BLUESTAR — Puxitatug samrotecan (AZD8205, B7-H4 ADC) in advanced/metastatic endometrial cancer¶
Abstract: LBA, Phase 1/2a · Session: Sunday April 12, 2026 · Simultaneous publication: TBD (likely JCO or Annals of Oncology)
Headline: First-in-human B7-H4 ADC clinical efficacy data — ORR 34.6% @ 2.0 mg/kg and 38.5% @ 2.4 mg/kg in endometrial cancer with mPFS 7.0 mo, opening a new ADC target class with read-through to breast and ovarian.
At a glance¶
- Sponsor: AstraZeneca (AZD8205)
- PI / presenter: Stephanie Gaillard, MD (Johns Hopkins)
- NCT ID: NCT05123482
- Phase / arms / N: Phase 1/2a basket; endometrial cohort dosed at 2.0 mg/kg and 2.4 mg/kg Q3W
- Primary endpoint(s): ORR (RECIST 1.1), safety, PK
- Indication / line: Advanced/metastatic endometrial cancer, post-platinum and post-IO
- Comparator: None (single-arm dose-cohort)
- Session: LBA, Day 3 (Apr 12)
Headline result¶
In the endometrial cohort, ORR 34.6% at the 2.0 mg/kg dose and 38.5% at 2.4 mg/kg, with 12-week DCR of 80.8% and 84.6% respectively. mPFS 7.0 months. Manageable safety profile consistent with topoisomerase-1 ADC class. First clinical efficacy data for the B7-H4 ADC class.
Mechanism / class¶
Puxitatug samrotecan (AZD8205) is an antibody-drug conjugate targeting B7-H4 (VTCN1) — a B7-family immune checkpoint ligand highly expressed in endometrial, breast (TNBC), and ovarian cancer with limited normal-tissue expression. Payload is a proprietary AstraZeneca topoisomerase-1 inhibitor with a cleavable linker. First-in-class B7-H4 ADC to report meaningful clinical efficacy.
Trial design deep dive¶
Study architecture: Phase 1/2a multi-tumor basket (BLUESTAR), with the endometrial cohort dosed at 2.0 mg/kg and 2.4 mg/kg Q3W as parallel dose-expansion arms. Single-arm design — no comparator at this stage. Patients required prior platinum and prior IO; B7-H4 IHC selection criteria TBD per cohort protocol (some BLUESTAR cohorts are all-comer with retrospective B7-H4 stratification, others enrich prospectively).
Endpoint definitions: ORR by RECIST 1.1 (primary, with both blinded independent central review and investigator-assessed readouts reported); safety per CTCAE v5; PK and immunogenicity as supporting endpoints. DCR at 12 weeks reported as an early activity signal. DOR and PFS as secondary; OS is exploratory at this Phase 1/2a stage.
Key subgroup analyses: Dose comparison (2.0 vs 2.4 mg/kg) is the core analysis informing the recommended Phase 2 dose — 38.5% ORR at the higher dose vs 34.6% at lower, against a tolerability gradient. B7-H4 IHC expression-by-response correlation TBD in poster appendix. Histology (endometrioid vs serous vs carcinosarcoma) and prior-line subgroups reserved for ASCO/ESMO follow-up.
Comparator context: No randomized comparator, but the 34–38% ORR sits favorably against current second-line endometrial benchmarks (~15–25% ORR for chemo, ~20–25% for dostarlimab in MMRp). The historical benchmark anchors the registration-enabling Phase 3 dose and comparator selection.
FDA implications: Pre-pivotal — establishes the B7-H4 ADC class and informs a registration-enabling Phase 3 design (TBD whether vs investigator-choice chemo or in combination). Project Optimus dose-optimization framework will apply.
Discussant takeaway¶
Per OncLive coverage: the 34.6–38.5% ORR in heavily pretreated endometrial sits comparable to current second-line standards and marks B7-H4 as a tractable ADC target. IHC-based patient selection (B7-H4+) is feasible and was used in BLUESTAR.
Practice-changing verdict¶
No (early-phase), but pivotal-design-defining. BLUESTAR establishes the B7-H4 ADC class clinically and sets the stage for a registration-enabling Phase 3 expansion. AZ presented the basket BLUESTAR design at AACR 2026 (one week after SGO) anchoring ovarian and breast cohorts.
Cross-links¶
- AACR 2026: B7-H4 ADC mechanism session (Apr 17–22, one week after SGO)
- Other SGO 2026: REJOICE-Ovarian01 (CDH6 ADC); mirvetuximab IMGN853-0420 (FRα ADC) — same ADC-class explosion
- ASCO 2026: Expected basket ovarian + breast cohort expansion
- SABCS 2026: Cross-tumor B7-H4 ADC follow-up in breast cohorts