Gene Expression-based Subtypes Research Articles

Evaluation of the prognostic potential of histopathological subtyping in high-grade serous ovarian carcinoma.

High-grade serous ovarian carcinoma (HGSOC) can be categorized into four gene expression-based subtypes, with supposedly distinct prognoses and treatment responses. Murakami et al. translated these gene expression-based subtypes into the histopathological mesenchymal, immunoreactive, solid and proliferative, and papilloglandular subtypes, showing differences in survival outcomes. Miyagawa et al. refined these criteria to improve the interobserver concordance. The current retrospective study evaluated the interobserver variability and the prognostic differences between the histopathologic subtypes using the criteria of both Murakami et al. and Miyagawa et al. in 208 HGSOC cases. The mesenchymal subtype was considered first, followed by the immunoreactive subtype. Non-conforming cases were categorized as solid and proliferative or papilloglandular. The mesenchymal subtype was identified in 122 patients (58.7%) for both criteria. Using the criteria of Murakami et al., 10 cases (4.8%) were immunoreactive, 26 (12.5%) solid and proliferative, and 50 (24%) papilloglandular, with a concordance rate of 62.5% (κ = 0.34, p < .001). Using the Miyagawa et al. criteria, 23 cases (11%) were immunoreactive, 20 (9.6%) solid and proliferative, and 43 (20.7%) papilloglandular. No survival differences were observed between the subtypes. The fair reproducibility of the histopathological subtype classification of HGSOC and the lack of survival differences among these subtypes indicate the need for further refinement of the criteria and exploration of their correlation with overall survival outcomes before clinical application.

Metastatic heterogeneity of the consensus molecular subtypes of colorectal cancer

Gene expression-based subtypes of colorectal cancer have clinical relevance, but the representativeness of primary tumors and the consensus molecular subtypes (CMS) for metastatic cancers is not well known. We investigated the metastatic heterogeneity of CMS. The best approach to subtype translation was delineated by comparisons of transcriptomic profiles from 317 primary tumors and 295 liver metastases, including multi-metastatic samples from 45 patients and 14 primary-metastasis sets. Associations were validated in an external data set (n = 618). Projection of metastases onto principal components of primary tumors showed that metastases were depleted of CMS1-immune/CMS3-metabolic signals, enriched for CMS4-mesenchymal/stromal signals, and heavily influenced by the microenvironment. The tailored CMS classifier (available in an updated version of the R package CMScaller) therefore implemented an approach to regress out the liver tissue background. The majority of classified metastases were either CMS2 or CMS4. Nonetheless, subtype switching and inter-metastatic CMS heterogeneity were frequent and increased with sampling intensity. Poor-prognostic value of CMS1/3 metastases was consistent in the context of intra-patient tumor heterogeneity.

Shared DNA-based copy number with divided methylation changes differentiate and clinically associate early stage pancreatic cancer tumors.

434 Background: There is no known molecular taxonomy of pancreatic cancer that can guide therapeutic strategies. Understanding the fundamental molecular mechanism underlying pancreatic cancer biology remains an unmet need. We explore the extent to which combinations of DNA-based molecular changes in copy number (CN) and methylation separate early stage PAAD tumors and associated with survival outcomes. Methods: We performed genome-wide combined cluster analyses on DNA-based CN and methylation changes using TCGA data. We examined cluster associations with clinical outcomes by comparing in months (mos), Kaplan--Meier estimated overall survival (OS) and disease-free interval (DFI) using a log-rank test. We performed additional comparisons among CN-Methylation derived clusters with respect to PAAD phenotypes. Results: Using 78 early stage pancreatic cancer tumors from TCGA with CN, methylation and clinical outcomes data, we identified two patient clusters with distinct gene copy number signatures that when combined with three methylation signatures, resulted in three additional clusters. Thus, the same gene CN signature, when combined with different methylation signatures, further differentiated tumors into sub-clusters with varying levels of associations with clinical outcome. Among them, analogous to current gene-expression based subtypes, we also identified an immune-rich subtype that was associated with improved overall survival (n=21, median OS=16mos; DFI=16mos), and an extracellular matrix (ECM)-rich with worse survival (n=19, median OS=12mos; DFI=8mos). Unlike previous expression subtypes, we identified another metabolic-enriched subtype with the same worse median OS and DFI, differentiated by methylation with the ECM-rich subtype. The improved OS cluster had a signature of CN neutral and increased methylation, while the poor cluster had a signature of CN gains and increased methylation among a set of genes distinct from the improved cluster. No significant differences in age, site, microsatellite instability and KRAS status among clusters were noted. Notably, in a multivariable model that included gene expression-based subtypes, only our DNA-level subtypes remained significantly associated with overall survival. Conclusions: While RNA-level changes often display large variations, DNA-level changes are more robust. Copy number changes appear to separate tumors into poor and improved prognosis clusters, while methylation appears to inform on the further separation of poor prognosis into various levels. A DNA-based molecular taxonomy for early stage pancreatic cancer could prove invaluable when used in combination with methylation-based circulating tumor DNA assays for clinical trial monitoring of tumor responses.

