Computational pathology to discriminate benign from malignant intraductal proliferations of the breast.

Fei Dong,Laleh Montaser-Kouhsari,Humayun Irshad,Eun-Yeong Oh,Stuart J Schnitt,Andrew H Beck,Nicholas C Jones,David C Wilbur,Nicole B Johnson,Nicholas W Knoblauch,Beverly Faulkner-Jones,Elena F Brachtel,Luigi K F Rao,Melinda F Lerwill,Anna Sapino

doi:10.1371/journal.pone.0114885

Abstract

The categorization of intraductal proliferative lesions of the breast based on routine light microscopic examination of histopathologic sections is in many cases challenging, even for experienced pathologists. The development of computational tools to aid pathologists in the characterization of these lesions would have great diagnostic and clinical value. As a first step to address this issue, we evaluated the ability of computational image analysis to accurately classify DCIS and UDH and to stratify nuclear grade within DCIS. Using 116 breast biopsies diagnosed as DCIS or UDH from the Massachusetts General Hospital (MGH), we developed a computational method to extract 392 features corresponding to the mean and standard deviation in nuclear size and shape, intensity, and texture across 8 color channels. We used L1-regularized logistic regression to build classification models to discriminate DCIS from UDH. The top-performing model contained 22 active features and achieved an AUC of 0.95 in cross-validation on the MGH data-set. We applied this model to an external validation set of 51 breast biopsies diagnosed as DCIS or UDH from the Beth Israel Deaconess Medical Center, and the model achieved an AUC of 0.86. The top-performing model contained active features from all color-spaces and from the three classes of features (morphology, intensity, and texture), suggesting the value of each for prediction. We built models to stratify grade within DCIS and obtained strong performance for stratifying low nuclear grade vs. high nuclear grade DCIS (AUC = 0.98 in cross-validation) with only moderate performance for discriminating low nuclear grade vs. intermediate nuclear grade and intermediate nuclear grade vs. high nuclear grade DCIS (AUC = 0.83 and 0.69, respectively). These data show that computational pathology models can robustly discriminate benign from malignant intraductal proliferative lesions of the breast and may aid pathologists in the diagnosis and classification of these lesions.

Highlights

The pathological classification of ductal carcinoma in situ (DCIS) versus usual ductal hyperplasia (UDH) on core biopsy of has major implications for patient management
The pathological grading of DCIS cases were obtained from the diagnostic pathology reports, and cases of DCIS were graded as low, intermediate or high based on the degree of nuclear atypia
We obtained strong performance for low-grade vs. high grade DCIS (AUC50.98), but only moderate performance for low grade vs. intermediate grade DCIS (AUC50.83) and for intermediate grade vs. high grade DCIS (AUC50.69). These results suggest the morphologic overlap between intermediate-and-low grade DCIS and between intermediate-and-high grade DCIS

Summary

Introduction

The pathological classification of ductal carcinoma in situ (DCIS) versus usual ductal hyperplasia (UDH) on core biopsy of has major implications for patient management. UDH is considered a benign proliferation, and patients with UDH carry only a small increased risk of developing subsequent breast cancer compared with patients without proliferative breast disease [1]. Clinical management includes the continuation of routine breast cancer screening. DCIS is a preinvasive malignant proliferation, and approximately 25% of patients diagnosed with DCIS on core biopsy are found to have invasive carcinoma upon surgical excision [2]. Primary treatment recommendations for DCIS include lumpectomy with or without whole breast radiation therapy and/or postoperative tamoxifen or total mastectomy with or without sentinel lymph node biopsy [3].

Objectives

Methods

Results

Discussion

Conclusion