Skip to main content

POST-NAT-BRCA

The Cancer Imaging Archive

Post-NAT-BRCA | Assessment of Residual Breast Cancer Cellularity after Neoadjuvant Chemotherapy using Digital Pathology

DOI: 10.7937/TCIA.2019.4YIBTJNO | Data Citation Required | 427 Views | 13 Citations | Image Collection

Location Species Subjects Data Types Cancer Types Size Status Updated
Breast Human 64 Histopathology, Demographic, Follow-Up, Molecular Test, Treatment, Diagnosis Breast Cancer 43.2GB Public, Complete 2019/10/01

Summary

Breast cancer (BC) is the second most commonly diagnosed cancer in the U.S. with more than 250,000 new cases of invasive breast cancers reported in 2017. The majority of women with locally advanced and a subset of patients with operable breast cancer will undergo systemic therapy prior to their surgery (neoadjuvant therapy/ NAT) to reduce the size of tumor(s) and possibly further undergo breast conserving surgery. The Post-NAT-BRCA dataset is a collection of representative sections from breast resections in patients with residual invasive BC following NAT. Histologic sections were prepared and digitized to produce high resolution, microscopic images of treated BC tumors. Also included, are clinical features and expert pathology annotations of tumor cellularity and cell types. The Residual Cancer Burden Index (RCBi), is a clinically validated tool for assessment of response to NAT associated with prognosis. Tumor cellularity is one of the parameters used for calculating the RCBi. In this dataset, tumor cellularity refers to a measure of residual disease after NAT, in the form of proportion of malignant tumor inside the tumor bed region; also annotated. (See MD Anderson RCB Calculator for a detailed description of tumor cellularity.) Malignant, healthy, lymphocyte and other labels were also provided for individual cells to aid development of cell segmentation algorithms.

Data Access

Version 1: Updated 2019/10/01

Title Data Type Format Access Points Subjects Studies Series Images License
Slide Images Histopathology SVS
Download requires IBM-Aspera-Connect plugin
 54 96 CC BY 3.0
Annotations Demographic, Follow-Up, Molecular Test, Treatment, Diagnosis XLSX CC BY 3.0
Related Datasets
No related Analysis Results found: Submit your proposal! No related Collections found
Legend: Analysis Results| Collections

Citations & Data Usage Policy

Data Citation Required: Users must abide by the TCIA Data Usage Policy and Restrictions. Attribution must include the following citation, including the Digital Object Identifier:

Data Citation

Martel, A. L., Nofech-Mozes, S., Salama, S., Akbar, S., & Peikari, M. (2019). Assessment of Residual Breast Cancer Cellularity after Neoadjuvant Chemotherapy using Digital Pathology [Data set]. The Cancer Imaging Archive. https://doi.org/10.7937/TCIA.2019.4YIBTJNO

Detailed Description

Content

The Post-NAT-BRCA dataset is composed of:

  • 96 whole slide images stored in an uncompressed .svs file format, standard for pathology slides. Slides were scanned at 20x objective on an Aperio slide scanner at Sunnybrook Health Sciences Centre.
  • An Excel (.xlsx) file containing clinical features for each patient including age, treatment, ER/PR/HER2 status etc. A key and detailed description of each column is provided in a separate tab titled “Definitions”. Each row in the spreadsheet corresponds to a single (.svs) slide and anonymised patient ID’s are provided in a separate column.
  • Manual annotations of tumor cellularity and cell labels, provided as Sedeen annotation files (.xml). Annotations are given in two directories, where the “WSI_train” folder contains WSIs annotated by a single rater and “WSI_test” was annotated by two raters.

Recommended Software

To browse whole slide images and annotations, we highly recommend you use Pathcore’s Sedeen Viewer which is available for free: https://pathcore.com/sedeen/
Please ensure that the “sedeen” folder is unzipped and placed into the same folder containing the .svs files. Sedeen Viewer loads annotations from this folder automatically when images are opened.

Upon opening WSIs in Sedeen, you will notice that annotations have been color-coded according to the following key:

  • Pink: Healthy (0% tumor cellularity)
  • Blue: Low tumor cellularity (0 – 30%)
  • Yellow: Medium tumor cellularity (31 – 70%)
  • Green: High tumor cellularity (70 – 100%)
  • White: Contains annotations at the cell level and labeled as:
    • Lymphocyte: TIL-E, TIL-S
    • Normal Epithelial: normal, UDH, ADH,
    • Malignant Epithelial: IDC, ILC, Muc C, DCIS 1, DCIS 2, DCIS 3, MC- E, MC – C, MC – M

Acknowledgements

We would like to acknowledge these individuals and institutions:

  • Canadian Cancer Society, grant #705772
  • National Cancer institute of the National Institutes of Health under #U24CA199374
  • Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

Related Publications

Publications by the Dataset Authors

The authors recommended the following as the best source of additional information about this dataset:

Publication Citation

Peikari, M., Salama, S., Nofech‐Mozes, S., & Martel, A. L. (2017). Automatic cellularity assessment from post‐treated breast surgical specimens. In Cytometry Part A (Vol. 91, Issue 11, pp. 1078–1087). Wiley. https://doi.org/10.1002/cyto.a.23244

No other publications were recommended by dataset authors.

