Aakash Desai: Can AI Improve How We Stratify Patients for Immunotherapy in Advanced NSCLC?
Aakash Desai, Associate Director, Phase 1 and Precision Oncology Program at UAB O’Neal Comprehensive Cancer Center, shared on LinkedIn:
“Can AI improve how we stratify patients for immunotherapy in advanced NSCLC?
A study published in JTO International Association for the Study of Lung Cancer suggests: YES!
Using samples from IMpower110 and IMpower150 (n > 1275), researchers validated AIM-PD-L1, a clone-agnostic AI tool for digital PD-L1 scoring in tumor cells.
Key findings:
- Strong agreement between AI and manual reads at both 1% and 50% TC thresholds
- AI identified PD-L1+ patients missed by manual scoring—and these patients had numerically better OS on atezolizumab
- Several human-interpretable features (HIFs), like PD-L1+ cell density in tumor epithelium, were linked to PFS/OS outcomes
- Potential for AI-based scoring to support more accurate, reproducible patient selection for anti–PD-(L)1 therapies
This work reflects a broader trend: AI/ML is reshaping biomarker analysis, decision support, and digital pathology. Looking ahead, such tools may reduce variability and accelerate precision in immunotherapy access.”
Title: Digital Versus Manual PD-L1 Scoring in Advanced Non-Small Cell Lung Cancer From the IMpower110 and IMpower150 Trials
Authors: Roy S. Herbst, Hen Prizant, Daniel Ruderman, Jake Conway, John Shamshoian, Hartmut Koeppen, Wei Zou, Filippo de Marinis, Giuseppe Giaccone, Jacek Jassem, David R. Spigel, Mark A. Socinski, Martin Reck, Luciana Molinero, Marcus Ballinger, David Shames, Michael Griffin, Limin Yu, Nishant Agrawal, Andrew Beck, Ilan Wapinski, Stephanie Hennek, Jennifer Giltnane, Minu K. Srivastava
More posts featuring Aakash Desai.
-
Challenging the Status Quo in Colorectal Cancer 2024
December 6-8, 2024
-
ESMO 2024 Congress
September 13-17, 2024
-
ASCO Annual Meeting
May 30 - June 4, 2024
-
Yvonne Award 2024
May 31, 2024
-
OncoThon 2024, Online
Feb. 15, 2024
-
Global Summit on War & Cancer 2023, Online
Dec. 14-16, 2023