Lin-Wei Wang

New breast cancer prognostic factors identified by computer-aided image analysis of HE stained histopathology images

Computer-aided image analysis (CAI) can help objectively quantify morphologic features of hematoxylin-eosin (HE) histopathology images and provide potentially useful prognostic information on breast cancer. We performed a CAI workflow on 1,150 HE images from 230 patients with invasive ductal carcinoma (IDC) of the breast. We used a pixel-wise support vector machine classifier for tumor nests (TNs)-stroma segmentation, and a marker-controlled watershed algorithm for nuclei segmentation. 730 morphologic parameters were extracted after segmentation, and 12 parameters identified by Kaplan-Meier analysis were significantly associated with 8-year disease free survival (P < 0.05 for all). Moreover, four image features including TNs feature (HR 1.327, 95%CI [1.001 - 1.759], P = 0.049), TNs cell nuclei feature (HR 0.729, 95%CI [0.537 - 0.989], P = 0.042), TNs cell density (HR 1.625, 95%CI [1.177 - 2.244], P = 0.003), and stromal cell structure feature (HR 1.596, 95%CI [1.142 - 2.229], P = 0.006) were identified by multivariate Cox proportional hazards model to be new independent prognostic factors. The results indicated that CAI can assist the pathologist in extracting prognostic information from HE histopathology images for IDC. The TNs feature, TNs cell nuclei feature, TNs cell density, and stromal cell structure feature could be new prognostic factors.

Quantum dots-based tissue and in vivo imaging in breast cancer researches: current status and future perspectives

As the most common malignant tumor for females, breast cancer (BC) is a highly heterogeneous disease regarding biological behaviors. Precisely targeted imaging on BC masses and biomarkers is critical to BC detection, treatment, monitoring, and prognostic evaluation. As an important imaging technique, quantum dots (QDs)-based imaging has emerged as a promising tool in BC researches owe to its outstanding optical properties. However, few reviews have been specifically devoted to discussing applications of QDs-based imaging in BC researches. This review summarized recent promising works in QDs-based tissue and in vivo imaging for BC studies. Physicochemical and optical properties of QDs and its potential applications were briefly described first. Then QDs-based imaging studies in BC were systematically reviewed, including tissue imaging for studying biomarkers interactions, and evaluating prognostic biomarkers, in vivo imaging for mapping axillary lymphatic system, showing BC xenograft tumor, and detecting BC metastases. At last, the future perspectives with special emphasis on the potential clinical applications have also been discussed. Potential applications of QDs-based imaging on clinical BC in the future are mainly focused on tissue study, especially in BC molecular pathology due to its optimal optical properties and quantitative information capabilities on multiple biomarkers.

Combined features based on MT1-MMP expression, CD11b + immunocytes density and LNR predict clinical outcomes of gastric cancer.

BackgroundGiven the complexity of tumor microenvironment, no single marker from cancer cells could adequately predict the clinical outcomes of gastric cancer (GC). The objective of this study was to evaluate the prognostic role of combined features including conventional pathology, proteinase and immune data in GC.MethodsIn addition to pathological studies, immunohistochemistry was used to assess membrane-type 1 matrix metalloproteinase (MT1-MMP) expression and CD11b + immunocytes density in three independent GC tissue microarrays containing 184 GC tissues. Separate and combined features were evaluated for their impact on overall survival (OS).ResultsWe found that traditional factors including tumor size, histological grade, lymph node status, serosa invasion and TNM stage were associated with OS (P < 0.05 for all). Moreover, statistically significant differences in OS were found among lymph node ratio (LNR) subgroups (P < 0.001), MT1-MMP subgroups (P = 0.015), and CD11b + immunocytes density subgroups (P = 0.031). Most importantly, combined feature (MT1-MMP positive, low CD11b + immunocytes density and high LNR) was found by multivariate analysis to be an independent prognostic factors for OS after excluding other confounding factors (HR = 3.818 [95%CI: 2.223-6.557], P < 0.001). In addition, this combined feature had better performance in predicting clinical outcomes after surgery long before recurrence had occurred (Area under the curve: 0.689 [95%CI: 0.609-0.768], P < 0.001).ConclusionsThese findings indicate that better information on GC prognosis could be obtained from combined clinico-pathological factors, tumor cells and the tumor microenvironment.

Evaluation of the staging systems for gastric cancer.

Some staging systems for gastric cancer (GC) have been developed as alternatives to the 6th and 7th TNM staging systems, including the Hybrid, tumor‐ratio‐metastasis (TRM), and Kiel staging systems. This study evaluated the overall performance of these systems for GC.

