CiPSC-derived autologous islets (RGB-5088)
Hangzhou Reprogenix Bioscience, Inc.
Her own cells — and full immunosuppression.
Islets grown from a person's own cells, chemically reprogrammed to pluripotency (CiPSC) and injected under the abdominal rectus sheath. This is the "woman cured of type 1 diabetes with her own cells" story. The result is real — one patient, sustained insulin independence from day 75, holding at one year. What it is not is a drug-free cure: she had already had a liver transplant and was on chronic systemic immunosuppression before the islets went in. Autologous cells answer rejection; they do not answer the autoimmunity that destroyed her beta cells in the first place. Now in a 10-person Phase 1 (RGB-5088) that explicitly recruits prior organ-transplant recipients.
The scorecard
One published patient achieved sustained insulin independence 75 days after transplant, with time-in-range rising from 43.2% to over 98% and HbA1c around 5%, still off insulin at one year. That is a genuine human result and we score it as one — but it is a single case, and the registered 10-person Phase 1 has no posted results.[1]
One year of graft function in one person. No multi-patient or multi-year data exists; the Phase 1 does not complete follow-up until December 2031.[1]
Autologous is not immunosuppression-free. The one human this worked in was already on chronic systemic immunosuppression for a previous liver transplant, which is precisely why she was chosen — the team could "leverage existing immunosuppressive therapy." No CiPSC islet graft has ever been tested in an untreated type 1 immune system, and the registered Phase 1 explicitly admits people who have had a liver or kidney transplant. Human drug-free evidence here is zero, so this scores with zimislecel and far below Sana's n=1 UP421 result — the only human drug-free graft on record.[3]
The cells are injected beneath the abdominal anterior rectus sheath — a site that can be imaged and, in principle, removed, which compares well with a non-retrievable portal-vein infusion. Against that, an autologous product means every recipient needs their own cells harvested, reprogrammed and manufactured before anything can be transplanted.[1]
Bespoke per-person manufacturing is the opposite of off-the-shelf. The Phase 1 takes 10 adults aged 18-60 with stimulated C-peptide under 0.3 ng/mL and at least one severe hypoglycemic event in the prior year, at a single hospital in Tianjin.[2]
One peer-reviewed single-patient report plus a recruiting single-arm, open-label Phase 1 (10 participants, started February 2025). Primary completion is estimated December 2027 and study completion December 2031.[2]
The full picture
What it is
RGB-5088 is an islet product grown from the recipient's own cells. The route is unusual and genuinely clever: rather than using genes to reprogram adult cells back to a stem-cell state, Deng Hongkui's group at Peking University uses a cocktail of small molecules — chemically induced pluripotent stem cells, CiPSCs — and then differentiates them into islets. The islets are injected beneath the abdominal anterior rectus sheath rather than into the liver's portal vein, so the graft can be imaged and, in principle, retrieved.1
In September 2024 the team published one-year results from a single patient in Cell: a woman with type 1 diabetes who became insulin-independent 75 days after transplant, whose time-in-range rose from 43.2% to over 98%, and whose HbA1c settled around 5%.1 This is the case that travelled the world as "woman cured of type 1 diabetes with her own reprogrammed cells."
The result is real, and we score it as a real human result. The headline is wrong.
The part the headlines left out
She was already on chronic systemic immunosuppression before the islets went in. She had previously received a liver transplant and was maintained on anti-rejection drugs for it. That was not an incidental detail of her history — it is why she was a candidate. As a peer-reviewed review of the case puts it: the first human trial "involved a patient with a history of liver transplantation, allowing the team to leverage existing immunosuppressive therapy."2
So the graft was placed into an immune system that had already been chemically disarmed. Nobody has yet shown a CiPSC islet graft surviving in an untreated type 1 immune system. That experiment has not been run.
The registered Phase 1 does not run it either. RGB-5088's trial, NCT06731218, lists as an inclusion criterion: type 1 diabetes patients "including those who have received organ transplantation such as liver and kidney" — that is, it explicitly recruits people who are already immunosuppressed.3 The trial is designed to answer whether the cells work. It is not designed to answer whether they survive without drugs.
Why "your own cells" is not the answer to immunosuppression
This is the misunderstanding worth being blunt about, because it is everywhere.
Transplanted islets face two enemies:
- Allo-rejection — the immune system attacking tissue from another person. Autologous cells genuinely solve this. The cells are yours; there is nothing foreign to reject.
