A novel model for Ki67 assessment in breast cancer

Conventional and digital Ki67 evaluation and their correlation with molecular prognosis and morphological parameters in luminal breast cancer

Digital counting methods were developed to decrease the high intra- and inter-observer variability of immunohistochemical markers such as Ki67, with most presenting a good correlation coefficient (CC). Since Ki67 is one of the major contributors to Oncotype DX, it is conceivable that Ki67 expression and the recurrence score (RS) obtained by the multigene panel are positively correlated. We decided first to test to what extent conventional and digital Ki67 quantification methods correlate in daily practice and, second, to determine which of these methods correlates better with the prognostic capacity of the Oncotype DX test. Both Ki67 evaluations were performed in 89 core biopsies with a diagnosis of estrogen receptor (ER) positive HER2-negative breast cancer (BC). Cases were, thus, classified twice for surrogate subtype: first by conventional analysis and then by digital evaluation. The Oncotype RS was obtained in 55 cases that were subsequently correlated to Ki67 evaluation by both methods. Conventional and digital Ki67 evaluation showed good concordance and correlation (CC = 0.81 (95% CI 0.73–0.89)). The correlation of Oncotype DX risk groups and surrogate derived subtypes was slightly higher for the digital technique (r_s = 0.46, p < 0.01) compared to the conventional method (r_s = 0.39, p < 0.01), even though both were statistically significant. In conclusion, we show that digital evaluation could be an alternative to conventional counting, and also has advantages for predicting the risk established by the Oncotype DX test in ER-positive BC. This study also supports the importance of an accurate Ki67 analysis which can influence the decision to submit ER-positive HER2-negative BC to prognostic molecular platforms.

Visual and digital assessment of Ki-67 in breast cancer tissue - a comparison of methods

Background

In breast cancer (BC) Ki-67 cut-off levels, counting methods and inter- and intraobserver variation are still unresolved. To reduce inter-laboratory differences, it has been proposed that cut-off levels for Ki-67 should be determined based on the in-house median of 500 counted tumour cell nuclei. Digital image analysis (DIA) has been proposed as a means to standardize assessment of Ki-67 staining in tumour tissue. In this study we compared digital and visual assessment (VA) of Ki-67 protein expression levels in full-face sections from a consecutive series of BCs. The aim was to identify the number of tumour cells necessary to count in order to reflect the growth potential of a given tumour in both methods, as measured by tumour grade, mitotic count and patient outcome.

Methods

A series of whole sections from 248 invasive carcinomas of no special type were immunohistochemically stained for Ki-67 and then assessed by VA and DIA. Five 100-cell increments were counted in hot spot areas using both VA and DIA. The median numbers of Ki-67 positive tumour cells were used to calculate cut-off levels for Low, Intermediate and High Ki-67 protein expression in both methods.

Results

We found that the percentage of Ki-67 positive tumour cells was higher in DIA compared to VA (medians after 500 tumour cells counted were 22.3% for VA and 30% for DIA). While the median Ki-67% values remained largely unchanged across the 100-cell increments for VA, median values were highest in the first 1-200 cells counted using DIA. We also found that the DIA100 High group identified the largest proportion of histopathological grade 3 tumours 70/101 (69.3%).

Conclusions

We show that assessment of Ki-67 in breast tumours using DIA identifies a greater proportion of cases with high Ki-67 levels compared to VA of the same tumours. Furthermore, we show that diagnostic cut-off levels should be calibrated appropriately on the introduction of new methodology.

Correlation of manual semi-quantitative and automated quantitative Ki-67 proliferative index with OncotypeDXTM recurrence score in invasive breast carcinoma

BACKGROUND

Ki-67 immunohistochemistry (IHC) staining is a widely used cancer proliferation assay; however, its limitations could be improved with automated scoring. The OncotypeDXTM Recurrence Score (ORS), which primarily evaluates cancer proliferation genes, is a prognostic indicator for breast cancer chemotherapy response; however, it is more expensive and slower than Ki-67.

OBJECTIVE

To compare manual Ki-67 (mKi-67) with automated Ki-67 (aKi-67) algorithm results based on manually selected Ki-67 "hot spots" in breast cancer, and correlate both with ORS.

METHODS

105 invasive breast carcinoma cases from 100 patients at our institution (2011-2013) with available ORS were evaluated. Concordance was assessed via Cohen's Kappa (κ).

RESULTS

57/105 cases showed agreement between mKi-67 and aKi-67 (κ 0.31, 95% CI 0.18-0.45), with 41 cases overestimated by aKi-67. Concordance was higher when estimated on the same image (κ 0.53, 95% CI 0.37-0.69). Concordance between mKi-67 score and ORS was fair (κ 0.27, 95% CI 0.11-0.42), and concordance between aKi-67 and ORS was poor (κ 0.10, 95% CI -0.03-0.23).

CONCLUSIONS

These results highlight the limits of Ki-67 algorithms that use manual "hot spot" selection. Due to suboptimal concordance, Ki-67 is likely most useful as a complement to, rather than a surrogate for ORS, regardless of scoring method.

The human-in-the-loop: an evaluation of pathologists' interaction with artificial intelligence in clinical practice

AIMS

One of the major drivers of the adoption of digital pathology in clinical practice is the possibility of introducing digital image analysis (DIA) to assist with diagnostic tasks. This offers potential increases in accuracy, reproducibility, and efficiency. Whereas stand-alone DIA has great potential benefit for research, little is known about the effect of DIA assistance in clinical use. The aim of this study was to investigate the clinical use characteristics of a DIA application for Ki67 proliferation assessment. Specifically, the human-in-the-loop interplay between DIA and pathologists was studied.

METHODS AND RESULTS

We retrospectively investigated breast cancer Ki67 areas assessed with human-in-the-loop DIA and compared them with visual and automatic approaches. The results, expressed as standard deviation of the error in the Ki67 index, showed that visual estimation ('eyeballing') (14.9 percentage points) performed significantly worse (P < 0.05) than DIA alone (7.2 percentage points) and DIA with human-in-the-loop corrections (6.9 percentage points). At the overall level, no improvement resulting from the addition of human-in-the-loop corrections to the automatic DIA results could be seen. For individual cases, however, human-in-the-loop corrections could address major DIA errors in terms of poor thresholding of faint staining and incorrect tumour-stroma separation.

