StrategY And Study Design

Discordant evidence is the norm in clinical development. So what should developers do when they encounter inconsistent results with an experimental therapy? The graph below depicts the state of evidence for an anti-tuberculosis drug candidate, moxifloxacin, as of 2011. After five phase 2 trials produced conflicting results, the tuberculosis research community faced the “go/no-go” decision point, where they had to decide if the evidence was compelling enough to initiate a pivotal (and expensive) phase 3 trial.

Bokeh Plot

What does this show?

This graph is intended to help model possible next steps for this drug’s development, with the blue nodes representing the options. Does the accumulating state of evidence justify another study? And if so, what type of study should that be? While the graph does not answer those questions (indeed, there are probably not objective answers to those questions), it can help to structure and broker that discussion across stakeholders. It also provides a systematic way of thinking about these decision points and optimal development strategies in general.