Automated Quantitative Analysis of p53, Cyclin D1, Ki67 and pERK Expression in Breast Carcinoma Does Not Differ from Expert Pathologist Scoring and Correlates with Clinico-Pathological Characteristics

Deep Learning-Based H-Score Quantification of Immunohistochemistry-Stained Images

Immunohistochemistry (IHC) is a well-established and commonly used staining method for clinical diagnosis and biomedical research. In most IHC images, the target protein is conjugated with a specific antibody and stained using diaminobenzidine (DAB), resulting in a brown coloration, whereas hematoxylin serves as a blue counterstain for cell nuclei. The protein expression level is quantified through the H-score, calculated from DAB staining intensity within the target cell region. Traditionally, this process requires evaluation by 2 expert pathologists, which is both time consuming and subjective. To enhance the efficiency and accuracy of this process, we have developed an automatic algorithm for quantifying the H-score of IHC images. To characterize protein expression in specific cell regions, a deep learning model for region recognition was trained based on hematoxylin staining only, achieving pixel accuracy for each class ranging from 0.92 to 0.99. Within the desired area, the algorithm categorizes DAB intensity of each pixel as negative, weak, moderate, or strong staining and calculates the final H-score based on the percentage of each intensity category. Overall, this algorithm takes an IHC image as input and directly outputs the H-score within a few seconds, significantly enhancing the speed of IHC image analysis. This automated tool provides H-score quantification with precision and consistency comparable to experienced pathologists but at a significantly reduced cost during IHC diagnostic workups. It holds significant potential to advance biomedical research reliant on IHC staining for protein expression quantification.

A simple digital image analysis system for automated Ki67 assessment in primary breast cancer

AIMS

Ki67 is a well-established immunohistochemical marker associated with cell proliferation that has prognostic and predictive value in breast cancer. Quantitative evaluation of Ki67 is traditionally performed by assessing stained tissue slides with light microscopy. Automated image analysis systems have become available and, if validated, could provide greater standardisation and improved precision of Ki67 scoring. Here, we aimed to evaluate the use of the Cognition Master Professional Suite (CogM) image analysis software, which is a simple system for scoring Ki67 in primary breast cancer samples.

METHODS AND RESULTS

Sections from 94 core-cut biopsies, 20 excision specimens and 29 pairs of core-cut biopsies and excision specimens were stained for Ki67 with MIB1 antibody and the Dako EnVision FLEX Detection System. Stained slides were scanned to convert them to digital data. Computer-based Ki67 scoring was performed with CogM. Manual Ki67 scoring assessment was conducted on previously stained sections from the same biopsies with a clinically validated system that had been calibrated against the risk of recurrence. A high correlation between manual and digital scores was observed [r_Cores  = 0.92, 95% confidence interval (CI) 0.87-0.94, P < 0.0001; r_Excisions  = 0.95, 95% CI 0.86-0.98, P < 0.0001] and there was no significant bias between them (P = 0.45). There was also a high correlation of Ki67 scores between paired core-cut biopsies and excision specimens when CogM was used (r = 0.78, 95% CI 0.59-0.89, P < 0.0001).

CONCLUSIONS

CogM image analysis allows for standardised automated Ki67 scoring that accurately replicates previously clinically validated and calibrated manual scores.

Brazilian berry extract (Myrciaria jaboticaba): A promising therapy to minimize prostatic inflammation and oxidative stress

BACKGROUND

Brazilian berry is a fruit popularly known as "Jaboticaba," rich in bioactive compounds with antioxidant and anti-inflammatory properties. Senescence and overweight are increasing worldwide and are considered risk factors to prostatic pathogenesis mainly due to oxidative and inflammatory processes induction. Thus, this study aimed to evaluate the effect of two increasing doses of the patented jaboticaba peel extract (PJE) on oxidative-stress and inflammation in the prostate of aging or high-fat-fed aging mice.