CONCLUSION

The findings indicate that the primary value of human-in-the-loop corrections is to address major weaknesses of a DIA application, rather than fine-tuning the DIA quantifications.

Assessment of Ki67 in Breast Cancer: Updated Recommendations From the International Ki67 in Breast Cancer Working Group

Ki67 immunohistochemistry (IHC), commonly used as a proliferation marker in breast cancer, has limited value for treatment decisions due to questionable analytical validity. The International Ki67 in Breast Cancer Working Group (IKWG) consensus meeting, held in October 2019, assessed the current evidence for Ki67 IHC analytical validity and clinical utility in breast cancer, including the series of scoring studies the IKWG conducted on centrally stained tissues. Consensus observations and recommendations are: 1) as for estrogen receptor and HER2 testing, preanalytical handling considerations are critical; 2) a standardized visual scoring method has been established and is recommended for adoption; 3) participation in and evaluation of quality assurance and quality control programs is recommended to maintain analytical validity; and 4) the IKWG accepted that Ki67 IHC as a prognostic marker in breast cancer has clinical validity but concluded that clinical utility is evident only for prognosis estimation in anatomically favorable estrogen receptor–positive and HER2-negative patients to identify those who do not need adjuvant chemotherapy. In this T1-2, N0-1 patient group, the IKWG consensus is that Ki67 5% or less, or 30% or more, can be used to estimate prognosis. In conclusion, analytical validity of Ki67 IHC can be reached with careful attention to preanalytical issues and calibrated standardized visual scoring. Currently, clinical utility of Ki67 IHC in breast cancer care remains limited to prognosis assessment in stage I or II breast cancer. Further development of automated scoring might help to overcome some current limitations.

Assessment of Ki67 in Breast Cancer: A Comparison Between the Eye-10 Method, Stepwise Counting Strategy, and International System of Ki67 Evaluation

Background & Objective:

Ki-67 evaluation is an essential tool to define luminal A and B breast cancers, which is not yet systematized. The International Ki67 in Breast Cancer Working Group suggests the counting of 500 or 1000 cancer cells, which is a time-consuming process. Therefore, novel methods, such as the Eye-10 method and stepwise counting strategy, are proposed to facilitate measurement.

Methods:

Immunohistochemical staining of Ki67 was performed on 100 hormone-receptor-positive invasive ductal carcinoma specimens. Ki67LI was evaluated for each case, and then results were dichotomized by a cut-off point of 20%. Next, for each sample, an expert pathologist visually assessed percentages of Ki67-positive cells in 10% intervals at a glance (Eye-10 method). Finally, by using a dynamic process with rejection regions, Ki67 was defined so if the estimate belonged to the upper or lower rejection region, the Ki67 status had been determined and if the rejection region could not be reached after counting the maximum number of 400 tumor cells, the specimen was regarded as equivocal (stepwise counting strategy).

Results:

The comparison between Eye-10 and Ki67LI revealed almost perfect agreement (kappa coefficient =0.889), and the concordance between the stepwise counting strategy and Ki67LI was substantial (kappa coefficient =0.639).

Conclusion:

Both two methods left some results in the gray/intermediate zone, which is unavoidable. Both methods are much faster and simpler than evaluation of Ki67LI and are also reliable. Regarding the gray zone in both methods, further improvements in the methodology, as well as more analytical studies, are needed.

Good staining quality ensuring the reproducibility of Ki67 assessment

AIMS

Although Ki67 labelling index (LI) is a prognostic and predictive marker in breast cancer, its accuracy and reproducibility must be validated before its clinical application. We aimed to evaluate the agreement of Ki67 LI in clinical practice in Taiwan.

METHODS

We conducted a Ki67 immunohistochemistry (IHC) proficiency test. The participants performed the Ki67 IHC test and measured the Ki67 LI of 10 cases of breast cancer tissue on a microarray slide. The staining quality was centrally reviewed based on the Ki67 staining of the tonsil surface epithelium.

RESULTS

Ki67 staining and counting methods are diverse in Taiwan. The reproducibility of Ki67 LI was poor to good (intraclass correlation coefficient: 0.581, 95% CI 0.354 to 0.802). The reproducibility and agreement in the high staining quality group were significantly higher than those in the low staining quality group. The majority of the Ki67 LIs derived from the low staining quality group were underestimated. Different counting methods did not reveal significant differences when determining Ki67 LI with microarray sections.

CONCLUSIONS

We suggest using the surface epithelium of the tonsil as external control and achieving optimal staining results that consist of a high positive parabasal layer, a low positive intermediate layer and a negative superficial layer. Good Ki67 staining quality can minimise the staining variations among different laboratories, and it is essential for the reproducibility of Ki67 LI.

The Role of CXCL13 and CXCL9 in Early Breast Cancer

BACKGROUND

Chemokines, cytokines in the immune microenvironment of tumors, may be associated with patient outcome. We assessed the impact of CXCL13 and CXCL9 on disease-free (DFS) and overall survival (OS), in an attempt to retrospectively evaluate both T and B cell function in the microenvironment of primary tumors from patients with breast cancer.

MATERIALS AND METHODS

Formalin-fixed paraffin-embedded tissue blocks from patients with intermediate/high-risk, early breast cancer, treated with sequential adjuvant epirubicin, paclitaxel, and cyclophosphamide methotrexate fluorouracil within a randomized trial, were tested for CXCL13 and CXCL9 messenger RNA expression; 557 patients with adequate tissue were eligible for the analysis.

