Small Biotechs:
Diagonal Tx: Clustering antibodies that mimic the action of the ligand and bypass the need for the ligand and receptor. This mutation that is created makes standard AI models not useful and so need a new method. This restores new ALK1 signaling in Hereditary Hemorrhagic Telangiectasia. It also treates LoF mutations in ALK1 or ENG which is on the cell surface.This leads to endothelial signaling causing PAH but in HHT- AVM’s. The bispecific force interaction of BRM1 with
HHT : Largest unaddressed bleeding disorder – also called nose bleed disease because it causes 95% nose bleed in the nasal cavity but there are other places too. Brain AVM and Lung AVM is also a problem. The patient needs new embolization to treat the new AVM (arteriovenous malformation)
Animal experiments show good treatment effect with DIAG723. Restores signaling in both endoglin knockout (HHT1( and ALK1 mutant (HHT2) patient derived cells. In humans, VEGFR1 is under ALK1 activation and can serve as a biomarker of disease. And DIAG723 can promote release of soluble VEGFR1 in NHP Plasma. The clinical trials starting in 2026 for HHT and then PAH (pulmonary arterial hypertension) in 2027. PAH -is coming from BMPRII downregulation or mutaitons coupled with excess of ploriferative ligand signaling leading to disease.
PAH model is the Sugen Hoypoxia model in mice. But in humans there is idiopathic and BMPR2 mutations too (https://pmc.ncbi.nlm.nih.gov/articles/PMC4137413/).
The binding of the antibody DIAG723) will have a long lasting effect and invitro is about 4-5 days which translates to twice a month in humans.
They are trying to mimic BMP9 which is a natural ligand in normal endothelial cells.
Platform can be used for many genetic diseases – which has LoF mutation or gain where multiple receptors are involved and where – mutation in ligand or receptor on the surface of the cell.
Revir:
Charcot Marie Tooth CMT disease studies. CMT1 is most prevalent but also CMT2 and intermediate. They aer demyelenating form.
ISR inhibiton is a therapeutic strategy for CMT2D and GARS1 mutation causes ISR activation in CMT2D. RTX-117 restores nerve and muscle function in CMT2D mouse model by lowering ISR in spinal cord, rescue of body weight, grip strength, beam walk capability latency and amplitude of NCV.
These are going through China-first execution strategy. – because China is good for rare neuro disease since it is enrolled in the central hospital. In CNS there is not much competition. HongKong for iPSC capability.
Many disease proteins are hard to be targeted by antibody. Small molecule RNA can modulate RNA splicing. Small molecules to promote inclusion of poison exons which lead ot NMD or truncated protein. They develop small molecules to enhance poison-exon inclusion.
U1 snRNA can bind to multiple splicing sites – 4 nucleosides per position. GGAgu, AGAgu, UGAgu, UGAgu. VoyageR identifies which splice sites show the storngest combination of potency, dose-dependence and clean selectivity.
Built a dose response transcriptome atlas from thousands of proprietary splicing focused compounds.
In Huntington disease – most HTT lowering therapies don’t halt somatic expansion (from CAG repeats that generates a mutated protein that causes neuron death and CAG repeat numbers continue to expand)
PMS1 is a genetic modifier of HD progression by aberrant MMR repair and is targeted by Revir compound. They are better than uniQure & PTC Therapeutics which do not target PMS1. They have great cellular potency.
RTX-317 can inhibit both pathogenic allele and normal too and hence a 50% inhibitor.
Splicing modulator – behaves like a small molecule, so watch out for safety. PK is still the same. This molecule needs a high Kpuu for better penetration.
Biotech is all about the molecule more than the platform.
TheriniBoi – Fibrin targeting immunotherapy for neuroinflammation and neurodegeneration.
Vascular integrity leads to chronic depositon of toxic fibrin triggering inflammation. C.
Fibrin is prominent in Retinal diseases and Alzhiemer’s disease.
Fibrin epitope is exposed by thrombin.
THN391 fibrin antibody targets neuroinflammation.
THN622 is a bispecific molecule is for fibrin and VEGF blocker.
The work started in 2002 by Katerina Akassoglou at UCSF. And FIH studies in 2023.
Fibrin acts as a pure antagonist blocking the inflammatory epitope.
Fibrin induces microglial-mediated neuronal loss and cognitive impairment in 5XFAD alzhiemer’s model – belief is the fibrin deposit is independent of plaque deposits.
Ph1 study shows safe and tolerable. No ARIA.
Will look for biomarkers for BBB breakdown, cytokines and neurodegeneration.
THN622 is being developed for opthalomogic indications
In mutliple sclerosis, fibrin activity in the CNS can also be targeted.
Cheiron: AI native platform is missing. Most work in Biopharma is cross domain and lives in disconnected systems such as PubMed vs other areas such as SEC.gov and Clinical Trial.gov and humans become the integration layer. There is lots of time in communicating etc… but also missed data and signals. It is more about a knowledge infrastructure problem. Building a LifeSciences Knowledge graph that integrates all knowledge that humans do it usually. Workflow specialized agents/pipelines – which tells you how you should you perform workflow.
Each agent traverses multiple domains and optimises it. Demo available at cheiron.bio.
They can provide deep citations. They have multiple agents to target different systems of record. When more artifacts are added – it gets added to the system of record.
Generation is a solved problem but retrieval is a tougher problem.
Aarvik therapeutics: ADC – increasing therapeutic index of ADC’s, 560 ADC progressed to the clinic. 14 approved and 155 about discontinued. Low (10%) success rate due to ADC efficacy. MTD limits efficacy. Many factors:
Tumor indication, target antigen, antibody, conjugation, payload.
1st Generation MMAE ADC : MED is the bottle neck. MTD and MED is only 2 fold.
1st Gen is 0.1 – 0.5 mg/kg and then 2-6 mg/kg. Then 2nd generation – 5-15 mg/kg. These help in dosing higher.
Traditional ADC – standard targets with high copy number – Her2, Trop2, Nectin-4, Fralpha, TF). However, there are other targets with <50000 copy number – there are many tumor antigens. How do you address that ?
MUTTA – multi epitope targeting tetravalent antibody. Multiple arms to the antibody to make it more efficacious.
Funded through friends and family.
Tried different formats for manufacturability. Tetravalent : Trastuzumab – Cetuximab like, Pertuzumab and EGFR novel epitope. Shows superior signaling blockade vs monospecific and bispecific biologics. MUTTA also internalizes inside the cell. Gets more payload into tumor cell faster – critical to achieving better efficacy in the clinic.
MUTTA antibodies have a similar clearance rate to Tarstuzumab.
To test selected targets that had failed in the clinic due to low copy number and showed good efficacy.
Screened 80 solid tumor targets…and selected 40 tumor targets to select several mutta ADC programs.
They show better efficacy than mono specific.
They cannot increase copy number but increasing valency helps
The exact positions in space of 4 targets is important and other targets are in the pipeline. They have many partners but not VC funded.
