Disease modification arrives in type 1 diabetes

For a century, treating type 1 diabetes has meant replacing what the immune system destroys. Insulin is lifesaving, and it is relentless; it manages the consequence of the disease. It does nothing about the cause: a slow autoimmune assault on the insulin-producing β-cells that is often well advanced by the time anyone notices. Teplizumab is the first therapy to step upstream of that, and in 2026 it moved further into routine pediatric care. The question now is how much it changes, and what it asks of the system next.

From managing glucose to interrupting the immune attack

Teplizumab is a humanized, Fc-nonbinding anti-CD3 monoclonal antibody; the prevailing account of its mechanism is that it shifts the T-cell balance toward immune tolerance, blunting the attack on β-cells.⁴ The pivotal evidence came from at-risk relatives with preclinical (Stage 2) disease: a single 14-day course delayed the median time to clinical diagnosis from about 24 to roughly 48 months — a hazard ratio of 0.41.¹ That result, in 2022, made teplizumab the first drug cleared to delay the onset of type 1 diabetes, and it reframed the disease as something one might intervene in before it fully arrives.

What the PROTECT trial showed

The 2026 expansion rests on a different and harder question: once a child is already diagnosed, can the drug protect what β-cell function remains? The PROTECT trial — 328 newly-diagnosed children and adolescents, two 12-day courses — answered the narrow version of that question yes. Stimulated C-peptide, a direct readout of surviving β-cell function, was significantly higher at 78 weeks on teplizumab than on placebo, and far more treated children held onto a clinically meaningful level.² Preservation of β-cell function was demonstrated.

But the trial's key secondary endpoints — insulin dose, HbA1c, time in target glucose range, serious hypoglycemia — did not differ significantly between groups at that point.² The FDA's 2026 clearance was an accelerated approval, granted on C-peptide as a surrogate "reasonably likely to predict" clinical benefit.³ Teplizumab preserves β-cell function; whether that preservation reaches the outcomes a child and family actually live is not yet shown, and confirming it is the obligation the accelerated pathway leaves open.

The binding constraint is detection

The strategy inverts here. Both the delay indication and the preserve indication only work early: in Stage 2, or within about six weeks of a Stage 3 diagnosis, before there is too little β-cell left to protect.³ Yet most type 1 diabetes is still found late, frequently at the point of diabetic ketoacidosis, when that window has closed. A disease-modifying drug therefore makes the long-academic project of staging type 1 diabetes operational: autoantibody screening is worth building only because there is finally something to do with an early answer. The rate-limiting step for impact has moved off the drug and onto detection infrastructure: who gets screened, how, and who pays for finding a disease years before it declares itself. Today that treatable window opens at Stage 2. Whether treating even earlier, in Stage 1, when more β-cell mass still survives, would help is unproven and would need its own trials, and it becomes a question worth asking only once screening can find children that early.

What it means

Teplizumab is a real first, and it warrants a careful reading. It converts type 1 diabetes from a condition you only manage into one you can, in a narrow and early window, modestly modify. The next decade of value will turn on whether the system can find the disease early enough to use the drug, and whether preserved β-cell function earns its accelerated approval by translating into outcomes patients actually experience. The drug has settled the narrow question and opened a larger one for the field.

← Back to Insights