Segmentation of Hematoxylin-Eosin stained breast cancer histopathological images based on pixel-wise SVM classifier

Hematoxylin-Eosin (HE) staining is the routine diagnostic method for breast cancer (BC), and large amounts of HE stained histopathological images are available for analysis. It is emergent to develop computational methods to efficiently and objectively analyze these images, with the aim of providing potentially better diagnostic and prognostic information for BC. This work focus on analyzing our in-house HE stained histopathological images of breast cancer tissues. Since tumor nests (TNs) and stroma morphological characteristics can reflect the biological behaviors of breast invasive ductal carcinoma (IDC), accurate segmentation of TNs and the stroma is the first step towards the subsequent quantitative analysis. We first propose a method based on the pixel-wise support vector machine (SVM) classifier for segmenting TNs and the stroma, then extract four morphological characters related to the TNs from the images and investigate their relationships with the patients’ 8-year disease free survival (8-DFS). The evaluation result shows that the classification based segmentation method is able to distinguish between TNs and stroma with 87.1% accuracy and 80.2% precision, suggesting that the proposed method is promising in segmenting HE stained IDC histopathological images. The Kaplan-Meier survival curves show that three morphological characters (number of TNs, total perimeter, and average area of TNs) in the images have statistical correlations with 8-DFS of the patients, illustrating that the segmented images can help to identify new morphological factors in IDC TNs for the prediction of BC prognosis.抽象创新点苏木素-伊红 (Hematoxylin-Eosin, HE) 染色组织病理图像分析是乳腺癌的常规诊断方法. 随着数字病理的发展, 病理实验室采集了大量数字化HE组织病理图像, 迫切需要开发基于计算机的高效客观的病理图像分析方法. 医学领域认为病理图像中癌巢和间质的形态学特征可以反映乳腺浸润性导管癌的生物学行为趋势, 因此精确分割癌巢和间质是计算机辅助分析的基础. 本文将图像分割问题看作是像素点的分类问题, 提出了一种基于像素级特征的支持向量机分类算法来识别癌巢/间质像素点, 从而实现癌巢-间质的分割. 基于此算法, 我们对本实验采集的HE 染色病理图像进行分割, 结果显示该算法在分割癌巢-间质时有87.1%的准确率和80.2%的精度. 我们提取出癌巢的4个形态学特征, Kaplan-Meier生存分析揭示其中三个癌巢形态学特征(癌巢数量、 癌巢总周长以及癌巢平均面积)与患者8年无病生存期 (8-DFS) 具有显著的统计相关性, 该结果表明该分割算法有助于鉴别乳腺浸润性导管癌新的病理学形态预后因子.

A Clinical Database of Breast Cancer Patients Reveals Distinctive Clinico-pathological Characteristics: a Study From Central China

BACKGROUND Breast cancer is the most common malignant tumor in females worldwide. Many differences exist in clinico-pathological characteristics of breast cancer patients between China and Western countries. This study aimed to analyze clinico-pathological characteristics of breast cancer from central China. METHODS Clinico- pathological information on breast cancer from three hospitals in central China was collected and analyzed. RESULTS From 1994 to 2012, 2,525 patients with a median age 50 years were included in this study. The 45-49-year age group and invasive ductal carcinoma not otherwise specified accounted for the highest proportions (19.1%, 480/2,525 and 81.0%, 1,982/2,446). Stages 0-I, II and III accounted for 28.0% (682/2,441), 48.4% (1,180/2,441), and 23.7% (578/2,441), respectively. Distribution of N stage showed that N0 accounted for 53.2% (1,344/2,525), and proportion of N0 rose from 51.1% (157/307) in 30-39-year age group to 64.3% (110/171) in ≥ 70-year age group, with an average increase of 2.1% in each age group. Modified radical mastectomy, radical mastectomy, breast-conserving surgery and simple mastectomy were performed for 71.8% (1,812/2,525), 18.0% (454/2,525), 5.2% (131/2,525) and 2.6% (66/2,525), respectively. Proportions of breast-conserving surgery in age ≤ 44-year group (68/132, 51.5%) and simple mastectomy in age ≥ 60-year group (57/89, 64.0%) were higher than in the other age groups. Breast cancers positive for estrogen receptor accounted for 53.0% (1,107/ 2,112). The comparisons among this study and other reports showed higher proportion of younger patients, lower proportion of breast- conserving surgery and positive estrogen receptor patients in China than western countries. CONCLUSIONS Clinico-pathological characteristics in this study demonstrated clear differences between the center of China than Western countries. Additional classification systems should be developed to guide grading of early breast cancer more accurately, especially for N0 patients. Invasive ductal carcinoma is a focus for intensive research.