Immune response characterization of endometrial cancer.

BackgroundThe comprehensive characterization and prognostic relevance of immune activation in endometrial cancer remain largely unknown.ResultsWe systematically reported a subset of endometrioid-type endometrial cancer characterized by multifaceted immune features such as low tumor purity, high leukocyte percentage, and striking CD8 lymphocytic infiltration with anti-tumor efficacy along with marked upregulation of immunosuppressive gene markers. We also showed that genes whose expression was significantly correlated with better survival were significantly enriched in the immune-related signaling pathways, suggesting that tumor-infiltrating lymphocytes give rise to a favorable prognosis in endometrial cancer. Furthermore, we showed that immune cell recruitment in this subset of tumors is likely due to the transcriptional activation of the STAT1 signaling network.MethodsWe obtained the multi-dimensional genomic data from publicly available databases and correlated them with the four gene expression-based subtypes we recently identified in endometrial cancer. Upstream regulator analysis was used to identify the most significantly enriched transcription regulators and Ingenuity pathway analysis was applied to determine enrichment of signaling pathways in survival-associated genes. Gene set enrichment analysis was performed on the 200-gene T-cell tumor infiltration gene signature comparing Cluster IV with the other three clusters combined. All statistical tests were two-sided, and a P value of less than 0.05 is considered significant across all analyses performed.ConclusionThis study helps to identify patients with immune activation who are likely to benefit from emerging immune checkpoint inhibitors.

The clinical heterogeneity of preeclampsia is related to both placental gene expression and placental histopathology

Preeclampsia is a life-threatening disorder of pregnancy, demonstrating a high degree of heterogeneity in clinical features such as presentation, disease severity, and outcomes. This heterogeneity suggests distinct pathophysiological mechanisms may be driving the placental disease underlying this disorder. Our group recently reported distinct clusters of placental gene expression in preeclampsia and control pregnancies, allowing for the identification of at least 3 clinically relevant gene expression-based subtypes of preeclampsia. Histopathological examination of a small number of samples from 2 of the gene expression-based subtypes revealed placental lesions consistent with their gene expression phenotype, suggesting that detailed placental histopathology may provide further insight into the pathophysiology underlying these distinct gene expression-based subtypes. The objective of the study was to assess histopathological lesions in the placentas of patients belonging to each identified gene expression-based subtype of preeclampsia, characterized in our previous study. Our goal was to further understand the pathophysiologies defining these gene expression-based subtypes by integrating gene expression with histopathological findings, possibly identifying additional subgroups of preeclampsia patients. Paraffin-embedded placental biopsies from patients included in the gene expression profiling study (n= 142 of 157, 90.4%) were sectioned, hematoxylin and eosin stained, and imaged. An experienced perinatal pathologist, blinded to gene expression findings and clinical information, assessed the presence and severity of histological lesions using a comprehensive, standardized data collection form. The frequency and severity scores of observed histopathological lesions were compared among gene expression-based subtypes as well as within each subtype using using Fisher exact tests, Kruskal-Wallis tests, and hierarchical clustering. The histological findings of the placental samples were visualized using t-distributed stochastic neighbor embedding and phylogenetic trees. Concordance and discordance between gene expression findings and histopathology were also investigated and visualized using principal component analysis. Several histological lesions were found to be characteristic of each gene expression-based preeclampsia subtype. The overall concordance between gene expression and histopathology for all samples was 65% (93 of 142), with characteristic placental lesions for each gene expression-based subtype complementing prior gene enrichment findings (ie, placentas with enrichment of hypoxia-associated genes showed severe lesions of maternal vascular malperfusion). Concordant samples were located in the central area of each gene expression-based cluster when viewed on a principal component analysis plot. Interestingly, discordant samples (gene expression and histopathology not reflective of one another) were generally found to lie at the periphery of the gene expression-based clusters and tended to border the group of patients with phenotypically similar histopathology. Our findings demonstrates a high degree of concordance between placental lesions and gene expression across subtypes of preeclampsia. Additionally, novel integrative analysis of scored placental histopathology severity and gene expression findings allowed for the identification of patients with intermediate phenotypes of preeclampsia not apparent through gene expression profiling alone. Future investigations should examine the temporal relationship between these 2 modalities as well as consider the maternal and fetal contributions to these subtypes of disease.