METHODS

PJE and/or high-fat diet (HFD) treatments started with 11-month-old mice and lasted 60 days. The levels or the immunoexpression of different inflammatory (nuclear factor κB [NFκB], CD3+, cyclooxygenase 2 [COX-2], toll-like receptor 4 [TLR4], phosphorylated signal transducers and activators of transcription 3 [pSTAT-3], tumor necrosis factor α [TNF-α], interleukin 6 [IL-6], and IL-1β) and oxidative-stress (catalase, superoxide dismutase 2 [SOD2], glutathione reductase [GSR], reduced glutathione, and glutathione peroxidase 3 [GPx3]) related molecules were analyzed by western-blotting, immunohistochemistry, and enzyme-linked immunosorbent assays.

RESULTS

Both PJE doses reduced the levels of oxidative-stress-related molecules (GPx3, GSR, catalase), lipid peroxidation (4-hydroxynonenal), inflammatory mediators (COX-2, TNF-α, and pSTAT-3) and CD3+ T cells number, which were associated with the maintenance of the glandular morphological integrity in aging and HFD-fed-aging mice. Nevertheless, only the high PJE dose reduced the NFκB and TLR4 levels in aging mice; and SOD2, IL-6, and IL-1β levels in HFD-aging mice. Aging itself promoted an oxidative inflammation in the prostate, interfering in the levels of the different oxidative-stress, lipid peroxidation, and inflammatory mediators evaluated, in association with high incidence of prostate epithelial and stromal damages. The HFD intake intensified aging alterations, showing an unfavorable prostatic microenvironment prone to oxidative and inflammatory damages.

CONCLUSIONS

PJE exerted a dose-dependent effect controlling inflammation and oxidative-stress in aging and HFD-fed aging mice prostate. This fact contributed to prostate microenvironment balance recovery, preserving the tissue architecture of this gland. Thus, the PJE emerges as a potential therapy to prevent inflammation and oxidative stress in the prostate.

An epigenetic increase in mitochondrial fission by MiD49 and MiD51 regulates the cell cycle in cancer: Diagnostic and therapeutic implications

Excessive proliferation and apoptosis-resistance are hallmarks of cancer. Increased dynamin-related protein 1 (Drp1)-mediated mitochondrial fission is one of the mediators of this phenotype. Mitochondrial fission that accompanies the nuclear division is called mitotic fission and occurs when activated Drp1 binds partner proteins on the outer mitochondrial membrane. We examine the role of Drp1-binding partners, mitochondrial dynamics protein of 49 and 51 kDa (MiD49 and MiD51), as drivers of cell proliferation and apoptosis-resistance in non-small cell lung cancer (NSCLC) and invasive breast carcinoma (IBC). We also evaluate whether inhibiting MiDs can be therapeutically exploited to regress cancer. We show that MiD levels are pathologically elevated in NSCLC and IBC by an epigenetic mechanism (decreased microRNA-34a-3p expression). MiDs silencing causes cell cycle arrest through (a) increased expression of cell cycle inhibitors, p27^Kip1 and p21^Waf1 , (b) inhibition of Drp1, and (c) inhibition of the Akt-mTOR-p70S6K pathway. Silencing MiDs leads to mitochondrial fusion, cell cycle arrest, increased apoptosis, and tumor regression in a xenotransplant NSCLC model. There are positive correlations between MiD expression and tumor size and grade in breast cancer patients and inverse correlations with survival in NSCLC patients. The microRNA-34a-3p-MiDs axis is important to cancer pathogenesis and constitutes a new therapeutic target.

Intravital imaging reveals systemic ezrin inhibition impedes cancer cell migration and lymph node metastasis in breast cancer

Background

Limited understanding of the cancer biology of metastatic sites is a major factor contributing to poor outcomes in cancer patients. The regional lymph nodes are the most common site of metastasis in most solid cancers and their involvement is a strong predictor of relapse in breast cancer (BC). We have previously shown that ezrin, a cytoskeletal–membrane linker protein, is associated with lymphovascular invasion and promotes metastatic progression in BC. However, the efficacy of pharmacological inhibition of ezrin in blocking cancer cell migration and metastasis remains unexplored in BC.

