Sherene Loi, Medical Oncologist, Lab Head of Breast Oncology Clinical Trials at Peter MacCallum Cancer Centre, Australia, shared on LinkedIn:
“Recently we published in the The Lancet Oncol our papers provide important advances for both tumour infiltrating lymphocytes (TILs) as a breast cancer biomarker and the use of AI in digital pathology.
In Catalina, we independently validated computational TIL assessment across 7 randomised trials in early TNBC. AI based TIL scores were strongly prognostic, supporting their potential for scalable assessment, particularly where routine expert pathologist scoring is not available.
In Aphinity, involving more than 4,000 patients with HER2+ early BC manual, digital and AI based measures of immune infiltration were prognostic. Importantly, AI also captured spatial immune features beyond conventional TIL scoring, while high TIL levels identified patients who appeared to derive greater benefit from pertuzumab.
Together, the studies show that AI does not simply need to reproduce what a pathologist sees. It can enable robust TIL assessment at scale and potentially extract additional spatial information from routine H and E slides.
These studies also reflect more than a decade of work by the International TIL Working Group, which has developed and validated standardised approaches to TIL assessment and now includes more than 1,200 members across 59 countries.
Congrats – Julia Dixon-Douglas Luis Lara Gonzalez Damien Drubay.
Thank you to the patients and families who participated in these trials and donated their breast cancer pathology slides for future research.
New How to Score TILs iPhone and Android App — Download Here.”
Title: Artificial intelligence-based tumour infiltrating lymphocyte quantification in patients with triple-negative breast cancer: an independent validation study
Authors: Julia R Dixon-Douglas, Damien Drubay, Roberto Salgado, Balazs Acs, Jeroen van der Laak, Yinyin Yuan, Mohamed Amgad, Lee Cooper, Yeman Hagos, Khalid AbdulJabbar, James Meakin, Bram van Ginneken, Haixi Yan, Jerôme Lemonnier, Frédérique Penault-Llorca, Magali Lacroix-Triki, Heikki Joensuu, Pirkko-Liisa Kellokumpu-Lehtinen, Sybille Loibl, Carsten Denkert, Giuseppe Viale, Marco Angelo Colleoni, Christos Sotiriou, Martine Piccart, Maria Vittoria Dieci, Sandra Demaria, Roswitha Kammler, Antonio C Wolff, Joseph A Sparano, Sylvia Adams, Sunil Badve, Robert J Gray, Giuseppe Curigliano, Anne Vincent Salomon, Torsten O Nielsen, Lajos Pusztai, Francesco Ciompi, Stefan Michiels, Sherene Loi
Title: Manual, digital, and AI tumour-infiltrating lymphocyte scoring: a secondary analysis of the APHINITY randomised trial
Authors: Luis E Lara González, Fabiola Giudici, Roberto Salgado, Haixi Yan, Federico Rojo, Laura Comerma, Stephan Wienert, Jose Palacios, Zuzana Kos, Sanne Lysbet de Haas, Alvaro Rodriguez-Lescure, Giuseppe Viale, Yue Zheng, Dongxia Gao, Astrid Kiermaier, Fabrice Andre, Sibylle Loibl, Martine J Piccart, Richard Gelber, David Cameron, Ian E Krop, Peter Savas, Torsten O Nielsen, Carsten Denkert, Stefan Michiels, Sherene Loi
Other articles featuring Sherene Loi on OncoDaily.