Letter to the Editor

Will Radiomics Replace Sentinel Lymph Node Biopsy?

10.4274/ejbh.galenos.2022.2022-2-3

  • Abdullah S. Eldaly
  • Ayman R. Fath
  • Sarah M. Mashaly

Received Date: 09.02.2022 Accepted Date: 19.02.2022 Eur J Breast Health 2022;18(2):203-204 PMID: 35445181

Keywords: Artificial Intelligence, Axillary lymph node metastases, cancer, machine learning, prediction, radiomics

Dear Editor,

Axillary lymph node metastasis is by far the most vital determinant of survival for breast cancer patients (1). Sentinel lymph node biopsy (SLNB) is the gold standard for axillary lymph node staging in clinically node-negative patients. However, despite the reported high sensitivity (44%–100%) and specificity (100%) of SLNB (2), it is associated with multiple morbidities, including sensory impairment, motor dysfunction, and lymphedema (3).

Radiomics is a new type of specialized artificial intelligence that extracts specific features from medical images to construct a disease-specific model, known as a radiomics signature, that is then used to predict disease status in other images (4). Radiomics workflow can be summarized in four cardinal steps: manual or automatic segmentation; feature extraction with specialized tools; selection of the most relevant features using machine learning methods; and radiomics analysis to build the predictive model (4).

Many studies have used radiomics to predict axillary lymph node status in breast cancer patients. Magnetic resonance imaging, computed tomography, and mammography are usually used as the image sources to build the predictive model. The results of radiomics are promising, with an accuracy, sensitivity, and specificity as high as 98% (5).

Although radiomics shows high accuracy in predicting axillary lymph node metastasis, it is not expected to replace SLNB in the near future. This is because the evidence from current radiomics studies is of modest quality. To date, almost all radiomics studies are retrospective in design and lack comparison with the gold standard. Moreover, most of the studies lack external validation and cost-effectiveness analysis.

We believe that replacing SLNB with radiomics in axillary lymph node staging in breast cancer is possible and will spare millions of patients unnecessary surgical interventions. However, implementing radiomics in breast cancer care requires robust evidence from randomized controlled trials. Whether or not the current evidence from the retrospective studies justifies clinical trials is yet to be determined. The answer to this question may be solved by conducting a meta-analysis of the existing literature.


Peer-review: Externally peer-reviewed.

Authorship Contributions

Concept: A.S.E; Design: A.R.F.; Data Collection and/or Processing: S.M.M.; Analysis and/or Interpretation: A.S.E.; Literature Search: A.R.F.; Writing: A.S.E.

Conflict of Interest: No conflict of interest declared by the authors.

Financial Disclosure: The authors declare that this study received no financial Disclosure.

  1. Beenken SW, Urist MM, Zhang Y, Desmond R, Krontiras H, Medina H, et al. Axillary lymph node status, but not tumor size, predicts locoregional recurrence and overall survival after mastectomy for breast cancer. Ann Surg 2003; 237: 738-739. (PMID: 12724640)
  2. Tew K, Irwig L, Matthews A, Crowe P, Macaskill P. Meta-analysis of sentinel node imprint cytology in breast cancer. Br J Surg 2005; 92: 1068-1080. (PMID: 16106479)
  3. McLaughlin SA, Wright MJ, Morris KT, Giron GL, Sampson MR, Brockway JP, et al. Prevalence of lymphedema in women with breast cancer 5 years after sentinel lymph node biopsy or axillary dissection: objective measurements. J Clin Oncol 2008; 26: 5213-5219. (PMID: 18838709)
  4. Mayerhoefer ME, Materka A, Langs G, Häggström I, Szczypiński P, Gibbs P, et al. Introduction to Radiomics. J Nucl Med 2020; 61: 488-495. (PMID: 32060219)
  5. Ashiba H, Nakayama R. Computerized evaluation scheme to detect metastasis in sentinel lymph nodes using contrast-enhanced computed tomography before breast cancer surgery. Radiol Phys Technol 2019; 12: 55-60. (PMID: 30499048)