Targeted DNA and RNA Sequencing of Paired Urothelial and Squamous Bladder Cancers Reveals Discordant Genomic and Transcriptomic Events and Unique Therapeutic Implications

BackgroundIntegrated molecular profiling has identified intrinsic expression-based bladder cancer molecular subtypes. Despite frequent histological diversity, robustness of subtypes in paired conventional (urothelial) and squamous components of the same bladder tumor has not been reported. ObjectiveTo assess the impact of histological heterogeneity on expression-based bladder cancer subtypes. Design, setting, and participantsWe performed clinically applicable, targeted DNA and/or RNA sequencing (multiplexed DNA and RNA sequencing [mxDNAseq and mxRNAseq, respectively]) on 112 formalin-fixed paraffin-embedded (FFPE) bladder cancer samples, including 12 cases with paired urothelial/squamous components and 21 bladder cancer cell lines. Outcome measurements and statistical analysisUnsupervised hierarchical and consensus clustering of target gene expression enabled derivation of basal/luminal molecular subtyping. Results and limitationAcross 21 bladder cancer cell lines, our custom mxRNAseq panel was highly concordant with whole transcriptome sequencing, and assessed targets robustly determined expression-based basal/luminal subtypes from The Cancer Genome Atlas data (in silico) and internally sequenced FFPE tissues. Frequent deleterious TP53 (56%) and activating hotspot PIK3CA (30%) somatic mutations were seen across 69 high-quality tissue samples. Potentially targetable focal ERBB2 (6%) or EGFR (6%) amplifications were also identified, and a novel subgene copy-number detection approach is described. Combined DNA/RNA analysis showed that focally amplified samples exhibit outlier EGFR and ERBB2 expression distinct from subtype-intrinsic profiles. Critically, paired urothelial and squamous components showed divergent basal/luminal status in three of 12 cases (25%), despite identical putatively clonal prioritized somatic genomic alterations. Limitations include lack of profiled paired normal tissues for formal somatic alteration determination, and the need for formal analytical and clinical validation. ConclusionsOur results support the feasibility of clinically relevant integrative bladder cancer profiling and challenge the intrinsic nature of expression subtypes in histologically diverse bladder cancers. Patient summaryA targeted RNA sequencing assay is capable of assessing gene expression-based subtypes in individual components of clinical bladder cancer tissue specimens. Different histological components of the same tumor may yield divergent expression profiles, suggesting that expression-based subtypes should be interpreted with caution in heterogeneous cancers.

PO-521 Gene expression-based subtypes of pancreatic ductal adenocarcinoma offer leads for targeted therapy

IntroductionOverall survival (OS) of patients with pancreatic ductal adenocarcinoma (PDAC) is extremely poor. Of patients eligible for surgery (20%), around 15% present with a recurrence within 6 months (m), while 10% survive over 5 years after diagnosis. Detailed clinicopathological and molecular knowledge of factors influencing survival may lead to better prognostic and/or predictive factors and preselection of individual patients for specific treatment strategies.Material and methodsFresh frozen PDAC resection specimens from the Academic Medical Centre Amsterdam (1993–2015) were histopathologically revised, and clinicopathological details were collected. From samples with a tumour cellularity of ≥30% (n=90), mRNA, miRNA, and DNA were used for NGS at multiple levels. Corresponding FFPE blocks were selected for tissue microarrays.Results and discussionsUnsupervised consensus clustering of gene expression profiles of 90 PDAC revealed four subgroups with divergent OS rates: secretory (14.7 m), epithelial (31.8 m), compound pancreatic (21.5 m), and mesenchymal (14.0 m) subgroups (p=0.002). The differences between the subgroups suggest that more targeted therapy may be feasible for PDAC.The epithelial and mesenchymal subgroups show an upregulation of genes related to DNA repair and hypoxia, respectively, suggesting sensitivity to hyperthermia, which is known to enhance cytotoxic radio-, and chemotherapy effects through frustration of tumour DNA repair and/or better oxygenation of hypoxic tumours. Immune-related gene expression was observed to be upregulated in the compound pancreatic and mesenchymal subtypes; differences were seen in expression of PDL-1, type I interferon, and MHC Class I, which may explain the difference in OS and suggests differential responses to immunotherapeutic agents. Finally, the epithelial subtype is characterised by prominent increases in metabolic signatures. Differences in metabolic flexibility suggest differences in response to mTOR inhibitors.To determine clinical relevance and identify biomarkers, we are currently characterising the subtypes at the miRNA, mutational, copy number, and immunohistochemical level.ConclusionIn our well-defined single-centre set of 90 PDAC, we identified four transcriptomics-based subgroups with different survival outcomes that show a correlation with altered DNA-repair-, hypoxia, immune-, and metabolic features. Complete characterisation of these liabilities may help to further identify prognosticators, and preselect patients for specific treatment strategies.

Prognostic value of Ki67 analysed by cytology or histology in primary breast cancer

AimsThe accuracy of biomarker assessment in breast pathology is vital for therapy decisions. The therapy predictive and prognostic biomarkers oestrogen receptor (ER), progesterone receptor, HER2 and Ki67 may act as...

