A packed day: 11 science picks span molecular glue/PROTAC degraders, hit-finding rigor, and a novel conjugate-release chemistry; 9 business stories include Bayer and AstraZeneca trial wins, a double Novartis Phase III setback, and fresh financing and licensing deals.
2026-09-09 · ⚗️Hypervirulent Klebsiella pneumoniae / siderophore-antibiotic conjugate · Organic Letters (ACS)
- Chemists conjugated the beta-lactam antibiotic aztreonam to aerobactin, a bacterial iron-scavenging siderophore, exploiting bacteria's own iron-uptake machinery as a 'Trojan horse' to smuggle the drug inside cells.
- A modular synthetic route enabled controlled mono- and di-functionalization of aerobactin, with late-stage click-chemistry (CuAAC) attachment of payloads, providing a flexible platform applicable beyond this single conjugate.
- The lead monovalent conjugate was markedly more potent than aztreonam alone against a hypervirulent, clinically concerning Klebsiella pneumoniae strain, a substantial improvement over the parent antibiotic.
- Competition and iron-dependent activity studies confirmed delivery proceeds through the aerobactin iron-acquisition pathway, supporting genuine siderophore-mediated uptake rather than incidental potency gains.
- The results establish aerobactin as a viable scaffold for delivering antibacterial payloads against multidrug-resistant, hypervirulent Enterobacteriaceae.
💡 PointsCall this the cefiderocol playbook done cleaner. Aerobactin's a natural Fe(III) chelator with dedicated Klebsiella uptake receptors, so hijacking it to smuggle aztreonam past the outer membrane is exactly the kind of Trojan-horse logic that's already proven clinically viable — nice to see it extended to a hypervirulent strain, which is a genuinely scary unmet-need bug. What I like more, honestly, is the platform: modular mono-/di-functionalization plus late-stage CuAAC means this isn't a one-off heroic synthesis, it's a conjugation chassis I could point at other payloads. And they didn't just show a potency bump and wave hands — the competition/iron-dependence data actually nail down siderophore-pathway uptake, not some off-target membrane-permeabilization artifact. That's the control every conjugate paper should run but often skips.
Modular Synthesis of Aerobactin–Aztreonam Conjugates with Enhanced Activity against Hypervirulent Klebsiella pneumoniae ↗
2026-09-09 · ⚗️Siglec-7 / bicyclic peptide inhibitor (DEL-adjacent hit-finding) · Journal of the American Chemical Society (ACS)
- Researchers used bicyclic genetically encoded peptide libraries (BiGEL) — phage-displayed peptides chemically stapled into two rings — to search for inhibitors of Siglec-7, a glycan-binding immune receptor considered notoriously hard to target with small molecules.
- Sequencing-based panning against Siglec-7 nominated hundreds of candidate peptides; a subset were confirmed as genuine binders by surface plasmon resonance and as functional disruptors of the Siglec-7:GD3 interaction by competitive ELISA, with machine learning models identifying further hits of similar potency.
- Structure-activity work showed the bicyclic ring architecture itself is essential for potency, and NMR studies located one optimized lead's binding site away from the canonical glycan-binding domain, pointing to a possible allosteric mechanism.
- Cell-based multivalent binding assays confirmed engagement with Siglec-7 but also revealed considerable nonspecific binding to receptor-negative cells, highlighting a selectivity challenge still to be resolved.
💡 PointsThe allosteric site is the real hook here — a bicyclic peptide binding away from the canonical glycan pocket on a target everyone assumed only the carbohydrate-recognition domain was druggable gives you a completely different SAR vector than trying to out-compete sialic acid for its own pocket. That's genuinely exciting for Siglec-7. But I'd want the nonspecific binding in the cell assays resolved before getting attached to any single lead — multivalent avidity effects can paper over a mediocre core affinity, and that's exactly the kind of artifact that kills a program downstream. Bicyclics themselves are the right chemotype to bet on: synthesizable, chemically stable enough for iterative SAR, and unlike DEL small molecules they're not fighting the flat, shallow topology that's stalled every prior GBP inhibitor effort.
Noncarbohydrate Inhibitors of Sialic Acid-Binding Immunomodulatory-Type Lectin-7 (Siglec-7) Discovered from Genetically Encoded Bicyclic Peptide Libraries ↗
2026-09-09 · ⚗️ALK / molecular glue degrader · Nature (Springer Nature)
- TRI-611 is an orally dosed molecular glue degrader designed to eliminate ALK fusion proteins that drive a subset of non-small cell lung cancers.
- Rather than blocking ALK's active site like conventional inhibitors, it recruits the E3 ligase adaptor cereblon to a distinct surface on ALK, forming a complex that tags the protein for degradation.
