Search

Radiotherapy (RT) is one of the primary treatments of head and neck squamous cell carcinoma (HNSCC), which has a high-risk of locoregional failure (LRF). Presently, there is no reliable predictive biomarker of radioresistance in HNSCC. Here, we found that mutations in NFE2L2, which encodes Nrf2, are associated with a significantly higher rate of LRF in patients with oral cavity cancer treated with surgery and adjuvant (chemo)radiotherapy but not in those treated with surgery alone. Somatic mutation of NFE2L2 led to Nrf2 activation and radioresistance in HNSCC cells. Tumors harboring mutant Nrf2E79Q were substantially more radioresistant than tumors with wild-type Nrf2 in immunocompetent mice, whereas the difference was diminished in immunocompromised mice. Nrf2E79Q enhanced radioresistance through increased recruitment of intratumoral polymorphonuclear myeloid-derived suppressor cells (PMN-MDSC) and reduction of M1-polarized macrophages. Treatment with the glutaminase inhibitor CB-839 overcame the radioresistance induced by Nrf2E79Q or Nrf2E79K. RT increased expression of PMN-MDSC–attracting chemokines, including CXCL1, CXLC3, and CSF3, in Nrf2E79Q-expressing tumors via the TLR4, which could be reversed by CB-839. This study provides insights into the impact of NFE2L2 mutations on radioresistance and suggests that CB-839 can increase radiosensitivity by switching intratumoral myeloid cells to an antitumor phenotype, supporting clinical testing of CB-839 with RT in HNSCC with NFE2L2 mutations.Significance:NFE2L2 mutations are predictive biomarkers of radioresistance in head and neck cancer and confer sensitivity to glutaminase inhibitors to overcome radioresistance.

Somatic DNA copy number alterations (CNAs) arise in tumor tissue because of underlying genomic instability. Recurrent CNAs that occur in the same genomic region across multiple independent samples are of interest to researchers because they may contain genes that contribute to the cancer phenotype. However, differences in copy number states between cancers are also commonly of interest, for example when comparing tumors with distinct morphologies in the same anatomic location. Current methodologies are limited by their inability to perform direct comparisons of CNAs between tumor cohorts, and thus they cannot formally assess the statistical significance of observed copy number differences or identify regions of the genome where these differences occur. We introduce the DiNAMIC.Duo R package that can be used to identify recurrent CNAs in a single cohort or recurrent copy number differences between two cohorts, including when neither cohort is copy neutral. The TCGA studies of lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) identified statistically significant CNAs in many known cancer-related genes, including gains of EGFR and losses of CDKN2A in each tumor type separately. By directly comparing the two cohorts, DiNAMIC.Duo detects statistically significant copy number differences for CDKN2A, thus suggesting that losses are more pronounced in LUSC; in contrast, differences for EGFR are not statistically significant, which suggests similar levels of gain. Existing methods that detect recurrent CNAs in a single cohort cannot make this distinction. Recent studies have leveraged TCGA data to find known cancer genes in chr3q, chr14q13, and chr20q11 that are differentially expressed in LUAD vs. LUSC. DiNAMIC.Duo identifies statistically significant copy number differences in these regions, which suggests that the observed expression changes may be driven by underlying differences in copy number. Citation Format: Vonn Walter, Hyo Young Choi, Xiaobei Zhao, Yan Gao, Jeremiah Holt, D. Neil Hayes. Detecting somatic DNA copy number differences with DiNAMIC.Duo [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2071.

Integrated analyses of multiple genomic datatypes are now common in cancer profiling studies. Such data present opportunities for numerous computational experiments, yet analytic pipelines are limited. Tools such as the cBioPortal and Regulome Explorer, although useful, are not easy to access programmatically or to implement locally. Here, we introduce the MVisAGe R package, which allows users to quantify gene-level associations between two genomic datatypes to investigate the effect of genomic alterations (e.g., DNA copy number changes on gene expression). Visualizing Pearson/Spearman correlation coefficients according to the genomic positions of the underlying genes provides a powerful yet novel tool for conducting exploratory analyses. We demonstrate its utility by analyzing three publicly available cancer datasets. Our approach highlights canonical oncogenes in chr11q13 that displayed the strongest associations between expression and copy number, including CCND1 and CTTN, genes not identified by copy number analysis in the primary reports. We demonstrate highly concordant usage of shared oncogenes on chr3q, yet strikingly diverse oncogene usage on chr11q as a function of HPV infection status. Regions of chr19 that display remarkable associations between methylation and gene expression were identified, as were previously unreported miRNA–gene expression associations that may contribute to the epithelial-to-mesenchymal transition. Significance: This study presents an important bioinformatics tool that will enable integrated analyses of multiple genomic datatypes. Cancer Res; 78(12); 3375–85. ©2018 AACR.