Spectral clustering using Nystr\xf6m approximation for the accurate identification of cancer molecular subtypes

A major challenge in clinical cancer research is the identification of accurate molecular subtype. While unsupervised clustering methods have been applied for class discovery, this clustering method remains a bottleneck in developing accurate method for molecular subtype discovery. In this analysis, we hypothesize that spectral clustering method could identify molecular subtypes in correlation with survival outcomes. We propose an accurate subtype identification method, Cancer Subtype Identification with Spectral Clustering using Nyström approximation (CSISCN), for the discovery of molecular subtypes, based on spectral clustering method. CSISCN could be used to improve gene expression-based identification of breast cancer molecular subtypes. We demonstrated that CSISCN identified the molecular subtypes with distinct clinical outcomes and was valid for the number of molecular subtypes. Furthermore, CSISCN identified molecular subtypes for improving clinical and molecular relevance which significantly outperformed consensus clustering and spectral clustering methods. To test the general applicability of the CSISCN, we further applied it on human CRC datasets and AML datasets and demonstrated superior performance as compared to consensus clustering method. In summary, CSISCN demonstrated the great potential in gene expression-based subtype identification.

Abstract 3407: Gene expression subtypes of high grade serous ovarian cancer in African American women

Abstract Ovarian cancer accounts for 5% of cancer deaths and is the fifth leading cause of cancer death in women in the United States. While incidence is higher in European American (EA) than African American (AA) women, five-year survival is worse for AA women (36%) than EA women (44%). Access to appropriate surgery and treatment is a major contributor but does not completely explain this disparity. The Cancer Genome Atlas (TCGA) identified four gene expression-based subtypes of the most common and lethal histotype, high grade serous carcinoma (HGSC): mesenchymal, proliferative, differentiated, and immunoreactive. We sought to characterize similarities and differences in gene expression-based subtypes arising in AA and EA women to determine whether there are underlying biologic features that may influence survival. We performed two distinct analyses, first using TCGA data and second using cases from the population-based African American Cancer Epidemiology Study (AACES). For both we summarized differential expression patterns for each subtype with moderated t statistic vectors for &amp;gt;10,000 genes using Significance Analysis of Microarrays. We calculated Pearson's correlations of these vectors to determine concordance of expression patterns between subtypes across EA and AA women. In TCGA, we observed correlations of subtype-specific expression patterns between the 24 AA and 475 EA tumors of 0.52-0.60 for each of the four subtypes. Thus, while analogous subtypes can be identified in AA and EA women, the magnitude of these correlations suggests that there are potential differences in gene expression patterns between AA and EA tumors that are assigned to the same subtype. We generated additional data from 58 AACES HGSC cases using the Affymetrix Human Transcriptome Array 2.0. Instead of assigning these tumors to previously-defined subtypes, we clustered samples to identify four subtypes de novo. We observed concordance with two of the TCGA subtypes; correlations for the mesenchymal-like and proliferative-like subtypes were 0.56-0.65. The mesenchymal-like subtype was more common in these AA women than in the TCGA EA women (33% versus 25%), and the proliferative-like subtype was marginally less common (14% versus 19%). Concordance for the differentiated-like subtype was considerably lower, at 0.21, and this subtype was less common in AA than EA women (19% versus 34%). Another subtype comprising 34% of the AA samples was only weakly correlated (-0.21-0.10) with any of the TCGA subtypes, suggesting that it is a novel subtype. The limited data available on HGSC in AA women suggest that at least two subtypes are comparable to those in EA women but differ in prevalence, and that there may be a novel subtype in AA women that does not strongly correspond to those described in EA women. Citation Format: Jennifer A. Doherty, Casey S. Greene, James E. Rudd, Laura J. Tafe, Anthony J. Alberg, Elisa V. Bandera, Jill Barnholtz-Sloan, Melissa Bondy, Michele L. Cote, Ellen Funkhouser, Patricia G. Moorman, Edward S. Peters, Ann G. Schwartz, Paul Terry, Rex Bentley, Andrew Berchuck, Jeffrey R. Marks, Joellen M. Schildkraut. Gene expression subtypes of high grade serous ovarian cancer in African American women. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3407.

Abstract 815: Patterns of metagene activation in ovarian cancer subtypes

Abstract High grade serous ovarian cancer (HGSC) is a complex and aggressive disease. Recently, three or four gene expression-based subtypes, which may be differentially associated with survival, have been reported in several populations. To identify the biological functions that define the subtypes, we determined the extent to which metagenes—linear combinations of gene expression vectors—were differentially activated across subtypes, could be reliably identified across populations, and showed consistent associations with survival. We previously clustered HGSC samples using gene expression data from TCGA, Tothill (GSE9891), Yoshihara (GSE32062), and Mayo (GSE74357) to identify subtypes across populations. We found subtype-specific genes within each population through differential expression analysis (p &amp;lt; 4.6×10-6). Using the intersection of differentially expressed genes for parallel subtypes across these four populations, we applied non-negative matrix factorization to identify metagenes. To determine whether the metagenes were consistently observed across populations, we performed leave-one-dataset-out cross validation. For each metagene, we performed gene set enrichment analysis against the National Cancer Institute pathway interaction database to annotate metagene pathways. We examined whether increasing tertiles of metagene activity, which we termed low, medium, and high activity, were associated with survival using a random effects meta-analysis of Cox regression estimates adjusting for age at diagnosis, tumor stage, tumor grade, and debulking status. Five metagenes were consistently identified and significantly associated with HGSC subtypes (p &amp;lt; 0.0001). Of these, a metagene weakly enriched for the CMYB pathway was associated with subtype 1; three metagenes (one significantly enriched with the IL12 pathway and the others weakly enriched with the FCER1 and CXCR4 pathways) were associated with subtype 2; and a metagene weakly enriched with the AVB3 Integrin pathway distinguished between all 3 subtypes. Neither the CYMB metagene nor the IL12 metagene was significantly associated with survival. High activity of the CXCR4 and AVB3 metagenes was associated with poorer survival (hazard ratios (HR) and 95% confidence intervals (CI) are, respectively: 1.21, 0.99-1.48 and 1.34, 1.09-1.64). In contrast, high activity of the FCER1 metagene was associated with improved survival (HR 0.76, 95% CI 0.62-0.93). Metagenes that are consistently and statistically significantly associated with subtype may be indicative of functional differences between HGSC subtypes. The contrast in hazard estimates for metagenes associated with subtype 2 may indicate that the metagenes capture survival signal distinct from the subtype association. Future work associating metagene activity with subtype uncertainty may better enable the refinement of subtype definitions and the development of subtype specific treatment strategies. Citation Format: James Rudd, Emily K. Shea, Gregory P. Way, Casey S. Greene, Jennifer A. Doherty. Patterns of metagene activation in ovarian cancer subtypes. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 815.

