Naoto T Ueno, Director of the University of Hawaii Cancer Center, shared a post by Eric Topol, Founder and Director of Scripps Research Translational Institute at The Scripps Research Institute, on X:
“This is a very nice paper on predicting immunotherapy outcomes in a pan-cancer setting. The results are impressive, and the approach is scientifically promising.
However, this is not yet ready for clinical use. The analysis is retrospective, even if the data were prospectively collected, and it relies on de-identified datasets. The next critical step is rigorous validation in truly prospective clinical studies using identifiable patient-level data to determine whether the model has sufficient real-world predictive power, particularly with respect to PPV and NPV.”
Quoting Eric Topol’s post:
“Using AI to improve cancer immunotherapy outcomes, via training from transcriptomes of 10,000 tumor samples, 33 cancer types.”
Title: Generalizable AI predicts immunotherapy outcomes across cancers and treatments
Authors: Wanxiang Shen, Intae Moon, Thinh H. Nguyen, Michelle M. Li, Yepeng Huang, Nitya Nair, Daniel Marbach, Marinka Zitnik
Other articles featuring Naoto T Ueno and Eric Topol on OncoDaily