- Autoimmunity — the thing that caused type 1 diabetes in the first place. Memory T cells that destroyed your original beta cells are still there, still primed, still specific to beta-cell antigens. A perfectly matched autologous graft is exactly the target they were trained on.
Autologous cell therapy answers the first problem and is silent on the second. It does not remove the need for immune protection in T1D; it removes one of the two reasons you needed it. That is a real contribution to the cell-supply problem — you can make islets for anyone, without a donor — and it is not a contribution to the protection problem.
For the same reason, the CiPSC case is not comparable to Sana's UP421 result. There, gene-edited donor islets survived 14 months in a man taking no immunosuppression at all. That is n=1 and at a sub-therapeutic dose, and it is still the only human on earth with transplanted insulin-producing cells and zero anti-rejection drugs. The CiPSC patient is not a second data point in that column. She is a data point about what happens when the immune system is already suppressed.
How to read the score
We score Immunosuppression-free at 20 — the same as zimislecel, which openly requires lifelong systemic immunosuppression. That will look harsh next to a therapy using the patient's own cells. It is deliberate: on this site the criterion measures evidence that a person can keep these cells without chronic systemic anti-rejection drugs, and for CiPSC islets that evidence is zero. A design intention is not a result. We will move the number the day a CiPSC graft is shown functioning in someone who is not on immunosuppression — and not before.
References
-
Wang S, Du Y, Zhang B, et al. "Transplantation of chemically induced pluripotent stem-cell-derived islets under abdominal anterior rectus sheath in a type 1 diabetes patient." Cell 187(22):6152-6164 (25 September 2024). One-year preliminary results from a single patient in first-in-human Phase 1 ChiCTR2300072200. https://doi.org/10.1016/j.cell.2024.09.004 ↩ ↩2
-
Lou Y. "Illuminating the future of diabetes treatment: Autologous CiPSC-derived islets take center stage." Cell Transplantation (2025). States plainly that the first human trial "involved a patient with a history of liver transplantation, allowing the team to leverage existing immunosuppressive therapy." https://doi.org/10.1177/09636897251366828 ↩
-
Hangzhou Reprogenix Bioscience. "A Single-Center, Single-Arm, Open-Label Phase I Clinical Trial Evaluating the Safety and Efficacy of RGB-5088 Islet Cell Injection in the Treatment of Type 1 Diabetes Mellitus." ClinicalTrials.gov NCT06731218. Registry checked 15 July 2026: recruiting; 10 participants; Tianjin First Center Hospital; key inclusion criterion "Type 1 diabetes patients (including those who have received organ transplantation such as liver and kidney)." https://clinicaltrials.gov/study/NCT06731218 ↩
Coming soon
ETA · Phase 1 recruiting; primary completion estimated December 2027, study completion December 2031. No approval timing anywhere.
- →First multi-patient RGB-5088 safety and insulin-independence readouts · Primary completion estimated December 2027
- →A second, larger Reprogenix study (NCT07464119, n=30, Peking University First Hospital) opens recruitment — registered at phase NA, still not recruiting, and not a test of drug-free survival either · Registered start March 2026 has passed; nobody dosed as of July 2026
- →The test that actually matters — a CiPSC islet graft in someone who is not already immunosuppressed · Not scheduled by anyone
Sources
- [1]Transplantation of chemically induced pluripotent stem-cell-derived islets under abdominal anterior rectus sheath in a type 1 diabetes patient · peer-reviewed · 2024-09-25 — Wang S, Du Y, ... Deng H, Shen Z. Cell 187(22):6152-6164. One-year results from a single patient in a first-in-human Phase 1 (ChiCTR2300072200): autologous CiPSC islets, insulin independence from day 75.
- [2]Phase I Clinical Study to Evaluate the Safety and Efficacy of RGB-5088 in Patients With Type 1 Diabetes Mellitus · registry · 2025-02-17 — Registry checked 15 July 2026: recruiting, 10 participants, Tianjin First Center Hospital. Inclusion criteria explicitly cover "Type 1 diabetes patients (including those who have received organ transplantation such as liver and kidney)."
- [3]Illuminating the future of diabetes treatment: Autologous CiPSC-derived islets take center stage · peer-reviewed · 2025-08-18 — Peer-reviewed review of the case: "The first human trial involved a patient with a history of liver transplantation, allowing the team to leverage existing immunosuppressive therapy."