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 χ2 (LR χ2) was higher for DIA that also added significantly more prognostic information to the manual scores (LR−Δχ2). 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.

Abstract P1-01-06: Digital image analysis outperforms manual scoring for breast cancer subclassification and prognostication

Abstract Introduction Categorization according to the four gene expression-based 'intrinsic' subtypes "Luminal A", "Luminal B", "HER2-enriched" and "Basal-like" is the method of choice for practical prognostic and predictive value in the heterogeneous spectrum of Breast Cancers. Gene expression tests are however not yet universally available, which has created an opportunity for routine immunohistochemical stains to act as surrogate markers (biomarkers) for the gene expression-based subtypes. As recommended by international expert consensus, the expressions of Estrogen receptor α (ER), Progesterone receptor (PR), Human Epidermal Growth factor Receptor 2 (HER2) and the proliferation-associated protein Ki67 are scored during the routine pathological work-up of breast cancer specimens. Thus, congruence of these biomarker tests to the gene expression tests are of utmost importance as discrepancies in classification induces dissimilar treatment decisions. In this study, we compare a novel system for Digital Image Analysis (DIA) with the manual scoring of biomarkers used in current clinicopathological routine and suggest methods to improve their congruence to gene expression assays, provide more robust prognoses for survival as well as reduce time consumption for pathologists. Methods 1 tissue micro array (TMA) cohort and 2 cohorts of primary breast cancer specimens (total n = 436) with &amp;gt;20 years survival data, were sectioned into physical glass slides and digitally scanned at ×20 and then reviewed for manual vs. DIA test congruence to PAM50 gene expression assays in terms of classification into the four intrinsic subtypes, and their prognostic power. This included the evaluation of 6 different methods for DIA biomarker testing. The DIA system used was the Visiopharm Integrator System (VIS) by Visiopharm A/S, Hoersholm, Denmark. In short, this system facilitates ER, PR and Ki67-testing by a 'sandwich' technology, in which each biomarker slide is aligned with an adjacent 3 μm slide stained with a pancytokeratin marker. Thus, non-tumor tissue is to a large extent automatically excluded from analysis and only cells that express cytokeratin are eligible for automatic detection of positivity or negativity of the respective biomarker. Results 60,8 % of the cases in the TMA cohort was classified in concordance with PAM50 (κ = 0,46) using DIA. Classification with regard to HER2-enriched, Basal-like and Luminal tumors without dichotomization of A and B subtype (thereby avoiding the impact of Ki67-scoring) was 80,9 % (κ = 0,63). In the whole slide cohorts, DIA performed with 95 % of the cases classified in concordance with PAM50 (κ = 0,90), compared to 67 % (κ = 0,59) for the manual method recommended by international consensus. In addition to this DIA produced better prognostication vs. the manual method in terms of hazard of all-cause death for each subtype. Conclusion The system for DIA evaluated here outperforms manual scoring in both predictive and prognostic value. It also has the potential to reduce time consumption for pathologists, as many of the steps in the workflow is either automatic or feasible to manage without pathological expertise. Based on the findings in this study however, TMA cannot be recommended for DIA scoring of Ki67. Citation Format: Stålhammar G, Rosin G, Kis L, Lippert M, Moelholm I, Grunkin M, Bergh J, Hartman J. Digital image analysis outperforms manual scoring for breast cancer subclassification and prognostication. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P1-01-06.