- The molecule is reported to be highly selective, sparing other cereblon-dependent targets and unrelated kinases, and it degrades both wild-type and kinase-inhibitor-resistant ALK fusion variants.
- A central finding is brain penetrance: TRI-611 produced tumor regressions in intracranial as well as subcutaneous models, addressing CNS metastases that limit many existing ALK-targeted therapies.
- Degradation was rapid and long-lasting, and combining TRI-611 with standard ALK inhibitors produced more durable responses in preclinical models.
💡 PointsTRI-611 recruiting cereblon to a distinct, non-canonical surface on ALK — rather than working off the usual IMiD-derived degron — is the real chemistry story here; if that ternary interface holds up, it's a fresh template for kinases that don't glue easily.
Wiping out kinase-inhibitor-resistant fusion variants is the killer selling point: degradation doesn't care about the affinity losses that gatekeeper/compound mutations inflict on ATP-competitive binders, so a mutant that's dead to a TKI can still form a productive ternary complex.
Getting CNS exposure out of a CRBN glue is the harder chemical win — these scaffolds usually sit well outside Ro5, so brain penetrance without sacrificing ternary complex cooperativity argues for real optimization, not luck.
Selectivity over other CRBN neosubstrates plus the TKI-combo data suggest this could slot in alongside inhibitors rather than just after them.
TRI-611, a selective, brain-penetrant molecular glue degrader of ALK ↗
2026-09-08 · ⚗️EP4 receptor (PTGER4) / oral small-molecule antagonist · Journal of Medicinal Chemistry (ACS)J. Med. Chem. 2026, DOI: 10.1021/acs.jmedchem.6c00863
- The EP4 receptor (EP4R), a prostaglandin E2 receptor subtype, is a promising immuno-oncology target because its signaling dampens antitumor immunity and drives tumor-associated inflammation.
- A scaffold hopping strategy moved medicinal chemists away from earlier EP4R antagonist chemotypes to a novel chemical series with improved potency and pharmacokinetics.
- Structure-activity relationship (SAR) studies then guided lead optimization, balancing target potency, selectivity, and drug-like properties.
- This process yielded DT-9081, an orally bioavailable small-molecule EP4R antagonist combining potent receptor antagonism, favorable ADME properties, and a clean preclinical safety profile.
- In vitro and in vivo pharmacology data support DT-9081's ability to modulate tumor immunity, and the compound has already completed a Phase I monotherapy trial in patients with solid cancers.
💡 PointsFinally, the paper behind the registry entry. Scaffold hopping off the earlier EP4 chemotypes almost certainly means they were fighting a liability — hERG, metabolic soft spot, take your pick — in the first series; jumping cores is usually the tell that a fix, not a patch, was needed. What's striking is the pace: hit-to-Phase-I told in one publication reads as a genuinely tractable SAR, not a rescue job. The harder sell is mechanism — EP4 antagonism doesn't kill tumor cells directly, it pulls the brake off PGE2-driven immune suppression, so efficacy will live or die on combination data, not monotherapy shrinkage. Worth tracking how DT-9081's selectivity stacks up against the other EP4 antagonists already in the clinic.
Discovery of DT-9081, a Potent and Orally Bioavailable Small Molecule Antagonist of the EP4 Receptor for the Treatment of Solid Cancers ↗
2026-09-08 · ⚗️PARP1 gene promoter G-quadruplex / small-molecule transcriptional silencer · Journal of Medicinal Chemistry (ACS)
- Rather than inhibiting or trapping the PARP1 enzyme like approved PARP inhibitors (PARPis), researchers targeted the PARP1 gene promoter itself to shut down PARP1 production at the transcriptional level, aiming to avoid the hematologic toxicity and resistance mechanisms associated with PARPi-mediated PARP1 trapping.
- The promoter region forms a non-canonical (3+1) hybrid G-quadruplex structure distinguished by an adenine bulge, a distinctive DNA fold not previously exploited as a drug target.
- Combining computational modeling with biophysical characterization, the team identified a small molecule that selectively engages this unusual G-quadruplex architecture.
- In multiple myeloma cells, the compound reduced PARP1 expression, disrupted downstream DNA repair signaling, and triggered DNA damage-associated apoptosis.
- Healthy blood cells were largely spared, pointing to a possible therapeutic window distinct from conventional PARP inhibitor toxicity profiles.
