Artificial intelligence‐based digital scores of stromal tumour‐infiltrating lymphocytes and tumour‐associated stroma predict disease‐specific survival in triple‐negative breast cancer

Triple‐negative breast cancer (TNBC) is known to have a relatively poor outcome with variable prognoses, raising the need for more informative risk stratification. We investigated a set of digital, artificial intelligence (AI)‐based spatial tumour microenvironment (sTME) features and explored their prognostic value in TNBC. After performing tissue classification on digitised haematoxylin and eosin (H&E) slides of TNBC cases, we employed a deep learning‐based algorithm to segment tissue regions into tumour, stroma, and lymphocytes in order to compute quantitative features concerning the spatial relationship of tumour with lymphocytes and stroma. The prognostic value of the digital features was explored using survival analysis with Cox proportional hazard models in a cross‐validation setting on two independent international multi‐centric TNBC cohorts: The Australian Breast Cancer Tissue Bank (AUBC) cohort (n = 318) and The Cancer Genome Atlas Breast Cancer (TCGA) cohort (n = 111). The proposed digital stromal tumour‐infiltrating lymphocytes (Digi‐sTILs) score and the digital tumour‐associated stroma (Digi‐TAS) score were found to carry strong prognostic value for disease‐specific survival, with the Digi‐sTILs and Digi‐TAS scores giving C‐index values of 0.65 (p = 0.0189) and 0.60 (p = 0.0437), respectively, on the TCGA cohort as a validation set. Combining the Digi‐sTILs feature with the patient's positivity status for axillary lymph nodes yielded a C‐index of 0.76 on unseen validation cohorts. We surmise that the proposed digital features could potentially be used for better risk stratification and management of TNBC patients. © 2023 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

[1]  L. Carey,et al.  Triple negative breast cancer: Pitfalls and progress , 2022, NPJ breast cancer.

[2]  A. Green,et al.  Standardization of the tumor-stroma ratio scoring method for breast cancer research , 2022, Breast Cancer Research and Treatment.

[3]  A. Madabhushi,et al.  Computational features of tumor-infiltrating lymphocyte architecture of residual disease after chemotherapy on H&E images as prognostic of overall and disease-free survival for triple-negative breast cancer. , 2021 .

[4]  Nasir M. Rajpoot,et al.  Simple non-iterative clustering and CNNs for coarse segmentation of breast cancer whole-slide images , 2021, Medical Imaging.

[5]  M. Dieci,et al.  Immune Infiltrates in Breast Cancer: Recent Updates and Clinical Implications , 2021, Cells.

[6]  X. Bian,et al.  Triple-negative breast cancer molecular subtyping and treatment progress , 2020, Breast Cancer Research.

[7]  Andrew H. Beck,et al.  Pitfalls in assessing stromal tumor infiltrating lymphocytes (sTILs) in breast cancer , 2020, npj Breast Cancer.

[8]  J. A. van der Laak,et al.  Histological subtypes in triple negative breast cancer are associated with specific information on survival. , 2020, Annals of Diagnostic Pathology.

[9]  H. Putter,et al.  The prognostic value of the tumor–stroma ratio is most discriminative in patients with grade III or triple‐negative breast cancer , 2020, International journal of cancer.

[10]  Ronnachai Jaroensri,et al.  Artificial intelligence in digital breast pathology: Techniques and applications , 2019, Breast.

[11]  S. Loi,et al.  Targeting immune pathways in breast cancer: review of the prognostic utility of TILs in early stage triple negative breast cancer (TNBC). , 2019, Breast.

[12]  Yeqi Bai,et al.  Automated brain histology classification using machine learning , 2019, Journal of Clinical Neuroscience.

[13]  M. Gurcan,et al.  Digital pathology and artificial intelligence. , 2019, The Lancet. Oncology.

[14]  Francesco Ciompi,et al.  Deep learning and manual assessment show that the absolute mitotic count does not contain prognostic information in triple negative breast cancer , 2019, Cellular Oncology.

[15]  F. Bertucci,et al.  Infiltrating stromal immune cells in inflammatory breast cancer are associated with an improved outcome and increased PD-L1 expression , 2019, Breast Cancer Research.

[16]  Jin Tae Kwak,et al.  Hover-Net: Simultaneous segmentation and classification of nuclei in multi-tissue histology images , 2018, Medical Image Anal..