Methods

We quantified ezrin expression in a BC tissue microarray (n = 347) to assess its correlation with risk of relapse. Next, we developed a quantitative intravital microscopy (qIVM) approach, using a syngeneic lymphatic reporter mouse tumor model, to investigate the effect of systemic ezrin inhibition on cancer cell migration and metastasis.

Results

We show that ezrin is expressed at significantly higher levels in lymph node metastases compared to matched primary tumors, and that a high tumor ezrin level is associated with increased risk of relapse in BC patients with regional disease. Using qIVM, we observe a subset of cancer cells that retain their invasive and migratory phenotype at the tumor-draining lymph node. We further show that systemic inhibition of ezrin, using a small molecule compound (NSC668394), impedes the migration of cancer cells in vivo. Furthermore, systemic ezrin inhibition leads to reductions in metastatic burden at the distal axillary lymph node and lungs.

Conclusions

Our findings demonstrate that the tumor ezrin level act as an independent biomarker in predicting relapse and provide a rationale for therapeutic targeting of ezrin to reduce the metastatic capacity of cancer cells in high-risk BC patients with elevated ezrin expression.

Electronic supplementary material

The online version of this article (10.1186/s13058-018-1079-7) contains supplementary material, which is available to authorized users.

Differential cyclooxygenase expression levels and survival associations in type I and type II ovarian tumors

BACKGROUND

High cyclooxygenase (COX)-2 expression in ovarian tumors has been associated with poor prognosis, but the role of COX-1 expression and its relation to survival is less clear. Here, we evaluated COX expression and associations with survival outcomes between type I (clear cell, mucinous, low grade endometrioid and low grade serous) and type II (high grade serous and high grade endometrioid) ovarian tumors.

METHODS

We developed and validated a new COX-1 antibody, and conducted immunohistochemical (IHC) staining for COX-1 and COX-2 on a tissue microarray (TMA) of 190 primary ovarian tumors. In addition to standard IHC scoring and H-scores to combine the percentage of positive cells and staining intensity, we also measured COX-1 and COX-2 mRNA expression by QPCR. High expression was defined as greater than or equal to median values. Clinical characteristics and disease outcomes were ascertained from medical records. Associations with disease-free survival (DFS) and overall survival (OS) were quantified by hazard ratios (HRs) and confidence intervals (CIs) from proportional hazards regression.

RESULTS

Type I tumors had high COX-2 expression, while type II tumors had high COX-1 expression. In multivariable adjusted regression models, higher COX-1 mRNA expression was associated with shorter DFS (HR: 6.37, 95% CI: 1.84-22.01) and OS (HR: 2.26, 95% CI: 1.04-4.91), while higher H-scores for COX-2 expression were associated with shorter DFS (HR: 1.92, 95% CI: 1.06-3.49). Stratified analysis indicated that COX-2 was significantly associated with DFS among cases with Type II tumors (HR: 1.93, 95% CI: 1.06-3.53).

CONCLUSIONS

These findings suggest that ovarian tumor type contributes to differences in COX expression levels and associations with survival.

The use of automated Ki67 analysis to predict Oncotype DX risk-of-recurrence categories in early-stage breast cancer

Ki67 is a commonly used marker of cancer cell proliferation, and has significant prognostic value in breast cancer. In spite of its clinical importance, assessment of Ki67 remains a challenge, as current manual scoring methods have high inter- and intra-user variability. A major reason for this variability is selection bias, in that different observers will score different regions of the same tumor. Here, we developed an automated Ki67 scoring method that eliminates selection bias, by using whole-slide analysis to identify and score the tumor regions with the highest proliferative rates. The Ki67 indices calculated using this method were highly concordant with manual scoring by a pathologist (Pearson’s r = 0.909) and between users (Pearson’s r = 0.984). We assessed the clinical validity of this method by scoring Ki67 from 328 whole-slide sections of resected early-stage, hormone receptor-positive, human epidermal growth factor receptor 2-negative breast cancer. All patients had Oncotype DX testing performed (Genomic Health) and available Recurrence Scores. High Ki67 indices correlated significantly with several clinico-pathological correlates, including higher tumor grade (1 versus 3, P<0.001), higher mitotic score (1 versus 3, P<0.001), and lower Allred scores for estrogen and progesterone receptors (P = 0.002, 0.008). High Ki67 indices were also significantly correlated with higher Oncotype DX risk-of-recurrence group (low versus high, P<0.001). Ki67 index was the major contributor to a machine learning model which, when trained solely on clinico-pathological data and Ki67 scores, identified Oncotype DX high- and low-risk patients with 97% accuracy, 98% sensitivity and 80% specificity. Automated scoring of Ki67 can thus successfully address issues of consistency, reproducibility and accuracy, in a manner that integrates readily into the workflow of a pathology laboratory. Furthermore, automated Ki67 scores contribute significantly to models that predict risk of recurrence in breast cancer.

