PRISE-hATG: SAB-142 in recent-onset and established type 1 diabetes

What is being tested and why it matters

Anti-thymocyte globulin (ATG) is an old transplant drug that temporarily depletes the T cells attacking the pancreas. Given at a low dose soon after diagnosis, it is one of the few immune therapies that has actually preserved the body's own insulin production in trials. Its catch is that the standard version is made in rabbits: the immune system learns to recognise the rabbit protein, so realistically you get one course, ever. SAB-142 is a fully human version, made in genetically engineered cattle, and the whole point of it is that it could be given more than once.¹

PRISE-hATG is the phase-3 test of that idea — and it is unusual in who it lets in. Most disease-modifying trials in type 1 diabetes shut the door around 100 days from diagnosis, on the assumption that after that there is too little beta-cell function left to save. This trial deliberately reaches past that line.¹

Who is in it

Participants are aged 5 to 40, have stage-3 type 1 diabetes, still have measurable insulin production (a stimulated C-peptide above 0.2 nmol/L) and carry at least one islet autoantibody. They are split by how long ago they were diagnosed:¹

• Cohort 2 — recent-onset: diagnosed more than 100 days but less than a year ago.
• Cohort 3 — established-onset: diagnosed one to two years ago.
• Cohort 1 — new-onset: not enrolled here. This cohort is made up of roughly 36 age-matched participants from the separate SAFEGUARD phase-2b trial, whose data are pooled in as an external comparison group.

That structure matters for reading the numbers. The registry lists an enrolment of 108, but that is the combined efficacy dataset. Only about 72 people are actually being recruited into this study and randomised 2:1 to SAB-142 or placebo; the other ~36 come from SAFEGUARD.¹

Dosing is an induction course at baseline followed by a maintenance course at month 6 — the repeat-dosing that a fully human ATG is supposed to make possible.¹ Anyone who has had teplizumab or another anti-CD3 antibody is excluded.¹ The four listed sites are all in the United States: UCSF Benioff, the Barbara Davis Center in Colorado, the University of Florida, and Riley Hospital for Children in Indianapolis.¹

Why the "established" cohort is the interesting part

The field has largely assumed the therapeutic window is narrow — that once you are a year or two out from diagnosis, there is nothing left worth protecting. That assumption has rarely been tested head-on, and it quietly excludes almost everyone living with type 1 diabetes today from disease-modifying trials. Cohort 3 tests it directly.¹

If SAB-142 preserves C-peptide in people 365 to 730 days from diagnosis, the population eligible for immune therapy gets dramatically larger. If it does not, that is still a useful answer — it would confirm the window really is as tight as the field believes.

The trial also plans a secondary analysis that splits participants into "responders" and "non-responders" identified in advance by a laboratory test, an explicit attempt at the responder-heterogeneity problem that has muddied readouts across this whole field.¹

What we do and don't know

Nothing about whether this works. The trial has not started. Registration is not evidence — it is a plan, and planned start dates slip. The human efficacy evidence for SAB-142 itself remains thin: a small phase-1 signal, with the phase-2b (SAFEGUARD) still running and not reporting before late 2027.¹ A phase-3 launching before the phase-2b reports is a statement of confidence by the sponsors, not a result.

The main measure is C-peptide — how much insulin the body makes of its own accord — during a standardised two-hour mixed-meal test at 12 months.¹ Estimated primary completion is March 2029.¹

If you are looking to take part: the trial is not yet recruiting as of July 2026. Check the registry entry below for the current status before contacting a site.

References

1. ClinicalTrials.gov. PRISE-hATG: A Phase 3 Study Evaluating SAB-142 for Delaying the Progression of Type 1 Diabetes (NCT07670650). University of Florida, with SAb Biotherapeutics and Breakthrough T1D. Registry record first posted 26 June 2026. https://clinicaltrials.gov/study/NCT07670650 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹²