Type 1 diabetes: artificial pancreas systems and stem-cell islets, where things really stand

Ten of twelve people in one trial stopped insulin after a single stem-cell infusion. That headline is real, and so is the price: lifelong immune-suppressing drugs.

Two different paths

In type 1 diabetes the immune system destroys the insulin-producing beta cells in the pancreas. Treatment has meant injecting or pumping insulin for life. Today, two approaches are changing that: technology that automates insulin delivery, and cell therapy that replaces the lost beta cells.

The artificial pancreas (automated insulin delivery)

A closed-loop system has three parts: a continuous glucose monitor under the skin that reads sugar levels every few minutes, an insulin pump, and software that decides how much insulin to give. The software predicts where glucose is heading over the next 30 to 60 minutes and adjusts background insulin and correction doses automatically, day and night.

These systems are called hybrid because the user still has to announce meals. Insulin given under the skin takes time to act, so the algorithm cannot catch up with a meal on its own. Some systems, such as the iLet, simplify this to "usual, more or less" rather than exact carbohydrate counting.

In real-world use, people typically spend much more time in the target glucose range, average HbA1c falls and dangerous lows become less common.

Replacing the cells: islet transplants

For decades, a small number of people received islets taken from deceased donors, infused into a vein that leads to the liver. In June 2023 the FDA approved Lantidra, the first commercial donor-islet product, for adults with severe, unmanageable low blood sugar. The limits are a shortage of donors, variable quality and the need for immune-suppressing drugs.

Stem-cell islets: the zimislecel results

Vertex grows islet cells from stem cells in the laboratory, so supply is no longer limited by donors. In the FORWARD trial, reported at the American Diabetes Association meeting and in the New England Journal of Medicine in June 2025, 12 adults received a full single infusion into the liver's portal vein.

• All 12 had no severe low-sugar episodes from day 90 to one year and reached the HbA1c target.
• 10 of the 12 no longer needed external insulin at one year.
• All participants produced their own insulin in response to meals.

Participants took immune-suppressing drugs throughout. One participant died of a fungal infection linked to those drugs. The trial has expanded into a larger pivotal stage of about 50 people.

Why it is not yet a cure

• Immunosuppression is lifelong. Stopping it means the immune system destroys the new cells, through both rejection and the original autoimmune attack.
• The drugs have serious side effects. In the Lantidra trials, infections and cancers were reported, and drugs such as tacrolimus can gradually damage the kidneys.
• The autoimmune disease remains. The cells are replaced, but the underlying immune problem is not fixed.

For these reasons, cell therapy is currently limited to adults whose severe hypoglycaemia is more dangerous than the drugs. For most people, a closed-loop system is the better option today.

Can the cells be protected without drugs?

This is the central research question.

• Protective devices: Vertex's VX-264 placed cells in an implanted pouch meant to shield them. It was discontinued in March 2025 after the cells produced too little insulin; scar tissue around the device appears to have starved them.
• Vascularised pouches: Sernova's Cell Pouch lets blood vessels grow in before cells are added and can be removed if needed. Results are early and still involve immunosuppression.
• Gene-edited "hypoimmune" cells: CRISPR Therapeutics' CTX211 removes the markers that immune cells recognise. Early trials are ongoing.
• Your own cells: in 2024 a Chinese team reported a woman who became insulin-free after receiving islets grown from her own reprogrammed fat cells. It is a single case, and she was already on immunosuppression for a liver transplant.

Cost and access

• United Kingdom: under NICE TA943, the NHS is rolling out closed-loop systems over five years for children and young people, people who are pregnant or planning pregnancy, and adults with HbA1c of 7.5% or higher or disabling hypoglycaemia.
• United States: closed-loop systems are generally covered by private insurance and Medicare, with deductibles and co-payments. Lantidra is offered only at specialist transplant centres.
• European Union: closed-loop systems are widely reimbursed through national health systems.
• Turkey: SGK began reimbursing glucose sensors for children in December 2024. Adult sensors are not routinely covered, and SGK's fixed contribution to insulin pumps is far below their market price, so families pay most of the cost. Neither Lantidra nor zimislecel is registered in Turkey.
• Stem-cell therapies: available only within clinical trials, paid for by the sponsor.

How to look for a trial

Search ClinicalTrials.gov, Breakthrough T1D's trial finder or major university hospitals. The zimislecel trial (NCT04786262) enrols adults aged 18 to 65 with type 1 diabetes for at least five years, no measurable own insulin, severe hypoglycaemia unawareness and good kidney function.

Questions patients ask