💡 PointsThe part that gets me: attacking PARP1 at the promoter instead of the active site. Every PARPi on the market — olaparib, niraparib, talazoparib — works by trapping or inhibiting the protein; the hematologic toxicity and trapping-driven resistance are baked into that mechanism. Silencing transcription via a G-quadruplex sidesteps the whole liability profile, at least on paper. That said, G4-targeting ligands are a rough class to work in — selectivity across the thousands of quadruplex-forming sequences in the genome is the perennial failure mode, not potency. And this (3+1) hybrid fold with an adenine bulge is a genuinely novel pocket, which cuts both ways: interesting SAR territory, but no playbook to lean on. Still early — a validated hit in MM cells, not a lead, let alone a candidate. Worth watching, not betting on yet.
Hit It before It Starts: Targeting a (3 + 1) Hybrid Non-canonical G-Quadruplex to Silence PARP1 in Cancer ↗
2026-09-08 · ⚗️GRP78 (BiP/HSPA5) / small-molecule inhibitor · Journal of Medicinal Chemistry (ACS)
- GRP78 (BiP/HSPA5) is a chaperone in the highly conserved HSP70 family, making isoform-selective inhibitors difficult to design.
- Using a Direct-to-Biology (D2B) strategy — synthesizing and screening compounds without intermediate purification — researchers identified compound 12 as a selective GRP78 inhibitor.
- Compound 12 binds the peptide-binding pocket in GRP78's substrate-binding domain and is selective for GRP78 over other tested HSP70 isoforms.
- It potently and selectively kills A549 lung cancer cells while showing minimal activity in other tested lung cancer cell lines.
- Cellular thermal shift assay (CETSA) confirmed cellular target engagement, and compound 12 modulates GRP78-linked unfolded protein response (UPR) stress signaling.
- The authors present compound 12 as a chemical probe for studying GRP78 biology and a starting point for GRP78-targeted therapeutics.
💡 PointsSubstrate-binding domain, not the ATPase pocket — that's the harder but more selective play, since most HSP70-family programs chase the nucleotide site and inherit its cross-reactivity problem. Getting selectivity for GRP78 over other HSP70 isoforms out of a peptide-binding groove is a real medicinal chemistry win; those pockets are shallow and promiscuous by design. What I'd want to see next: the SAR trail from D2B — crude-plate hits jumping straight to CETSA-confirmed engagement is fast, but I'd want a clean resynthesis and dose-response before trusting potency off unpurified reactions. Still, as a probe-grade starting point with A549-selective killing, compound 12 is a legitimate template for follow-on optimization, not another ATPase-site me-too.
Development of Isoform-Selective GRP78 Inhibitors from a Direct-to-Biology Strategy ↗
2026-09-08 · ⚗️HDAC8 / PROTAC (compact 2-in-1 design) · Journal of Medicinal Chemistry (ACS)
- Researchers describe a "2-in-1" PROTAC design that merges the CRBN-recruiting ligand into the capping group of an HDAC8 warhead, rather than linking two separate ligands.
- This lets one structural motif engage both HDAC8 and the CRBN E3 ligase, unlike conventional heterobifunctional PROTACs, whose separate POI- and E3-binding units plus linker often yield large, poorly drug-like molecules.
- Relative to the team's earlier HDAC8 degrader YX862, the new compact degraders show improved selectivity, lower molecular weight, and better overall drug-like properties.
- The compounds also achieved effective HDAC8 degradation in vivo, indicating therapeutic potential beyond cell-based activity.
- The study proposes a broader design framework for building smaller, more efficient PROTACs by integrating target engagement and E3 ligase recruitment into a single motif.
💡 PointsThe move that catches my eye: folding the CRBN ligand into the HDAC8 warhead's capping group instead of bolting on a full separate E3 unit plus linker. That's a direct attack on the MW/logP bloat that tanks oral exposure in most heterobifunctional degraders — not incremental linker trimming, a structural collapse of the whole ternary-complex scaffold. Going from YX862 to these analogs with better selectivity AND lower MW (rarely a package deal) says the capping vector is actively shaping the HDAC8–CRBN interface, not just shedding atoms. In vivo degradation data is the sanity check I always want before trusting a compact-PROTAC claim. I'd want SAR around that fused cap next — how much substituent tolerance before CRBN engagement drops off. Feels like a real template for smaller, more drug-like degraders, not just this target.
A 2-in-1 Strategy for the Discovery of Compact HDAC8 Protac Degraders ↗
2026-09-05 · ⚗️Mutant p53 (R280K/R273H) / covalent small-molecule stabilizer · ChemMedChem (Wiley)ChemMedChem 2026, 21 (17)
- p53 is the tumor suppressor most frequently inactivated in cancer, and hotspot mutations such as R280K and R273H destabilize the folded DNA-binding domain, making mutant p53 reactivation a long-standing 'undruggable' target class.