RESULTS

CXCL13 was correlated with CXCL9 (rho = 0.52; P < .001). High-expressing CXL13 and CXCL9 tumors had higher Ki67 and tumor infiltrating lymphocyte density (P-values < .001). High CXCL9 expression was an unfavorable prognosticator for OS among all patients (hazard ratio [HR], 1.73; P = .021), whereas it showed favorable significance for both DFS and OS in patients with triple negative disease (HR, 0.29; P = .027 and HR, 0.32; P = .045). High CXCL13 conferred longer DFS and OS among patients with luminal-human epidermal growth factor receptor 2 disease (HR, 0.31; P = .013 and HR, 0.25; P = .005). Patients with low CXCL13 and high CXCL9 expression had shorter DFS and OS compared with those with high expression of both chemokines (HR, 1.63; P = .006 and HR, 1.61; P = .016).

CONCLUSIONS

Both biomarkers were associated with poor prognosis characteristics and with tumor infiltrating lymphocyte density. High CXCL9 conferred an improved prognosis in the triple negative subtype, whereas high CXCL13 was associated with improved outcome in the luminal-human epidermal growth factor receptor 2 subtype. Chemokines can be associated with breast cancer subtype and outcome. These data should be evaluated prospectively.

Cyclin D1 differential activation and its prognostic impact in patients with advanced breast cancer treated with trastuzumab

Introduction

We sought to determine the level of activation of the critical components of the cyclin D1-mediated pathway and to evaluate their prognostic significance across the different molecular subtypes of advanced breast cancer.

Patients and methods

The study population comprised 219 female patients with advanced breast cancer who had been found to have human epidermal growth factor receptor 2 (HER2)-positive disease by local testing and were all treated with trastuzumab-based regimens. For all tumours, central testing for HER2 was performed, and cyclin D1 gene (CCND1) amplification, mRNA and protein expression were assessed by FISH, quantitative real-time-PCR and immunohistochemistry, respectively. Prognostic impact on clinical endpoints was evaluated with Cox regression analyses.

Results

After central testing, only 134 (61.2%) of 219 patients were confirmed to have HER2 gene amplification by FISH and/or 3+ HER2 protein expression by immunohistochemistry. After a median follow-up time of 136.0 months (95% CI 123.3 to 148.9), 105 (78.4%) HER2-positive patients and 76 (89.4%) HER2-negative patients had died, while 80% of the former and 87.1% of the latter had experienced a disease relapse. Patients with positive oestrogen receptor/progesterone receptor status presented with higher cyclin D1 mRNA expression. In the HER2-negative subgroup, patients with negative cyclin D1 protein expression were at higher risk of progression (HR= 1.66, 95%CI 1.01 to 2.72, Wald’s p=0.045). Among de novo metastatic patients, the risk of progression was higher for patients with non-amplified CCND1 tumours (HR= 2.00, 95% CI 1.03 to 3.90, p=0.041).

Conclusion

Aberrant activation of the cyclin D1-mediated pathway appears to reduce the risk of progression in HER2-negative tumours, but not in HER2-positive ones.

Practical approaches to automated digital image analysis of Ki-67 labeling index in 997 breast carcinomas and causes of discordance with visual assessment

The Ki-67 labeling index (LI) is an important prognostic factor in breast carcinoma. The Ki-67 LI is traditionally calculated via unaided microscopic estimation; however, inter-observer and intra-observer variability and low reproducibility are problems with this visual assessment (VA) method. For more accurate assessment and better reproducibility with Ki-67 LI, digital image analysis was introduced recently. We used both VA and automated digital image analysis (ADIA) (Ventana Virtuoso image management software) to estimate Ki-67 LI for 997 cases of breast carcinoma, and compared VA and ADIA results. VA and ADIA were highly correlated (intraclass correlation coefficient 0.982, and Spearman’s correlation coefficient 0.966, p<0.05). We retrospectively analyzed cases with a greater than 5% difference between VA and ADIA results. The cause of these differences was: (1) tumor heterogeneity (98 cases, 56.0%), (2) VA interpretation error (32 cases, 18.3%), (3) misidentification of tumor cells (26 cases, 14.9%), (4) poor immunostaining or slide quality (16 cases, 9.1%), and (5) Estimation of non-tumor cells (3 cases, 1.7%). There were more discrepancies between VA and ADIA results in the group with a VA value of 10–20% compared to groups with <10% and ≥20%. Although ADIA is more accurate than VA, there are some limitations. Therefore, ADIA findings require confirmation by a pathologist.

Evaluation of the Insulin-like Growth Factor Receptor Pathway in Patients with Advanced Breast Cancer Treated with Trastuzumab

BACKGROUND

Trastuzumab is a monoclonal antibody against HER2-positive breast cancer. Despite improving the natural history of the disease, there is a number of patients who are resistant to it, whereas all patients will eventually develop resistance and disease will progress. Inconsistent preclinical data show that the IGF-R pathway may contribute to either de novo or acquired resistance to trastuzumab.

MATERIALS AND METHODS

In total, 227 trastuzumab-treated metastatic breast cancer patients were evaluated for IGF-1, IGF-1R, GLP-1R, Akt1, Akt2 Akt3 mRNA expression, and IGF-1Rα, IGF-1Rβ, IGF-2R protein expression.

RESULTS

Only 139 patients were truly HER2-positive by central assessment. Among HER2-positive patients, high Akt2 and GLP-1R mRNA expression showed a trend towards higher and lower risk of progression, respectively (HR=1.83, 95%CI=0.90-3.72, p=0.094 and HR=0.62, 95%CI=0.36-1.06, p=0.079), while high Akt1 and GLP-1R mRNA expression presented a trend towards unfavorable survival (HR=1.67, 95%CI=0.93-2.99, p=0.086 and HR=1.67, 95%CI=0.94-2.96, p=0.080). Among HER2-negative patients, high GLP-1R mRNA expression and negative stromal IGF-1Rβ protein expression showed a trend towards worse survival (HR=2.31, 95%CI=0.87-6.13, p=0.094 and HR=2.03, 95%CI=0.94-4.35, p=0.071, respectively). In the multivariate analyses, HER2-positive patients with high Akt1 and GLP-1R mRNA expression had a worse survival (HR=1.86, 95%CI=1.01-3.43, p=0.045 and HR=1.83, 95%CI=0.99-3.41, p=0.055, respectively).

