Axial - Observations #54
Build with Axial: https://axial22.axialvc.com/
Axial partners with great founders and inventors. We invest in early-stage life sciences companies such as Appia Bio, Seranova Bio, Delix Therapeutics, Simcha Therapeutics, among others often when they are no more than an idea. We are fanatical about helping the rare inventor who is compelled to build their own enduring business. If you or someone you know has a great idea or company in life sciences, Axial would be excited to get to know you and possibly invest in your vision and company . We are excited to be in business with you - email us at info@axialvc.com
Observations #54
A set of ideas and observations from a week’s worth of work analyzing businesses and technologies.
Why can’t we use 1-2 cells for CAR-T?
From my perspective, the most important question to answer in cell therapy is why can’t we use a countable (credit to Justin Farlow from Serotiny on this concept) number of cells? Key drivers to address this problem are clonal expansion, maintaining memory-like states, controlling tonic signalling, exhaustion profiles:
Clonal expansion - https://www.nature.com/articles/s41467-019-13880-1
Tonic signalling - https://stm.sciencemag.org/content/13/591/eabe7378
Exhaustion - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6944329/
With some of the top labs in the field
Roybal Lab at UCSF -
https://www.roybal-lab.org/
Mackall Lab at Stanford - https://med.stanford.edu/mackalllab.html
Marson Lab at UCSF -
https://marsonlab.org/
Turtle Lab at Fred Hutch - https://research.fredhutch.org/turtle/en.html?_ga=2.70861840.1227956745.1619843671-1359929626.1618775908
Like any complex product, whether it’s a bag of cells or a consumer beverage, formulation is everything. Understanding how cells (the active ingredient) behave and interact with each other and their environment will improve delivery and patient outcomes. Clonal expansion determines which CAR-T clones expand and why others disappear. Research has linked differential transcriptional profiles to clonal kinetics post-infusion; this is still a very active field of research. The other part is to increase the proportion of memory T-cells (or at least memory-like state) pre-infusion where editing often induces a lower fraction of memory T-cells. The Marson Lab has done some of the best work in this part of immunology. For tonic signalling, scFv aggregation plays an important role in regulating CAR-T activity - this is intimately linked to exhaustion where too much tonic signalling leads to early exhaustion of T-cells.
The next 2-3 years, the action is likely on honing in on these 4 areas to precisely engineer specific cell types to set up precise formulations of different cell therapies in 5-10 years. So what are some new opportunities in this arc (please email me if you have an interest in building new companies or already have one in this space)?:
Bring precision medicine to cell therapies by measuring clonality of a patient’s cells
Sourcing allogeneic cells from patients with strong immune systems. More broadly, how to reduce donor variability?
Cell sorting needed along with new biomarkers
Standardize preconditioning for patients
How long do you want the cell to be in the body? Where to place the cell? From which donor?
An achievable vision for cell therapies is to deliver the treatment in a countable number of cells. The rules for each immune cell are still being written along with manufacturing (hopefully it can be outpatient one day).
Precision medicine in dermatology
A general opportunity is to bring precision medicine, which has made the most progress in cancer, to other diseases. There is an interesting opportunity in dermatology due to pretty easy delivery, clear endpoints, and a large number of diverse patients.
Ease of biomarker discovery and iteration: Interestedly, cytokines, chemokines, and other inflammatory markers on the skin diffuse into the blood after shedding of endothelial cells into circulation. There are opportunities to correlate the concentration of these biomarkers on the skin to blood as well as localizing biomarkers to a given disease. This would establish different endophenotypes (phenotype associated with a disease) across dermatological diseases; there is early work here in eczema. Skin biopsies are relatively non-invasive allowing for much faster testing iterations.
Patient stratification: Genomics is very powerful to stratify not predict. New biomarkers can stratify based on disease severity and develop pathway-specific medicines
Match drugs to pathway(s): One of the canonical examples of pharmacogenomics is thiopurine C-methyltransferase (TMPT), a metabolizer of the drugs 6-mercaptopurine and azathioprine. Genetic variants of TMPT have been used to inform treatment of patients with immune-related diseases and kidney transplant rejection, respectively. Given that most diseases in dermatology are multi-axis immune diseases going from sequence to clinical relevance will be pretty difficult. Correlating patient immune profiles to different subtypes is a fertile area of work.
Precision medicine in dermatology can easily get started but de-convoluting the multi-dimensional patient responses and actually establishing a biomarker to a treatment is still a major barrier. There is high patient need in eczema (AD; image below), psoriasis, acne, rosacea, vitiligo, hives, among others - https://www.rxlist.com/conditions_15_uncommon_skin_diseases/article.htm
Graph Neural Networks in biology
Petar Veličković, a senior research scientist at DeepMind, recently published an incredibly useful presentation on the application of GNNs in comp bio - https://petar-v.com/talks/GNNBio-Imperial.pdf
Iron Lung phase of sleep technology
Technology really hasn’t touched sleep. What is the iPhone equivalent for the other half of our days? It really seems like sleep products right now are in the equivalent “Iron Lung” phase. Iron Lungs were predominately used to treat polio patients but once a vaccine was produced and modern ventilators introduced, the Iron Lung became obsolete. We use seemingly archaic or non-existent (i.e. bed softness) methods to improve our sleep. We may even be in the pre-Iron Lung phase where we still need better technology to improve our sleep. Eight has a compelling product but there is so much left to build.