- The study describes RVJB59, a small molecule from a tryptophanol-derived oxazoloisoindolinone series, designed to covalently engage mutant p53 protein rather than bind it only reversibly.
- Biophysical thermal-stability assays reported for this compound series show increased melting stability of mutant p53 DNA-binding domains, consistent with partial restoration of a more folded, wild-type-like conformation.
- Combining covalent engagement with thermostabilization supports a mutant-reactivation strategy aimed at restoring p53 tumor-suppressor function in cancers driven by the R280K and R273H mutations.
💡 PointsMutant p53 reactivation is the target class every oncology med chemist wants a win on — most p53 lesions are destabilizing loss-of-function, not a clean druggable pocket, so 'fix the fold' beats 'block the site.' RVJB59's angle — covalent engagement plus measurable thermal stabilization of R280K and R273H DBDs — is mechanistically the right read: lock a reactive residue, nudge the folding equilibrium back toward a wild-type-like conformation. My first questions are the covalent-chemist ones: which residue, what warhead, how clean is selectivity across the proteome — thermal-shift data alone won't settle that. Worth benchmarking against PC14586/rezatapopt's reversible pocket-binding approach for R175H; different mutant, different chemotype, same underlying bet. Reads as solid hit-to-lead chemical biology, not anything near clinic-ready.
Covalent Targeting and Thermostabilization of Oncogenic R280K and R273H Mutants p53 by Small Molecule RVJB59 ↗
2026-08-31 · ⚗️Targeted drug-conjugate chemistry (binding-triggered payload release beyond ADCs) · Nature
- Researchers introduce a binding-to-release (BTR) strategy that triggers drug-conjugate payload release directly upon target binding, rather than requiring receptor-mediated endocytosis.
- The approach uses a new bioconjugation reaction, phosphorus(V)-phenol exchange (PhoPEx) — inspired by SuFEx click chemistry — engineered so that chemistry within the binding pocket itself cleaves the drug free once the conjugate engages its target.
- Because only a minority of drug targets support efficient endocytosis, this binding-triggered release could extend conjugate-based therapies to otherwise poorly internalizing targets.
- The authors describe proof-of-concept work with a conjugate targeting fibroblast activation protein (FAP), evaluated from in vitro settings through patient-derived tissue specimens.
💡 PointsThe clever bit is topological: you need a nucleophilic residue (Ser/Tyr/Lys) sitting close enough to the P(V) center once the conjugate is bound, so this only works when pocket geometry cooperates — not a universal fix, more a case-by-case design problem per target. SuFEx proved robust and bioorthogonal enough to become a real click-chemistry platform; whether PhoPEx's release kinetics and selectivity hold up outside FAP is the real open question — proof-of-concept in patient tissue is promising but still one target. What I like: it decouples release from endocytosis and lysosomal processing entirely, which is the actual bottleneck for most ADC-style approaches against non-internalizing or slowly-trafficking surface receptors (GPCRs, some ion channels, stromal markers like FAP itself). If the chemistry generalizes, it's a genuine expansion of addressable target space for conjugates, not just a linker tweak.
A binding-to-release strategy for targeted anticancer drug delivery ↗
2026-09-10 · 💻Structure-based hit-finding / virtual screening methodology (metadynamics) · Journal of Chemical Information and Modeling (ACS)
- Virtual screening often yields false positives from fast, approximate scoring functions, while accurate free-energy calculations to filter them remain too costly for routine high-throughput use.
- The authors introduce CTMD (c(t)-based metadynamics), which ranks binding stability using the nonequilibrium reversible-work estimator c(t) from short, independent metadynamics runs, without requiring converged binding free energies.
- Across diverse targets and chemotypes, CTMD shows robust early enrichment while staying fast, transferable, and largely tuning-free.
- Benchmarking AI co-folding, the authors find Boltz-2's enrichment tracks similarity to its training data and persists even after active-site modifications, pointing to memorization rather than true binding physics.
- CTMD is offered as an open-source middle ground between quick docking/co-folding and costly free-energy methods, cutting wasted discovery-campaign resources.
💡 PointsA triage tier between docking and full FEP is exactly what's been missing from my workflow. Standard docking scores are too noisy to prioritize synthesis, and running FEP on every virtual-screening hit is a non-starter for budget and turnaround — CTMD's short, independent metadynamics runs with the c(t) estimator could slot right into that gap and actually change which compounds we send to synthesis, without a cluster's worth of compute. Open-source and largely tuning-free means it's adoptable without a specialist babysitting it.
The Boltz-2 finding is the one I'd flag to my team before anyone leans on AI co-folding for real decisions: enrichment tracking training-set similarity even after active-site mutation means it may be pattern-matching known chemotypes, not scoring genuine binding physics. That's a trap for exactly the novel scaffolds we're chasing.