CONCLUSION

This study revealed a crosstalk between the IGF-R pathway and HER2. There was evidence that high Akt1 and GLP-1R mRNA expression might affect survival among HER2-positive metastatic breast cancer patients treated with trastuzumab.

Evaluation of the prognostic value of CD3, CD8, and FOXP3 mRNA expression in early-stage breast cancer patients treated with anthracycline-based adjuvant chemotherapy

BACKGROUND

Tumor-infiltrating lymphocytes (TILs) have been shown to be of prognostic value in several cancer types. In early breast cancer, TILs have a prognostic utility, as well, especially in HER2-positive and triple-negative breast cancer. TILs presence is broadly associated with improved survival; however, there is controversy regarding TILs subpopulations.

PATIENTS AND METHODS

Early-stage breast cancer patients treated with anthracycline-based chemotherapy within two randomized trials were included in the study. We evaluated, by qRT-PCR, 826 tumor tissue samples for mRNA expression of CD3, CD8, and FOXP3 for potential prognostic significance in terms of disease-free survival (DFS) and overall survival (OS).

RESULTS

After a median follow-up of 133.0 months, 255 patients (30.9%) had died and 314 (38.0%) had disease progression. In the univariate analysis, high CD3 and CD8 mRNA expression was found to be of favorable prognostic value for DFS (P = 0.007 and P = 0.016, respectively). In multivariate analyses, the association of high CD8 mRNA expression with increased DFS was retained (HR = 0.77, 95% CI 0.60-0.998, Wald's P = 0.048), whereas that of high CD3 mRNA expression was of marginal statistical significance (HR = 0.77, 95% CI 0.59-1.01, P = 0.059). Moreover, a significant interaction was observed between HER2 status and CD3 mRNA expression with respect to DFS (interaction P = 0.032). In the HER2-positive subgroup, the hazard ratio associated with high CD3 mRNA expression was of greater magnitude (HR = 0.48, 95% CI 0.30-0.76, P = 0.002) compared with the hazard ratio presented above, for the entire cohort. No significant findings were observed for FOXP3 in terms of DFS, while none of the studied markers were of prognostic value for OS.

CONCLUSIONS

High CD3 and CD8 mRNA expression in early-stage breast cancer patients is of prognostic value for decreased risk of relapse and, in the future, could potentially be of importance in deciding the most appropriate therapeutic strategy in light of the recent immune-related treatment developments.

Associations of angiogenesis-related proteins with specific prognostic factors, breast cancer subtypes and survival outcome in early-stage breast cancer patients. A Hellenic Cooperative Oncology Group (HeCOG) trial

Several studies support an important role of angiogenesis in breast cancer growth and metastasis. The main objectives of the study were to investigate the immunohistochemical expression of vascular endothelial growth factor (VEGF) family ligands (VEGF-A and VEGF-C) and receptors (VEGFR1, VEGFR2 and VEGFR3) in breast cancer and their associations with clinicopathological parameters, cancer subtypes/subgroups and patient outcome. Formalin-fixed paraffin-embedded tumor tissue samples were collected from early-stage breast cancer patients treated with anthracycline-based chemotherapy within a randomized trial. Immunohistochemistry was performed on serial 2.5 μm thick tissue sections from tissue microarray blocks. High VEGF-A, VEGF-C, VEGFR1, VEGFR2 and VEGFR3 protein expression was observed in 11.8% (N = 87), 80.8% (N = 585), 28.1% (N = 202), 64.6% (N = 359) and 71.8% (N = 517) of the cases, respectively. Significant associations were observed among all proteins (all p-values <0.05), with the exception of the one between VEGF-C and VEGFR1 (chi-square test, p = 0.15). Tumors with high VEGF-A protein expression, as compared to tumors with low expression were more frequently ER/PgR-negative (33.3% vs. 20.8%, chi-square test, p = 0.009) and HER2-positive (44.8% vs. 20.6%, p<0.001). In addition, tumors with high VEGFR1 expression, were more frequently HER2-positive (32.8% vs. 19.6%, p<0.001), while tumors with high VEGFR3 expression were more frequently ER/PgR-negative (24.9% vs. 17.0%, p = 0.024) and HER2-positive (26.9% vs. 14.8%, p = 0.001). High VEGF-A and VEGF-C protein expression was associated with increased DFS in the entire cohort (HR = 0.57, 95% CI 0.36–0.92, Wald’s p = 0.020 and HR = 0.71, 95% CI 0.52–0.96, p = 0.025, respectively), as well as in specific subtypes/subgroups, such as HER2-positive (VEGF-A, HR = 0.32, 95% CI 0.14–0.74, p = 0.008) and triple-negative (VEGF-C, HR = 0.44, 95% CI 0.21–0.91, p = 0.027) patients. High vs. low VEGFR1 expression was an unfavorable factor for DFS in triple-negative patients (HR = 2.74, 95% CI 1.26–5.98, p = 0.011), whereas the opposite was observed among the ER/PgR-positive patients (HR = 0.69, 95% CI 0.48–0.98, p = 0.041). Regarding OS, high VEGF-C protein expression was associated with increased OS in the entire cohort (HR = 0.64, 95% CI 0.46–0.89, Wald’s p = 0.008), as well as in in specific subtypes/subgroups, such as ER/PgR-negative (HR = 0.37, 95% CI 0.20–0.71, p = 0.003) and triple-negative (HR = 0.42, 95% CI 0.19–0.90, p = 0.026) patients. In conclusion, high expression of angiogenesis-related proteins is associated with adverse clinicopathological parameters in early-stage breast cancer patients and may be surrogate markers of biologically distinct subgroups of ER/PgR-negative or triple-negative tumors with superior outcome. Further validation of our findings in independent cohorts is needed.

Interlaboratory variability of Ki67 staining in breast cancer

BACKGROUND

Postanalytic issues of Ki67 assessment in breast cancers like counting method standardisation and interrater bias have been subject of various studies, but little is known about analytic variability of Ki67 staining between pathology labs. Our aim was to study interlaboratory variability of Ki67 staining in breast cancer using tissue microarrays (TMAs) and central assessment to minimise preanalytic and postanalytic influences.