Activating mutations in NRF2 frequently occur in human lung cancers, especially the E79Q mutation. However, how the activating mutation contributes to lung tumor development is not well understood. We used a new genetically engineered mouse model (GEMM), LSL-Nrf2E79Q/+ with Cre-inducible expression of this mutation within the endogenous Nrf2 locus, to understand its contribution to lung tumor development. We compared tumor development in p53fl/fl;p16fl/fl;LSL-Nrf2+/+ (WT) versus p53fl/fl;p16fl/fl;Nrf2E79Q/+ (Het) mice after intranasal instillation of Adenoviral Cre. While we did not observe a significant difference in lung adenocarcinoma (LUAC) between the Het and WT mice, the Het mice showed significantly higher frequencies of bronchial neuroendocrine tumors that resembled small cell lung carcinomas (SCLC) compared to WT. Additionally, the latency of SCLC lesions was lower in Het mice compared to WT. Surprisingly, LUACs in both groups were negative for NRF2 by IHC labeling despite recombination of the mutant Nrf2 allele. While all lesions in the WT and Het mice displayed positive ASCL1 and CGRp (NE markers) IHC labeling, only the SCLCs in Het mice expressed NRF2. This is the first study showing activation of Nrf2E79Q mutation drives development of small cell lung carcinomas. Interestingly, our data also suggest that the E79Q NRF2 activating mutation may inhibit the progression of LUAD. Citation Format: Samera Hamad, Stephanie Montgomery, Brittany Bowman, Ryan Murphy, Scott Randell, Trudy Oliver, D. Neil Hayes, M. Ben Major, Bernard Weissman. Activation of NRF2 signaling accelerates development of small cell lung cancer in mice [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2939.

Few targeted therapies exist for triple negative breast cancer (TNBC), an aggressive and deadly subtype. Obesity exacerbates poor outcomes in TNBC due to elevated invasion and metastasis leading to increased mortality. Obesity has paradoxically been shown to improve immune checkpoint therapies in other cancers, yet the underlying mechanisms are unclear. Despite being an inflammatory state, obesity increases immune checkpoint ligands PD-1 and PD-L1 which drives immunosuppression. Thus, we tested the hypothesis that obesity-driven changes to the immune milieu will improve ICB efficacy in TNBC. Syngeneic tumors were generated via orthotopic engraftment of E0771 basal-like TNBC cell line into age-matched immune-competent C57BL/6J female littermates on obesogenic or low fat diets. Mice were treated with anti-mouse anti-PD-1 or isotype control (IgG2a) every 3rd day until sacrifice. Obese mice gained almost 2-fold great body weight and 3.6-fold greater adiposity compared to lean mice and immunotherapy did not impact weights or body composition. Obesity led to immunosuppression systemically in bone marrow and spleen in tumor-free mice, which was exacerbated in tumor-bearing mice. Obese mice had significantly greater tumor progression and fewer regressed tumors at endpoint vs. lean mice. Anti-PD-1 significantly reduced tumor progression in obese mice with 4.2-fold reduction in volume and 5.7-fold reduction in tumor weights in obese mice vs. isotype controls. Anti PD-1 significantly reduced immunosuppressive cells including M2-like tumor associated macrophages and monocytic and granulocytic myeloid derived suppressor cells (MDSC) and raised anti-tumor M1-like macrophages, cytotoxic CD8+ T cells, and dendritic cells. Last, the microbiome has potent effects on responses to anti-tumor therapies such as chemotherapy and immune checkpoint blockade. Obesity is a major regulator of the gut microbiome. We found that beneficial bacteria belonging to genus Bifidobacterium was higher in lean compared to obese mice, which could limit tumor progression and lead to greater regression through robust anti-tumor immunity. In obese mice, beneficial bacteria belonging to genus Ruminococcus, Adlercreutzia, Corpococcus that promote anti-tumor immunity increased with anti PD-1 immunotherapy. In sum, we show for the first time that obesity-induced systemic and microenvironmental immunosuppression augmented tumor incidence and tumor progression. Furthermore, anti-PD-1 immune checkpoint blockade successfully reduced tumor progression in obese mice through reprogramming not only the TME but systemic immune milieu as well. Immunosuppressive targets unique to the obese TME could be targeted in concert with checkpoint inhibitors in future interventions to enhance durable anti-tumor immunity. Citation Format: Ajeeth K Pingili, Mehdi Chaib, Laura M Sipe, Emily J Miller, Bin Teng, Rahul Sharma, Qusai AI Abdallah, Deidra Daria, Radhika Sekhri, Heejoon Jo, Mims S Tahliyah, D. Neil Hayes, Joseph F Pierre, Liza Makowski. Immune checkpoint blockade reprograms tumor microenvironment and systemic immune landscape in obesity associated triple negative breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS17-04.

The most basic differential expression analysis involves the identification of genes that exhibit distinct expression patterns in two cohorts. Surprisingly, even though there are numerous computational methods for detecting and assessing the statistical significance of somatic DNA copy number (CN) gains and losses in a single cohort, there are no tools for detecting and assessing the statistical significance of CN differences between two cohorts. Instead, typically each cohort is analyzed separately and distinct findings are presented as evidence of CN differences. Cyclic shift testing was originally introduced as a method to identify and assess the statistical significance of recurrent somatic CN gains and losses. Subsequent work explored theoretical statistical underpinnings of cyclic shift testing. Here we extend the previous results by showing that cyclic shift testing can be applied to detect and assess the statistical significance of CN differences between two cohorts. Human papilloma virus (HPV) infection is a risk factor for head and neck squamous cell carcinoma (HNSC), and it is known that HPV+ and HPV- HNSC have distinct mutational and gene expression profiles. We apply cyclic shift testing to detect and assess the statistical significance of CN differences between HPV+ and HPV- HNSC using data from The Cancer Genome Atlas HNSC cohort. Our approach detects statistically significant CN differences in regions that are known to contain HNSC drivers (7p, 9p21, 11q13, 11q14-qter), as well as regions of chr14, chr16, and others that are less well characterized. Citation Format: Vonn Walter, Hyo Young Choi, Xiaobei Zhao, Jose Zevallos, D. Neil Hayes. Detecting somatic DNA copy number differences with cyclic shift testing [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5460.

SWI/SNF chromatin remodeling complexes regulate critical cellular processes, including cell-cycle control, programmed cell death, differentiation, genomic instability, and DNA repair. Inactivation of this class of chromatin remodeling complex has been associated with a variety of malignancies, including lung, ovarian, renal, liver, and pediatric cancers. In particular, approximately 10% of primary human lung non–small cell lung cancers (NSCLC) display attenuations in the BRG1 ATPase, a core factor in SWI/SNF complexes. To evaluate the role of BRG1 attenuation in NSCLC development, we examined the effect of BRG1 silencing in primary and established human NSCLC cells. BRG1 loss altered cellular morphology and increased tumorigenic potential. Gene expression analyses showed reduced expression of genes known to be associated with progression of human NSCLC. We demonstrated that BRG1 losses in NSCLC cells were associated with variations in chromatin structure, including differences in nucleosome positioning and occupancy surrounding transcriptional start sites of disease-relevant genes. Our results offer direct evidence that BRG1 attenuation contributes to NSCLC aggressiveness by altering nucleosome positioning at a wide range of genes, including key cancer-associated genes. Cancer Res; 74(22); 6486–98. ©2014 AACR.

NRF2 is a transcription factor that mediates stress responses. Oncogenic mutations in NRF2 localize to one of its two binding interfaces with KEAP1, an E3 ubiquitin ligase that promotes proteasome-dependent degradation of NRF2. Somatic mutations in KEAP1 occur commonly in human cancer, where KEAP1 may function as a tumor suppressor. These mutations distribute throughout the KEAP1 protein but little is known about their functional impact. In this study, we characterized 18 KEAP1 mutations defined in a lung squamous cell carcinoma tumor set. Four mutations behaved as wild-type KEAP1, thus are likely passenger events. R554Q, W544C, N469fs, P318fs, and G333C mutations attenuated binding and suppression of NRF2 activity. The remaining mutations exhibited hypomorphic suppression of NRF2, binding both NRF2 and CUL3. Proteomic analysis revealed that the R320Q, R470C, G423V, D422N, G186R, S243C, and V155F mutations augmented the binding of KEAP1 and NRF2. Intriguingly, these “super-binder” mutants exhibited reduced degradation of NRF2. Cell-based and in vitro biochemical analyses demonstrated that despite its inability to suppress NRF2 activity, the R320Q “superbinder” mutant maintained the ability to ubiquitinate NRF2. These data strengthen the genetic interactions between KEAP1 and NRF2 in cancer and provide new insight into KEAP1 mechanics. Cancer Res; 74(3); 808–17. ©2013 AACR.

Integrated analyses of multiple genomic datatypes are now common in cancer profiling studies. Such data present opportunities for numerous computational experiments, yet analytic pipelines are limited. Tools such as the cBioPortal and Regulome Explorer, while useful, are not easy to access programmatically or implement locally. The MVisAGe R package allows users to quantify gene-level associations between two genomic datatypes in order to investigate the effect of genomic alterations (e.g. DNA copy number changes on gene expression). Visualizing Pearson/Spearman correlation coefficients according to the genomic positions of the underlying genes provides a powerful yet novel tool for conducting exploratory analyses. We demonstrate its utility by analyzing three publicly available datasets from squamous tumors. Our approach highlights canonical oncogenes in chr11q13 that displayed the strongest associations between expression and copy number, including CCND1 and CTTN, genes not identified by copy number analysis in the primary reports. We demonstrate highly concordant usage of shared oncogenes on chr3q, yet strikingly diverse oncogene usage on chr11q as a function of HPV infection status. Regions of chr19 that display remarkable associations between methylation and gene expression were identified, as were previously unreported miRNA-gene expression associations that may contribute to the epithelial-to-mesenchymal transition. Citation Format: Vonn Walter, Ying Du, Ludmila Danilova, Michele Hayward, D. Neil Hayes. Integrated analysis with MVisAGe identifies concordant and discordant genomic alterations of driver genes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4254.

Tumor heterogeneity influences the clinical outcome of patients with cancer, and the diagnostic method to measure the tumor heterogeneity needs to be developed. We analyzed genomic features on pairs of primary and multiple metastatic lymph nodes from six patients with lung cancer using whole-exome sequencing and RNA sequencing. Although somatic single-nucleotide variants were shared in primary lung cancer and metastases, tumor evolution predicted by the pattern of genomic alterations was matched to anatomic location of the tumors. Four of six cases exhibited a branched clonal evolution pattern. Lymph nodes with acquired somatic variants demonstrated resistance to the cancer treatment. In this study, we demonstrated that multiple biopsies and sequencing strategies for different tumor regions are required for a comprehensive understanding of the landscape of genetic alteration and for guiding targeted therapy in advanced primary lung cancer. Cancer Res; 76(22); 6568–76. ©2016 AACR.

Objective: To assess the expression of Apela and evaluate its co-expression with cancer stem cell markers and angiogenesis markers in high-grade serous ovarian cancer and further elucidate the role of this poorly characterized gene. Background: High-grade serous ovarian cancer (HGSOC) presents a major therapeutic challenge due to advanced stage at diagnosis, aggressive metastatic behavior, and therapeutic resistance. A critical barrier to progress in early diagnosis and treatment is the lack of understanding of the mechanisms that underlie ovarian cancer tumorigenesis. Apela (Apelin early ligand A), also called Toddler and Elabela, is a novel ligand for the G-coupled protein receptor Apelin-R, which has been implicated in cancer tumorigenesis and angiogenesis through activation of the PIK3-AKT (phosphatidylinositol-4,5-bisphosphate 3-kinase and protein-kinase B) pathway. This signaling pathway plays an important role in embryonic stem cell growth and self-renewal. Current research indicate that Apela may be involved in cancer stem cells and angiogenesis pathways as well. Markers of angioneogenesis (Angptl4 [angiopoietin-like protein 4]) and cancer cell stemness (Nestin and CD133/PROM1), were selected for study with Apela because of known expression in HGSOC and other studies showing co-expression in other high-grade malignancies with Apela. Methods: RNA was extracted from established ovarian cancer cell lines (OVCAR3, OVCAR8, CAOV4, OV90, and SKOV3) and expression of Apela, Apelin (another ligand for the Apelin-R), Apelin-R and ANGPTL-4 was determined by qPCR and normalized to actin expression. Formalin-fixed paraffin-embedded (FFPE) slides of HGSOC tissue and normal adjacent ovarian tissue were subjected to RNAscope Multiplex Fluorescent in situ hybridization (ISH) using individual gene probes for Angptl4, Nestin, CD133 and Apela to detect expression at the mRNA level. Immunohistochemistry (IHC) was then performed to confirm Angptl4, Nestin, CD133 and Apela co-expression at the protein level in the FFPE HCSOC tissue. All images were analyzed by confocal microscopy. The TCGA database was analyzed to determine levels of upregulation of Apela in ovarian cancer. An ovarian cancer microarray using 10 paired matched controls was utilized to further validate the presence of the Results: Apela expression can be detected in FFPE tissue at the RNA level and the same pattern of protein expression is seen with IHC. Apela and Angptl4 co-localize with the cancer stem cell markers nestin and CD133. Apela is upregulated in HGSOC when examining the TCGA database compared to other forms of cancer. Conclusions: Apela is expressed at relatively low levels in established ovarian cancer cell lines. However, Apela is highly localized in ovarian cancer tissue, and is co-expressed with markers of angiogenesis and stem cells, possibly representing a stem cell niche that promotes tumorigenesis. Citation Format: Laura R. Daily, Debolina Ganguly, D Neil Hayes, Yinan Wang, Michelle M. Sims, Jinjun Cheng, Adam C. ElNaggar, Lawrence M. Pfeffer. The role of Apela in high-grade serous ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3071.

Targeted exome sequencing and variant calling involves comparing a patient’s tumor and normal sequencing to identify mutated loci within clinically actionable (CA) genes. However sequencing matched normal (MN) increases the marginal cost greatly relative to the amount of genomic information provided since most bases are unchanged. MN may not be available or in some cases poor quality of the MN sequencing may hinder the interpretation of results. Without the MN control, with the expected mix of sequencing errors, artifacts, and mis-mapped regions, germline (GM) variants will further contaminate the list of variant calls (VCs) with up to 10,000 alternate loci. Some reports address TO sequencing challenges and these have significant limitations. In general, more controls leads to both higher computational costs and risk of selecting low quality controls. We aim to return several dozen VCs using a reasonable number of controls. We propose using information obtained from pooling high quality normal samples and GM/somatic databases and applying a set of filtering criteria for a cohort of tumor only (TO) samples to exclude the majority of false positive (FP) VCs while retaining those that are CA. Through benchmarking paired tumor normal (PTN) samples, we aim to find the minimum controls needed to reach an adequate sensitivity (SENS), specificity (SPEC), and final number of VCs. We selected a cohort of patients with clinically validated mutations from a clinical trial of NGS in cancer (LCCC1108: NCT01457196). From a cohort we selected a subset of 100 patients with a spectrum of clinical and genomic parameters relevant to a robust method. All patients were PTN samples as well as CLIA confirmation of relevant CA mutations. We developed a set of filtering and annotations processes to allow VC annotation of the cohort which did not include the MN and compared to results obtained from PTN sequencing. Samples were sequenced using hybrid capture technology and Illumina brand paired-end sequencing. VCs were combined using Strelka, UNCseqR, and Cadabra and annotated with Oncotator. SENS and SPEC are used to quantify the pipeline’s performance by treating the filtered PTN VCs and clinically validated as gold standard. Overlaps in the tumor and pooled normal cohort (PNC) samples were avoided. With respect to a larger set of high quality mutations called by the PTN pipeline, the TO method documented on average 95% SENS and >99% SPEC for variant detection. The method showed a tradeoff between SENS and FP rate allowing for small drops in SENS for large decreases in the numbers of FP (4-5% decrease in SENS from 99% to 95%) resulted in 90% decrease in FP VCs with a final average of 70 filtered VCs per sample. Results suggest that analyzing PTN targeted exome NGS samples as TO and obtaining a disjoint PNC selected based on sequencing quality metrics can achieve high SENS and SPEC with a reasonable number of filtered VCs. Also benchmarking aids in excluding controls from the PNC that may miss CA calls. Citation Format: Paul Little, D. Neil Hayes, Joel Parker, Jose Zevallos. Statistical considerations in tumor only variant calling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3584. doi:10.1158/1538-7445.AM2017-3584

Insertions and deletions in the transcriptome can potentially have significant impact on function and can be clinically actionable. A number of methods have recently been developed that improve indel detection in DNA by utilizing realignment and/or localized assembly to aid in discovering mutations that are more difficult to detect than single nucleotide polymorphisms. While significant effort has been put into these methods in the context of DNA discovery, RNA-Seq has not received the same attention. For example, the widely used Genome Analysis Toolkit (GATK) recommends removal of splice junctions from all reads prior to proceeding with variant calling methods that are very similar to those used for DNA. We hypothesize that utilizing splice junctions to augment localized assembly across exon-exon boundaries can improve read alignments resulting in improved variant detection in RNA-Seq data. We have developed an update to the Assembly Based ReAligner (ABRA2) that makes use of splice junction information to aid in realignment of reads. We assessed indel detection performance using the BEERs RNA-Seq simulator. Two million simulated reads of length 100nt were generated across 1000 human genes. In order to test more complex indel detection, the simulator was modified to generate an even distribution of 1391 indels of length 1 to 100. The simulated reads were aligned using STAR and subsequently realigned using ABRA2. Freebayes and GATK were run against the non-realigned STAR output resulting in indel detection sensitivity of 15% and 18% with precision of 97% and 88% respectively. Freebayes was then run against the ABRA2 realignments resulting in an approximately 4 fold increase in indel detection sensitivity of 67% while maintaining precision of 97%. In frame deletions in Epidermal Growth Factor Receptor (EGFR) have oncogenic potential, can be indicators for Gefitinib or Erlotinib treatment, and are frequently detected in lung cancer via DNA sequencing. Among 514 TCGA Lung Adenocarcinoma RNA-Seq samples, ABRA2 revealed 73 in frame coding indels in EGFR ranging in length from 3 to 24 bases, including 58 deletions in exon 19 of length 15 bases. This represents a 3 fold increase relative to what was initially reported from DNA by TCGA. Further, in the absence of ABRA2, only 5 in frame coding indels are detected in RNA, the largest of which is a 9 base deletion. We have presented here ABRA2, a new version of the Assembly Based ReAligner that is capable of accurately realigning RNA-Seq reads containing variations that are currently not well handled by widely used aligners and variant callers, thus improving accuracy of variant detection in RNA-Seq. Citation Format: Lisle E. Mose, D. Neil Hayes, Charles M. Perou, Joel S. Parker. Improved indel detection in RNA-seq data via assembly based re-alignment reveals expressed Epidermal Growth Factor Receptor indels in Lung Adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3592. doi:10.1158/1538-7445.AM2017-3592

Often, the problem in cancer research is figuring out how a gene or pathway works in regulating cellular transformation. The question of what RAS activates or PTEN inhibits have been classic dilemmas of modern cancer biology. In these cases, biochemical and genetic studies have provided us with a fairly clear picture of the cancer relevant functions of these genes. For LKB1, a more recently identified human tumor suppressor gene, however, the problem is different. This serine-threonine kinase that is conserved from yeast to mammals seems to play a role in many diverse cellular pathways. Therefore, although elegant functional and genetic approaches have established critical roles for LKB1 in the regulation of metabolism, motility, polarity, and the cell cycle, the role(s) responsible for its true tumor suppressor function(s) is unknown. One is reminded of an Agatha Christie murder mystery where nearly every character in the book has reason to be suspected of committing the crime—there are too many suspects for how LKB1 might repress lung cancer. [Cancer Res 2008;68(10):3562–5]

High dimensional assays, such as DNA microarrays for gene expression, have helped delineate the complexity and diversity of most major cancer types. The Cancer Genome Atlas (TCGA), with its goal to comprehensively genomically characterize over twenty different types of tumors, gives us a wealth of data to start looking at similarities across tumor types. Much of the current data is now generated using Massively Parallel Sequencing (MPS) approaches, which yields tens of millions of “reads” per sample. Illumina Hiseq-based mRNA-sequencing data for twelve tumor types representing over 3,500 samples were analyzed for patterns unique to each tumors, and patterns present across multiple tumor types. Tumor types tested included rectal adenocarcinoma, colon adenocarcinoma, lung squamous cell carcinoma and adenocarcinoma, uterine corpus endometrial carcinoma, ovarian serous cystadenocarcinoma, kidney clear cell, breast ductal and lobular carcinoma, glioblastoma multiforme, bladder carcinomas, acute myeloid leukemia, and head and neck squamous cell carcinoma. Tissue-specific expression was the dominant signature; however, there were similarities across sets of tumors including a common “squamous” pattern that was populated by tumors from three distinct anatomic sites. Pathway or Module-based analysis also provided methods to comparisons across tissues. Cross tissue analysis of mRNA-seq data provides interesting observations of common tumor biology with potential insights into treatment. Citation Format: Katherine A. Hoadley, Cheng Fan, Matthew D. Wilkerson, Lisle E. Mose, Stuart R. Jefferys, J. Todd Auman, Joel S. Parker, D. Neil Hayes, Charles M. Perou. Multi-tumor analysis of TCGA data identifies expression commonalities across tumor types. [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 SY12-03. doi:10.1158/1538-7445.AM2013-SY12-03

Background: Clinical cancer samples frequently have significant contamination by stromal cells that increase the difficulty of mutation calling. Histopathologic light microscopy (LM) estimation of purity is commonly used to qualify samples for molecular testing. With large datasets generated by current technologies, more accurate estimations of purity may be facilitated by use of SNP chip (SNP) or sequencing (seq) data. We present a computational framework that considers a variety of technical factors, including percent (%) tumor, seq quality, and mutation type that impact the detection and evaluation of somatic mutations reported to patients. Methods: LCCC1108 (“The development of a tumor molecular analyses program and its uses to support treatment decisions”) is a registry study for cancer patients that performs targeted sequencing with a panel of >200 clinically actionable genes for clinical use. A subset of patients have additional analyses performed including DNA SNP and RNA seq. Any clinically actionable variant is orthogonally confirmed in a CLIA-certified lab. Results: As of 12/1/2013, 816 patients have enrolled. 350 cases have both molecular and LM estimates of tumor purity. Agreement between molecular estimates of % tumor and LM review was low (Pearson's correlation coefficient 0.16 for the most favorable comparison). No clear etiology could explain the low agreement, although examination of extreme cases in which LM approached 100% tumor and molecular estimates approached zero generally included tumors with low rates of molecular alterations (e.g. adenoid cystic carcinoma, chronic myelogenous leukemia). Additionally, independent assessment of % tumor by molecular analysis such as SNP vs. seq also had low agreement (0.13), although a clear pattern could be seen in the disagreement for certain cases. Methods that rely heavily on copy number(CN) alterations often under-predicted estimates of % tumor relative to methods that relied on the mutant allele fraction (MAF) in cases where there were few CN alterations but where point mutations predominated. Excluding these outliers, the agreement was high and molecular estimates of % tumor more reproducible. Most molecular tumor estimates performed more poorly or failed as the % stromal infiltration rose above 50%. Seq estimates of % tumor tended to be noisier for coverage Conclusion: In clinical practice, accurate estimates of % tumor spanning the clinical spectrum are critical for somatic mutation calling. Pure analytic approaches to estimates of % tumor may fail in tumors with low rates of somatic mutation and CN alteration. Ultimately, a combination of methods will be required to describe samples for optimal variant detection. Citation Format: Heejoon Jo, Nirali M. Patel, Karen E. Weck, David A. Eberhard, D. Neil Hayes, Joel S. Parker, Michele Hayward, Ashley Salazar, Juneko E. Grilley-Olson. Correlating molecular and histopathologic tumor purity: An analysis of 816 patients. [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 5598. doi:10.1158/1538-7445.AM2014-5598

Background: While the importance of RAS and mTOR signaling pathways in non-small cell lung cancer (NSCLC) is well known, the activation status of key components in the pathway as a function of histology and oncogene mutation status has been not completely described. The purpose of this study is to investigate phosphorylation status of signal transduction pathway members and to elucidate their prognostic significance in relation to adenocarcinoma and squamous cell carcinoma. Methods: Phosphorylated extracellular signal-regulated kinases 1 and 2 (pERK), MEK 1 and 2 (pMEK), and mammalian target of rapamycin (pmTOR) were immunohistochemically examined using tissue microarray samples of 269 surgically resected primary NSCLC. Thyroid transcription factor-1 (TTF-1) expression and mutation status of EGFR and KRAS were also evaluated. The relationships between 5-year progression free survival (PFS) and phospho-protein expression were investigated by univariate analyses of log-rank test and multivariate analyses using Cox regression model. Results: A total of 98 adenocarcinomas and 134 squamous cell carcinomas were included in the analyses. In adenocarcinoma, pERK expression was significantly related to pMEK (P = 0.001), pmTOR (P < 0.001), and TTF-1 (P = 0.013) expression. In squamous cell carcinoma, pERK expression was significantly related to pMEK (P < 0.001) expression. EGFR or KRAS mutations were significantly associated with TTF-1 expression (P = 0.014) in adenocarcinoma. However, EGFR and KRAS mutation status were not significantly associated with pERK, pMEK, and pmTOR expressions in adenocarcinoma. By univariate analyses, high pERK expression was associated with better 5-year PFS (P = 0.018) in adenocarcinoma and high pMEK expression was associated with worse 5-year PFS (P = 0.043) in squamous cell carcinoma. In multivariate analyses, pERK (P = 0.040) and pMEK (P = 0.048) expression were independent prognostic factors. Conclusions: There were different patterns of phospho-protein expression in adenocarcinoma and squamous cell carcinoma. pERK expression was a good prognostic factor only in adenocarcinoma and pMEK expression was a poor prognostic factor only in lung squamous cell carcinoma. Citation Format: Sang-Won Um, Jingxiang Chen, Carrie Lee, Norman E. Sharpless, Patrick J. Roberts, D. Neil Hayes. Differential phosphorylation of signaling targets as a function of RAS mutation status and tumor histology in non-small cell lung cancer. [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 2400. doi:10.1158/1538-7445.AM2013-2400