Early-Enrichment Hit Discovery via Reversible-Work c(t) Estimation in Metadynamics (CTMD) ↗
2026-09-09 · 💻Structure-based virtual screening / hit-finding methodology (methods paper, not target-specific) · chemRxiv
- The preprint asks a foundational methods question: given the size of a typical structure-based virtual screening benchmark, how large a difference in AUROC or enrichment between two protocols can actually be distinguished from statistical noise.
- Rather than assuming benchmark comparisons are inherently reliable, the authors first directly measure these resolution limits.
- They then pre-registered six candidate interventions to docking/scoring protocols and tested each against the measured resolution limits, rather than reporting post-hoc 'wins.'
- The pre-registration design is meant to guard against selective reporting, a known issue where small, benchmark-specific gains get presented as general improvements.
- This is a methodology/rigor study, not a new-inhibitor or target-specific paper; it addresses how virtual screening improvements should be evaluated.
- Practical implication: many published claims of improved virtual screening performance may fall within the noise floor of typical benchmarks, so hit-finding teams should treat single-benchmark improvement claims cautiously.
💡 PointsEvery VS vendor pitch should have to clear this paper's bar. If a 'protocol improvement' can't beat the measured noise floor of AUROC/enrichment on a benchmark our size, it's not an improvement — it's a coin flip dressed up in a webinar slide. Six pre-registered interventions is a high bar; realistically most of us won't pre-register anything before a campaign, but I can at least ask collaborators to report effect sizes against a resolution limit, not just a headline AUROC bump. Sharpest implication for me: the co-folding hype (Boltz-2 and friends) needs the same treatment — if those models are just pattern-matching PDB-adjacent chemotypes, their 'wins' over classic docking/FEP on public benchmarks could be memorization, not generalization. I'm not swapping FEP for a black box on this evidence alone.
What actually moves virtual screening performance: two measured resolution limits and six pre-registered interventions ↗
2026-09-10 · 🏢pipeline progress · Endpoints News · TwoStep Therapeutics
- TwoStep Therapeutics, a 2024 Stanford spinout, announced an oversubscribed $62.5M Series A financing on September 10, 2026, alongside FDA clearance of the investigational new drug (IND) application for its lead candidate, TS-104.
- The round was led by Insight Partners and Medical Excellence Capital, with pharma-linked corporate venture arms M Ventures (Merck KGaA) and Pfizer Ventures co-leading; existing investors NFX, 2048 Ventures, and Stanford University also participated, bringing total funding raised to $71.2M.
- TS-104 is a first-in-class multi-specific peptide-drug conjugate (PDC) that pairs a polyspecific integrin-binding peptide, engineered to bind five tumor-associated integrins across solid tumors, with the cytotoxic payload MMAE, also used in several approved antibody-drug conjugates.
- The company positions this peptide-based platform as an alternative to antibody-based ADCs for reaching solid tumor targets.
- The IND clearance opens the way for a first-in-human Phase 1 trial, with patient enrollment expected later this year.
💡 PointsPDCs are pitching themselves as the cheaper, more tumor-penetrant cousin of ADCs — same MMAE payload, smaller delivery vehicle, potentially better access to solid-tumor stroma where antibodies struggle. That's a direct pitch at the ADC franchise now dominating oncology dealmaking.
The investor list matters more than the dollar figure: M Ventures and Pfizer Ventures co-leading, not just writing a check, reads as two ADC-heavy pharmas hedging their own franchises and buying an early look at a modality that could either complement or cannibalize their antibody platforms. Oversubscribed Series A plus IND clearance inside two years of founding is a fast validation cycle for a still-unproven conjugate class — worth watching whether Phase 1 tolerability holds up before any partnering thesis gets real.
Exclusive: Cancer biotech TwoStep gets $62.5M to take twist on ADCs into clinic ↗
2026-09-09 · 🏢pipeline/regulatory progress · FiercePharma · Bayer
- Hyrnuo (sevabertinib) is Bayer's oral, HER2-selective tyrosine kinase inhibitor (TKI).
- The FDA granted accelerated approval expanding Hyrnuo to first-line use in adults with HER2 (ERBB2)-mutated, locally advanced or metastatic non-squamous NSCLC.
- The nod is based on Phase I/II SOHO-01 data showing a 75% objective response rate in treatment-naive patients, with responses lasting over six months in 73% of cases.
- It builds on Hyrnuo's November 2025 accelerated approval for previously treated patients with the same HER2-mutated tumor type.
- Continued approval depends on the ongoing Phase III SOHO-02 trial versus standard of care in treatment-naive patients.
