Evaluation of the Use of Single- and Multi-Magnification Convolutional Neural Networks for the Determination and Quantitation of Lesions in Nonclinical Pathology Studies.

Digital pathology platforms with integrated artificial intelligence have the potential to increase the efficiency of the nonclinical pathologist's workflow through screening and prioritizing slides with lesions and highlighting areas with specific lesions for review. Herein, we describe the comparison of various single- and multi-magnification convolutional neural network (CNN) architectures to accelerate the detection of lesions in tissues. Different models were evaluated for defining performance characteristics and efficiency in accurately identifying lesions in 5 key rat organs (liver, kidney, heart, lung, and brain). Cohorts for liver and kidney were collected from TG-GATEs open-source repository, and heart, lung, and brain from internally selected R&D studies. Annotations were performed, and models were trained on each of the available lesion classes in the available organs. Various class-consolidation approaches were evaluated from generalized lesion detection to individual lesion detections. The relationship between the amount of annotated lesions and the precision/accuracy of model performance is elucidated. The utility of multi-magnification CNN implementations in specific tissue subtypes is also demonstrated. The use of these CNN-based models offers users the ability to apply generalized lesion detection to whole-slide images, with the potential to generate novel quantitative data that would not be possible with conventional image analysis techniques.

The EMT Transcription Factor ZEB2 Promotes Proliferation of Primary and Metastatic Melanoma While Suppressing an Invasive, Mesenchymal-Like Phenotype.

Epithelial-to-mesenchymal transition (EMT)-inducing transcription factors (TF) are well known for their ability to induce mesenchymal states associated with increased migratory and invasive properties. Unexpectedly, nuclear expression of the EMT-TF ZEB2 in human primary melanoma has been shown to correlate with reduced invasion. We report here that ZEB2 is required for outgrowth for primary melanomas and metastases at secondary sites. Ablation of Zeb2 hampered outgrowth of primary melanomas in vivo, whereas ectopic expression enhanced proliferation and growth at both primary and secondary sites. Gain of Zeb2 expression in pulmonary-residing melanoma cells promoted the development of macroscopic lesions. In vivo fate mapping made clear that melanoma cells undergo a conversion in state where ZEB2 expression is replaced by ZEB1 expression associated with gain of an invasive phenotype. These findings suggest that reversible switching of the ZEB2/ZEB1 ratio enhances melanoma metastatic dissemination. SIGNIFICANCE: ZEB2 function exerts opposing behaviors in melanoma by promoting proliferation and expansion and conversely inhibiting invasiveness, which could be of future clinical relevance. GRAPHICAL ABSTRACT: http://cancerres.aacrjournals.org/content/canres/80/14/2983/F1.large.jpg.

Low expression of pro-apoptotic proteins Bax, Bak and Smac indicates prolonged progression-free survival in chemotherapy-treated metastatic melanoma.

Despite the introduction of novel targeted therapies, chemotherapy still remains the primary treatment for metastatic melanoma in poorly funded healthcare environments or in case of disease relapse, with no reliable molecular markers for progression-free survival (PFS) available. As chemotherapy primarily eliminates cancer cells by apoptosis, we here evaluated if the expression of key apoptosis regulators (Bax, Bak, Bcl-2, Bcl-xL, Smac, Procaspase-9, Apaf-1, Procaspase-3 and XIAP) allows prognosticating PFS in stage III/IV melanoma patients. Following antibody validation, marker expression was determined by automated and manual scoring of immunohistochemically stained tissue microarrays (TMAs) constructed from treatment-naive metastatic melanoma biopsies. Interestingly and counter-intuitively, low expression of the pro-apoptotic proteins Bax, Bak and Smac indicated better prognosis (log-rank p < 0.0001, p = 0.0301 and p = 0.0227 for automated and p = 0.0422, p = 0.0410 and p = 0.0073 for manual scoring). These findings were independently validated in the cancer genome atlas (TCGA) metastatic melanoma cohort (TCGA-SKCM) at transcript level (log-rank p = 0.0004, p = 0.0104 and p = 0.0377). Taking expression heterogeneity between the markers in individual tumour samples into account allowed defining combinatorial Bax, Bak, Smac signatures that were associated with significantly increased PFS (p = 0.0002 and p = 0.0028 at protein and transcript level, respectively). Furthermore, combined low expression of Bax, Bak and Smac allowed predicting prolonged PFS (> 12 months) on a case-by-case basis (area under the receiver operating characteristic curve (ROC AUC) = 0.79). Taken together, our results therefore suggest that Bax, Bak and Smac jointly define a signature with potential clinical utility in chemotherapy-treated metastatic melanoma.

