half-life, dosed intermittently in early Alzheimer's patients, and found it safe with early signs of a neuroprotective biomarker effect.
- 4.Quantitative calibration of a spatial QSP model identifies fibroblast impact on HCC immunotherapy
A spatial computational model of liver cancer, calibrated directly against real tumor imaging data, shows that fibroblasts physically wall off tumors from immune cells and predicts which patients will respond to combination immunotherapy.
Paper 1 · Nature Medicine
An anti-PMEL antibody-drug conjugate with a Gq/11 inhibitor payload in GNAQ/GNA11-mutant melanomas: a phase 1 trial
A phase 1 trial of DYP688, an ADC that delivers a Gq/11 signaling inhibitor otherwise too toxic to dose systemically, showed 19.7% objective response and 82% disease control in GNAQ/GNA11-mutant melanomas.
Core finding
DYP688, an antibody-drug conjugate targeting the melanocyte lineage antigen PMEL and delivering the Gq/11 inhibitor SDZ475 as payload, produced objective responses in 13 of 66 patients (19.7%) with metastatic uveal melanoma or other GNAQ/GNA11-mutant melanomas, with tumor reduction in 71.2% and a median progression-free survival of 7.2 months. Grade 3 or higher treatment-related adverse events occurred in only 7.6% of patients.
What is novel
The payload is not a conventional cytotoxin but a direct signaling pathway inhibitor that causes dose-limiting toxicity when given systemically. Targeted ADC delivery makes this otherwise undeliverable mechanism clinically viable, with paired tumor biopsies confirming on-target pathway suppression after treatment.
Limitations
This is a phase 1 dose-escalation study with no maximum tolerated dose identified; uveal melanoma is a rare indication, and durability of response beyond roughly seven months has not yet been established.
Why it matters in context
Uveal melanoma has had limited systemic therapy options; the approved bispecific T-cell engager tebentafusp targets the same PMEL biology through immune redirection rather than direct pathway inhibition, and is restricted to HLA-A*02:01-positive patients. DYP688 offers a mechanistically distinct, HLA-independent alternative built on the same validated target.
Paper 2 · Nature Communications
Induced ubiquitination of the partially disordered estrogen receptor alpha via a 14-3-3 directed molecular glue-PROTAC
Researchers built a PROTAC that recruits a degradation enzyme to a disordered region of estrogen receptor alpha by routing through its structured partner protein 14-3-3, revealing an entirely new protein interface by cryo-EM.
Core finding
A molecular glue-PROTAC series chemically links a 14-3-3-directed molecular glue to a VHL-recruiting ligand, enabling induced ubiquitination of the 14-3-3/estrogen receptor alpha complex through the receptor's disordered F-domain. Cryo-EM of the most effective compound revealed a previously unseen interface between the VHL ligase complex and 14-3-3, and linker chemistry could be tuned to direct ubiquitination toward one protein or the other.
What is novel
Most targeted protein degraders require a well-defined binding pocket on the target. This approach instead uses a partner protein's surface as the recruitment platform, opening a path to degrading proteins, like many transcription factors and RNA-binding proteins, that lack any pocket of their own.
Limitations
Despite efficient ubiquitination of both proteins in biochemical assays, the compounds did not measurably reduce estrogen receptor alpha levels in breast cancer cells, likely because the underlying molecular glue engages multiple different 14-3-3 complexes throughout the cell rather than concentrating on this one target.
Why it matters in context
This work sits within a growing field of induced-proximity chemistry that extends targeted protein degradation beyond single well-folded proteins toward multi-protein complexes and intrinsically disordered regions, an active frontier as the field looks past first-generation PROTAC targets.
Paper 3 · Nature Medicine
Immunotherapy with a short-lived anti-PD-L1 antibody in Alzheimer's disease: a phase 1b, randomized, double-blind trial
A phase 1b trial tested an anti-PD-L1 antibody deliberately engineered for a short, roughly four-day half-life, dosed intermittently in early Alzheimer's patients, and found it safe with early signs of a neuroprotective biomarker effect.
Core finding
IBC-Ab002, an anti-PD-L1 antibody engineered with Fc-effector silencing and reduced neonatal Fc receptor binding for rapid clearance, was safely administered to 40 early Alzheimer's patients across five ascending dose cohorts over 48 weeks. At the highest dose, cerebrospinal fluid markers of neuronal and synaptic damage showed directional improvement, though the study was not powered to reach statistical significance.
What is novel
Rather than pursuing the field's usual goal of extending antibody half-life, this antibody was purpose-built for rapid clearance, enabling brief, pulsed immune checkpoint activation instead of sustained exposure, based on preclinical evidence that transient rather than continuous blockade drives the beneficial neuroinflammatory effect.
Limitations
The primary endpoint was safety and tolerability, not efficacy; per-cohort sample sizes were small (five to six patients at the highest dose for biomarker analysis), and no biomarker change reached statistical significance.
Why it matters in context
This represents a distinct mechanistic approach to Alzheimer's immunotherapy compared with amyloid-targeting antibodies, acting instead on peripheral immune checkpoint biology to support the brain's capacity for self-repair, a strategy with broader potential applicability across neurodegenerative conditions where neuroinflammation contributes to progression.
Paper 4 · PNAS
Quantitative calibration of a spatial QSP model identifies fibroblast impact on HCC immunotherapy
A spatial computational model of liver cancer, calibrated directly against real tumor imaging data, shows that fibroblasts physically wall off tumors from immune cells and predicts which patients will respond to combination immunotherapy.
Core finding
Researchers built a spatial quantitative systems pharmacology model of hepatocellular carcinoma and calibrated its cell-movement parameters against real spatial proteomic imaging data from 401 treatment-naive patients. The calibrated model reproduced fibroblast-driven exclusion of immune cells from tumors and predicted therapy response rates that closely matched an independent clinical trial cohort treated with combination immunotherapy.
What is novel
Prior computational tumor models were typically fit only to clinical outcome data. This framework instead calibrates directly against spatial tissue architecture using a Bayesian statistical approach, allowing the model to capture how fibroblasts physically organize around tumor cells and block immune cell access.
Limitations
Calibration and validation were performed at the population level using different patient cohorts for pre- and post-treatment comparisons rather than matched longitudinal samples from the same patients, and the fibroblast model currently represents only one of several known fibroblast subtypes.
Why it matters in context
This extends a line of spatial tumor modeling work by incorporating fibroblast biology, a cell type increasingly implicated in immunotherapy resistance, and demonstrates that spatial architecture can be used prospectively to identify biomarkers of treatment response before large clinical trials are run.
Primary papers
- [1] An anti-PMEL antibody-drug conjugate with a Gq/11 inhibitor payload in GNAQ/GNA11-mutant melanomas: a phase 1 trial (Nature Medicine)
- [2] Induced ubiquitination of the partially disordered estrogen receptor alpha via a 14-3-3 directed molecular glue-PROTAC (Nature Communications)
- [3] Immunotherapy with a short-lived anti-PD-L1 antibody in Alzheimer's disease: a phase 1b, randomized, double-blind trial (Nature Medicine)
- [4] Quantitative calibration of a spatial QSP model identifies fibroblast impact on HCC immunotherapy (PNAS)