- The move deepens Hyrnuo's rivalry with Boehringer Ingelheim's Hernexeos, which won its own first-line nod in February 2026, as the two oral TKIs vie for the small HER2-mutated NSCLC market.
💡 PointsBayer is catching up, not leading. Boehringer's Hernexeos already claimed first-line HER2-mutated NSCLC in February 2026 — Hyrnuo's approval today closes that gap rather than opening a new front. Worth noting: this is accelerated approval built on single-arm Phase I/II SOHO-01 data (75% ORR, treatment-naive), with confirmatory Phase III SOHO-02 still reading out. Continued approval — and Bayer's ability to actually compete for first-line share — hinges on that trial beating standard of care head-to-head, not just replicating early response rates. Commercially, first-line access matters more than second-line ever will in a niche indication like this, since most patients get one shot at frontline therapy. If SOHO-02 disappoints, Bayer loses the label it just won; if it reads out clean, expect a genuine two-TKI market with prescribers picking on tolerability and durability data rather than brand.
FDA clears Bayer’s Hyrnuo in first-line lung cancer, amping up Hernexeos rivalry ↗
2026-09-09 · 🏢partnership · AnswersNews · Elix / Axcelead Drug Discovery Partners
- Elix, an AI drug-discovery company, and Axcelead Drug Discovery Partners (Axcelead DDP), a Takeda-origin drug-discovery CRO, signed a joint research agreement on targeted protein degraders (TPDs), announced September 9, 2026.
- The collaboration will build a new AI drug-discovery model by combining Axcelead DDP's structural and proprietary experimental/evaluation data with Elix's AI-based molecular design and prediction technology.
- The resulting model will be applied to Axcelead DDP's TPD drug-discovery programs, aiming to improve efficiency and success rates versus conventional approaches.
- The partners say the approach could open up previously hard-to-drug targets by pairing Elix's AI platform with Axcelead DDP's drug-discovery experience and experimental infrastructure.
- The two companies also plan to develop new drug-discovery support solutions based on knowledge and technology gained through this joint research.
- No financial or milestone terms were disclosed in available reporting.
💡 PointsLook at what each side actually brings before dismissing this as another AI-plus-CRO logo swap: Axcelead supplies Takeda-pedigree structural and proprietary experimental/assay data on TPD targets, Elix supplies the AI design/prediction layer, and each fills a gap the other genuinely lacks. Positioning-wise, Elix is staking a claim specifically in TPD, a messier modality than standard small-molecule design where good training data is scarce, which is a smarter wedge than competing head-on with the bigger AI-discovery names on general hit-finding. For Axcelead, it's a way to differentiate against larger domestic CROs without building AI in-house. The phrase to watch is 'new drug-discovery support solutions' — that reads as platform/licensing ambition beyond servicing Axcelead's own pipeline. No financial terms disclosed, so there's no way yet to gauge how committed either side really is.
ElixとAxcelead DDP、タンパク質分解誘導薬の共同研究／帝人、信州大と細胞・遺伝子治療製品製造技術の共同研究 など｜製薬業界きょうのニュースまとめ読み（2026年9月9日） ↗
2026-09-09 · 🏢pipeline progress / corporate spinout · STAT News · Kura Oncology / Caspian Therapeutics / Eli Lilly
- Kura Oncology spun off Caspian Therapeutics, a separately financed company pursuing menin inhibition for diabetes; Kura retains about 50% ownership, and CEO Troy Wilson chairs Caspian's board.
- Caspian launched with $50 million in financing led by BVF Partners, joined by Eli Lilly, T1D Fund, Invus, Montanova and Kura itself.
- Its lead candidate, KO-7246, targets the same menin-KMT2A interaction as Kura's FDA-approved leukemia drug ziftomenib, but is optimized for chronic metabolic disease rather than AML.
- Menin represses genes that limit pancreatic beta-cell growth; blocking it expanded beta-cell mass and boosted insulin production in preclinical type 1 and type 2 diabetes models.
- Caspian plans to file an IND and begin early trials assessing KO-7246's effects on beta-cell function and insulin output.
💡 PointsThe spinout structure is the real story here: Kura offloads capital-intensive diabetes development onto a separately financed entity while keeping ~50% upside — a capital-efficient way to pursue a second indication without diluting the oncology balance sheet. Lilly's participation is the tell: not just funding, but a diabetes-franchise incumbent buying an option on a genuinely novel beta-cell-mass-expansion mechanism, distinct from GLP-1s and existing insulin-secretagogue classes. That novelty cuts both ways, though — menin inhibition is proliferative by design, and driving beta-cell expansion in pancreatic tissue carries an obvious tumorigenicity/hyperplasia overhang preclinical models won't fully resolve; watch how the IND-enabling tox package addresses it. Broader signal: oncology target biology migrating into metabolic disease is a repeatable playbook other biotechs with approved oncology assets may start mining.