Pilot study of bevacizumab in combination with docetaxel and cyclophosphamide as adjuvant treatment for patients with early stage HER-2 negative breast cancer, including analysis of candidate circulating markers of cardiac toxicity: ICORG 08-10 trial.

BACKGROUND: Combining bevacizumab and chemotherapy produced superior response rates compared with chemotherapy alone in metastatic breast cancer. As bevacizumab may cause hypertension (HTN) and increase the risk of cardiac failure, we performed a pilot study to evaluate the feasibility and toxicity of a non-anthracycline-containing combination of docetaxel with cyclophosphamide and bevacizumab in early stage breast cancer patients. METHODS: Treatment consisted of four 3-weekly cycles of docetaxel and cyclophosphamide (75/600 mg/m2). Bevacizumab was administered 15 mg/kg intravenously on day 1, and then every 3 weeks to a total of 18 cycles of treatment. Serum biomarker concentrations of vascular endothelial growth factor (VEGF), cardiac troponin-I (cTnI), myeloperoxidase (MPO), and placental growth factor (PlGF) were quantified using enzyme-linked immunosorbent assay (ELISA) in 62 patients at baseline and whilst on treatment to determine their utility as biomarkers of cardiotoxicity, indicated by left ventricular ejection fraction (LVEF). RESULTS: A total of 106 patients were accrued in nine sites. Median follow up was 65 months (1-72 months). Seventeen protocol-defined relapse events were observed, accounting for an overall disease-free survival (DFS) rate of 84%. The DFS rates for hormone receptor positive (HR+) and triple-negative (TN) patients were 95% versus 43%, respectively. The median time to relapse was 25 (12-54) months in TN patients versus 38 (22-71) months in HR+ patients. There have been 13 deaths related to breast cancer . The overall survival (OS) rate was 88%. The 5-year OS rate in HR+ versus TN was 95% versus 57%. None of the measured biomarkers predicted the development of cardiotoxicity. CONCLUSIONS: We observed a low relapse rate in node-positive, HR+ patients; however, results in TN breast cancer were less encouraging. Given the negative results of three large phase III trials, it is unlikely that this approach will be investigated further. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT00911716.

A Machine Learning Platform to Optimize the Translation of Personalized Network Models to the Clinic.

PURPOSE: Dynamic network models predict clinical prognosis and inform therapeutic intervention by elucidating disease-driven aberrations at the systems level. However, the personalization of model predictions requires the profiling of multiple model inputs, which hampers clinical translation. PATIENTS AND METHODS: We applied APOPTO-CELL, a prognostic model of apoptosis signaling, to showcase the establishment of computational platforms that require a reduced set of inputs. We designed two distinct and complementary pipelines: a probabilistic approach to exploit a consistent subpanel of inputs across the whole cohort (Ensemble) and a machine learning approach to identify a reduced protein set tailored for individual patients (Tree). Development was performed on a virtual cohort of 3,200,000 patients, with inputs estimated from clinically relevant protein profiles. Validation was carried out in an in-house stage III colorectal cancer cohort, with inputs profiled in surgical resections by reverse phase protein array (n = 120) and/or immunohistochemistry (n = 117). RESULTS: Ensemble and Tree reproduced APOPTO-CELL predictions in the virtual patient cohort with 92% and 99% accuracy while decreasing the number of inputs to a consistent subset of three proteins (40% reduction) or a personalized subset of 2.7 proteins on average (46% reduction), respectively. Ensemble and Tree retained prognostic utility in the in-house colorectal cancer cohort. The association between the Ensemble accuracy and prognostic value (Spearman ρ = 0.43; P = .02) provided a rationale to optimize the input composition for specific clinical settings. Comparison between profiling by reverse phase protein array (gold standard) and immunohistochemistry (clinical routine) revealed that the latter is a suitable technology to quantify model inputs. CONCLUSION: This study provides a generalizable framework to optimize the development of network-based prognostic assays and, ultimately, to facilitate their integration in the routine clinical workflow.

High-throughput oncogene mutation profiling shows demographic differences in BRAF mutation rates among melanoma patients.