Comparison of immunohistochemistry with PCR for assessment of ER, PR, and Ki-67 and prediction of pathological complete response in breast cancer

Background

Proliferation may predict response to neoadjuvant therapy of breast cancer and is commonly assessed by manual scoring of slides stained by immunohistochemistry (IHC) for Ki-67 similar to ER and PgR. This method carries significant intra- and inter-observer variability. Automatic scoring of Ki-67 with digital image analysis (qIHC) or assessment of MKI67 gene expression with RT-qPCR may improve diagnostic accuracy.

Methods

Ki-67 IHC visual assessment was compared to the IHC nuclear tool (AperioTM) on core biopsies from a randomized neoadjuvant clinical trial. Expression of ESR1, PGR and MKI67 by RT-qPCR was performed on RNA extracted from the same formalin-fixed paraffin-embedded tissue. Concordance between the three methods (vIHC, qIHC and RT-qPCR) was assessed for all 3 markers. The potential of Ki-67 IHC and RT-qPCR to predict pathological complete response (pCR) was evaluated using ROC analysis and non-parametric Mann-Whitney Test.

Results

Correlation between methods (qIHC versus RT-qPCR) was high for ER and PgR (spearman´s r = 0.82, p < 0.0001 and r = 0.86, p < 0.0001, respectively) resulting in high levels of concordance using predefined cut-offs. When comparing qIHC of ER and PgR with RT-qPCR of ESR1 and PGR the overall agreement was 96.6 and 91.4%, respectively, while overall agreement of visual IHC with RT-qPCR was slightly lower for ER/ESR1 and PR/PGR (91.2 and 92.9%, respectively). In contrast, only a moderate correlation was observed between qIHC and RT-qPCR continuous data for Ki-67/MKI67 (Spearman’s r = 0.50, p = 0.0001). Up to now no predictive cut-off for Ki-67 assessment by IHC has been established to predict response to neoadjuvant chemotherapy. Setting the desired sensitivity at 100%, specificity for the prediction of pCR (ypT0ypN0) was significantly higher for mRNA than for protein (68.9% vs. 22.2%). Moreover, the proliferation levels in patients achieving a pCR versus not differed significantly using MKI67 RNA expression (Mann-Whitney p = 0.002), but not with qIHC of Ki-67 (Mann-Whitney p = 0.097) or vIHC of Ki-67 (p = 0.131).

Conclusion

Digital image analysis can successfully be implemented for assessing ER, PR and Ki-67. IHC for ER and PR reveals high concordance with RT-qPCR. However, RT-qPCR displays a broader dynamic range and higher sensitivity than IHC. Moreover, correlation between Ki-67 qIHC and RT-qPCR is only moderate and RT-qPCR with MammaTyper® outperforms qIHC in predicting pCR. Both methods yield improvements to error-prone manual scoring of Ki-67. However, RT-qPCR was significantly more specific.

Electronic supplementary material

The online version of this article (doi:10.1186/s12885-017-3111-1) contains supplementary material, which is available to authorized users.

