← News

The first human dose, informed by a QSP model

FDA's June 2026 draft guidance describes how a quantitative systems pharmacology model can inform a MABEL-based first-in-human dose, and when a start above MABEL may be justified.

In June 2026, CDER's Office of Clinical Pharmacology published a draft guidance on dose selection for the minimum anticipated biological effect level (MABEL) in first-in-human trials, informed by quantitative systems pharmacology (QSP) models. It applies to drugs for which a MABEL approach is recommended. That can include, but is not limited to, drugs that cause T-cell activation and cytokine release. FDA notes that the approach has the potential to reduce reliance on animal toxicology studies.

The draft sits within FDA's wider set of actions to speed development from the IND to the pivotal trial. That package also included an Expedited IND Pilot, whose application window closed on 30 October 2026.

What the IND should show

The guidance is written consistent with ICH M15. The IND package should state the model's intended context of use, include a model risk assessment, and present sensitivity and uncertainty analyses. In other words, the model's purpose, its risk and its uncertainty are documented up front, as for any other model-informed decision.

Where time can be won

MABEL is, by definition, a low-end dose. The guidance states that in some cases, starting at a minimal active dose higher than the absolute MABEL may be justified, and that QSP models can help determine that minimal active dose. In life-threatening disease, this is where the model earns its place.

Where the guidance asks for caution

  • When the target or mechanism of action is novel, additional safety margins should be considered.
  • For a novel target without precedent, when the dose estimates disagree, sponsors should select the most conservative (lowest) starting dose.
  • The protocol should plan to update the model with data from the first participants.

Our read: in oncology, the bigger time gain may come less from faster IND review than from fewer patients treated at doses too low to work. For human-specific targets, the practical question is how the first-in-human dose is justified today: classic MABEL estimates, or a mechanistic model whose context of use and risk are assessed before the IND.

Sources

Download the slides (PDF)Discuss it on LinkedIn
Cover slide: The first human dose, informed by a QSP model
The slides, as a PDF

Is this question in front of your program?

A 15-minute call is enough to tell whether modeling can move it. A fixed proposal follows within 48 hours.