Abstract 1928: High-grade serous ovarian cancer subtypes are similar across diverse populations

Abstract The most common and lethal type of invasive epithelial ovarian cancer is high grade serous (HGSC). Three to four gene expression-based HGSC subtypes have been identified in prior studies. In contrast to most previous studies, which have assessed the performance of survival classifiers in validation sets, we sought to determine the degree of similarity of gene expression patterns in subtypes between populations using systematic unsupervised clustering within populations. We analyzed publically-available mRNA expression data from studies with &amp;gt;200 HGSC tumors: The Cancer Genome Atlas (TCGA, US, n = 519, Affymetrix HT U133a), Tothill et al. (GSE9891, Australia, n = 242, Affymetrix U133 Plus 2.0) and Yoshihara et al. (GSE32062, Japan, n = 258, Agilent G4112a). We restricted analyses to the 12,249 genes shared across all datasets and selected from these the union of the 1,500 most variant genes per population (2,824). Using these datasets, we performed k-means clustering within each population for k = 3 and k = 4. We compared each cluster to all other clusters using Significance Analysis of Microarrays, which outputs an F score for all 12,249 genes, measuring cluster-specific differential expression. We then calculated the correlation of the resulting F score vectors across populations and within populations across both numbers of centroids (k = 3 or k = 4). We identified analogous clusters by high F score correlations and determined each cluster's similarity to the TCGA subtypes based on cluster-specific differentially expressed genes. We observed high concordance of gene expression patterns for clusters across populations and across k-means runs, suggesting that analogous clusters exist in most analyses. For k = 3, F score correlations across populations for clusters 1, 2 and 3, respectively, ranged between 0.77-0.85, 0.80-0.90, and 0.66-0.72. For k = 4, F score correlations for clusters 1-4 were, respectively: 0.76-0.85, 0.82-0.85, 0.65-0.78, and 0.52-0.78. Across k = 3 and k = 4, correlations for cluster 1 within TCGA, Tothill, and Yoshihara were 0.99, 1.00 and 1.00, and correlations for cluster 2 were 0.96, 0.98, and 0.95, respectively. Correlations for cluster 3 were less strong: 0.56, 0.88, and 0.60, respectively. For k = 4, cluster 4 was composed mainly of samples that belonged to cluster 3 for k = 3; 88% for TCGA, 54% for Tothill, and 95% for Yoshihara. When compared to TCGA subtypes, cluster 1 corresponded most strongly to mesenchymal, cluster 2 to proliferative, cluster 3 to differentiated, and cluster 4 to immunoreactive. Our observation of highly correlated gene expression patterns between clusters across populations, across platforms, and across the number of centroids provides strong evidence that at least three biological HGSC subtypes exist. The mesenchymal-like and proliferative-like subtypes are particularly consistent across populations, and could be uniquely targeted for treatment. Citation Format: Gregory P. Way, James Rudd, Casey S. Greene, Jennifer A. Doherty. High-grade serous ovarian cancer subtypes are similar across diverse populations. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1928. doi:10.1158/1538-7445.AM2015-1928

Evidence for a Functional Link between Chromosomal Breakpoints and Aberrant DNA Methylation in Cancer

Evidence for a Functional Link between Chromosomal Breakpoints and Aberrant DNA Methylation in Cancer

Abstract P6-05-27: p53, c-Myc and EGFR protein interactions confer worse prognosis in patients with triple negative breast cancer (TNBC)

Abstract Background - aim: TNBC is a heterogeneous disease comprising different molecular subtypes that are mostly defined by gene expression profiling and are related to patient outcome. Herein, we investigated the prognostic significance of molecular subtype-related, surrogate protein markers. Methods: Immunohistochemistry was applied on tissue microarrays containing a series of 330 centrally assessed TNBC from early high-risk breast cancer patients treated with anthracyclines and/or taxanes. The following markers were examined (in parentheses: cut-offs for low/high and negative/positive): Ki67 (14%); p53 (10%); c-Myc (10%); P-cadherin (10%); E-cadherin (10%); Claudin-7 (H-score median: 105); CK5 (1%); EGFR (1%); androgen receptor (AR [1%]). Results: The majority of tumors (269/327, 81.5%) were grade-III ductal carcinomas of no special type. Tumors were frequently positive for Ki67 (239/314, 76.1%), E-cadherin (193/271, 71.2%), P-cadherin (249/302, 82.5%) and expressed markers of the basal phenotype (CK5+ and/or EGFR+, 268/316, 84.8%). Grade-III, Ki67-high carcinomas and tumours positive for E- and P-cadherin were more prevalent among those of basal phenotype (Fisher's exact p&amp;lt;0.001). Tumors expressing low levels of Claudin-7 were infrequently positive for AR and EGFR proteins (p = 0.015 and p = 0.003, respectively). AR positivity was more frequent among CK5 negative, grade I-II, and Ki67-low tumors (p = 0.021, p = 0.05 and p = 0.011 respectively). When examined individually, none of these markers was related to patient outcome. An interaction was noticed between p53 and c-Myc immunopositivity (Wald's p = 0.028 for disease-free [DFS] and p = 0.005 for overall survival [OS]). Among patients with c-Myc negative tumors, patients with p53 positive tumors (n = 63) had shorter median DFS (46.8 months) and OS (56 months) (HR 1.9; 95%CI 1.0-3.9; p = 0.049 and HR 2.28; 95%CI 1.0-4.8; p = 0.031, respectively) as compared to those with p53 negative tumors (n = 56; median DFS 69 months, OS 70 months). Similarly, in the absence of p53, c-Myc expression correlated with shorter median OS (44.5 months) (HR 2.1; 95%CI 1.0-4.5; p = 0.038) but showed only a trend for shorter DFS as compared to c-Myc negative tumors. p53 protein status was a discriminating factor for EGFR negative tumors (Wald's p = 0.052), whereby p53 positive tumors (n = 54) were associated with reduced DFS as compared to p53 negative tumors (n = 68; median DFS: 47 vs. 68.8 months; HR 2.1; 95%CI 1.1-4.0; p = 0.025). Accordingly, EGFR immunopositivity conferred a trend for shorter DFS within p53 negative cases. Conclusions: In the current study, it was not possible to assign prognostic significance to individual proteins that have been introduced as surrogate markers for gene expression-based molecular subtypes in TNBC. By contrast, significant interactions involving the status of p53 and c-Myc or EGFR were observed. It seems that aberrant p53 adversely influences tumor behaviour in the absence of c-Myc and EGFR, while the opposite is also effective but to a lesser extent. These findings are hypothesis generating and need to be pursued in sufficiently powered studies for validating the apparently mutually exclusive adverse effects of p53, c-Myc and EGFR on tumor behaviour. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-05-27.