METHODS

Thirty European pathology labs stained serial slides of a TMA set of breast cancer tissues with Ki67 according to their routine in-house protocol. The Ki67-labelling index (Ki67-LI) of 70 matched samples was centrally assessed by one observer who counted all cancer cells per sample. We then tested for differences between the labs in Ki67-LI medians by analysing variance on ranks and in proportions of tumours classified as luminal A after dichotomising oestrogen receptor-positive cancers into cancers showing low (<14%, luminal A) and high (≥14%, luminal B HER2 negative) Ki67-LI using Cochran's Q.

RESULTS

Substantial differences between the 30 labs were indicated for median Ki67-LI (0.65%-33.0%, p < 0.0001) and proportion of cancers classified as luminal A (17%-57%, p < 0.0001). The differences remained significant when labs using the same antibody (MIB-1, SP6, or 30-9) were analysed separately or labs without prior participation in external quality assurance programs were excluded (p < 0.0001, respectively).

CONCLUSION

Substantial variability in Ki67 staining of breast cancer tissue was found between 30 routine pathology labs. Clinical use of the Ki67-LI for therapeutic decisions should be considered only fully aware of lab-specific reference values.

An Advanced Deep Learning Approach for Ki-67 Stained Hotspot Detection and Proliferation Rate Scoring for Prognostic Evaluation of Breast Cancer

Being a non-histone protein, Ki-67 is one of the essential biomarkers for the immunohistochemical assessment of proliferation rate in breast cancer screening and grading. The Ki-67 signature is always sensitive to radiotherapy and chemotherapy. Due to random morphological, color and intensity variations of cell nuclei (immunopositive and immunonegative), manual/subjective assessment of Ki-67 scoring is error-prone and time-consuming. Hence, several machine learning approaches have been reported; nevertheless, none of them had worked on deep learning based hotspots detection and proliferation scoring. In this article, we suggest an advanced deep learning model for computerized recognition of candidate hotspots and subsequent proliferation rate scoring by quantifying Ki-67 appearance in breast cancer immunohistochemical images. Unlike existing Ki-67 scoring techniques, our methodology uses Gamma mixture model (GMM) with Expectation-Maximization for seed point detection and patch selection and deep learning, comprises with decision layer, for hotspots detection and proliferation scoring. Experimental results provide 93% precision, 0.88% recall and 0.91% F-score value. The model performance has also been compared with the pathologists’ manual annotations and recently published articles. In future, the proposed deep learning framework will be highly reliable and beneficial to the junior and senior pathologists for fast and efficient Ki-67 scoring.

Loss of DLG5 promotes breast cancer malignancy by inhibiting the Hippo signaling pathway

Discs Large Homolog 5 (DLG5) plays an important role in the maintenance of epithelial cell polarity. Recent research showed that DLG5 is decreased in Yes-associated protein (YAP)-overexpressing cells. However, the exact relationship between DLG5 and YAP is not clear. In this study, we showed that loss of DLG5 promoted breast cancer cell proliferation by inhibiting the Hippo signaling pathway and increasing nuclear YAP expression. Furthermore, depletion of DLG5 induced epithelial-mesenchymal transition (EMT) and disrupted epithelial cell polarity, which was associated with altered expression of Scribble, ZO1, E-cadherin and N-cadherin and their mislocalization. Interestingly, we first reported that loss of DLG5 inhibited the interaction of Mst1 and Lats1 with Scribble, which was crucial for YAP activation and the transcription of TEA domain (TEAD) family members. In summary, loss of DLG5 expression promoted breast cancer malignancy by inactivating the Hippo signaling pathway and increasing nuclear YAP.

Increased Eps15 homology domain 1 and RAB11FIP3 expression regulate breast cancer progression via promoting epithelial growth factor receptor recycling

Recent research indicates that the C-terminal Eps15 homology domain 1 is associated with epithelial growth factor receptor–mediated endocytosis recycling in non-small-cell lung cancer. The aim of this study was to determine the clinical significance of Eps15 homology domain 1 gene expression in relation to phosphorylation of epithelial growth factor receptor expression in patients with breast cancer. Primary breast cancer samples from 306 patients were analyzed for Eps15 homology domain 1, RAB11FIP3, and phosphorylation of epithelial growth factor receptor expression via immunohistochemistry. The clinical significance was assessed via a multivariate Cox regression analysis, Kaplan–Meier curves, and the log-rank test. Eps15 homology domain 1 and phosphorylation of epithelial growth factor receptor were upregulated in 60.46% (185/306) and 53.92% (165/306) of tumor tissues, respectively, as assessed by immunohistochemistry. The statistical correlation analysis indicated that Eps15 homology domain 1 overexpression was positively correlated with the increases in phosphorylation of epithelial growth factor receptor ( r = 0.242, p < 0.001) and RAB11FIP3 ( r = 0.165, p = 0.005) expression. The multivariate Cox proportional hazard model analysis demonstrated that the expression of Eps15 homology domain 1 alone is a significant prognostic marker of breast cancer for the overall survival in the total, chemotherapy, and human epidermal growth factor receptor 2 (−) groups. However, the use of combined expression of Eps15 homology domain 1 and phosphorylation of epithelial growth factor receptor markers is more effective for the disease-free survival in the overall population, chemotherapy, and human epidermal growth factor receptor 2 (−) groups. Moreover, the combined markers are also significant prognostic markers of breast cancer in the human epidermal growth factor receptor 2 (+), estrogen receptor (+), and estrogen receptor (−) groups. Eps15 homology domain 1 has a tumor suppressor function, and the combined marker of Eps15 homology domain 1/phosphorylation of epithelial growth factor receptor expression was identified as a better prognostic marker in breast cancer diagnosis. Furthermore, RAB11FIP3 combines with Eps15 homology domain 1 to promote the endocytosis recycling of phosphorylation of epithelial growth factor receptor.

