We say “first do no harm,” then make toxicity a secondary outcome …

It’s more complicated than this, because we do first assess any intervention for safety. So, as a clinical trial process, we do adhere to the “first, do no harm” principle. However, for the clinician who will then apply those results to an individual patient, they need to know not just that the intervention bypassed early phase trials, but also how likely it is that the medication will cause certain side effects. This information comes from later-phase trials (the interventional distribution is required). For this reason, side effects should not be relegated to secondary outcomes in later-phase trials.

The most important paper of #ASCO26 just dropped in Nature Med! Or any other @asco or @myESMO meetings to be honest. When you hear someone present “the toxicities were acceptable”, please respond “downplaying toxicities is not acceptable”. Thank you @NatureMedicine for publishing… pic.twitter.com/t2HeHtb4z4

— Bishal Gyawali, MD, PhD, FASCO (@oncology_bg) June 1, 2026