X-Ray of Excised Cancerous Breast Tissue Does Not Affect Clinical Biomarker Expression

CONTEXT

College of American Pathologists (CAP) and the American Society of Clinical Oncology have emphasized the need to reduce preanalytic variables for evaluating predictive biomarker expression in breast cancer. Postoperative x-ray of excised breast tissue is commonplace, yet is a variable that has not been investigated previously. We asked whether such radiation affects expression of relevant biomarkers.

DESIGN

A previous study found that human breast cancers grown in mice demonstrate the same immunohistochemical and molecular profiles as the original tumors. Thirteen patient-derived xenografts were harvested fresh and divided for specimen radiography and a matched nonirradiated control, while following CAP/ASCO guidelines for cold ischemia time and fixation. Samples were processed in a tissue microarray for immunohistochemistry. Estrogen receptor (ER), progesterone receptor (PR), p53, and Ki67 staining was evaluated using an optimized scoring algorithm performed on digitally scanned slides. Samples were also scored manually by a blinded pathologist using the H-score method, and HER2 by the CAP/ASCO 2013 protocol. Histologic scores were compared by analysis of variance.

RESULTS

There was no significant difference in quantity or intensity of staining between irradiated and nonirradiated samples for estrogen receptor (P=0.28), p53 (P=0.96), and Ki67 (P=0.94). A small but statistically significant difference was observed for PR (P=0.0058). HER2 staining was similarly unchanged in the 1 tumor exhibiting 3+ staining.

CONCLUSIONS

Our study demonstrates that x-ray of breast carcinomas does not significantly affect the expression of predictive biomarkers, with the exception of PR for unclear reasons. It also highlights the utility of the patient-derived xenograft model for biomarker studies.

Ki67 Heterogeneity in Estrogen Receptor-Positive Breast Cancers: Which Tumor Type Has the Most Heterogeneity?

Heterogeneity of Ki67 expression, often seen in breast cancer, can make evaluation of the expression of this marker difficult and give rise to confusion when considering adjuvant treatments for patients. Herein, we investigated estrogen receptor-positive breast cancers to reveal the tumor characteristics associated with Ki67 heterogeneity. Surgical specimens from 85 invasive ductal carcinomas of no special type and 13 invasive lobular carcinomas were examined. We first calculated the differences between Ki67 expression in a hot spot and those in 4 random fields on the same slide. We then evaluated Ki67 heterogeneity within the tumor, based on these differences. Among clinicopathological factors, solid-tubular carcinoma, an architectural growth pattern subtype of invasive ductal carcinoma, correlated with high Ki67 heterogeneity (P < .05). Our results indicate that we might need to be aware of histological patterns when selecting appropriate microscopic fields for evaluating Ki67 expression.

Standardized Ki67 Diagnostics Using Automated Scoring--Clinical Validation in the GeparTrio Breast Cancer Study

PURPOSE

Scoring proliferation through Ki67 immunohistochemistry is an important component in predicting therapy response to chemotherapy in patients with breast cancer. However, recent studies have cast doubt on the reliability of "visual" Ki67 scoring in the multicenter setting, particularly in the lower, yet clinically important, proliferation range. Therefore, an accurate and standardized Ki67 scoring is pivotal both in routine diagnostics and larger multicenter studies.

EXPERIMENTAL DESIGN

We validated a novel fully automated Ki67 scoring approach that relies on only minimal a priori knowledge on cell properties and requires no training data for calibration. We applied our approach to 1,082 breast cancer samples from the neoadjuvant GeparTrio trial and compared the performance of automated and manual Ki67 scoring.

RESULTS

The three groups of autoKi67 as defined by low (≤ 15%), medium (15.1%-35%), and high (>35%) automated scores showed pCR rates of 5.8%, 16.9%, and 29.5%, respectively. AutoKi67 was significantly linked to prognosis with overall and progression-free survival P values P(OS) < 0.0001 and P(PFS) < 0.0002, compared with P(OS) < 0.0005 and P(PFS) < 0.0001 for manual Ki67 scoring. Moreover, automated Ki67 scoring was an independent prognosticator in the multivariate analysis with P(OS) = 0.002, P(PFS) = 0.009 (autoKi67) versus P(OS) = 0.007, PPFS = 0.004 (manual Ki67).

