Multiscale deep learning framework captures systemic immune features in lymph nodes predictive of triple negative breast cancer outcome in large‐scale studies

The suggestion that the systemic immune response in lymph nodes (LNs) conveys prognostic value for triple‐negative breast cancer (TNBC) patients has not previously been investigated in large cohorts. We used a deep learning (DL) framework to quantify morphological features in haematoxylin and eosin‐stained LNs on digitised whole slide images. From 345 breast cancer patients, 5,228 axillary LNs, cancer‐free and involved, were assessed. Generalisable multiscale DL frameworks were developed to capture and quantify germinal centres (GCs) and sinuses. Cox regression proportional hazard models tested the association between smuLymphNet‐captured GC and sinus quantifications and distant metastasis‐free survival (DMFS). smuLymphNet achieved a Dice coefficient of 0.86 and 0.74 for capturing GCs and sinuses, respectively, and was comparable to an interpathologist Dice coefficient of 0.66 (GC) and 0.60 (sinus). smuLymphNet‐captured sinuses were increased in LNs harbouring GCs (p < 0.001). smuLymphNet‐captured GCs retained clinical relevance in LN‐positive TNBC patients whose cancer‐free LNs had on average ≥2 GCs, had longer DMFS (hazard ratio [HR] = 0.28, p = 0.02) and extended GCs' prognostic value to LN‐negative TNBC patients (HR = 0.14, p = 0.002). Enlarged smuLymphNet‐captured sinuses in involved LNs were associated with superior DMFS in LN‐positive TNBC patients in a cohort from Guy's Hospital (multivariate HR = 0.39, p = 0.039) and with distant recurrence‐free survival in 95 LN‐positive TNBC patients of the Dutch‐N4plus trial (HR = 0.44, p = 0.024). Heuristic scoring of subcapsular sinuses in LNs of LN‐positive Tianjin TNBC patients (n = 85) cross‐validated the association of enlarged sinuses with shorter DMFS (involved LNs: HR = 0.33, p = 0.029 and cancer‐free LNs: HR = 0.21 p = 0.01). Morphological LN features reflective of cancer‐associated responses are robustly quantifiable by smuLymphNet. Our findings further strengthen the value of assessment of LN properties beyond the detection of metastatic deposits for prognostication 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]  Jakob Nikolas Kather,et al.  Artificial intelligence in histopathology: enhancing cancer research and clinical oncology , 2022, Nature Cancer.

[2]  Cheryl Gillet,et al.  Deep Multi-Scale U-Net Architecture and Label-Noise Robust Training Strategies for Histopathological Image Segmentation , 2022, 2022 IEEE 22nd International Conference on Bioinformatics and Bioengineering (BIBE).

[3]  A. W. Lund,et al.  Tumor-draining lymph nodes: At the crossroads of metastasis and immunity , 2021, Science Immunology.

[4]  Manfredo Atzori,et al.  Multi_Scale_Tools: A Python Library to Exploit Multi-Scale Whole Slide Images , 2021, Frontiers in Computer Science.

[5]  J. Hartman,et al.  What do we still need to learn on digitally assessed biomarkers? , 2021, EBioMedicine.

[6]  Shuoyu Xu,et al.  A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer , 2021, EBioMedicine.

[7]  A. Purushotham,et al.  Systemic immune reaction in axillary lymph nodes adds to tumor-infiltrating lymphocytes in triple-negative breast cancer prognostication , 2021, NPJ breast cancer.

[8]  A. Purushotham,et al.  Systemic immune reaction in axillary lymph nodes adds to tumor-infiltrating lymphocytes in triple-negative breast cancer prognostication , 2021, npj Breast Cancer.

[9]  R. Molina,et al.  Learning from crowds in digital pathology using scalable variational Gaussian processes , 2021, Scientific Reports.

[10]  A. Grigoriadis,et al.  Immune Crosstalk Between Lymph Nodes and Breast Carcinomas, With a Focus on B Cells , 2021, Frontiers in Molecular Biosciences.

[11]  Andreas Keller,et al.  Swarm Learning for decentralized and confidential clinical machine learning , 2021, Nature.

[12]  G. Litjens,et al.  Deep learning in histopathology: the path to the clinic , 2021, Nature Medicine.

[13]  G. Curigliano,et al.  Immune analysis of lymph nodes in relation to the presence or absence of tumor infiltrating lymphocytes in triple-negative breast cancer. , 2021, European journal of cancer.

[14]  Jinhuai Liu,et al.  MSU-Net: Multi-Scale U-Net for 2D Medical Image Segmentation , 2021, Frontiers in Genetics.

