The prognostic significance of tumor-infiltrating lymphocytes (TILs) in breast cancer has been recognized for over a decade. Although histology-based scoring recommendations exist to standardize visual TILs assessment, interobserver agreement and reproducibility are hampered by heterogeneous infiltration patterns, highlighting the importance of computational approaches. Despite advances to automate TILs quantification, adoption of computational models has been hindered by lack of consensus on scoring methods and lack of large-scale benchmarks. To address these limitations, we launched the international TIGER challenge, a public competition to build open-source computational TILs (cTILs) models in digital pathology. Here, we present the largest comprehensive multi-centric validation of multiple cTILs methods on surgical resections and biopsies using 3,708 Triple Negative Breast Cancer (TNBC) and human epidermal growth factor receptor 2 positive (HER2+) breast cancers from clinical practice and phase 3 clinical trials. We report benchmarks on image analysis performance of each method and show the strong agreement of cTILs with panels of pathologists. We show the positive association of cTILS with response after neoadjuvant therapy in HER2-positive, superior to visually scored TILs. We also show that cTILs add independent information to clinical variables in surgically resected TNBC but not in HER2-positive disease and breast biopsies.

FC was Chair of the Scientific and Medical Advisory Board of TRIBVN Healthcare, France, and received advisory board fees from TRIBVN Healthcare, France, in the last five years. He is a shareholder of Aiosyn BV, the Netherlands. MB is medical advisor at Aiosyn BV. JvdL was a member of the advisory boards of Philips, the Netherlands and ContextVision, Sweden, and received research funding from Philips, the Netherlands, ContextVision, Sweden, and Sectra, Sweden in the last five years. He is chief scientific officer (CSO) and shareholder of Aiosyn BV, the Netherlands. RS reports non-financial support from Merck, Case 45 and Bristol Myers Squibb; research support from Merck, Puma Biotechnology and Roche; personal fees from Roche, BMS, AstraZeneca, Daiichi Sankyo and Exact Sciences for advisory boards. SL receives research funding from Novartis, Bristol Myers Squibb, Puma Biotechnology, AstraZeneca/Daiichi Sankyo, Roche-Genentech and Seattle Genetics. SL has acted as consultant to Roche-Genentech, MSD, Gilead Sciences, AstraZeneca/Daiichi Sankyo, Bristol Myers Squibb, Novartis, Eli Lilly, Amaroq Therapeutics, Mersana Therapeutics, Domain Therapeutics, BioNTech, Bicycle Therapeutics, Exact Sciences. SL holds stock of Bicycle Therapeutics. LCB reports receiving personal fees from Roche, MSD, AstraZeneca and Diaceutics, non-financial support from Roche, MSD, AstraZeneca and Phillips and held advisory roles for Roche and AstraZeneca. MvB served on advisory boards for AstraZeneca and Sakura, and received research funding from Roche Diagnostics, all unrelated to the present work. She also received funding from the not-for-profit organization 'Foundation Against Cancer' (grant number 2019-089) and the Fondation Saint-Luc. HJ reports grants or contracts from Mersana Therapeutics and Defense Therapeutics, consulting fees from Orion Pharma, patents planned, issues or pending with Sartar Therapeutics, advisory board role in Orion Pharma, position in the board of Maud Kuistila Foundation and stock ownership of Orion Pharma and Sartar Therapeutics. AM is co-founder and CSO of Pathcore. NMR is the co-founder, CEO, and CSO, as well as a shareholder of Histofy Ltd. He is also the GSK Chair of Computational Pathology and is in receipt of research funding from GSK and AstraZeneca. The authors affiliated with VUNO Inc. conducted this research while employed at VUNO Inc., and their contributions to this study were part of their professional responsibilities at the company. SX, ZJ, FX and JK are all employees of Bio-totem Pte Ltd, and SX is also a shareholder of Bio-totem. AMT is a co-founder and Chief Scientific Officer of Spotlight Pathology Ltd. CD reports other support from MSD Oncology, personal fees from Daiichi Sankyo and AstraZeneca, and grants from Myriad Genetics and German Breast Group outside the submitted work; he also reports a patent for VMscope digital pathology software with royalties paid, a patent for WO2020109570A1 issued, a patent for WO2015114146A1 issued, and a patent for WO2010076322A1 issued. PJ reports grants and travel expenses from Gilead Sciences GmbH outside the submitted work. MvR is currently employed at Ellogon.ai, Netherlands. All other authors declare no conflict of interest.

This project has received funding from the Dutch Cancer Society (PROACTING Project Number 11917), from the Alpe dHuZes / Dutch Cancer Society Fund, Grant Number KUN 2014-7032, from the European Union's Horizon 2020 research and innovation programme under grant agreement No 825292 (ExaMode, https://www.examode.eu/), from the Dutch Research Council NWO-TTW VIDI project (IGNITE), project number 18388, from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement No 945358 (BigPicture, https://www,bigpicture.eu/), from the Breast Cancer Research Foundation (BCRF, TIL-challenge). Research at the Netherlands Cancer Institute is supported by institutional grants from the KWF and the Dutch Ministry of Health, Welfare and Sport. Sherene Loi is supported by the National Breast Cancer Foundation of Australia Endowed Chair and the Breast Cancer Research Foundation, New York.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

The requirement for ethical approval to use data from the RUMC–TNBC and RUMC–HER2+ cohorts was waived by the institutional review board (case number 2015–1711) of the Radboud University Medical Center (Radboudumc). For the FinHer trial, an ethics committee at Helsinki University Hospital approved the study and study participants provided written informed consent (trial identifier: ISRCTN76560285). For DIGITILS, the institutional ethics committee from Cliniques universitaires Saint-Luc (Brussels, Belgium) approved this study (file number: RETRO–TNBC–15–2019/03JUL/297 for the TNBC cohort, file name: RETRO–HER2–15 for the HER2+ cohort). The approval to use slides from the GBG was given at two institutions: Charite ethic committee approval number: EA1/139/05; Philipps-University Marburg ethic committee approval number: 38/20. The use of the slides from SCDC for the study was approved by the Ethics Committee for Clinical Research of the Provinces of Verona and Rovigo under number 25046. The use of the slides from NKI for the study was approved by the institutional review board of the Netherlands Cancer Institute under number CFMPB737.

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes