David A. Fenstermacher

An Introduction to MAMA (Meta-Analysis of MicroArray data) System

Analyzing microarray data across multiple experiments has been proven advantageous. To support this kind of analysis, we are developing a software system called MAMA (meta-analysis of microarray data). MAMA utilizes a client-server architecture with a relational database on the server-side for the storage of microarray datasets collected from various resources. The client-side is an application running on the end users computer that allows the user to manipulate microarray data and analytical results locally. MAMA implementation will integrate several analytical methods, including meta-analysis within an open-source framework offering other developers the flexibility to plug in additional statistical algorithms

Tissue-specific RMA models to incrementally normalize Affymetrix GeneChip data

Gene expression classifiers have been used to predict diagnosis of disease, patient prognosis and patient response to therapy. Although there have been remarkable successes in this area, a particular goal of personalized medicine is the ability predict a response from a single gene expression microarray. One aspect of this problem is the normalization of microarrays. Affymetrix GeneChip microarrays are typically processed using model-based algorithms that require all of the data in order to adequately estimate the model. We experiment with the RMA normalization procedure in an incremental fashion, adding new chips to an existing normalization model. Focusing on tissue-specific normalization models, we generate datasets that have very small differences from a batch normalization. Through several large datasets of patient samples, we provide evidence that RMA models of normalization converge to a common model in homogenous samples. These results offer the promise of maintaining large data warehouses of patient microarray samples without the requirement of constant renormalization.

Abstract A58: JAK1 deficiency as a novel mechanism of tumor immune escape in uterine cancer

Recognition of MHC class I antigens on the surface of tumor cells is an essential process in the effector phase of CD8+ cytotoxic T lymphocyte (CTL)-mediated anti-tumor response. Presentation of MHC class I antigens on tumor cells is regulated by the JAK1/JAK2-mediated interferon-γ signaling pathway. We analyzed the targeted exome sequencing data of 3,274 human tumors from 48 different tissues in the Total Cancer Care (TCC) tumor bank and found JAK1 truncating mutations in 36 of 635 gynecologic tumors. The highest JAK1 truncating mutation rate (9.5%) was found in endometrial cancer. These truncating mutations result in the loss of the JAK1 protein tyrosine kinase (PTK) domain that is located in the C-terminal region. JAK1 truncating mutations in cancer cell lines of the Cancer Cell Line Encyclopedia (CCLE) databank also occur most often in endometrial cancer cells. Analysis of mutation sites identified three mutation hot spots. Re-sequencing of cancer cell lines, primary tumors, and matched normal tissues confirmed JAK1 mutations and revealed that these JAK1 mutations are somatic. Besides JAK1, recurrent JAK2 loss-of-function (LOF) mutations were also found in TCC tumors but at lower rates. Similar to JAK1, the highest rate of LOF JAK2 mutations was found in endometrial/uterine cancer. No JAK1 or JAK2 truncating mutation was identified in 130 samples from hematological malignancies. Functional assays showed that JAK1 deficient cancer cells were defective in interferon-γ-induced expression of tumor antigen processing machinery proteins LMP2 and TAP1 and cell surface expression of HLA molecules. These data identify recurrent JAK1 deficiency in endometrial cancer that impairs tumor antigen presentation by the MHC class I complex to CTLs. While PTKs have been perceived mostly as oncogenes and PTK inhibitors, including JAK inhibitors, have been developed for cancer therapy, our findings suggest that JAK1 is a novel tumor suppressor and that LOF JAK1 mutations allow tumor immune escape. Our findings also raise caution about the use of JAK inhibitors as therapeutic agents in solid tumors. Citation Format: Yuan Ren, Yonghong Zhang, Richard Z. Liu, David A. Fenstermacher, Kenneth L. Wright, Jamie K. Teer, Jie Wu. JAK1 deficiency as a novel mechanism of tumor immune escape in uterine cancer. [abstract]. In: Proceedings of the AACR Special Conference: Tumor Immunology and Immunotherapy: A New Chapter; December 1-4, 2014; Orlando, FL. Philadelphia (PA): AACR; Cancer Immunol Res 2015;3(10 Suppl):Abstract nr A58.

Abstract 1560: Loss-of-function JAK1 mutations reveal a new role of protein tyrosine kinase mutations in human cancer

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA Protein tyrosine kinases (PTKs) have been perceived mostly as proto-oncogenes. PTK mutations that have been studied in human cancer usually have increased kinase activities to drive malignant phenotypes. Thus, these PTK mutations are gain-of-function (GOF) mutations. By analyzing the targeted exome sequencing data of 3,274 human tumors from 48 different tissues in the Total Cancer Care (TCC) tumor bank, we have found JAK1 truncating mutations in 36 of 635 gynecologic tumors. The highest JAK1 truncating mutation rate (9.5%) was found in endometrial cancer. These truncating mutations result in the loss of the JAK1 PTK domain. Thus, they are loss-of-function (LOF) mutations. JAK1 truncating mutations in the Cancer Cell Line Encyclopedia (CCLE) databank also occur most often in endometrial cancer cell lines. Analysis of mutation sites identified three mutation hot spots. Re-sequencing of cancer cell lines, primary tumors, and matched normal tissues confirmed JAK1 mutations and revealed that these JAK1 mutations are somatic. JAK1 and JAK2 mediate interferon (IFN)-γ-regulated tumor immune surveillance. Recurrent JAK2 LOF mutations were also found in TCC tumors but at lower rates. Similar to JAK1, the highest rate of LOF JAK2 mutations was found in endometrial/uterine cancer. No JAK1 or JAK2 truncating mutation was identified in 130 samples from hematological malignancies. Functional assays show that JAK1-deficient cancer cells are defective in IFN-γ-induced expression of tumor antigen processing machinery proteins LMP2 and TAP1 and cell surface expression of HLA molecules. These data identify recurrent JAK1 deficiency in endometrial cancer that could impair tumor antigen presentation by the MHC class I complex to cytotoxic T lymphocytes. Our findings suggest that JAK1 is a PTK tumor suppressor and reveal a novel mechanism of tumor immune evasion via LOF mutations in a PTK. Citation Format: Yuan Ren, Yonghong Zhang, Richard Z. Liu, David A. Fenstermacher, Kenneth L. Wright, Jamie K. Teer, Jie Wu. Loss-of-function JAK1 mutations reveal a new role of protein tyrosine kinase mutations in human cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 1560. doi:10.1158/1538-7445.AM2014-1560

Abstract 3170: Identification of novel genetic alterations driving melanoma tumorigenesis.

