Aakash Desai, Assistant Professor and Associate Director of Phase 1 and Precision Oncology Program at the UAB O’Neal Comprehensive Cancer Center, shared a post on LinkedIn:
“Translation of ADCs targeting TROP-2 and cMET into NSCLC clinical practice is currently hindered by subjective visual scoring and inconsistent reproducibility in biomarker assessment. New publication in AACR Journals shows AI can enable this biomarker quantification.
Key findings:
- Primary endpoint: Recapitulated pathologist annotations for TROP-2 with r = 0.98-0.99
- Key efficacy: Zero-shot generalization to cMET with r = 0.99 and CCC = 0.96 ▪ Safety signal: Not applicable (computational pathology study)
How this fits: Foundation model-based scoring produces expert-level, scalable biomarker quantification.
The resulting TME phenotypes – TROP-2-high immune-deserted vs cMET-high, immune-active – reveal therapeutic implications for combining ADCs with immunotherapies or kinase inhibitors.
What’s next: These findings suggest that automated, modular AI pipelines could potentially standardize ADC biomarker quantification across different protein targets in non-small cell lung cancer research.”
Title: ADC target profiling in NSCLC: Generalizable AI separates TROP-2 and cMET phenotypes
Authors: Philipp Anders, Marvin Sextro, Katja Lingelbach, Kai Standvoss, Suhas Pandhe, Sandip Ghosh, Cornelius Böhm, Stephan Tietz, Rosemarie Krupar, Lars Tharun, Marie-Lisa Eich, Julika Ribbat-Idel, Evelyn Ramberger, Xizi Liang, Verena Aumiller, Sabine Merkelbach-Bruse, Alexander Quaas, Nikolaj Frost, Georg Schlachtenberger, Matthias Heldwein, Ulrich Keilholz, Khosro Hekmat, Jens-Carsten Rückert, Reinhard Büttner, Christian Grohe, David Horst, Maximilian Alber, Lukas Ruff, Frederick Klauschen, Gabriel Dernbach, Philipp Seegerer, Simon Schallenberg
Read the Full Article.
Other articles featuring Aakash Desai on OncoDaily.
