Application of automated image analysis reduces the workload of manual screening of sentinel lymph node biopsies in breast cancer. Issue 6 (22nd September 2017)

• Record Type: Journal Article
• Authors: Holten‐Rossing, Henrik; Talman, Maj‐Lis Møller; Jylling, Anne Marie Bak; Lænkholm, Anne‐Vibeke; Kristensson, Martin; Vainer, Ben
• Is Part Of: Histopathology. Volume 71:Issue 6(2018)
• Page Start: 866
• Page End: 873
• Publication Date: 2017-09-22
• Subjects: Breast cancer -- image analysis -- sentinel lymph node biopsy; Histology, Pathological -- Periodicals; 611.018
• DOI: 10.1111/his.13305
• Languages: English
• ISSNs: 0309-0167

Abstract

Abstract : Aims: Breast cancer is one of the most common cancer diseases in women, with >1.67 million cases being diagnosed worldwide each year. In breast cancer, the sentinel lymph node (SLN) pinpoints the first lymph node(s) into which the tumour spreads, and it is usually located in the ipsilateral axilla. In patients with no clinical signs of metastatic disease in the axilla, an SLN biopsy (SLNB) is performed. Assessment of metastases in the SLNB, when using a conventional microscope, is performed by manually observing a metastasis and measuring its size and/or counting the number of tumour cells. This is done essentially to categorize the type of metastasis as macrometastasis, micrometastasis, or isolated tumour cells, which is used to determine which treatment the breast cancer patient will benefit most from. The aim of this study was to evaluate whether digital image analysis can be applied as a screening tool for SNLB assessment without compromising the diagnostic accuracy. Materials and results: Consecutive SLNBs from 135 patients with localized breast cancer receiving surgery in the period February to August 2015 were collected and included in this study. Of the 135 patients, 35 were received at the Department of Pathology, Rigshospitalet, Copenhagen University Hospital, 50 at the Department of Pathology, Zealand University Hospital, and 50 at the Department of Pathology, Odense University Hospital. Formalin‐fixed paraffin‐embedded tissue sections were analysed by immunohistochemistry with the BenchMark ULTRA Ventana platform. Rigshospitalet used a mixture of cytokeratin (CK) 7 and CK19, Zealand University Hospital used pancytokeratin AE1/AE3 and Odense used pancytokeratin CAM5.2 for detection of epithelial tumour cells. Slides were stained locally. SLNB sections were assessed in a conventional microscope according to national guidelines for SLNBs in breast cancer patients. The immunohistochemically stained sections were scanned with a Hamamatsu NanoZoomer‐XR digital whole slide scanner, and the images were analysed with Visiopharm's software by use of a custom‐made algorithm for SLNBs in breast cancer. The algorithm was optimized to the CK antibodies and the local laboratory conditions, on the basis of staining intensity and background staining. Conventional microscopy was used as the gold standard for assessment of positive tumour cells, and the results were compared with those from digital image analysis. The algorithm showed a sensitivity of 100% (that is, no false‐negative slides were observed), including 67.2%, 19.2% and 56.1% of the slides from the three pathology departments being negative, respectively. This means that, on average, the workload could have been decreased by 58.2% by use of the digital SLNB algorithm as a screening tool. Conclusions: The SLNB algorithm showed a sensitivity of 100% regardless of the antibody used for immunohistochemistry and the staining protocol. No false‐negative slides were observed, which proves that the SLNB algorithm is an ideal screening tool for selecting those slides that a pathologist does not need to see. The implementation of automated digital image analysis of SLNBs in breast cancer would decrease the workload in this context for examining pathologists by almost 60%.