Long term follow up and retrospective study on 533 gastric cancer cases

BackgroundGastric cancer (GC) is the third leading cause of cancer death in China and the outcome of GC patients is poor. The aim of the research is to study the prognostic factors of gastric cancer patients who had curative intent or palliative resection, completed clinical database and follow-up.MethodsThis retrospective study analyzed 533 GC patients from three tertiary referral teaching hospitals from January 2004 to December 2010 who had curative intent or palliative resection, complete clinical database and follow-up information. The GC-specific overall survival (OS) status was determined by the Kaplan-Meier method, and univariate analysis was conducted to identify possible factors for survival. Multivariate analysis using the Cox proportional hazard model and a forward regression procedure was conducted to define independent prognostic factors.ResultsBy the last follow-up, the median follow-up time of 533 GC patients was 38.6 mo (range 6.9-100.9 mo), and the median GC-specific OS was 25.3 mo (95% CI: 23.1-27.4 mo). The estimated 1-, 2-, 3- and 5-year GC-specific OS rates were 78.4%, 61.4%, 53.3% and 48.4%, respectively. Univariate analysis identified the following prognostic factors: hospital, age, gender, cancer site, surgery type, resection type, other organ resection, HIPEC, LN status, tumor invasion, distant metastases, TNM stage, postoperative SAE, systemic chemotherapy and IP chemotherapy. In multivariate analysis, seven factors were identified as independent prognostic factors for long term survival, including resection type, HIPEC, LN status, tumor invasion, distant metastases, postoperative SAE and systemic chemotherapy.ConclusionsResection type, HIPEC, postoperative SAE and systemic chemotherapy are four independent prognostic factors that could be intervened for GC patients for improving survival.

Quantum dot-based multispectral fluorescent imaging to quantitatively study co-expressions of Ki67 and HER2 in breast cancer.

Both Ki67 and HER2 are key prognostic molecules for invasive breast cancer (BC), but the individual relative impacts on prognosis of these molecules are not known. This study was aimed at establishing a quantum dot (QD)-based double-color in-situ quantitative imaging technique to study the co-expressions of Ki67 and HER2, and delineate the individual impacts of these molecules on prognosis. The QD-based fluorescent immunostaining technique could simultaneously image the co-expressions of Ki67 and HER2 in BC specimens, with the former stained as clear red fluorescence in cancer cell nucleus, and the latter as bright green fluorescence on cancer cell membrane. Both Ki67 and HER2 expressions were significantly correlated with 8-year disease free survival (8-DFS) (P<0.05). However, the two molecules had different weights in terms of negative impacts on clinical prognosis. The median 8-DFS was statistically significantly shorter in High-Ki67 High-HER2 subgroup than Low-Ki67 High-HER2 subgroup (11.7 vs. 60.1months, P<0.05), shorter in High-Ki67 Low-HER2 subgroup than Low-Ki67 Low-HER2 subgroup (16.4 vs. 96.0months, P<0.01), shorter in High-Ki67 High-HER2 subgroup than Low-Ki67 Low-HER2 subgroup (11.7 vs. 96.0months, P<0.01), but there were no statistically significant differences in median 8-DFS between High-Ki67 Low-HER2 subgroup and High-Ki67 High-HER2 subgroup (11.7 vs. 16.4months, P=0.586). The hazard ratio (HR) of Ki67 negative impact on 8-DFS was about 3 fold of that of HER2 (HR 4.493 vs. 1.481). This study demonstrated that QD-based fluorescent imaging technique could help the quantitative study on the co-expressions of Ki67 and HER2 in BC, and Ki67 has a greater negative impact on BC prognosis than HER2.

Computer-aided prognosis on breast cancer with hematoxylin and eosin histopathology images: A review:

With the advance of digital pathology, image analysis has begun to show its advantages in information analysis of hematoxylin and eosin histopathology images. Generally, histological features in hematoxylin and eosin images are measured to evaluate tumor grade and prognosis for breast cancer. This review summarized recent works in image analysis of hematoxylin and eosin histopathology images for breast cancer prognosis. First, prognostic factors for breast cancer based on hematoxylin and eosin histopathology images were summarized. Then, usual procedures of image analysis for breast cancer prognosis were systematically reviewed, including image acquisition, image preprocessing, image detection and segmentation, and feature extraction. Finally, the prognostic value of image features and image feature–based prognostic models was evaluated. Moreover, we discussed the issues of current analysis, and some directions for future research.

Quantum dots-based double imaging combined with organic dye imaging to establish an automatic computerized method for cancer Ki67 measurement.

As a widely used proliferative marker, Ki67 has important impacts on cancer prognosis, especially for breast cancer (BC). However, variations in analytical practice make it difficult for pathologists to manually measure Ki67 index. This study is to establish quantum dots (QDs)-based double imaging of nuclear Ki67 as red signal by QDs-655, cytoplasmic cytokeratin (CK) as yellow signal by QDs-585, and organic dye imaging of cell nucleus as blue signal by 4′,6-diamidino-2-phenylindole (DAPI), and to develop a computer-aided automatic method for Ki67 index measurement. The newly developed automatic computerized Ki67 measurement could efficiently recognize and count Ki67-positive cancer cell nuclei with red signals and cancer cell nuclei with blue signals within cancer cell cytoplasmic with yellow signals. Comparisons of computerized Ki67 index, visual Ki67 index, and marked Ki67 index for 30 patients of 90 images with Ki67 ≤ 10% (low grade), 10% < Ki67 < 50% (moderate grade), and Ki67 ≥ 50% (high grade) showed computerized Ki67 counting is better than visual Ki67 counting, especially for Ki67 low and moderate grades. Based on QDs-based double imaging and organic dye imaging on BC tissues, this study successfully developed an automatic computerized Ki67 counting method to measure Ki67 index.