CONCLUSIONS

The computer-assisted Ki67 scoring approach presented here offers a standardized means of tumor cell proliferation assessment in breast cancer that correlated with clinical endpoints and is deployable in routine diagnostics. It may thus help to solve recently reported reliability concerns in Ki67 diagnostics.

The impact of FGFR1 and FRS2α expression on sorafenib treatment in metastatic renal cell carcinoma

Background

Angiogenesis plays a role in tumor growth and is partly mediated by factors in both the fibroblast growth factor (FGF) and vascular endothelial growth factor (VEGF) pathways. Durable clinical responses with VEGF tyrosine kinase inhibitors (TKIs) may be limited by intrinsic tumor resistance. We hypothesized that FGF signaling may impact clinical responses to sorafenib.

Methods

Nephrectomy material was available from 40 patients with metastatic renal cell carcinoma (RCC) enrolled in a phase II clinical trial of sorafenib ± interferon (ClinicalTrials.gov Identifier NCT00126594). Fibroblast growth factor receptor 1 (FGFR1) and fibroblast growth factor receptor substrate 2 alpha (FRS2α) expression was assessed by in situ hybridization and immunofluorescence, respectively. The relationship between fibroblast growth factor pathway marker levels and progression-free survival (PFS) was analyzed using Kaplan-Meier and Cox proportional hazards regression methods.

Results

Univariate analysis indicated that more intense FGFR1 staining was associated with shorter PFS (log-rank P = 0.0452), but FRS2α staining was not significantly associated with PFS (log-rank P = 0.2610). Multivariate Cox proportional hazards regression models were constructed for FGFR1 and FRS2α individually, adjusting for baseline Eastern Cooperative Oncology Group performance status, treatment arm and anemia status. When adjusted for each of these variables, the highest intensity level of FGFR1 (level 3 or 4) had increased progression risk relative to the lowest intensity level of FGFR1 (level 1) (P = 0.0115). The highest intensity level of FRS2α (level 3 or 4) had increased progression risk relative to the lowest intensity level of FRS2α (level 1) (P = 0.0126).

Conclusions

Increased expression of FGFR1 and FRS2α was associated with decreased PFS among patients with metastatic RCC treated with sorafenib. The results suggest that FGF pathway activation may impact intrinsic resistance to VEGF receptor inhibition.

Electronic supplementary material

The online version of this article (doi:10.1186/s12885-015-1302-1) contains supplementary material, which is available to authorized users.

Increased growth of colorectal liver metastasis following partial hepatectomy

Nearly 50 % of colorectal cancer (CRC) patients develop liver metastases with liver resection being the only option to cure patients. Residual micrometastases or circulating tumor cells are considered a cause of tumor relapse. This work investigates the influence of partial hepatectomy (PH) on the growth and molecular composition of CRC liver metastasis in a syngeneic rat model. One million CC531 colorectal tumor cells were implanted via the portal vein in WAG/Rij rats followed by a 30 % PH a day later. Control groups either received tumor cells followed by a sham-operation or were injected with a buffer solution followed by PH. Animals were examined with magnetic resonance imaging (MRI) and liver tissues were processed for immunolabeling and PCR analysis. One-third PH was associated with an almost threefold increase in relative tumor mass (MRI volumetry: 2.8-fold and transcript levels of CD44: 2.3-fold). Expression of molecular markers for invasiveness and aggressiveness (CD49f, CXCR4, Axin2 and c-met) was increased following PH, however with no significant differences when referring to the relative expression levels (relating to tumor mass). Liver metastases demonstrated a significantly higher proliferation rate (Ki67) 2 weeks following PH and cell divisions also increased in the surrounding liver tissue. Following PH, the stimulated growth of metastases clearly exceeded the compensation in liver volume with long-lasting proliferative effects. However, the distinct tumor composition was not influenced by liver regeneration. Future investigations should focus on the inhibition of cell cycle (i.e. systemic therapy strategies, irradiation) to hinder liver regeneration and therefore restrain tumor growth.