Relevant impact of central pathology review on nodal classification in individual breast cancer patients

Abstract

Pathology review changed the N-classification in 24%, mainly upstaging, with potentially clinical relevance for individual patients. The association of isolated tumor cells and micrometastases with outcome remained unchanged. Quality control should include nodal breast cancer staging.

Deep learning reveals cancer metastasis and therapeutic antibody targeting in whole body

“…In our data, a single human annotator missed around 29% of all metastases. This is in line with previous studies where human experts missed 1 in 4 breast cancer metastases in histopathology 27 , an issue that even increases substantially if humans work under time pressure 28 . Motivated by this, deep learning based approaches for cancer and metastasis detection recently gained substantial momentum for various imaging modalities, also beyond microscopy [29][30][31][32] .…”

“…Furthermore, this analysis revealed that the most prevalent source of annotation error was overlooked metastases (false negatives). Here, around 29% of metastases were missed in the human annotation, in line with previous studies 27,28 . To effectively identify all missed metastases in the entire data set, our deep learning algorithm (see next section) was trained on the status quo of the annotations and applied to the data set with high sensitivity.…”

Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast Cancer

“…A retrospective study showed that pathology review by experts changed the nodal status in 24% of patients. 5 Furthermore, SLN assessment is tedious and time-consuming. It has been shown that deep learning algorithms could identify metastases in SLN slides with 100% sensitivity, whereas 40% of the slides without metastases could be identified as such.…”

AGO Recommendations for Diagnosis and Treatment of Patients with Early Breast Cancer: Update 2013

“…According to ASCO-CAP guidelines, discordances for central versus local immunohistochemical staining of hormone receptors (HR) and human epidermal growth factor receptor 2 (HER2) are reported in about 20%, major discrepancies in grading for 40% [3,4,5]. Furthermore, in 2012, Mirror trialists reported an upgrade of 22% of pN0 cases to pN1 in central pathology [6]. In the context of these data, and because of the lack of consideration of HER2 over-expression as a prognostic and predictive factor, the AGO guidelines have downgraded the available version 8.0 of Adjuvant!…”