Because of advances in targeted therapies, the clinical evaluation of cutaneous melanoma is increasingly based on a combination of traditional histopathology and molecular pathology. Therefore, it is necessary to expand our knowledge of the molecular events that accompany the development and progression of melanoma to optimize clinical management. The central objective of this study was to increase our knowledge of the mutational events that complement melanoma progression. High-throughput genotyping was adapted to query 159 known single nucleotide mutations in 33 cancer-related genes across two melanoma cohorts from Ireland (n=94) and Belgium (n=60). Results were correlated with various clinicopathological characteristics. A total of 23 mutations in 12 genes were identified, that is--BRAF, NRAS, MET, PHLPP2, PIK3R1, IDH1, KIT, STK11, CTNNB1, JAK2, ALK, and GNAS. Unexpectedly, we discovered significant differences in BRAF, MET, and PIK3R1 mutations between the cohorts. That is, cases from Ireland showed significantly lower (P<0.001) BRAF(V600E) mutation rates (19%) compared with the mutation frequency observed in Belgian patients (43%). Moreover, MET mutations were detected in 12% of Irish cases, whereas none of the Belgian patients harbored these mutations, and Irish patients significantly more often (P=0.027) had PIK3R1-mutant (33%) melanoma versus 17% of Belgian cases. The low incidence of BRAF(V600E)(-) mutant melanoma among Irish patients was confirmed in five independent Irish cohorts, and in total, only 165 of 689 (24%) Irish cases carried mutant BRAF(V600E). Together, our data show that melanoma-driving mutations vary by demographic area, which has important implications for the clinical management of this disease.

Investigation of molecular alterations of AKT-3 in triple-negative breast cancer.

AIMS: Triple-negative breast cancer (TNBC) is responsible for a disproportionate number of breast cancer (BC) deaths, owing to its intrinsic aggressiveness and a lack of treatment options, especially targeted therapies. Thus, there is an urgent need for the development of better targeted treatments for TNBC. Molecular alteration of AKT-3 was previously reported in oestrogen receptor (ER)-positive BC. AKT-3 has also been suggested to play a role in hormone-unresponsive BC. The aim of this study was to investigate molecular alterations of AKT-3 in TNBC, to perform associated survival analysis, and to compare these findings with the incidence of AKT-3 molecular alterations in ER-positive BC. RESULTS: Our study revealed AKT-3 amplification and deletions in 11% (9/82) and 13% (11/82) of TNBCs, respectively. In contrast, 1% (2/209) of ER-positive BCs were found to have AKT-3 amplifications and deletions. A higher prevalence of AKT-3 copy number gains was observed in TNBC [26% (21/82)] than in ER-positive BC [9% (19/209)]. AKT-3 amplification together with Akt-3 protein expression was negatively associated with recurrence-free survival in TNBC. Furthermore, a negative association between high AKT-3 copy number and recurrence-free survival was observed. CONCLUSION: AKT-3 amplification could represent a potentially relevant oncogenic event in a subset of TNBCs that may, in turn, select cells sensitive to Akt-3 inhibitors.

Tumor profiling using protein biomarker panels in malignant melanoma: application of tissue microarrays and beyond.

Despite advances in our knowledge of the disease, malignant melanoma remains an unpredictable entity. The revolution in molecular biological techniques, such as DNA sequencing and gene-expression profiling, has uncovered many potential protein targets and biomarkers relevant to melanoma progression. Successful clinical application would be aided significantly by downstream proteomic validation of those candidate markers using a combination of immunohistochemistry and tissue microarrays. Yet, research in this context seems to lag behind the output of genomic data relating to melanoma. In this article, we look at the strengths and pitfalls of tissue microarrays in malignant melanoma. We will show how tissue microarrays have become a vital step in the transition from molecular techniques to useful clinical assays and interventions and look at likely future developments for advances in this field.

P-Rex1 is required for efficient melanoblast migration and melanoma metastasis.

Metastases are the major cause of death from melanoma, a skin cancer that has the fastest rising incidence of any malignancy in the Western world. Molecular pathways that drive melanoblast migration in development are believed to underpin the movement and ultimately the metastasis of melanoma. Here we show that mice lacking P-Rex1, a Rac-specific Rho GTPase guanine nucleotide exchange factor, have a melanoblast migration defect during development evidenced by a white belly. Moreover, these P-Rex1(-/-) mice are resistant to metastasis when crossed to a murine model of melanoma. Mechanistically, this is associated with P-Rex1 driving invasion in a Rac-dependent manner. P-Rex1 is elevated in the majority of human melanoma cell lines and tumour tissue. We conclude that P-Rex1 has an important role in melanoblast migration and cancer progression to metastasis in mice and humans.

