基于INSEMA组织学图像的深度学习预测乳腺癌前哨淋巴结状态
研究基于INSEMA试验及TCGA、曼海姆和雷根斯堡大学医院队列的原发性乳腺肿瘤苏木精/伊红染色图像,训练ResNet 50深度学习图像分析算法预测前哨淋巴结状态,并与仅用临床数据的逻辑回归比较。结果显示,各图像分析算法在包括INSEMA留出集在内的测试集上受试者工作特征曲线下面积均未显著优于随机,而曼海姆队列拟合的逻辑回归在INSEMA和雷根斯堡队列上保持优于随机的性能。将图像分析模型输出纳入逻辑回归未进一步改善INSEMA上的性能。
要点
- 基于INSEMA试验组织学图像的深度学习算法未能预测前哨淋巴结状态,测试集AUC未显著优于随机。
- 仅用临床数据的逻辑回归在曼海姆队列拟合后,于INSEMA和雷根斯堡队列保持优于随机的性能。
- 将图像分析模型输出加入逻辑回归,未进一步改善INSEMA上的预测性能。
核心结论
- 采用基于深度学习的组织学切片图像分析,无法在INSEMA试验及其他以管腔型为主的队列中预测未见病例的前哨淋巴结状态。
研究设计与样本
- 使用在ImageNet上预训练的ResNet 50架构及既往成功策略,训练深度学习图像分析算法,在INSEMA试验及三个独立队列(TCGA、曼海姆和雷根斯堡大学医院队列)以管腔型为主的原发性乳腺肿瘤苏木精/伊红染色图像上预测前哨淋巴结状态,并与仅用临床数据的逻辑回归比较性能。
- 以盲法评估算法在INSEMA留出集上的性能。
主要发现
- 各图像分析算法在包括INSEMA留出测试集在内的测试集上,受试者工作特征曲线下面积均未显著优于随机。
- 曼海姆队列拟合的逻辑回归在INSEMA和雷根斯堡队列上保持优于随机的性能。
- 将图像分析模型输出纳入逻辑回归后,在INSEMA上未进一步改善性能。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Sentinel lymph node (SLN) 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 randomised Intergroup Sentinel Mamma (INSEMA) trial address the potential for further de-escalation of axillary surgery. Therefore, it would be helpful to accurately predict the pretherapeutic sentinel status using medical images. METHODS: Using a ResNet 50 architecture pretrained on ImageNet and a previously successful strategy, we trained deep learning (DL)-based image analysis algorithms to predict sentinel status on hematoxylin/eosin-stained images of predominantly luminal, primary breast tumours from the INSEMA trial and three additional, independent cohorts (The Cancer Genome Atlas (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. RESULTS: None 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. CONCLUSIONS: Employing DL-based image analysis on histological slides, we could not predict SLN status for unseen cases in the INSEMA trial and other predominantly luminal cohorts.
引用
- Marmé F, Krieghoff-Henning E, Gerber B, et al. Deep learning to predict breast cancer sentinel lymph node status on INSEMA histological images. Eur J Cancer. 2023. PMID: 37890350. doi:10.1016/j.ejca.2023.113390
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。