Two cycles of blinatumomab in place of chemotherapy: four-year event-free survival 83% against 70%

Design

randomised, 1:1, open-label trial with planned interim analysis (AIEOP-BFM ALL 2017)

Population

709 children with newly diagnosed high-risk B-cell acute lymphoblastic leukaemia, randomised after consolidation

Primary outcome

event-free survival (resistance to protocol treatment, relapse, second cancer or death)

Effect

4-year event-free survival 83.0% (95% CI 77.4-87.4) vs 70.3% (63.8-75.9); HR 0.51 (0.35-0.73); treatment-related infection 23.9% vs 69.4%; neurotoxic events 12.0% vs 3.2%

AIEOP-BFM ALL 2017 randomised 709 children with newly diagnosed high-risk B-cell acute lymphoblastic leukaemia, after consolidation, to two cycles of blinatumomab or two cycles of conventional chemotherapy. At a planned interim analysis with median follow-up of 2.9 years, estimated four-year event-free survival was 83.0% with blinatumomab and 70.3% with chemotherapy, a hazard ratio of 0.51.

The toxicity comparison runs the same way rather than the opposite way, which is what makes this unusual. Treatment-related infection occurred in 23.9% of the blinatumomab group against 69.4% of the chemotherapy group. Life-threatening adverse events were 0.5% against 4.7%. A substitution that improves efficacy while removing most of the infective burden of two chemotherapy cycles is not the trade-off oncology usually deals in.

The agent has its own toxicities and they need naming rather than folding into a favourable summary. Neurotoxic events occurred in 12.0% against 3.2%, and cytokine release syndrome of grade 2 or higher in 1.1%. Both are manageable with experience, and both require a unit set up to recognise them.

The constraint in India is delivery as much as cost. Blinatumomab is given by continuous infusion across each cycle, which needs reliable venous access, an infusion pump and a service able to support a child on continuous therapy — infrastructure that is uneven outside major centres.

• Identify high-risk B-cell ALL at the point where the protocol branches; this replaced two specific post-consolidation cycles
• Confirm the unit can support continuous infusion before committing a family to the regimen
• Brief nursing staff on neurotoxicity recognition; it was four times more common than with chemotherapy
• Set out both toxicity profiles at consent, not the net comparison
• Note that this is an interim analysis reporting a four-year estimate from 2.9 years of follow-up

The statistics, in plain English

The hazard ratio of 0.51 (95% CI 0.35-0.73) means events at roughly half the rate, with the interval well below 1.0. The four-year figures are Kaplan-Meier estimates projected from a median follow-up of 2.9 years, so they extrapolate beyond the observed data; their intervals, 77.4-87.4 and 63.8-75.9, reflect that uncertainty. Trials reported at a planned interim analysis that crosses an efficacy boundary tend to overstate the effect size somewhat, because they stop at a favourable moment in the random fluctuation of the estimate. The direction of benefit here is not in doubt; the precise 13-point gap probably is.

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.