乳腺癌
医学
前哨淋巴结
淋巴结
癌症
哨兵节点
肿瘤科
病理
内科学
作者
Frederik Marmé,Eva Krieghoff-Henning,Bernd Gerber,Max Schmitt,Dirk-Michael Zahm,Dirk Olaf Bauerschlag,Helmut Forstbauer,Guido Hildebrandt,Beyhan Ataseven,Tobias Brodkorb,Carsten Denkert,Angrit Stachs,David Krug,Jörg Heil,Michael Golatta,Thorsten Kühn,Valentina Nekljudova,Timo Gaiser,Rebecca Schönmehl,Christoph Brochhausen,Sibylle Loibl,Toralf Reimer,Titus J. Brinker
标识
DOI:10.1016/j.ejca.2023.113390
摘要
BackgroundSentinel lymph node status is a clinically important prognostic biomarker in breast cancer and is used to guide therapy, especially for hormone receptor-positive, HER2-negative cases. However, invasive lymph node staging is increasingly omitted before therapy and studies such as the randomized INSEMA trial address the potential for further de-escalation of axillary surgery. Therefore, it would be helpful to accurately predict the pre-therapeutic sentinel status using medical images.MethodsUsing a ResNet 50 architecture pre-trained on ImageNet and a previously successful strategy, we trained deep learning-based image analysis algorithms to predict sentinel status on Hematoxylin/Eosin-stained images of predominantly luminal, primary breast tumors from the INSEMA trial and three additional, independent cohorts (TCGA and cohorts from the University hospitals of Mannheim and Regensburg), and compared their performance with that of a logistic regression using clinical data only. Performance on an INSEMA hold-out set was investigated in a blinded manner.ResultsNone of the generated image analysis algorithms yielded significantly better than random areas under the receiver operating characteristic curves on the test sets, including the hold-out test set from INSEMA. In contrast, the logistic regression fitted on the Mannheim cohort, retained a better than random performance on INSEMA and Regensburg. Including the image analysis model output in the logistic regression did not improve performance further on INSEMA.ConclusionsEmploying deep learning-based image analysis on histological slides, we could not predict sentinel lymph node status for unseen cases in the INSEMA trial and other predominantly luminal cohorts.
科研通智能强力驱动
Strongly Powered by AbleSci AI