Executive summary
PRISE-hATG evaluates SAB-142, a fully human anti-thymocyte globulin, in people aged 5–40 with Stage 3 type 1 diabetes (T1D) roughly 100 days to two years after diagnosis. The question is whether SAB-142 preserves residual beta-cell function over 12 months, measured by stimulated C-peptide during a mixed-meal tolerance test (MMTT).
The rationale rests on three lines of evidence:
- Disease biology. Residual beta-cell function varies between people, and baseline C-peptide does not fully describe the later rate of decline.
- Therapeutic rationale. Rabbit-derived ATG studies show immunomodulation can alter immune-cell populations and may preserve C-peptide under some regimens, but results have not been consistent.
- Early SAB-142 data. Sponsor-reported Phase 1 findings include preliminary safety, pharmacodynamic, and exploratory C-peptide observations in a very small cohort. They support further study but do not establish efficacy.
PRISE also raises a design question: how convincingly can outcomes from external SAFEGUARD participants be combined with outcomes from participants recruited directly into PRISE?
Preliminary assessment. The clinical rationale is plausible, but efficacy remains unproven in these populations. Public evidence identifies the key scientific and design questions. It is not enough to calculate a reliable screening-failure rate or predict the treatment effect.
Study design
| Attribute | Publicly reported information |
|---|---|
| Registry | NCT07670650, Phase 3, multicenter |
| Population | Stage 3 T1D, ages 5–40 |
| Disease duration | About 100 days to two years after diagnosis (prospective participants) |
| Allocation | 2:1 to SAB-142 or placebo, as described by the sponsor |
| Treatment | Induction at baseline, maintenance at Month 6 |
| Follow-up | Through Month 12 |
| Primary measure | Change from baseline in log-transformed C-peptide AUC+1 during a two-hour MMTT |
Three disease-duration cohorts
Cohort 1: new-onset
36 age-matched external participants from SAFEGUARD, under 100 days from diagnosis.
Cohort 2: recent-onset
36 newly recruited participants, more than 100 days and under one year from diagnosis.
Cohort 3: established
36 newly recruited participants, one to two years from diagnosis.
The structure asks whether an effect can be shown in participants whose residual beta-cell function and disease trajectories may differ from those newly diagnosed. The protocol and statistical analysis plan (SAP) are needed to confirm the estimand, analysis populations, missing-data methods, and whether all cohorts contribute identically to the primary analysis.
Scientific rationale: residual beta-cell function
C-peptide marks endogenous insulin secretion. Interpreting it takes attention to both a participant’s starting level and the rate of change over time.
Trajectories are heterogeneous
Besser et al. (2018) analyzed serial MMTT data from 442 people diagnosed before age 18, followed up to six years. Average C-peptide level and rate of decline were not strongly correlated, and younger age at diagnosis was associated with faster decline. Two participants with the same baseline may therefore follow different paths.
Limits: a pediatric cohort with historical data. It does not establish PRISE eligibility rates or predict response to SAB-142.
Beta-cell function has more than one dimension
Gitelman et al. (2023) found that beta-cell glucose sensitivity in pooled placebo-arm data captures change that conventional C-peptide measures do not fully summarize. Galderisi et al. (2025) found that trajectories differ between slower and faster progressors in Stage 2 T1D. Both support careful characterization of the target population. Neither validates an enrichment model for SAB-142.
What earlier ATG trials show
Rabbit-derived ATG studies give biological and clinical context. They are not direct evidence for SAB-142.
| Study | Regimen and population | What it showed |
|---|---|---|
| START (Gitelman, 2016) | Rabbit ATG 6.5 mg/kg, 58 participants aged 12–35, within 100 days of diagnosis | Prespecified two-year endpoints not met. A post hoc signal in older participants is hypothesis-generating only. Notable adverse-event burden. |
| Low-dose ATG (Haller, 2019) | Low-dose ATG alone, ATG plus G-CSF, or placebo in recent-onset T1D | ATG alone gave higher stimulated C-peptide at two years than placebo. The combination arm missed its prespecified criterion. |
Reading the class evidence. Outcomes depend on the specific regimen and study context. Prior ATG studies support investigating the hypothesis, not assuming that SAB-142 preserves C-peptide or has a particular safety profile.
SAB-142 Phase 1 evidence
Sponsor disclosures describe safety, pharmacokinetic, immunogenicity, and exploratory pharmacodynamic findings. The established-T1D cohort had six adults: four received 2.5 mg/kg and two received placebo. The sponsor reported no C-peptide decline at Day 120 in the treated participants, no drug-related serious adverse events or serum sickness, and transient rather than sustained lymphocyte reduction.
That is a useful early signal. It does not establish statistically supported C-peptide preservation, efficacy in children or adolescents, efficacy in the PRISE duration groups, long-term safety, a dose-response relationship, or which participants are likely to benefit. Decisive evidence has to come from controlled trial data.
