Skip to content
type1.science
Phase 2CompletedNCT04774224

BANDIT: Baricitinib (JAK inhibitor) in recent-onset type 1 diabetes

What this study tests

A phase 2 randomized, placebo-controlled trial testing whether baricitinib — an oral JAK1/JAK2 inhibitor already used for rheumatoid arthritis — can protect the body's own insulin production when started within 100 days of a type 1 diabetes diagnosis. Over 48 weeks it preserved C-peptide versus placebo. This was an early disease-modifying signal, not evidence of a cure; the checked US label has no T1D indication.

Editorial review: .

Registry checked: 2026-09-17. Registry’s own update: 2024-12-03.

Most recent recorded citation date: 2026-08-21. Only explicit date metadata is included; an undated citation may be newer. This does not mean every claim was reviewed on that date.

Trial status, labels and access can change between reviews. How we review the evidence · How to read the evidence

Full sources ↓Full discussion ↓Report an issue →

Evidence at a glance

Who can enter the study?
Ages 10–30, T1D diagnosed within 100 days before starting study drug, at least one islet autoantibody, and stimulated C-peptide >0.2 nmol/L or random C-peptide >0.3 nmol/L. Exclusions included systemic immunomodulatory treatment, DVT/pulmonary embolism history, specified renal/liver abnormalities, serious infection, most malignancies and pregnancy; the registry gives the full criteria. Registered eligibility ↗Enrollment criteria describe who may enter. They do not establish who was analyzed or the denominator for a reported result.
Reported benefit and results
91 randomized (60 baricitinib, 31 placebo). At week 48, median stimulated mean C-peptide was 0.65 versus 0.43 nmol/L/min (P=0.001). Lower insulin dose and glucose variability were exploratory secondary findings, without multiplicity adjustment; HbA1c was similar. Adverse events occurred in 47/60 versus 26/31; none of the seven serious events was attributed to study drug. In the published off-treatment follow-up, 88 completed week 96; the C-peptide difference was significant at week 72 (P=0.015) but not at week 96 (P=0.336).Read the result sources and their limitations →
Important harms
Read the reported results, safety discussion and original sources. A separate harms summary is not available; this does not establish safety.
Research access and approval
Study regions: Australia. A trial listing does not establish product approval or current recruitment at a particular site. Check the study’s current entry requirements.
What remains uncertain?
Read the full discussion and original sources for study limitations. Planned endpoints and completion dates are not results.

Research status alone does not establish approval, clinical benefit or local availability.

Australian sites in the registry

  • Women's and Children's Hospital Adelaide, North Adelaide, South Australia — Site status not reported
  • St Vincent's Hospital Melbourne, Fitzroy, Victoria — Site status not reported
  • Royal Melbourne Hospital, Parkville, Victoria — Site status not reported
  • Royal Children's Hospital Melbourne, Parkville, Victoria — Site status not reported

Overall recruitment does not imply availability at every site. Confirm eligibility and local recruitment with the study team.

Primary endpoints

  • Mixed-meal-stimulated mean C-peptide at week 48: 2-hour concentration–time area under the curve divided by 120 minutes (published primary outcome)

The full picture

What was tested and why it matters

In type 1 diabetes, the immune system mistakenly attacks the insulin-making beta cells in the pancreas. A key relay system inside immune cells, called the JAK-STAT pathway, helps drive that attack.1 Baricitinib is a pill that blocks two of those relays (JAK1 and JAK2). It is already approved for rheumatoid arthritis, and the protocol cites prevention and reversal findings with JAK1/2 inhibitors in diabetic mouse models as a preclinical rationale.1 The BANDIT trial asked a simple, important question: could this existing oral drug, given soon after diagnosis, slow the immune-driven loss of insulin-making cells in people?2

The trial measured preservation of remaining insulin secretion, rather than replacing insulin treatment. It did not establish a reduction in long-term diabetes complications.34

Who it was for

The trial enrolled people aged 10 to 30 who had been diagnosed with type 1 diabetes within the previous 100 days and still had measurable insulin-making function.2 It ran across centres in Australia.2

How it was designed

BANDIT was a phase 2, double-blind, randomized, placebo-controlled trial.3 Ninety-one participants were assigned in a 2-to-1 ratio to baricitinib 4 mg once daily (60 people) or matching placebo (31 people) for 48 weeks.3 Participants, investigators and outcome assessors were masked; study pharmacists held the treatment assignments.2 The main measure was the body's own insulin output — C-peptide — during a standardized "mixed-meal" test at 48 weeks.3

Key results

At 48 weeks, median stimulated mean C-peptide was 0.65 nmol/L/min with baricitinib versus 0.43 with placebo (P=0.001), supporting preservation of remaining insulin secretion. Mean daily insulin doses were 0.41 versus 0.52 U/kg/day and CGM glucose variability was 29.6% versus 33.8%. These secondary analyses were exploratory and not adjusted for multiple comparisons; variability was also lower in the baricitinib group at baseline. Average HbA1c was similar, and a time-in-range advantage was not demonstrated at week 48.3