[15]  D. Kerr,et al.  Designing deep learning studies in cancer diagnostics , 2021, Nature Reviews Cancer.

[16]  P. Fasching,et al.  Pembrolizumab for Early Triple-Negative Breast Cancer. , 2020, The New England journal of medicine.

[17]  W. M. Smit,et al.  High-Dose Chemotherapy With Hematopoietic Stem Cell Transplant in Patients With High-Risk Breast Cancer and 4 or More Involved Axillary Lymph Nodes: 20-Year Follow-up of a Phase 3 Randomized Clinical Trial. , 2020, JAMA oncology.

[18]  M. Futakuchi,et al.  Detection of lung cancer lymph node metastases from whole-slide histopathological images using a two-step deep learning approach. , 2019, The American journal of pathology.

[19]  A. Madabhushi,et al.  Artificial intelligence in digital pathology — new tools for diagnosis and precision oncology , 2019, Nature reviews. Clinical oncology.

[20]  R. Gelber,et al.  Axillary dissection versus no axillary dissection in patients with breast cancer and sentinel-node micrometastases (IBCSG 23-01): 10-year follow-up of a randomised, controlled phase 3 trial. , 2018, The Lancet. Oncology.

[21]  Meyke Hermsen,et al.  1399 H&E-stained sentinel lymph node sections of breast cancer patients: the CAMELYON dataset , 2018, GigaScience.

[22]  Loïc Le Folgoc,et al.  Attention U-Net: Learning Where to Look for the Pancreas , 2018, ArXiv.

[23]  Yi Li,et al.  Cancer Metastasis Detection With Neural Conditional Random Field , 2018, ArXiv.

[24]  O. Schilling,et al.  Morphology of Immunomodulation in Breast Cancer Tumor Draining Lymph Nodes Depends on Stage and Intrinsic Subtype , 2018, Scientific Reports.

[25]  A. Tutt,et al.  Histological scoring of immune and stromal features in breast and axillary lymph nodes is prognostic for distant metastasis in lymph node‐positive breast cancers , 2018, The journal of pathology. Clinical research.

[26]  A. Giuliano,et al.  Effect of Axillary Dissection vs No Axillary Dissection on 10-Year Overall Survival Among Women With Invasive Breast Cancer and Sentinel Node Metastasis: The ACOSOG Z0011 (Alliance) Randomized Clinical Trial , 2017, JAMA.

[27]  A. Ruddell,et al.  Tumor Regulation of Lymph Node Lymphatic Sinus Growth and Lymph Flow in Mice and in Humans , 2017, The Yale journal of biology and medicine.

[28]  M. Takeya,et al.  Possible functions of CD169‐positive sinus macrophages in lymph nodes in anti‐tumor immune responses , 2017, Cancer science.

[29]  Peter Bankhead,et al.  QuPath: Open source software for digital pathology image analysis , 2017, Scientific Reports.

[30]  Melissa M. Sprachman,et al.  SCS macrophages suppress melanoma by restricting tumor-derived vesicle–B cell interactions , 2016, Science.

[31]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[32]  S Michiels,et al.  Guidelines for time-to-event end point definitions in breast cancer trials: results of the DATECAN initiative (Definition for the Assessment of Time-to-event Endpoints in CANcer trials)†. , 2015, Annals of oncology : official journal of the European Society for Medical Oncology.

[33]  Trevor Darrell,et al.  Fully convolutional networks for semantic segmentation , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[34]  S. Rodenhuis,et al.  Genomic patterns resembling BRCA1- and BRCA2-mutated breast cancers predict benefit of intensified carboplatin-based chemotherapy , 2014, Breast Cancer Research.

[35]  Ronald N. Germain,et al.  A Spatially-Organized Multicellular Innate Immune Response in Lymph Nodes Limits Systemic Pathogen Spread , 2012, Cell.

[36]  I. Narama,et al.  Histopathology of Incidental Findings in Cynomolgus Monkeys (Macaca Fascicularis) Used in Toxicity Studies , 2012, Journal of toxicologic pathology.

[37]  D G Altman,et al.  REporting recommendations for tumour MARKer prognostic studies (REMARK) , 2005, British journal of cancer.

[38]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[39]  F. Batista,et al.  The who, how and where of antigen presentation to B cells , 2009, Nature Reviews Immunology.

[40]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[41]  T. Sørensen,et al.  A method of establishing group of equal amplitude in plant sociobiology based on similarity of species content and its application to analyses of the vegetation on Danish commons , 1948 .