SAFEGUARD and the external cohort
SAFEGUARD (NCT07187531) is the ongoing registrational Phase 2b study of SAB-142 in Stage 3 T1D diagnosed within 100 days. The sponsor’s update plans 159 participants aged 5–40, with topline data expected in the second half of 2027. PRISE extends the program to longer disease duration and would borrow 36 SAFEGUARD participants for new-onset data.
The sponsor states that sampling and analytical methods are harmonized. That helps, but it does not make participants from separate studies exchangeable. A complete evaluation would test five things:
Questions for protocol and statistical review
- Measurement. Are MMTT procedures, timing, assays, and sample handling equivalent?
- Population. How do age, baseline C-peptide, duration, and other prognostic factors differ?
- Treatment. Are dose, timing, background care, and exposure aligned?
- Follow-up. Are assessment windows, dropout patterns, and missing-data handling comparable?
- Statistics. How are external participants incorporated, and how is between-study uncertainty handled?
Public disclosures cannot show that pooling introduces bias, or that it avoids it.
Eligibility: what is public
The registry establishes the broad population: ages 5–40, Stage 3 T1D, and for prospective cohorts more than 100 days to two years from diagnosis. It also reports a random C-peptide screening threshold above 0.2 nmol/L, with one retest permitted.
| Measure | Purpose |
|---|---|
| Random C-peptide above 0.2 nmol/L | Entry criterion: decides whether a candidate can proceed |
| Stimulated C-peptide AUC, two-hour MMTT | Primary endpoint: change in beta-cell function over follow-up |
Other domains the record points to include autoantibody status, age and anthropometric criteria, safety laboratories, infections, concomitant medications, other autoimmune conditions, reproductive precautions, vaccination, and ability to attend visits. Exact thresholds, exceptions, and retest rules require the current protocol, so this review does not convert the public criteria into a screening algorithm.
What the recruitment literature adds
Kinney et al. (2023) studied five TrialNet Stage 1/2 prevention trials. Between 8.5% and 49.8% of eligible participants proceeded to screening, depending on the trial. Age, sex, family relationship, site volume, and site location were associated with uptake.
Those are presymptomatic prevention trials, so the percentages should not be applied to PRISE. The more useful lesson is that three different metrics need three different denominators:
Screening uptake
Of those eligible or approached, how many proceed to screening.
Screen failure
Of those screened, how many do not meet requirements or cannot proceed.
Enrollment conversion
Of a defined upstream group, how many ultimately enroll.
A screening-uptake study cannot supply a PRISE screen-failure estimate, and eligibility criteria alone cannot produce one.
Evidence gaps
Clinical evidence
- Magnitude and uncertainty of the SAB-142 effect on stimulated C-peptide
- Whether the effect differs by age or disease duration, and how long it lasts beyond 12 months
- Adverse events and immune effects with broader exposure and longer follow-up
Trial design and analysis
- How the three cohorts and the external SAFEGUARD participants enter the primary estimand
- Assumptions supporting cross-study comparability
- Handling of missing data, protocol deviations, and intercurrent events
Recruitment and eligibility
- Counts at each stage: approached, prescreened, screened, eligible, randomized, treated
- Which criteria drive the most screen failures, and how borderline results and retests change eligibility
- Variation by site, age group, disease duration, and referral pathway
- Share of eligible patients who decline or cannot complete visits
None of these can be answered reliably from the registry and literature alone. They need protocol documents, trial results, site-level recruitment data, or participant-level data.
Preliminary assessment
SAB-142 has a plausible immunomodulatory rationale supported by prior ATG research and early human observations. The rabbit-ATG literature is mixed, and the Phase 1 C-peptide findings are too limited to show efficacy. SAFEGUARD remains the key study in newly diagnosed Stage 3 T1D, while PRISE extends the program to longer duration.
Overall conclusion. Public evidence supports continued investigation of SAB-142. Efficacy, cross-cohort comparability, and PRISE-specific recruitment performance remain unresolved. The next step is to verify the detailed protocol and SAP and identify what recruitment data actually exist, then build a quantitative screening-risk assessment with explicit denominators.
References
- PRISE-hATG, ClinicalTrials.gov NCT07670650
- SAFEGUARD, ClinicalTrials.gov NCT07187531
- SAB BIO, Q2 2026 financial results and clinical-program update
- SAB BIO, March 2026 Phase 1 C-peptide update; December 2025 Phase 1 safety update
- Besser et al. (2018), C-peptide loss and growth-curve analysis
- Galderisi et al. (2025), beta-cell function trajectories in Stage 2 T1D
- Gitelman et al. (2023), beta-cell glucose sensitivity in recent-onset Stage 3 T1D
- Gitelman et al. (2016), two-year randomized rabbit-ATG trial
- Haller et al. (2019), low-dose ATG trial
- Kinney et al. (2023), barriers to screening in T1D prevention trials