Impact of tissue sampling on accuracy of Ki67 immunohistochemistry evaluation in breast cancer

Background

Gene expression studies have identified molecular subtypes of breast cancer with implications to chemotherapy recommendations. For distinction of these types, a combination of immunohistochemistry (IHC) markers, including proliferative activity of tumor cells, estimated by Ki67 labeling index is used. Clinical studies are frequently based on IHC performed on tissue microarrays (TMA) with variable tissue sampling. This raises the need for evidence-based sampling criteria for individual IHC biomarker studies. We present a novel tissue sampling simulation model and demonstrate its application on Ki67 assessment in breast cancer tissue taking intratumoral heterogeneity into account.

Methods

Whole slide images (WSI) of 297 breast cancer sections, immunohistochemically stained for Ki67, were subjected to digital image analysis (DIA). Percentage of tumor cells stained for Ki67 was computed for hexagonal tiles super-imposed on the WSI. From this, intratumoral Ki67 heterogeneity indicators (Haralick’s entropy values) were extracted and used to dichotomize the tumors into homogeneous and heterogeneous subsets. Simulations with random selection of hexagons, equivalent to 0.75 mm circular diameter TMA cores, were performed. The tissue sampling requirements were investigated in relation to tumor heterogeneity using linear regression and extended error analysis.

Results

The sampling requirements were dependent on the heterogeneity of the biomarker expression. To achieve a coefficient error of 10 %, 5–6 cores were needed for homogeneous cases, 11–12 cores for heterogeneous cases; in mixed tumor population 8 TMA cores were required. Similarly, to achieve the same accuracy, approximately 4,000 nuclei must be counted when the intratumor heterogeneity is mixed/unknown. Tumors of low proliferative activity would require larger sampling (10–12 TMA cores, or 6,250 nuclei) to achieve the same error measurement results as for highly proliferative tumors.

Conclusions

Our data show that optimal tissue sampling for IHC biomarker evaluation is dependent on the heterogeneity of the tissue under study and needs to be determined on a per use basis. We propose a method that can be applied to determine the sampling strategy for specific biomarkers, tissues and study targets. In addition, our findings highlight the benefit of high-capacity computer-based IHC measurement techniques to improve accuracy of the testing.

Electronic supplementary material

The online version of this article (doi:10.1186/s13000-016-0525-z) contains supplementary material, which is available to authorized users.

St Gallen 2015 subtyping of luminal breast cancers: impact of different Ki67-based proliferation assessment methods

Ki67 has been proposed as prognostic proliferation marker in luminal breast cancer (BC), but little is known on the influence of Ki67 assessment methods on subtyping into luminal A- and B-like tumors. Our aim was to study the influence of different Ki67-labeling index (Ki67-LI) assessment methods on the proportion of BCs classified as luminal A-like. 280 early BCs were subtyped according to the St Gallen 2015 definitions into 71 % luminal (HER2 negative), 6 % luminal B-like (HER2 positive), 13 % triple negative, 1 % HER2 positive (nonluminal), and 9 % special type. Digitized whole slides were counted manually on the screen. We used nine defined counting methods to assess the Ki67-LI (including the International Ki67 in Breast Cancer Working Group recommendations), and compared the resulting medians and the proportions of cancers classified as luminal A-like according to the formerly used cut-off <20 %. Methods assessing hot spots and tumor periphery resulted in significantly higher Ki67-LI medians than those measuring an average proliferation (27.45 % vs 16.96 %, p < 0.0001). Substantially lower median Ki67-LI were found when assessing 1020 compared to counting 100, 200, 300 cells (17.65 vs 33 %, vs 28 %, vs 24.33 %, respectively; p < 0.0001), or 510 cells (20.59 %, p = 0.019). Applying a standard Ki67-LI cut-off <20 % to define low proliferation for all methods, the proportion of luminal A-like cancers varied between 13 and 44 %. The proportion of BCs classified as luminal A-like is highly influenced by the Ki67-LI assessment method. As a consequence, the selection of a specific Ki67-LI assessment method may have a direct effect on the proportion of patients considered having low-risk disease and thus influence therapeutic decision making. This calls for a standardized assessment method.

Clinicopathological Significance of the Proliferation Markers Ki67, RacGAP1, and Topoisomerase 2 Alpha in Breast Cancer

Objectives. The aims of this study are to evaluate expressions of Ki67, RacGAP1 (MgcRacGAP) and topoisomerase 2 alpha (TOP2a), the markers related with cell proliferation that have been proposed to affect the prognosis in the literature and correlate the results with clinicopathological parameters of breast cancer patients. Methods. Ki67, RacGAP1, and TOP2a antibodies were applied immunohistochemically to the tissue micrarray blocks of 457 female breast cancer patients. The results were correlated with clinical, prognostic, histopathological features, and other immunohistochemical findings (estrogen receptor [ER], progesterone receptor [PR], HER2, cytokeratin [CK]5/6, CK14, epidermal growth factor receptor [EGFR] and vimentin), statistically. Results. Ki67 expression demonstrated direct correlation with TOP2a expression, mitotic count, tumor grade, geographic necrosis, basal-like phenotype. RacGAP1 expression was directly correlated with TOP2a expression, nipple invasion, and number of metastatic lymph nodes, and it was inversely correlated with PR expression. TOP2a expression was directly correlated with vimentin and Ki67 expressions, mitotic count, tumor grade, and geographic necrosis, and nipple invasion, and negatively correlated with ER and PR expressions. Higher TOP2a and Ki67 expressions were correlated with shorter overall survival. Higher TOP2a expression and RacGAP1 positivity were directly correlated with shorter disease-free survival. Conclusion. This study showed that the overexpressions of Ki67, RacGAP1, and TOP2a affect the prognosis adversely, thus to develop target therapies against RacGAP1 and TOP2a as well as using Ki67 as a part of routine pathology practice might be beneficial in breast cancer therapy and prediction of prognosis.

