[aggie]

"Small n" is not always a problem. They are not trying to tease out a small effect size between two treatments. This is a strong signal -- literally children can now hear when they couldn't before. A small sample size does not allow statistical noise to produce that effect.

[justrealist]

Yeah the FDA is also being pretty lenient about allowing single-arm trials for gene therapy for this reason. The effect size needs to be obvious, but there's no reason to subject 3/6 deaf children to being a control if the risk is relatively low and the odds of success are high.

(also, for many of theses diseases, there's no plan B -- if it's not fixed in early childhood, you can never interpret sound).