single-arm trial

In most trials participants are split at random, with one group receiving the treatment and another receiving a placebo or the existing standard of care. A single-arm trial has no such second group. Everyone enrolled gets the treatment, and the results are compared against something outside the study: how patients with the condition have fared historically, or data from a separate group of patients.

Regulators do not use the phrase as their formal term. The international guideline on choice of control group speaks instead of an externally controlled trial, defined as one in which the control group consists of patients who are not part of the same randomised study as the group receiving the investigational agent. Its warning is direct: the inability to control bias restricts use of the external control design to situations in which the effect of treatment is dramatic and the usual course of the disease highly predictable.

So this is weaker evidence by design, and everyone involved knows it. Without a comparison group enrolled at the same time, an improvement could reflect the treatment, or better care generally, or which patients happened to be enrolled. It is accepted anyway when a condition is rare, serious and has no good alternative, and the US Food and Drug Administration says plainly that while single-arm trials have been commonly used to support accelerated approval, a randomised controlled trial is the preferred approach. A result from one is a reason to keep going, not a settled answer.