Proceedings: AACR 104th Annual Meeting 2013; Apr 6-10, 2013; Washington, DC Recent high-throughput sequencing efforts have provided a comprehensive view of cancer genomes, revealing their complexity and heterogeneity. However, most of these studies remain descriptive without further functional and clinical validation of the candidate alterations, mainly due to the overwhelming number of somatic alterations present. To discover novel driver mutations of melanoma, massively parallel whole genome sequencing (WGS) was used to characterize 15 metastatic melanomas derived from 13 patients. A large number of somatic alterations were discovered in these tumors and hybridization capture-based validation confirmed 17,361 tier 1 point mutations, 84 tier 1 indels, and 411 somatic structural variants. As a pilot study to exploit this genomic data in order to identify novel genetic alterations driving melanoma tumorigenesis, we performed mutation proximity analysis to select candidates for further analysis. In this study, we addressed possible roles of DBC1 (Deleted in Bladder Cancer 1) and RASA1 (RAS p21 protein activator 1), which showed previously undocumented neighboring mutations. RASA1 is a GTPase activation protein that acts as a suppressor of RAS function. RASA1 has been implicated in actin filament polymerization, vascular development, cellular apoptosis, and cell motility. Our whole-genome analyses of melanomas identified two somatic missense mutations, targeting highly conserved neighboring Y472 and L481 in or around the PH domain in RASA1 in two samples. DBC1, also called BRINP, DBCCR1, and FAM5A, was previously shown to undergo loss of heterozygosity at 9q32-q33 in bladder cancer, and methylation silencing in bladder, breast, and lung cancers. Ectopic expression of DBC1 in bladder and lung cancer cells was reported to cause cell death and to inhibit cell proliferation, respectively. We observed six DBC1 missense mutations by whole genome analyses in four patients, including 2 neighboring mutations targeting S688 and S690. The shRNA-mediated knock down of DBC1 and RASA1 in melanocyte derived from Ink4a/Arf deletion/BRAF mutation background promoted proliferation, soft agar colony formation, and invasion. Ectopic expression of wild type DBC1 and RASA1 in human melanoma cell lines SKmel28 and WM983C (all with BRAFV600E), respectively, decreased soft agar colony formation, supporting their tumor suppressive roles. Various mutant forms of RASA1 and DBC1 were addressed for their roles. Interestingly, loss of RASA1 conferred decreased sensitivity to BRAF inhibitor Vemurafenib. In order to address the mutation frequency of DBC1 and RASA1, we analyzed additional melanoma samples and observed mutation rates of 21% for DBC1 (20/96 patients) and 9% for RASA1 (20/221). Therefore, our findings support that DBC1 and RASA1 play roles in melanoma suppression and the utility of genomic data for the identification of novel genes involved in tumorigenesis. Citation Format: Minjung Kim, Li Ding, Nathan Dees, Krishna L. Kanchi, Hyeran Sung, David Fenstermacher, Malachi Griffith, Gerry Linette, Lynn Cornelius, Vernon K. Sondak, James J. Mule, Richard K. Wilson, Jeffrey S. Weber. Identification of novel genetic alterations driving melanoma tumorigenesis. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3170. doi:10.1158/1538-7445.AM2013-3170

Abstract 2905: Validation techniques for PCA-based gene expression signatures.

Background: Many gene expression signatures exist for measuring the biological state of a profiled tumor. One way to derive a gene-signature of a biological event is to perturb cell lines to mimic the event. The signature is applied to tumor data to predict the magnitude of the biological effect in tumors. Examples include pathway activation, metastasis scores, and chemo-resistance. Principal Component Analysis (PCA) summarizes a gene signature into a score, however many published gene signatures capture proliferation, rather than the intended process, as the first PCA component (PC) when applied to tumor datasets Venet et al. (Venet, 2011, PLoS Comput. Biol.). Determining why this happens and how this effect can be detected is an important for utilizing these signatures. Aim: To develop a set of tools to determine if a derived gene signature works as intended and is robustly represented when applied to a tumor dataset. Results: Differences observed between a cell line experiment and tumor data set is a problem of calibration transfer. The signature is derived on data with controlled variation and then applied to data often exhibiting much larger variation. PCA is a powerful tool for summarizing a gene signature into a score, but there are several pitfalls, particularly when the variation is larger than expected. We have developed a visual tool for evaluating the application of a signature, PCA characteristics are measured against thousands of random gene signatures to determine the significance of the findings. Four key concepts are measured using this tool. Coherence: Elements of a signature should be correlated beyond chance; if a common mechanism is measured, there should be coherence to the gene signature. This is the amount of variance explained in the first PC. Uniqueness: The general direction of the data can drive most of the observed signal. A gene signature possessing a unique direction (relative to the entire dataset) provides confidence that it is measuring a true effect. Robustness: If a signature is designed to measure a single biological effect, then this signal should be sufficiently strong and distinct compared to other signals within the signature. This is measured by calculating the ratio between the explained variance of the first and second PCA components. Transferability. The derived PCA gene signature score should describe the same biology in the tumor data set as it does in the cell line data. This can be verified by comparing the cell-line based PCA model with the tumor-based PCA model. If the sign and relative importance (loadings) of individual genes are similar between the two PCA models, this is an indication that the calibration transfer was successful. Conclusions: We have developed a technique for validating PCA-based gene signatures work as intended when applied to tumor data. Application of this technique can identify instances in which signature scores do not represent the desired biological effect. Citation Format: Anders E. Berglund, Eric A. Welsh, David A. Fenstermacher, Steven A. Eschrich. Validation techniques for PCA-based gene expression signatures. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2905. doi:10.1158/1538-7445.AM2013-2905