基于临床和病理特征的机器学习框架预测乳腺癌新辅助化疗反应。

Machine Learning Frameworks to Predict Neoadjuvant Chemotherapy Response in Breast Cancer Using Clinical and Pathological Features.

机构信息

Division of Medical Oncology, Department of Medicine, University of Toronto, ON, Canada.

Temerty Centre for AI Research and Education in Medicine, University of Toronto, ON, Toronto, Canada.

出版信息

JCO Clin Cancer Inform. 2021 Jan;5:66-80. doi: 10.1200/CCI.20.00078.

DOI:10.1200/CCI.20.00078

PMID:33439725

Abstract

PURPOSE

Neoadjuvant chemotherapy (NAC) is used to treat locally advanced breast cancer (LABC) and high-risk early breast cancer (BC). Pathological complete response (pCR) has prognostic value depending on BC subtype. Rates of pCR, however, can be variable. Predictive modeling is desirable to help identify patients early who may have suboptimal NAC response. Here, we test and compare the predictive performances of machine learning (ML) prediction models to a standard statistical model, using clinical and pathological data.

METHODS

Clinical and pathological variables were collected in 431 patients, including tumor size, patient demographics, histological characteristics, molecular status, and staging information. A standard multivariable logistic regression (MLR) was developed and compared with five ML models: k-nearest neighbor classifier, random forest (RF) classifier, naive Bayes algorithm, support vector machine, and multilayer perceptron model. Model performances were measured using a receiver operating characteristic (ROC) analysis and statistically compared.

RESULTS

MLR predictors of NAC response included: estrogen receptor (ER) status, human epidermal growth factor-2 (HER2) status, tumor size, and Nottingham grade. The strongest MLR predictors of pCR included HER2+ versus HER2- BC (odds ratio [OR], 0.13; 95% CI, 0.07 to 0.23; < .001) and Nottingham grade G3 versus G1-2 (G1-2: OR, 0.36; 95% CI, 0.20 to 0.65; < .001). The area under the curve (AUC) for the MLR was AUC = 0.64. Among the various ML models, an RF classifier performed best, with an AUC = 0.88, sensitivity of 70.7%, and specificity of 84.6%, and included the following variables: menopausal status, ER status, HER2 status, Nottingham grade, tumor size, nodal status, and presence of inflammatory BC.

CONCLUSION

Modeling performances varied between standard versus ML classification methods. RF ML classifiers demonstrated the best predictive performance among all models.

摘要

目的

新辅助化疗（NAC）用于治疗局部晚期乳腺癌（LABC）和高危早期乳腺癌（BC）。病理完全缓解（pCR）的预后价值取决于 BC 亚型。然而，pCR 的发生率可能存在差异。预测模型是理想的，可以帮助早期识别可能对 NAC 反应不佳的患者。在这里，我们使用临床和病理数据测试和比较了机器学习（ML）预测模型与标准统计模型的预测性能。

方法

收集了 431 例患者的临床和病理变量，包括肿瘤大小、患者人口统计学特征、组织学特征、分子状态和分期信息。开发了标准多变量逻辑回归（MLR）模型，并与 5 种 ML 模型进行了比较：k-最近邻分类器、随机森林（RF）分类器、朴素贝叶斯算法、支持向量机和多层感知器模型。使用接收者操作特征（ROC）分析测量模型性能，并进行了统计学比较。

结果

NAC 反应的 MLR 预测因子包括：雌激素受体（ER）状态、人表皮生长因子-2（HER2）状态、肿瘤大小和诺丁汉分级。预测 pCR 的最强 MLR 预测因子包括 HER2+BC 与 HER2-BC（优势比[OR]，0.13；95%置信区间，0.07 至 0.23；<0.001）和诺丁汉分级 G3 与 G1-2（G1-2：OR，0.36；95%置信区间，0.20 至 0.65；<0.001）。MLR 的曲线下面积（AUC）为 AUC=0.64。在各种 ML 模型中，RF 分类器表现最佳，AUC=0.88，灵敏度为 70.7%，特异性为 84.6%，并包括以下变量：绝经状态、ER 状态、HER2 状态、诺丁汉分级、肿瘤大小、淋巴结状态和炎症性 BC 的存在。