Metallothionein 1E is methylated in malignant melanoma and increases sensitivity to cisplatin-induced apoptosis.

DNA methylation plays a major role in cancer by silencing tumour suppressor genes. In melanoma, only a discrete number of methylated genes have been identified so far. After the treatment of melanoma cells with a DNA methyltransferase inhibitor and subsequent transcriptomic profiling, we had identified earlier a cohort of melanoma progression-associated genes regulated by methylation. Here, we identified which of these genes are directly methylated in melanoma cell lines and tissues. First, we examined 16 genes by bisulphite sequencing in the WM793 isogenic cell line model series. Five of these genes (CYBA, FABP5, MT1E, TSPY1 and TAC1) displayed increased methylation in several invasive cell lines compared with the parental WM793 cells, indicating their involvement in progression. Next, we analyzed several matched primary/metastatic tumours using methylation-specific PCR, which revealed that MT1E (one of the five genes assessed) was methylated in the largest proportion of tumours. Examination of a larger cohort of samples showed that 1 of 17 (6%) of the benign naevi, 16 of 43 (37%) primary tumours and 6 of 13 (46%) of the metastases displayed MT1E methylation. In addition, ectopic over-expression of MT1E mediated sensitization to cisplatin-induced apoptosis. Overall, these studies suggest that MT1E is a potential tumour suppressor gene, whose loss may promote resistance to apoptosis-inducing therapies.

Topoisomerase I amplification in melanoma is associated with more advanced tumours and poor prognosis.

In this study, we used array-comparative genomic hybridization (aCGH) and fluorescent in situ hybridization (FISH) to examine genetic aberrations in melanoma cell lines and tissues. Array-comparative genomic hybridization revealed that the most frequent genetic changes found in melanoma cell lines were amplifications on chromosomes 7p and 20q, along with disruptions on Chr 9, 10, 11, 12, 22 and Y. Validation of the results using FISH on tissue microarrays (TMAs) identified TOP1 as being amplified in melanoma tissues. TOP1 amplification was detected in a high percentage (33%) of tumours and was associated with thicker, aggressive tumours. These results show that TOP1 amplification is associated with advanced tumours and poor prognosis in melanoma. These observations open the possibility that TOP1-targeted therapeutics may be of benefit in a particular subgroup of advanced stage melanoma patients.

Selective expression of Syntaxin-7 protein in benign melanocytes and malignant melanoma.

To search for proteins expressed in human melanocytes and melanoma, we employed an antibody-based proteomics strategy to screen for protein expression in tissue microarrays containing normal tissues, cancer tissues and cell lines. Syntaxin-7 (STX7) was identified as a novel protein, not previously characterized in cells of melanocytic lineage, displaying a cell type-specific protein expression pattern. In tumor tissues, STX7 was expressed in malignant melanoma and lymphoma. The protein was further characterized regarding subcellular localization, specificity, tissue distribution pattern and potential as a diagnostic and prognostic marker using cell lines and tissue microarrays containing normal skin, melanocytic nevi and primary and metastatic melanoma. STX7 was expressed in normal melanocytes, various benign melanocytic nevi, atypical nevi and malignant melanoma. Analysis in two independent melanoma cohorts demonstrated STX7 expression in nearly all investigated tumors, although at varying levels (> 90% positive tumors). The expression level of STX7 protein was inversely correlated to tumor stage, suggesting that decreased expression of STX7 is associated with more aggressive tumors. In conclusion, we present protein profiling data for a novel protein showing high sensitivity and specificity for cells of the melanocytic lineage. The presented antibody-based proteomics approach can be used as an effective strategy to identify novel tumor markers and evaluate their potential clinical relevance.

Identification and functional validation of therapeutic targets for malignant melanoma.

