Analysis of computational tumor-infiltrating lymphocytes in breast cancer from the results of the TIGER challenge
- Mart van Rijthoven
- Witali Aswolinskiy
- Leslie Tessier
- Roberto Salgado
- Jeroen van der Laak
- Francesco Ciompi
- Maschenka Balkenhol
- Joep M. A. Bogaerts
- Damien Drubay
- Laura Comerma Blesa
- Dieter Peeters
- Elisabeth Specht Stovgaard
- Anne-Vibeke Lænkholm
- Harry Haynes
- Ligia Craciun
- Denis Larsimont
- Mohamed T. Amgad
- Lee AD Cooper
- Cyril de Kock
- Valerie Dechering
- Johannes Lotz
- Nick Weiss
- Mieke van Bockstal
- Christine Galant
- Esther Lips
- Hugo M. Horlings
- Jelle Wesseling
- Lennart Mulder
- Sandra van den Belt
- Karsten Weber
- Paul Jank
- Carsten Denkert
- Enrico Munari
- Giuseppe Bogina
- Chris Russ
- Alex Lemm
- Sherene Loi
- Julia Dixon-Douglas
- Stefan Michiels
- Rogier Donders
- Scott Maurits
- Miriam Groeneveld
- Anne Mickan
- James Meakin
- Bram van Ginneken
- Heikki Joensuu
- Ming Fan
- Daehong Lee
- Jaehyung Ye
- Kangwon Byun
2026-05-15
Tumor-infiltrating lymphocytes (TILs) is a recognized prognostic biomarker in breast cancer. However, poor interobserver agreement and limited reproducibility highlight the need for computational approaches. Despite advances, adoption of computational models has been hindered by lack of standardized methods and robust benchmarks. To address this, we launched TIGER, an international competition to build open-source computational TILs (cTILs) models. Here, we present a multi-centric analysis of cTILs methods on resections and biopsies from 3,708 human epidermal growth factor receptor 2-positive (HER2+) or triple-negative breast cancers (TNBC) from clinical practice and phase 3 trials. We report benchmarks on image analysis performance, show strong agreement of cTILs with pathologists, and demonstrate positive association of cTILs with neoadjuvant therapy response in HER2+, superior to visually scored TILs. We also show that cTILs add independent prognostic information to clinical variables in TNBC resections. Data, methods and benchmarks are publicly available: https://tiger.grand-challenge.org/ .