Molecular Portrait of the Normal Human Breast Tissue and Its Influence on Breast Carcinogenesis

Normal human breast tissue consists of epithelial and nonepithelial cells with different molecular profiles and differentiation grades. This molecular heterogeneity is known to yield abnormal clones that may contribute to the development of breast carcinomas. Stem cells that are found in developing and mature breast tissue are either positive or negative for cytokeratin 19 depending on their subtype. These cells are able to generate carcinogenesis along with mature cells. However, scientific data remains controversial regarding the monoclonal or polyclonal origin of breast carcinomas. The majority of breast carcinomas originate from epithelial cells that normally express BRCA1. The consecutive loss of the BRCA1 gene leads to various abnormalities in epithelial cells. Normal breast epithelial cells also express hypoxia inducible factor (HIF) 1α and HIF-2α that are associated with a high metastatic rate and a poor prognosis for malignant lesions. The nuclear expression of estrogen receptor (ER) and progesterone receptor (PR) in normal human breast tissue is maintained in malignant tissue as well. Several controversies regarding the ability of ER and PR status to predict breast cancer outcome remain. Both ER and PR act as modulators of cell activity in normal human breast tissue. Ki-67 positivity is strongly correlated with tumor grade although its specific role in applied therapy requires further studies. Human epidermal growth factor receptor 2 (HER2) oncoprotein is less expressed in normal human breast specimens but is highly expressed in certain malignant lesions of the breast. Unlike HER2, epidermal growth factor receptor expression is similar in both normal and malignant tissues. Molecular heterogeneity is not only found in breast carcinomas but also in normal breast tissue. Therefore, the molecular mapping of normal human breast tissue might represent a key research area to fully elucidate the mechanisms of breast carcinogenesis.

Digital image analysis outperforms manual biomarker assessment in breast cancer

In the spectrum of breast cancers, categorization according to the four gene expression-based subtypes 'Luminal A,' 'Luminal B,' 'HER2-enriched,' and 'Basal-like' is the method of choice for prognostic and predictive value. As gene expression assays are not yet universally available, routine immunohistochemical stains act as surrogate markers for these subtypes. Thus, congruence of surrogate markers and gene expression tests is of utmost importance. In this study, 3 cohorts of primary breast cancer specimens (total n=436) with up to 28 years of survival data were scored for Ki67, ER, PR, and HER2 status manually and by digital image analysis (DIA). The results were then compared for sensitivity and specificity for the Luminal B subtype, concordance to PAM50 assays in subtype classification and prognostic power. The DIA system used was the Visiopharm Integrator System. DIA outperformed manual scoring in terms of sensitivity and specificity for the Luminal B subtype, widely considered the most challenging distinction in surrogate subclassification, and produced slightly better concordance and Cohen's κ agreement with PAM50 gene expression assays. Manual biomarker scores and DIA essentially matched each other for Cox regression hazard ratios for all-cause mortality. When the Nottingham combined histologic grade (Elston-Ellis) was used as a prognostic surrogate, stronger Spearman's rank-order correlations were produced by DIA. Prognostic value of Ki67 scores in terms of likelihood ratio χ(2) (LR χ(2)) was higher for DIA that also added significantly more prognostic information to the manual scores (LR-Δχ(2)). In conclusion, the system for DIA evaluated here was in most aspects a superior alternative to manual biomarker scoring. It also has the potential to reduce time consumption for pathologists, as many of the steps in the workflow are either automatic or feasible to manage without pathological expertise.

Expression of cell cycle markers is predictive of the response to primary systemic therapy of locally advanced breast cancer

We aimed to analyze to what extent expression of four cell cycle regulation markers-minichromosome maintenance protein (MCM2), Ki-67, cyclin A, and phosphohistone-H3 (PHH3)-predict response to primary systemic therapy in terms of pathological complete remission (pCR). In search of an accurate and reproducible scoring method, we compared computer-assisted (CA) and routine visual assessment (VA) of immunoreactivity. We included 57 patients with breast cancer in the study. The cell cycle markers were detected using immunohistochemistry on pre-therapy core biopsy samples. Parallel CA (validated by manual labeling) and standard VA were performed and compared for diagnostic agreement and predictive value for pCR. CA and VA results were dichotomized based on receiver operating characteristic analysis defined optimal cut-off values. "High" was defined by staining scores above the optimal cut-off, while "low" had staining scores below the optimal cut-off. The CA method resulted in significantly lower values for Ki-67 and MCM2 compared to VA (mean difference, -3.939 and -4.323). Diagnostic agreement was highest for cyclin A and PHH3 (-0.586 and -0.666, respectively). Regardless of the method (CA/VA) used, all tested markers were predictive of pCR. Optimal cut-off-based dichotomization improved diagnostic agreement between the CA and VA methods for every marker, in particular for MCM2 (κ = 1, p < 0.000). Cyclin A displayed excellent agreement (κ = 0.925; p < 0.000), while Ki-67 and PHH3 showed good agreement (κ = 0.789, p < 0.000 and κ = 0.794, p < 0.000, respectively). We found all cell cycle markers (Ki-67, MCM2, cyclin A, and PHH3) predictive of pCR. Diagnostic agreement between CA and VA was better at lower staining scores but improved after optimal cut-off-based dichotomization.

Bimodality of intratumor Ki67 expression is an independent prognostic factor of overall survival in patients with invasive breast carcinoma

Proliferative activity, assessed by Ki67 immunohistochemistry (IHC), is an established prognostic and predictive biomarker of breast cancer (BC). However, it remains under-utilized due to lack of standardized robust measurement methodologies and significant intratumor heterogeneity of expression. A recently proposed methodology for IHC biomarker assessment in whole slide images (WSI), based on systematic subsampling of tissue information extracted by digital image analysis (DIA) into hexagonal tiling arrays, enables computation of a comprehensive set of Ki67 indicators, including intratumor variability. In this study, the tiling methodology was applied to assess Ki67 expression in WSI of 152 surgically removed Ki67-stained (on full-face sections) BC specimens and to test which, if any, Ki67 indicators can predict overall survival (OS). Visual Ki67 IHC estimates and conventional clinico-pathologic parameters were also included in the study. Analysis revealed linearly independent intrinsic factors of the Ki67 IHC variance: proliferation (level of expression), disordered texture (entropy), tumor size and Nottingham Prognostic Index, bimodality, and correlation. All visual and DIA-generated indicators of the level of Ki67 expression provided significant cutoff values as single predictors of OS. However, only bimodality indicators (Ashman's D, in particular) were independent predictors of OS in the context of hormone receptor and HER2 status. From this, we conclude that spatial heterogeneity of proliferative tumor activity, measured by DIA of Ki67 IHC expression and analyzed by the hexagonal tiling approach, can serve as an independent prognostic indicator of OS in BC patients that outperforms the prognostic power of the level of proliferative activity.