Abstract 5064: Cytogenetic and somatic characterization of myeloma tumors from a multi-ethnic cohort

Abstract Previous epidemiological studies have suggested that African Americans (AA) are twice as likely to be diagnosed with and to die from Multiple Myeloma (MM) as compared to European Americans (EA). In recent years, however, mortality rates have become more comparable between these groups. Although socioeconomic factors could play a major role to this potential disparity, biological factors may also be responsible for influencing these differences in either incidence and/or mortality. Here we examined a multi-ethnic cohort of MM tumors to assess potential differences in the frequency of molecular events in tumors derived from either African American (AA) or European American (EA) patients. First, we compared the frequency of 14q32 translocation breakpoints at the IgH locus among a series of data from 115 AA MM patients from three studies and EA MM from the Eastern Cooperative Oncology Group (ECOG). Furthermore, we analyzed matching genome-wide copy number and transcriptional data among 47 AA MM tumors and 196 EA MM tumors to determine differences in the frequency of molecular events in tumors from these populations. Univariate analyses were conducted using the Fisher's Exact Test and p-values were calculated to measure statistical significance. We observed statistically significant differences in the frequency of IgH translocations and somatic copy number alterations. Although we did not detect differences for specific translocation subtypes, we saw a statistically significant difference in the overall frequency of IgH translocations between the two groups, which occurred more frequently in EA patients (p = 0.032). When assessing genomic copy number events previously shown to be associated with poor outcome in MM, chromosome 13 monosomy was more frequent in hyperdiploid MM from AA patients and significant after correcting for multiple testing using the Bonferoni correction method (p = 0.001). Moreover, when using the Benjamini-Hochberg multiple testing correction method, chromosome 13 monosomy remained significantly different among hyperdiploid AA patients (p = 0.0121). Chromosome 1q amplification was also significant among hyperdiploid tumors from EA patients when using this correction method (p = 0.036). No differences in the frequency of hyperdiploid status or gene expression-based subtypes were detected when comparing tumors from AA and EA patients. Our study represents the first comprehensive assessment of the frequency and distribution of molecular alterations in MM tumors from both AA and EA patients and could help shed light on possible biological features associated with population differences in incidence and outcome in MM. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5064. doi:1538-7445.AM2012-5064

Abstract 3347: A proneural signature profile predicts response to Notch inhibition in glioma stem cells

Abstract The Cancer Genome Atlas Network described genomic abnormalities and gene expression-based molecular subtypes of GBM that showed a strong relationship between subtypes, genomic alterations and different neural lineages. Future studies are required to elucidate the intricate relationship between tumor subtypes and treatment response. We have thus embarked on a comprehensive effort to detecting expression signatures that are associated with response to the therapy and these signatures may allow prospective selection of patients with high likelihood of responding to therapy. Notch signaling pathway is an evolutionarily conserved pathway that plays an important role in multiple cellular and developmental processes including cell fate decision, differentiation, proliferation, survival, angiogenesis and migration. Recent studies indicate that the Notch signaling pathway regulates normal stem cells in the brain, and glioma stem Cells (GSCs) with high Notch activity. We investigated the effects of Notch pathway inhibition on GSCs growth using commercially available inhibitors. Drug cytotoxicity test on 16 GSCs show differential growth response to Notch inhibitors stratifying GSCs into two groups: responders (6 cell lines) vs non-responders (10 cell lines). Gene expression clustering identified a responder gene signature enriched in 1: proneural genes such as Olig2, Sox-9, NKX2-2, and 2: Notch signaling component such as Notch-1, Notch-3, Hes-3 and Hes-5. These signature were validated by western blot analysis. Treatment with Notch inhibitors reduced neurosphere formation in vitro accompanied with attenuated Notch intracellular domain (NICD), Hes-1, Hes-3, and Hes-5. In addition, the expression of nestin-the stem cell marker was reduced and an increased expression of lineage differentiation markers TuJ1 and GFAP was observed following Notch inhibition. In conclusion, we have identified a responder signature for a group of glioma stem cells that can be targeted by Notch inhibitors and this may allow the selection of patients with high likelihood of responding to this therapy. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3347. doi:1538-7445.AM2012-3347