Kura cancer drug discovery spurs launch of a diabetes biotech ↗
2026-09-09 · 🏢partnership · Fierce Biotech · Schrödinger / Tectora Therapeutics
- On September 9, 2026, Schrödinger announced a licensing and collaboration agreement with Tectora Therapeutics, a new immunology/inflammation biotech it co-founded with NEA and RA Capital Management.
- Schrödinger licensed two early-stage small-molecule programs, SDGR-4594 and SDGR-8139, to Tectora for an equity stake plus potential future milestone payments and royalties.
- Tectora concurrently closed a $55 million Series A financing from NEA and RA Capital to fund preclinical work on the programs.
- Tectora will use Schrödinger's computational drug discovery platform to advance the licensed assets toward clinical candidates.
- The deal lets Schrödinger monetize and advance internal pipeline assets via an external company while retaining equity upside.
💡 PointsSchrödinger is monetizing its own discovery engine here, not just licensing it to others — a meaningful shift for a company whose story has long been 'platform, not pipeline.' Spinning SDGR-4594 and SDGR-8139 out to a VC-backed shell instead of advancing them in-house is capital-efficiency logic: Schrödinger keeps equity and royalty upside without funding preclinical burn off its own balance sheet, while NEA and RA Capital absorb the binary risk. The equity-plus-milestones structure, instead of a straight license fee, signals that early discovery-stage assets are hard to price for cash right now — everyone's betting on the come. Watch for more computational-discovery shops (Recursion, Insitro, Isomorphic) following this playbook: spin out rather than sell or self-fund, and let Series A investors underwrite the translational risk.
Schrödinger births new immunology and inflammation biotech, Tectora Therapeutics ↗
2026-09-08 · 🏢trial result · Fierce Biotech · AstraZeneca
- AstraZeneca reported full results from two Phase III trials, OBERON and TITANIA, testing tozorakimab, an anti-IL-33 monoclonal antibody, added to standard inhaled therapy for chronic obstructive pulmonary disease (COPD).
- Across both trials, tozorakimab substantially cut moderate-to-severe exacerbation rates versus placebo in former smokers, with meaningful reductions also seen in a broader population including current smokers.
- The antibody is engineered to engage two separate IL-33 signaling pathways at once, a mechanism AstraZeneca positions as differentiating it from earlier single-pathway IL-33 drugs.
- This marks tozorakimab's third positive Phase III readout; results were presented at the European Respiratory Society Congress in Barcelona and published simultaneously in the New England Journal of Medicine.
- AstraZeneca is targeting US regulatory approval, expected in early 2027.
💡 PointsThree-for-three in Phase III is the real headline here — a track record like that de-risks tozorakimab in a way single-pathway IL-33 rivals (think itepekimab, astegolimab) can't yet match, strengthening AstraZeneca's hand with regulators and payers alike. The dual-pathway engineering pitch is the commercial wedge: as COPD biologics get crowded, "we hit both arms of IL-33 signaling" is a differentiation story sales teams can actually use. Strategically, this is AstraZeneca building a respiratory biologics franchise in a market that's long been underserved beyond inhalers — COPD's patient pool dwarfs severe asthma's. And running NEJM publication simultaneously with the ERS presentation isn't routine; it's a credibility move, front-loading peer-reviewed validation well ahead of the 2027 filing to pre-empt scrutiny.
AstraZeneca's faith in IL-33 inhibitor continues to pay off with latest phase 3 COPD win ↗
2026-09-08 · 🏢trial result (failure) · Endpoints News · Novartis / Avidity Biosciences
- Del-desiran (delpacibart etedesiran), the antibody-oligonucleotide conjugate (AOC) Novartis picked up through its roughly $12B acquisition of Avidity Biosciences, failed to separate from placebo on the primary endpoint of the Phase 3 HARBOR trial.
- The 159-patient trial tested del-desiran against placebo in adults with myotonic dystrophy type 1 (DM1), a progressive muscle-wasting disease with no approved treatments.
- The primary endpoint, change in video hand opening time (vHOT) over 54 weeks, did not show a statistically significant improvement over placebo.
- Novartis said secondary and exploratory endpoints showed some evidence of clinical activity, and the company is reviewing the full dataset and plans to engage regulators on a path forward.
- Endpoints News framed the miss as Novartis' second major late-stage trial disappointment within days (following a setback for its Lp(a) drug pelacarsen), reporting that Novartis shares sank roughly 10% on the news.