A methodology for comprehensive breast cancer Ki67 labeling index with intra-tumor heterogeneity appraisal based on hexagonal tiling of digital image analysis data

Digital image analysis (DIA) enables higher accuracy, reproducibility, and capacity to enumerate cell populations by immunohistochemistry; however, the most unique benefits may be obtained by evaluating the spatial distribution and intra-tissue variance of markers. The proliferative activity of breast cancer tissue, estimated by the Ki67 labeling index (Ki67 LI), is a prognostic and predictive biomarker requiring robust measurement methodologies. We performed DIA on whole-slide images (WSI) of 302 surgically removed Ki67-stained breast cancer specimens; the tumour classifier algorithm was used to automatically detect tumour tissue but was not trained to distinguish between invasive and non-invasive carcinoma cells. The WSI DIA-generated data were subsampled by hexagonal tiling (HexT). Distribution and texture parameters were compared to conventional WSI DIA and pathology report data. Factor analysis of the data set, including total numbers of tumor cells, the Ki67 LI and Ki67 distribution, and texture indicators, extracted 4 factors, identified as entropy, proliferation, bimodality, and cellularity. The factor scores were further utilized in cluster analysis, outlining subcategories of heterogeneous tumors with predominant entropy, bimodality, or both at different levels of proliferative activity. The methodology also allowed the visualization of Ki67 LI heterogeneity in tumors and the automated detection and quantitative evaluation of Ki67 hotspots, based on the upper quintile of the HexT data, conceptualized as the "Pareto hotspot". We conclude that systematic subsampling of DIA-generated data into HexT enables comprehensive Ki67 LI analysis that reflects aspects of intra-tumor heterogeneity and may serve as a methodology to improve digital immunohistochemistry in general.

The region-of-interest size impacts on Ki67 quantification by computer-assisted image analysis in breast cancer

Therapeutic decision-making in breast cancer depends on histopathologic biomarkers and is influenced by the Ki67 proliferation index. Computer-assisted image analysis (CAIA) promises to improve Ki67 quantification. Several commercial applications have been developed for semiautomated CAIA-based Ki67 quantification, many of which rely on measurements in user-defined regions of interest (ROIs). Because of intratumoral proliferative heterogeneity, definition of the ROI is an important step in the analytical procedure. This study explores the ROI size impacts on Ki67 quantification. Whole-slide sections of 100 breast cancers were immunostained with the anti-Ki67 antibody 30-9 and were analyzed on the iScan Coreo digital pathology platform using a Food and Drug Administration-cleared Ki67 quantification software version v5.3 (Virtuoso; Ventana, Tucson, TX). For each case, the Ki67 labeling index (LI) was determined in multiple ROIs of gradually increasing size centered around a high-proliferation area. The spatial Ki67 decline was modeled with nonlinear regression. Depending on the ROI size, the median Ki67 LI varied between 55% and 15%. The proportion of tumors classified as Ki67 low according to the St Gallen 2013/2015 cutoff increased from 2% to 56%, as the ROI size increased from 50 to 10,000 cells captured. The interrater reliability of conventional Ki67 assessment versus CAIA-based Ki67 quantification was also dependent on the ROI size and varied between slight and almost perfect agreement (Cohen κ = 0.06-0.85). In conclusion, the ROI size is a critically important parameter for semiautomated Ki67 quantification by CAIA. Ki67 LIs determined on platforms like iScan Coreo/Virtuoso require an ROI size adjustment, for which we offer a downloadable data transformation tool.

Quantum dots-based quantitative and in situ multiple imaging on ki67 and cytokeratin to improve ki67 assessment in breast cancer

Background

As a marker for tumor cell proliferation, Ki67 has important impacts on breast cancer (BC) prognosis. Although immunohistochemical staining is the current standard method, variations in analytical practice make it difficult for pathologists to manually measure Ki67 index. This study was to develop a fluorescent spectrum-based quantitative analysis of Ki67 expression by quantum-dots (QDs) multiple imaging technique.

Methods

A QDs-based in situ multiple fluorescent imaging method was developed, which stained nuclear Ki67 as red signal and cytoplasmic cytokeratin (CK) as green signal. Both Ki67 and CK signals were automatically separated and quantified by professional spectrum analysis software. This technique was applied to tissue microarrays from 240 BC patients. Both Ki67 and CK values, and Ki67/CK ratio were obtained for each patient, and their prognostic value on 5-year disease free survival was assessed.

Results

This method simultaneously stains nuclear Ki67 and cytoplasmic CK with clear signal contrast, making it easy for signal separation and quantification. The total fluorescent signal intensities of both Ki67 sum and CK sum were obtained, and Ki67/CK ratio calculated. Ki67 sum and Ki67/CK ratio were each attributed into two grades by X-tile software based on the best P value principle. Multivariate analysis showed Ki67 grade (P = 0.047) and Ki67/CK grade (P = 0.004) were independent prognostic factors. Furthermore, area under curve (AUC) of ROC analysis for Ki67/CK grade (AUC: 0.683, 95%CI: 0.613–0.752) was higher than Ki67 grade (AUC: 0.665, 95%CI: 0.596–0.734) and HER-2 gene (AUC: 0.586, 95%CI: 0.510–0.661), but lower than N stage (AUC: 0.760, 95%CI: 0.696–0.823) and histological grade (AUC: 0.756, 95%CI: 0.692–0.820) on predicting the risk for recurrence.

Conclusions

A QDs-based quantitative and in situ multiple imaging on Ki67 and CK was developed to improve Ki67 assessment in BC, and Ki67/CK grade had better performance than Ki67 grade in predicting prognosis.