Abstract PD07-06: Basal-Like Molecular Subtype Predicts Response to Neoadjuvant Chemotherapy with Epirubicine, Cyclophosphamide, and Docetaxel in Patients with Primary Breast Cancer

Abstract Background: Gene expression profiling provides an opportunity to predict response to specific regimens of neo-adjuvant chemotherapy (NAC) in breast cancer patients. Materials and Methods: In a prospective cohort study of 32 women with primary invasive breast cancer with a measurable lesion, we obtained a tumor specimen by high speed core biopsy before and after 4 cycles of NAC with epirubicine 90mg/m2 and cyclophosphamide 600mg/m2 every 3 weeks, followed by 4 cycles of docetaxel 100mg/m2. Total RNA was extracted from tumor specimens and the whole transcriptome was quantified with Agilent's 44K single color microarray interrogating 44000 unique human genes. Tumor lesions were ultrasonographically measured to assess response using Sinn criteria. Data analysis was performed by GeneSpring v11 and IBM SPSS v18. Results: Using three single sample predictors, 10 tumors were classified as basal-like and 22 tumors were classified as non basal-like. We found that gene expression-based molecular subtype (basal-like vs. non basal-like) (p=0.003), but not tumor grade (p=0.07), estrogen receptor (p=0.1), progesterone receptor (p=0.6), and HER2 status (p=0.4) predicted response to NAC. Specifically, 7/10 basal-like tumors responded to NAC, whereas 19/22 non basal-like tumors did not respond. Comparing gene expression signatures before and after 4 cycles of NAC, we found that all patients with an initial non basal-like tumor retained this tumor type, whereas 5/7 basal-like tumors, including all responders, lost this molecular subtype. Using regression models based on centroid predicition gene sets, complete prediction of response to NAC based on the initial tumor biopsy as well as on the change of gene expression between two tumor biopsies was achieved with a 21 gene list (p=0.000008) and a 23 gene list (p=0.000007), respectively. Of note, both the expression and upregulation of a single gene, ie HER4, predicted response to NAC in 26/32 (81%; p=0.002) and in 23/25 (92%; P&amp;lt;0.001) patients, respectively. Conclusions: Basal-like molecular subtype and therapy-induced HER4 gene upregulation predict response to NAC with epirubicine, cyclophosphamide, and docetaxel. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr PD07-06.

Abstract B13: OncoPredictor: An integrative genomics platform to systematically predict responsive tumor populations from in vitro drug response data

Abstract A key challenge in drug development is identifying clinically relevant genomic biomarkers from preclinical data. Here, we developed OncoPredictor, a cellular screening and bioinformatics platform to (1) evaluate multiple types of genomic biomarkers for association with in vitro drug response and (2) analyze identified biomarkers in clinical tumor populations, thereby suggesting potential drug development strategies. We examined data for 18 targeted, anticancer agents tested across more than 200 cancer cell lines for which detailed genomic data were collected. To support association analysis of drug response and genomic biomarkers, a cell line biomarker catalog was constructed including gene and pathway mutations, copy number aberrations and gene expression changes, in addition to tumor-derived gene expression molecular subtypes. Similarly, to support the rapid translation of in vitro biomarkers to clinical tumor populations, a tumor biomarker catalog was developed, characterizing the frequency of biomarkers across the Oncomine database, which includes more than 40,000 genomic profiles from tumor specimens. Notably, our analysis uncovered several known and novel drug- biomarker-tumor population associations providing a series of cancer treatment hypotheses. For example, within Ras/Raf mutant cell lines, overexpression of a novel biomarker associated with sensitivity to CI-1040, a MEK inhibitor, and identified a sub-population of Ras/Raf mutant colorectal cancer and melanoma that may be more likely to benefit from MEK inhibitors. Similarly, we showed that a growth factor activation network associated with a dramatic response to sunitinib and identified a fraction of glioblastoma and particular types of sarcoma with activation, suggesting that these indications may benefit from a tumor intrinsic response to sunitinib. Finally, for everolimus, an mTOR inhibitor, we found that a gene expression-based molecular subtype of breast cancer were particularly sensitive and demonstrated that the while subtype-positive breast cancers have a good prognosis, 10% of breast cancers that metastasize are subtype-positive, suggesting a defined clinical population that could benefit from mTOR inhibitors. In summary, our platform provides a comprehensive and clinically relevant framework to discover and apply genomic biomarkers of drug response. Citation Information: Clin Cancer Res 2010;16(14 Suppl):B13.