💡 PointsThe AOC platform itself isn't dead — secondary and exploratory signals suggest the biology worked, just not enough on this endpoint — but a primary miss on the lead asset from a $12B acquisition is about the worst possible first data readout for validating a new modality. Novartis bought Avidity for the platform, not just del-desiran, so the real question is whether this de-risks or re-risks the rest of the inherited AOC pipeline (cardiac, muscular dystrophy programs). Pair it with pelacarsen's stumble the same week and this reads less like bad luck than a pattern: aggressive, high-multiple platform bets hitting rocky Phase 3 data almost immediately. That's a harder story to sell investors than either miss alone, and it may cool appetite for the next big platform acquisition — Novartis' or anyone else's.
Updated: Novartis sinks after second disastrous trial failure, this time of Avidity-acquired drug ↗
2026-09-07 · 🏢trial discontinuation (futility) · Endpoints News · Novo Nordisk
- Novo Nordisk is stopping two Phase 3 trials, HERMES and ATHENA, of ziltivekimab early after a data monitoring committee flagged futility.
- Ziltivekimab is an antibody that targets interleukin-6 (IL-6), an inflammatory cytokine, and the trials were testing it in heart failure patients with inflammation.
- The committee concluded neither trial was likely to produce a different result from the earlier Phase 3 ZEUS trial, which missed its cardiovascular endpoint in July.
- HERMES tracked time to cardiovascular death or heart failure-related hospitalization/urgent visits; ATHENA measured broader heart failure outcomes including quality of life.
- Novo Nordisk is continuing a separate Phase 3 trial, ARTEMIS, testing ziltivekimab as an acute post-heart-attack treatment, with a readout expected in the first half of 2027.
- The repeated failures raise doubts about IL-6 inhibition as a viable cardiovascular treatment strategy.
💡 PointsThree strikes on cardiovascular endpoints now — ZEUS in July, HERMES and ATHENA today — and Novo isn't spinning this as a design problem anymore; the DMC futility call is about as clear a kill signal as you get.
Sunk cost here isn't trivial: ziltivekimab was Novo's flagship diversification bet outside GLP-1s, and this guts the broader inflammation-in-CV thesis that Novartis also struggled to validate with canakinumab post-CANTOS. IL-6 as a druggable node for heart failure now looks structurally shaky, not just underpowered.
ARTEMIS surviving to 2027 reads less like conviction and more like sunk-cost inertia — acute post-MI biology differs, but a quiet wind-down before readout wouldn't surprise me if enrollment softens.
Zoom out and this underscores just how thin Novo's non-GLP-1 pipeline really is, right as obesity/diabetes competition intensifies.
Updated: Novo Nordisk cans two more trials of IL-6 heart drug for futility ↗
2026-09-04 · 🏢trial result (failure) · STAT News · Novartis / Ionis Pharmaceuticals
- Novartis' pelacarsen missed the primary endpoint in the Phase 3 Lp(a)HORIZON cardiovascular outcomes trial, which enrolled 8,323 patients with elevated lipoprotein(a) and established cardiovascular disease.
- Pelacarsen is an antisense oligonucleotide, licensed from Ionis Pharmaceuticals, designed to suppress hepatic production of Lp(a), a lipid-protein particle linked to elevated cardiovascular risk.
- The drug achieved substantial lowering of Lp(a) levels in treated patients, but this biomarker reduction did not translate into a statistically significant drop in cardiovascular death, heart attack, stroke, or urgent revascularization versus placebo.
- The result is the first major late-stage setback for the emerging Lp(a)-lowering drug class, raising questions for rival programs from Amgen and Eli Lilly that use different modalities but showed larger Lp(a) reductions in earlier-stage studies.
- Novartis and Ionis had held global rights under a licensing and collaboration deal, with Ionis having discovered and advanced pelacarsen through early development before Novartis took it into Phase 3.
💡 PointsPelacarsen just became the class's first hard-outcomes readout — and it missed, despite deep Lp(a) lowering. That's the uncomfortable part: this was the cleanest test yet of the Lp(a) causal hypothesis, and it didn't confirm it. Olpasiran (Amgen) and lepodisiran (Lilly) post larger percentage reductions in early studies, but bigger knockdown of the same biomarker doesn't guarantee outcomes benefit — their own outcomes trials now carry real doubt, not just execution risk. For Novartis, this was a marquee asset licensed from Ionis to anchor its cardiovascular pipeline; the miss removes a near-term blockbuster case and puts pressure on how the Ionis partnership economics get valued going forward. Biggest read-through: Lp(a) may need rethinking as a validated drug target, not just a biomarker to chase.
Novartis' experimental cardiovascular drug fails a pivotal study ↗