Adverse events were reported by 47/60 baricitinib and 26/31 placebo recipients. Seven serious events occurred (two in one baricitinib recipient, five in three placebo recipients); none was attributed to study drug. One baricitinib recipient developed shingles. This small, short trial cannot establish long-term safety or exclude rare harms.3 The current US label has boxed warnings for serious infections, mortality, malignancy, cardiovascular events and thrombosis, plus a June 2026 warning about hypoglycemia in people with diabetes. Some comparative class-warning evidence comes from another JAK inhibitor in older adults with rheumatoid arthritis; those rates cannot simply be transferred to BANDIT participants.7

The August 2026 follow-up publication reports outcomes after stopping the 48-week course: 88 of the 91 randomized participants completed week 96 (58 previously assigned baricitinib, 30 placebo). Mean C-peptide was significantly greater with baricitinib at week 72 (P=0.015) but not at week 96 (P=0.336); insulin dose, HbA1c and CGM measures did not differ significantly during follow-up. The authors conclude the benefits wane over the 48 weeks after stopping treatment — a waning-benefit finding, not proof of equivalence or an established indefinite treatment regimen.6

What it means and what's next

BANDIT showed a beta-cell preservation signal with an oral JAK inhibitor over 48 weeks.3 It was not the first oral drug trial to report preservation: an imatinib trial reported an earlier, time-limited signal in 2021.5 Neither finding establishes routine T1D use or a cure. The protocol proposed further combination and earlier-stage studies.1 BARICADE-PRESERVE and BARICADE-DELAY are now testing new-onset preservation and earlier-stage delay respectively; their separate registry pages describe current recruitment. A 2025 secondary analysis explored routine clinical measures to predict treatment response; these remain research tools requiring further validation.4

Sources

  1. [1]
    Investigating the efficacy of baricitinib in new onset type 1 diabetes mellitus (BANDIT)-study protocol for a phase 2, randomized, placebo controlled trial. · Peer-reviewed study · 2022-05-23

    Waibel M, Thomas HE, Wentworth JM, et al. Investigating the efficacy of baricitinib in new onset type 1 diabetes mellitus (BANDIT) — study protocol for a phase 2, randomized, placebo controlled trial. Trials (2022).

  2. [2]
    ClinicalTrials.gov. A Phase 2, Randomised, Placebo Controlled Study Investigating the Efficacy of Baricitinib in New Onset Type 1 Diabetes Mellitus (BANDIT). *ClinicalTrials.gov* (NCT04774224) · Trial registry

    ClinicalTrials.gov. A Phase 2, Randomised, Placebo Controlled Study Investigating the Efficacy of Baricitinib in New Onset Type 1 Diabetes Mellitus (BANDIT). ClinicalTrials.gov (NCT04774224).

    ClinicalTrials.gov. BANDIT study design — randomized allocation, quadruple masking (participant, care provider, investigator, outcomes assessor). ClinicalTrials.gov (NCT04774224).

  3. [3]
    Baricitinib and β-Cell Function in Patients with New-Onset Type 1 Diabetes. · Peer-reviewed study · 2023-12-06

    Waibel M, Wentworth JM, So M, et al. Baricitinib and β-Cell Function in Patients with New-Onset Type 1 Diabetes. N Engl J Med (2023);389(23):2140-2150.

  4. [4]
    β-Cell Function Derived From Routine Clinical Measures Reports and Predicts Treatment Response to Immunotherapy in Recent-Onset Type 1 Diabetes. · Peer-reviewed study · 2025-05-27

    So M, Vogrin S, Waibel M, Kay TWH, Wentworth JM. β-Cell Function Derived From Routine Clinical Measures Reports and Predicts Treatment Response to Immunotherapy in Recent-Onset Type 1 Diabetes. Diabetes Care (2025);48(8):1370-1376.

  5. [5]
  6. [6]
    β-Cell Function and Diabetes Outcomes 1 Year After Stopping Oral Baricitinib Immunotherapy for Type 1 Diabetes · Peer-reviewed study · 2026-08-21 — 88/91 completed the off-treatment week-96 visit; no statistically significant C-peptide difference at week 96. This is follow-up of BANDIT, not an independent replication.

    So M, et al. β-Cell Function and Diabetes Outcomes 1 Year After Stopping Oral Baricitinib Immunotherapy for Type 1 Diabetes. Diabetes Care (21 August 2026).

  7. [7]
    OLUMIANT (baricitinib) US prescribing information · Regulatory decision — Revised June 2026; boxed warnings and section 5.8 hypoglycemia warning. No T1D indication.

    OLUMIANT US prescribing information, revised June 2026.