Publication Citation

Peikari, M., Salama, S., Nofech‐Mozes, S., & Martel, A. L. (2017). Automatic cellularity assessment from post‐treated breast surgical specimens. In Cytometry Part A (Vol. 91, Issue 11, pp. 1078–1087). Wiley. https://doi.org/10.1002/cyto.a.23244

Research Community Publications

The Collection authors suggest the below will give context to this dataset:

  • Akbar, S., Peikari, M., Salama, S., Panah, A. Y., Nofech-Mozes, S., Martel, A. L. (2019) Automated and Manual Quantification of Tumour Cellularity in Digital Slides for Tumour Burden Assessment. Sci Rep, 9, 14099. https://doi.org/10.1038/s41598-019-50568-4

TCIA maintains a list of publications which leverage TCIA data. If you have a manuscript you’d like to add please contact TCIA’s Helpdesk.

  1. Zhang, X., Zhu, X., Tang, K., Zhao, Y., Lu, Z., & Feng, Q. (2022). DDTNet: A dense dual-task network for tumor-infiltrating lymphocyte detection and segmentation in histopathological images of breast cancer. In Medical Image Analysis (Vol. 78, p. 102415). Elsevier BV. https://doi.org/10.1016/j.media.2022.102415
  2. Diao, J. A., Wang, J. K., Chui, W. F., Mountain, V., Gullapally, S. C., Srinivasan, R., Mitchell, R. N., Glass, B., Hoffman, S., Rao, S. K., Maheshwari, C., Lahiri, A., Prakash, A., McLoughlin, R., Kerner, J. K., Resnick, M. B., Montalto, M. C., Khosla, A., Wapinski, I. N., … Taylor-Weiner, A. (2021). Human-interpretable image features derived from densely mapped cancer pathology slides predict diverse molecular phenotypes. In Nature Communications (Vol. 12, Issue 1). Springer Science and Business Media LLC. https://doi.org/10.1038/s41467-021-21896-9
  3. Ciga, O., Xu, T., & Martel, A. L. (2022). Self supervised contrastive learning for digital histopathology. In Machine Learning with Applications (Vol. 7, p. 100198). Elsevier BV. https://doi.org/10.1016/j.mlwa.2021.100198

TCIA maintains a list of publications that leveraged this dataset. If you have a manuscript you’d like to add please contact TCIA’s Helpdesk.

Other Publications Using this Data

The Collection authors suggest the below will give context to this dataset:

  • Akbar, S., Peikari, M., Salama, S., Panah, A. Y., Nofech-Mozes, S., Martel, A. L. (2019) Automated and Manual Quantification of Tumour Cellularity in Digital Slides for Tumour Burden Assessment. Sci Rep, 9, 14099. https://doi.org/10.1038/s41598-019-50568-4

TCIA maintains a list of publications which leverage TCIA data. If you have a manuscript you’d like to add please contact TCIA’s Helpdesk.

  1. Zhang, X., Zhu, X., Tang, K., Zhao, Y., Lu, Z., & Feng, Q. (2022). DDTNet: A dense dual-task network for tumor-infiltrating lymphocyte detection and segmentation in histopathological images of breast cancer. In Medical Image Analysis (Vol. 78, p. 102415). Elsevier BV. https://doi.org/10.1016/j.media.2022.102415
  2. Diao, J. A., Wang, J. K., Chui, W. F., Mountain, V., Gullapally, S. C., Srinivasan, R., Mitchell, R. N., Glass, B., Hoffman, S., Rao, S. K., Maheshwari, C., Lahiri, A., Prakash, A., McLoughlin, R., Kerner, J. K., Resnick, M. B., Montalto, M. C., Khosla, A., Wapinski, I. N., … Taylor-Weiner, A. (2021). Human-interpretable image features derived from densely mapped cancer pathology slides predict diverse molecular phenotypes. In Nature Communications (Vol. 12, Issue 1). Springer Science and Business Media LLC. https://doi.org/10.1038/s41467-021-21896-9
  3. Ciga, O., Xu, T., & Martel, A. L. (2022). Self supervised contrastive learning for digital histopathology. In Machine Learning with Applications (Vol. 7, p. 100198). Elsevier BV. https://doi.org/10.1016/j.mlwa.2021.100198