Dosing ATG to monitored exposure improved GVHD-free, relapse-free survival in a phase 3 trial

Design

Randomised, multicentre phase 3 trial, TDM-based targeted dosing vs fixed 10 mg/kg

Population

204 patients having unmanipulated haploidentical haematopoietic stem cell transplantation

Primary outcome

365-day graft-versus-host-disease-free and relapse-free survival

Effect

66.7% vs 50.0%; HR 0.666 (95% CI 0.4456 to 0.9954), P=0.048

Anti-thymocyte globulin is standard prophylaxis against graft-versus-host disease, but its pharmacokinetics vary widely between patients, and weight-based fixed dosing may not match exposure to need. This multicentre, randomised phase 3 trial in haploidentical stem cell transplantation assigned 204 patients 1:1 to a dose targeted by therapeutic drug monitoring or to a fixed dose of 10 mg/kg.

The primary endpoint, one-year graft-versus-host-disease-free and relapse-free survival, was 66.7% with targeted dosing and 50.0% with fixed dosing (hazard ratio 0.666, 95% CI 0.4456 to 0.9954; P=0.048). Moderate to severe chronic graft-versus-host disease at one year was 9.8% vs 22.5% (P=0.026). Grade 3 to 5 infections were 44.1% vs 70.6% (P<0.001). More patients in the targeted arm reached optimal ATG exposure, with better CD4 T-cell reconstitution.

The primary result is borderline, with an interval whose upper limit nearly reaches 1, and the trial was in a specific transplant setting. It supports individualised dosing where ATG assays and pharmacokinetic expertise exist; it does not show an overall survival gain.

• Where assays and pharmacological support exist, consider exposure-guided ATG dosing in haploidentical transplant.
• Targeted dosing reduced chronic graft-versus-host disease and severe infection in this trial.
• The primary benefit was borderline (upper confidence limit 0.9954), so confirmation matters.
• Plan access to drug monitoring before adopting the strategy.
• Overall survival was not reported in the abstract.

Why it matters

It shows that matching drug exposure to the patient, not the weight, changed clinical outcomes in a drug used in routine transplant practice.

Don't overread it

The primary endpoint was borderline, the setting was haploidentical transplantation, and overall survival was not reported in the abstract.

The statistics, in plain English

A hazard ratio of 0.666 with an interval of 0.4456 to 0.9954 only just excludes no effect, and P=0.048 is close to the 0.05 threshold. A trial of 204 patients can shift with a few events, so the secondary results are supportive rather than independent proof.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.