Despite remarkable effort, malignant melanoma still remains a potent killer. Millions of dollars have been spent on clinical trials that have not succeeded in achieving significant patient benefit. The thorough validation of drug targets at an earlier stage is, therefore, an essential step in the development of new therapies. Since the development of microarray experiments, putative drug targets are being identified in a high-throughput manner. Though high-throughput functional validation methods are currently being established, a more specific, pre-clinical analysis of promising target genes remains inevitable. For this, a broad range of increasingly sophisticated functional models is available. In vitro, the microenvironment of skin can be simulated through various two or three-dimensional models. In vivo assays range from xenograft studies to the establishment of transgenic organisms. Here, we provide a summary of functional interrogation approaches in melanoma research, focusing on the application of these strategies to the development of new effective therapies.

Real-time quantitative reverse transcriptase-polymerase chain reaction analysis of melanoma progression-associated genes.

BACKGROUND: Melanoma is an aggressive disease that spreads quickly and is resistant to most therapeutic agents. In an effort to provide insight into the molecular basis of melanoma progression, the expression of 94 genes in 20 metastatic melanomas using a high-throughput real-time quantitative RT-PCR assay was analysed. MATERIALS AND METHODS: A TaqMan low density array (LDA) was designed containing probes/primers directed towards a cohort of genes previously found to be differentially expressed in an isogenic cell line model of melanoma progression. For each sample, cDNA was prepared and added to the quantitative assay. The resulting data were then analysed for correlations with clinical data. RESULTS: Clustering analysis divided the melanomas into two major subgroups based on gene expression patterns. When analysed individually, several genes were associated with overall survival, depth and type of the primary tumour. CONCLUSION: We have identified a selection of genes linked to melanoma progression and patient outcome.

Common critical pathways in embryogenesis and cancer.

Cancer may arise because the developmental programs that create the dramatic alterations in form and structure in embryonic development are potentially corrupted. The cells in our bodies retain memories of these processes and cancer can occur later in life if imperfections occur in the fidelity of these pathways. This article is particularly interested in the phenomenon of epithelial to mesenchymal transition, which occurs in embryogenesis. Also reviewed are the small molecules and pathways that are involved both in homeostasis in adult epithelium and embryogenesis in utero. There are five such pathways in particular selected for review in this article: the Wnt pathway, Hedgehog, Notch, PAR and Bone morphogenetic peptide/TGF beta. These are usually conserved throughout mammalian evolution. Though they have been arbitrarily separated in this article they are not exclusive from one another. Their pathologically altered expression is found especially frequently in childhood tumours where they may recapitulate their developmental role, and in tumours that resemble primitive precursor cells. These pathways are important for selecting cell fates, cellular rearrangements, cytological context and morphologic design in embryology as well as participating in epithelial function in adults.

Multiple markers for melanoma progression regulated by DNA methylation: insights from transcriptomic studies.

The incidence of melanoma is increasing rapidly, with advanced lesions generally failing to respond to conventional chemotherapy. Here, we utilized DNA microarray-based gene expression profiling techniques to identify molecular determinants of melanoma progression within a unique panel of isogenic human melanoma cell lines. When a poorly tumorigenic cell line, derived from an early melanoma, was compared with two increasingly aggressive derivative cell lines, the expression of 66 genes was significantly changed. A similar pattern of differential gene expression was found with an independently derived metastatic cell line. We further examined these melanoma progression-associated genes via use of a tailored TaqMan Low Density Array (LDA), representing the majority of genes within our cohort of interest. Considerable concordance was seen between the transcriptomic profiles determined by DNA microarray and TaqMan LDA approaches. A range of novel markers were identified that correlated here with melanoma progression. Most notable was TSPY, a Y chromosome-specific gene that displayed extensive down-regulation in expression between the parental and derivative cell lines. Examination of a putative CpG island within the TSPY gene demonstrated that this region was hypermethylated in the derivative cell lines, as well as metastatic melanomas from male patients. Moreover, treatment of the derivative cell lines with the DNA methyltransferase inhibitor, 2'-deoxy-5-azacytidine (DAC), restored expression of the TSPY gene to levels comparable with that found in the parental cells. Additional DNA microarray studies uncovered a subset of 13 genes from the above-mentioned 66 gene cohort that displayed re-activation of expression following DAC treatment, including TSPY, CYBA and MT2A. DAC suppressed tumor cell growth in vitro. Moreover, systemic treatment of mice with DAC attenuated growth of melanoma xenografts, with consequent re-expression of TSPY mRNA. Overall, our data support the hypothesis that multiple genes are targeted, either directly or indirectly, by DNA hypermethylation during melanoma progression.