Good question @SSilverstein. With the small sample size, it's...

Good question @SSilverstein.

With the small sample size, it's hard to say whether (1) the proportion (2/7 or ~29%) is clinically meaningful; (2) we'll see the same pattern in a larger sample, and (3) most importantly, whether this will affect the efficacy of the drug over the long term. The authors cited a study which found that anti-drug anti-bodies do not impact drug effects in diabetes.

IMO, it demonstrates how efficacious Exenatide was in lowering ICP over 12 weeks: despite the fact that almost 1/3 of the treatment group developed antibodies, there was still a significant difference in ICP between the treatment and placebo arms (and even in such a small sample).