Adaptive platform trial: golimumab in stage 1 (presymptomatic) T1D

What is being tested, and why it matters

Type 1 diabetes (T1D) starts long before anyone needs insulin. The immune attack on the insulin-making cells can be detected years earlier through islet autoantibodies in the blood, and it moves through recognised stages: autoantibodies with normal blood sugar (stage 1), autoantibodies with abnormal blood sugar but no symptoms (stage 2), and finally clinical diabetes (stage 3).

Almost every disease-modifying drug tried so far has been given at stage 2 or stage 3 — after the glucose has already started to drift. This trial goes earlier. It asks whether golimumab, an antibody that blocks TNF (a signalling molecule the immune system uses to drive inflammation), can be given while glucose tolerance is still completely normal and keep it that way for longer.¹

That is a meaningful step for this particular drug. Golimumab has already been shown, in the T1GER trial, to preserve insulin production in young people newly diagnosed with T1D — a placebo-controlled phase 2 study in 84 participants aged 6 to 21 in which C-peptide production was better preserved at 52 weeks on golimumab than on placebo.² This new study is the first attempt to take it from treating the disease to trying to postpone it.

The platform design

This is not a single conventional trial but the first substudy of an adaptive platform trial run by the NIH's National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK).¹ A platform trial keeps one shared protocol, one shared control group and one shared endpoint, and swaps candidate drugs in and out of it. Golimumab is the drug it is starting with; the structure exists so that further agents can be added without building a new trial from scratch each time.

The registry lists NIDDK as sponsor with no named collaborator. The eligibility text refers to TrialNet's immunisation requirements, which strongly suggests it runs through the TrialNet network, but TrialNet is not formally listed on the registry record — so treat that as an inference, not a fact.¹

Who it is for

People aged 2 to 44 who weigh at least 15 kg, have either confirmed IA-2A alone or two or more islet autoantibodies, still have normal glucose tolerance on an oral glucose tolerance test, and have never been treated with insulin.¹ In practice these are people found through autoantibody screening — relatives enrolled in programmes like TrialNet's Pathway to Prevention, or children picked up by general-population screening.

Enrolment is about 250 people. We say "about" deliberately: the registry header gives 255, while the study's own detailed description says 225 randomised (150 to golimumab, 75 to placebo). The registry has not reconciled the two figures, and we are not going to pick one silently.¹

How it is designed

Participants are randomised 2-to-1 to golimumab or matching placebo. Golimumab is injected under the skin, with loading doses at weeks 0 and 2 and monthly dosing after that, at a weight-based dose (100 mg then 50 mg for those 15-40 kg; 200 mg then 100 mg for those 40 kg and over). Dosing continues for as long as a participant is enrolled — for the earliest enrollees that could be around six years, since the plan is roughly three years of recruitment plus about three years of follow-up.¹

The primary endpoint is time from randomisation to confirmed dysglycemia or clinically diagnosed T1D, using American Diabetes Association criteria.¹ In the registry's entry criteria, dysglycemia means a fasting plasma glucose of at least 100 mg/dL, a 2-hour glucose of at least 140 mg/dL (7.8 mmol/L), or a 30-, 60- or 90-minute glucose of at least 200 mg/dL (11.1 mmol/L) on an oral glucose tolerance test. (One caution: the registry pairs 100 mg/dL with "6.1 mmol/L", which is not a correct conversion — 100 mg/dL is 5.6 mmol/L. We quote the mg/dL figures, which are the standard thresholds, and flag the registry's unit conversion as unreliable.)¹

There is one further oddity worth naming rather than papering over: the detailed description scopes the primary outcome to "persons less than 18 years of age", even though eligibility runs from 2 to 44. We do not know which is intended.¹

Where things stand

The record was first posted on 6 July 2026 and the trial is not yet recruiting. The estimated start date is 8 August 2026, with primary completion in September 2032. Only one site — Joslin Diabetes Center in Boston — is listed so far, so this is not yet a broad multi-site network, whatever it may become.¹

What to watch for

Nothing here is a result yet. This is a registration, and registrations slip. The reasons to pay attention anyway: it targets the earliest point in the disease anyone has yet tried to treat with a drug; it re-tests an agent that already has a positive signal at diagnosis, which is a better starting position than most prevention candidates have; and the platform structure means a negative result for golimumab would not be the end of the effort, only the end of one arm. The counterweight is that long-term TNF blockade in healthy-glucose children is a serious ask, and six years is a long time to take an immunosuppressive drug on the strength of a blood test.

References

1. National Institute of Diabetes and Digestive and Kidney Diseases. Adaptive Platform Trial to Delay Progression From Normoglycemia to Dysglycemia in Presymptomatic Type 1 Diabetes: Golimumab Substudy (NCT07683026). ClinicalTrials.gov, first posted 6 July 2026. https://clinicaltrials.gov/study/NCT07683026 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

2. Quattrin T, Haller MJ, Steck AK, et al. Golimumab and Beta-Cell Function in Youth with New-Onset Type 1 Diabetes (T1GER). New England Journal of Medicine 2020;383:2007-2017. https://doi.org/10.1056/NEJMoa2006136 ↩