Researchers can now replace insulin-making cells and watch them respond to glucose inside the body. The central questions are becoming clearer: how do we protect those cells, keep them working for years, and produce enough for millions of people?
The biology
Type 1 diabetes is an autoimmune condition. The immune system mistakenly attacks and destroys the beta cells in the pancreas that make insulin. These cells live in clusters called islets and act as a living glucose sensor.
As the beta cells disappear, the body makes little or no insulin. Glucose builds up in the bloodstream while the body's cells cannot access it efficiently for energy. People with Type 1 diabetes replace the missing insulin through injections or a pump. Cell replacement aims to restore the living glucose-sensing system itself.
A therapy has to become reliable, repeatable, and manufacturable at enormous scale.
IDF Diabetes Atlas 2025
State-of-the-art therapies
Scientists can collect insulin-making islets from a donor or grow new ones from stem cells. Once transplanted, these cells can sense glucose and release insulin minute by minute, much like the cells lost in type 1 diabetes.
Recent work is testing two connected ideas: how well replacement cells can control glucose, and how to keep the immune system from destroying them.
Sana, Vertex, and the University of Chicago / Eledon reached the problem from different directions. Together, their results show both how far the field has moved and what it needs to solve next.
Sana's gene-edited donor islets have survived and produced glucose-responsive insulin for 14 months in one person without immunosuppressive drugs. The dose was small, so this result is about immune protection and cell function, not insulin independence.
Vertex's lab-grown islet cells restored insulin production in all 12 full-dose participants followed for at least a year. Ten stopped taking insulin. This approach currently uses immunosuppressive drugs.
In a University of Chicago Medicine study, donor islets were paired with Eledon Pharmaceuticals' tegoprubart, a newer anti-rejection drug. In the June 2026 update, all 12 participants had achieved insulin independence. Stable graft function was observed across the cohort, with a median follow-up of 8 months and the longest follow-up reaching 22 months.
The work ahead
The biology has moved quickly. Making the result durable and available at population scale is now the central challenge.
Replacement cells need to sense glucose and release the right amount of insulin for years, not months.
The cells must be protected from transplant rejection and the autoimmune response that caused type 1 diabetes, ideally without lifelong immunosuppression.
Producing billions of mature, consistent beta cells requires reliable processes, strict quality control, and much more capacity.
Cells need a safe home with enough oxygen, a way to monitor their health, and a practical path to replacement if their function declines.
AI and mathematics
There are many ways to alter a beta cell, and the effects of each change are hard to predict in isolation. A single gene edit can affect multiple biological pathways, while the outcome also depends on the state of the cell, how it was produced, and how the immune system responds.
The number of possible combinations quickly becomes too large to test experimentally. Mathematics can help narrow that space. Graph theory can model relationships between genes and cell states. Combinatorics can help explore combinations of edits efficiently. Spectral methods can identify structure that persists across different conditions. Statistics can distinguish reproducible biological effects from experimental noise.
Why now
Models such as GPT-5.6 Sol and Claude Fable are expanding what AI can contribute to mathematics and scientific discovery. AI-generated work is already making substantial progress on long-standing mathematical problems.
Many problems in computational biology have a similar mathematical core: understanding complex networks, searching enormous spaces of possibilities, and predicting how a system will respond to change. Better mathematical reasoning could make more of these problems tractable, including questions about how cells function and interact with the immune system.
Mathematics offers a glimpse of what could come next for biology. We believe many computational biology problems can see similar breakthroughs as AI improves, with experiments turning promising predictions into evidence.
Our working papers
Explore working papers on mathematics and computational biology.
A computational extension of Kobaisi et al.'s insulin signal-peptide engineering identifies a two-residue CHGA leader variant with lower predicted HLA-I presentation counts and an unchanged predicted cleavage boundary. We have not tested the proposed variant in the laboratory.
Read the paperAcross two public human beta-cell CRISPR screens, we show how repeated guide identity and shared controls can create apparent reproducibility without a reliable gene-level phenotype.
Read the paperWe prove the conjectured higher-alphabet spectral bound for Hamming graphs, strengthen it to arbitrary supports in the minimal dimension, and extend the result to unequal coordinate weights.
Read the paperWe establish a spectral compression principle, construct new obstructions to Hamming-ball optimality, and classify the first two compressed defect layers exactly.
Read the paperWe are building the quantitative foundation that cell replacement needs to become a routine therapy, alongside the groups manufacturing the cells and running the trials. We bring the mathematics and the models, and we set the experimental agenda together.
Get in touch