Stem-cell-derived islets (zimislecel / VX-880)

Type 1 diabetes involves autoimmune destruction of insulin-producing beta cells in pancreatic islets and usually requires ongoing insulin replacement. Zimislecel (formerly VX-880) aims to replace those lost cells rather than just dose around their absence. Vertex grows fully differentiated, insulin-producing islet cells from pluripotent stem cells — a potentially scalable source of cells — and infuses them as a single dose into the hepatic portal vein (a vessel carrying blood into the liver), where they engraft and secrete insulin in response to blood glucose.¹ Because the cells come from a donor stem-cell line rather than the recipient, this is allogeneic: patients also take chronic immune-suppressing drugs to stop rejection.¹ This approach aims to reduce reliance on scarce deceased-donor pancreases; it is not an established cure.¹

The clinical evidence

The published early clinical data come from the FORWARD study (VX-880-101, ClinicalTrials.gov NCT04786262), a Phase 1/2/3 trial whose interim Phase 1/2 results were published in the New England Journal of Medicine in June 2025.¹ In Part A, participants received a half dose; in Parts B and C they received a full single dose of 0.8 billion cells, all with glucocorticoid-free immunosuppression.¹ Among the 14 participants followed at least 12 months, every one had C-peptide below the assay’s detection threshold at baseline — and after infusion, all showed detectable C-peptide consistent with engraftment and insulin production. All analyses were interim and not prespecified; the study was open-label and had no randomized control group. Vertex designed the study with its steering committee, collected and analyzed the data, and participated in manuscript preparation.¹

Among the 12 full-dose recipients, all were free of severe hypoglycemic events during the evaluated days 90–365, all reached HbA1c below 7%, and all spent more than 70% of the day in the target glucose range of 70-180 mg/dL (3.9-10 mmol/L).¹ Ten of the 12 (83%) were completely off injected insulin at one year,¹ with a mean reduction in daily insulin dose of 92% across those 12.¹ Glucose-responsive C-peptide persisted through day 365. The findings are from a small selected cohort, so they do not establish multi-year durability, comparative superiority or outcomes in the broader T1D population.¹

Eligibility and safety

The published study allowed adults aged 18–65 with at least five years of insulin dependence, impaired hypoglycemia awareness and at least two severe hypoglycemic events in the previous year despite appropriate care. The current master-protocol registry also lists stable diabetes treatment and at least three months of consistent CGM use; additional eligibility conditions apply.¹²

In the safety analysis of 14 recipients, serious neutropenia occurred in 3 and serious acute kidney injury in 2. Two participants died: one of cryptococcal meningitis and one of progression of pre-existing neurocognitive impairment. Investigators considered no serious adverse event related to zimislecel itself, but attributed the cryptococcal meningitis to immunosuppressive medication. That patient had also received prolonged high-dose systemic glucocorticoids prohibited by the protocol and experienced a skull-base injury during sinus surgery. The death occurred about 19.5 months after infusion; the other death occurred about 30 months after infusion. These risks must be considered as part of the entire transplantation regimen, not excluded because they were not attributed to the cells.¹

Phase and proximity to availability

Zimislecel is in the pivotal Phase 3 portion of FORWARD. The September 2026 master protocol estimates 57 participants across the VX-880 and newly added VX-017 programmes; that total is not the size of the zimislecel pivotal cohort.² Vertex reports Regenerative Medicine Advanced Therapy (RMAT) and Fast Track designations from the U.S. FDA, PRIME designation from the European Medicines Agency, and an Innovation Passport under the UK MHRA's Innovative Licensing and Access Pathway. These development designations do not confer marketing approval.³

Vertex had guided to marketing applications in 2026, but that timing is now in doubt. Dosing in FORWARD was paused pending an internal manufacturing analysis; in its first-quarter 2026 results Vertex said it had completed the analysis and restarted dosing, without repeating the old filing target.⁴ Vertex's 3 August 2026 update says FORWARD continues to enroll and dose and that updated timelines for both zimislecel and VX-017 are expected later in 2026.⁵ The trial registry now describes a recruiting master protocol for VX-880 (Phase 1/2/3) and VX-017 (Phase 1/2), last updated 1 September 2026. It estimates 57 participants in total, primary completion on 31 December 2027 and final completion on 31 December 2031.² The August company update did not provide a filing date. These development timelines are separate from the published early efficacy and safety findings.

What's coming

Future approval, its timing and any eventual label remain uncertain. The studied population has severe hypoglycemia and the regimen requires immunosuppression; a future indication cannot be assumed from these early data. Vertex’s August 2026 statement expects updated development timelines later in 2026.¹⁵

The real prize is removing the immunosuppression requirement — and Vertex has already lost one attempt at it. Its device-encapsulated version, VX-264, was discontinued in March 2025 after Vertex reported that the product was generally well tolerated but failed its C-peptide efficacy endpoint; Vertex booked a $379 million intangible-asset impairment on the programme (see our VX-264 record).³⁶ Killing VX-264 did not, however, take encapsulation off Vertex's list. The same March 2025 update named three research routes it was pursuing toward dropping systemic immunosuppression: alternative immunosuppressive regimens, gene-edited hypoimmune stem-cell-derived islets, and novel encapsulation devices.³

VX-017 is a distinct clinical-development programme within the same registered master protocol. Its Phase 1/2 arm now specifies hepatic portal-vein infusion; the checked registry contains no VX-017 results and its public summary does not establish immune evasion or a detailed immunosuppression regimen.²⁵ “Regardless of blood type” is the disclosed product property; it is not evidence that VX-017 avoids immunosuppression.