[17]  Joseph P. Romano,et al.  MULTIPLE DATA SPLITTING FOR TESTING By , 2019 .

[18]  N. Borcherding,et al.  The clinical promise of immunotherapy in triple-negative breast cancer , 2018, Cancer management and research.

[19]  R. Tollenaar,et al.  The prognostic value of tumour–stroma ratio in primary breast cancer with special attention to triple-negative tumours: a review , 2018, Breast Cancer Research and Treatment.

[20]  Rajarsi R. Gupta,et al.  Spatial Organization and Molecular Correlation of Tumor-Infiltrating Lymphocytes Using Deep Learning on Pathology Images. , 2018, Cell reports.

[21]  Nasir M. Rajpoot,et al.  Context-Aware Learning using Transferable Features for Classification of Breast Cancer Histology Images , 2018, ICIAR.

[22]  S. Gallus,et al.  Clinical and pathological factors influencing survival in a large cohort of triple-negative breast cancer patients , 2018, BMC Cancer.

[23]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..

[24]  J. Balko,et al.  Triple-negative breast cancer: challenges and opportunities of a heterogeneous disease , 2016, Nature Reviews Clinical Oncology.

[25]  N. Rajpoot,et al.  Locality Sensitive Deep Learning for Detection and Classification of Nuclei in Routine Colon Cancer Histology Images , 2016, IEEE Trans. Medical Imaging.

[26]  I. Fernández,et al.  Tumor-Infiltrating Lymphocytes in Triple Negative Breast Cancer: The Future of Immune Targeting , 2016, Clinical Medicine Insights. Oncology.

[27]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[28]  Nassir Navab,et al.  Fast Training of Support Vector Machines for Survival Analysis , 2015, ECML/PKDD.

[29]  T. Nielsen,et al.  The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: recommendations by an International TILs Working Group 2014. , 2015, Annals of oncology : official journal of the European Society for Medical Oncology.

[30]  Dong Yu,et al.  Deep Learning: Methods and Applications , 2014, Found. Trends Signal Process..

[31]  K. Hess,et al.  The prognostic impact of age in patients with triple-negative breast cancer , 2013, Breast Cancer Research and Treatment.

[32]  R. Vink,et al.  The prognostic value of tumour-stroma ratio in triple-negative breast cancer. , 2012, European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology.

[33]  Matthew W. Conklin,et al.  2012 Landes Bioscience. Do not distribute. Why the stroma matters in breast cancer Insights into breast cancer patient outcomes through the examination of stromal biomarkers , 2012 .

[34]  Aedín C. Culhane,et al.  survcomp: an R/Bioconductor package for performance assessment and comparison of survival models , 2011, Bioinform..

[35]  Andrew H. Beck,et al.  Systematic Analysis of Breast Cancer Morphology Uncovers Stromal Features Associated with Survival , 2011, Science Translational Medicine.

[36]  M. Pencina,et al.  On the C‐statistics for evaluating overall adequacy of risk prediction procedures with censored survival data , 2011, Statistics in medicine.

[37]  Hein Putter,et al.  Tumor–stroma ratio in the primary tumor is a prognostic factor in early breast cancer patients, especially in triple-negative carcinoma patients , 2011, Breast Cancer Research and Treatment.

[38]  N Harbeck,et al.  Triple-negative breast cancer--current status and future directions. , 2009, Annals of oncology : official journal of the European Society for Medical Oncology.

[39]  Debra L Winkeljohn.  Triple-negative breast cancer. , 2008, Clinical journal of oncology nursing.

[40]  Balaji Krishnapuram,et al.  On Ranking in Survival Analysis: Bounds on the Concordance Index , 2007, NIPS.

[41]  S. Narod,et al.  Triple-Negative Breast Cancer: Clinical Features and Patterns of Recurrence , 2007, Clinical Cancer Research.

[42]  Ian O Ellis,et al.  Prognostic markers in triple‐negative breast cancer , 2007, Cancer.

[43]  F. Harrell,et al.  Prognostic/Clinical Prediction Models: Multivariable Prognostic Models: Issues in Developing Models, Evaluating Assumptions and Adequacy, and Measuring and Reducing Errors , 2005 .