Personalized cancer medicine, the matching of therapies to a given patient's somatic alterations, depends on highly accurate and complete identification of patients’ somatic alterations, or their mutome. Advances in sequencing technologies (exome sequencing, RNAseq, and whole genome sequencing) have provided a means to examine large portions of the genetic content of patients’ cancers. Computational tools have arisen that make somatic mutation predictions utilizing particular sequencing assays; however, each sequencing assay has limitations and existing mutation detection tools exhibit less than ideal agreement when analyzing the same data. The task of identifying all somatic mutations in one patient's cancer remains a challenge to personalized cancer medicine. Typically, somatic mutation detection is performed utilizing DNA sequencing. Because RNA sequencing is often a component of genome characterization projects along with DNA sequencing, we sought to evaluate the possible added value of RNA sequencing in somatic mutation detection. We have developed an original computational method, UNCeqR, that makes patient-specific somatic mutation predictions utilizing RNA sequencing combined with DNA sequencing. DNA mutations and RNA mutations are statistically modeled separately and results are combined in a meta-analytic fashion, resulting in up to three predictions for a locus: DNA-only, RNA-only, and DNA+RNA. In addition to de novo genomewide mutation predictions, UNCeqR can query specific a priori mutations. UNCeqR was applied to The Cancer Genome Atlas (TCGA) lung squamous cell carcinoma sequencing data, consisting of Ilumina RNAseq and Illumina exome sequencing. Of annotated exons, 20% had very low to zero coverage in RNA and 5% had very low to zero coverage in DNA, indicating that both sequencing assays add new genomic territory for mutation detection. Limiting to regions with both DNA and RNA coverage, 56% of mutations detected from DNA were also predicted by RNA, providing an independent validation of these mutations. To evaluate if mutation detection using DNA+RNA is superior to detection using DNA-only, cancer specimen DNA and RNA reads were randomly split into subsamples. UNCeqR was executed on each of the subsamples and mutation agreement was compared among pairs of subsamples within regions of DNA and RNA coverage. Compared with the DNA-only method, DNA+RNA mutation detection exhibited a 42% relative increase in percent agreement across subsamples and a 230% relative increase in the number of mutations detected. Therefore, RNA sequencing adds positive value to somatic mutation detection via UNCeqR. 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 3975. doi:1538-7445.AM2012-3975

Background: Head and neck squamous cell carcinoma (HNSCC) is a complex and heterogeneous disease for which the underlying genomic aberrations are unknown. Previous studies have suggested that chr11q13 is a region of interest in HNSCC. Independent studies suggest that genes from this region may be important – for example, CTTN, CCND1, MYEOV, and ANO1 – but to our knowledge no studies have focused on the integration of copy number data and potential targets, or investigated the prevalence of copy number aberrations (CNAs) and clinical covariates. Human papillomavirus-associated HNSCC (HPV-HNSCC) is of particular interest, and it is known that HPV-HNSCC is most commonly found in the oropharynx. However, to our knowledge no studies have examined the association between CNAs of chr11q13 and patient outcome for patients with oropharynx tumors. Techniques: After receiving informed consent, we obtained tumor samples from an incident surgical series of HNSCC patients at the University of North Carolina Hospital. Copy number (CN) and gene expression (GE) assays were performed using Affymetrix Genome-Wide SNP6.0 and Agilent 44K Gene Expression platfoms, respectively. CN values were subsequently analyzed for amplifications in region of chr11q13 using an amplification threshold equivalent to at least a 50% increase in copy number. Results: Samples from 162 patients were obtained, making this the largest genomic HNSCC series ever reported to our knowledge. After passing quality control on both platforms, CN, GE, and clinical data was available for 86 patients in the integrated analysis. Data from the remaining 76 samples was also reviewed, and for these samples either CN, GE, or clinical information is available. We estimate that amplifications of chr11q13 are found in approximately 30% of all HNSCC patients. The consensus region covers seven genes, and four of these genes – CTTN, CCND1, MYEOV, and ANO1 – have expression values that are statistically significantly correlated with copy number, supporting existing literature both for the region and these genes. If attention is restricted to patients with oropharynx tumors, we find that patients with chr11q13 amplifications have an increased risk of recurrence or death when compared to patients without chr11q13 amplifications. These amplifications are highly overlapping with smoking status in oropharynx patients, suggesting a potential mechanism for the worse outcome seen in HPV positive smokers relative to nonsmokers (N Eng J Med 2007;356:1944-1956). Conclusion: Integrated genomic analysis unmasks frequent copy number gains in chr11q13. Copy number values in chr11q13 are associated with expression of known regional oncogenes. We suggest a potential explanation for the worse clinical outcomes seen in HPV positive smokers. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3939. doi:10.1158/1538-7445.AM2011-3939

Background: Epithelial ovarian cancer is the 5th leading cause of cancer death in women. Our objective was to identify key genetic events important for the pathogenesis of this lethal disease. Methods: Levels of microRNA (miRNA) expression were examined in specimens of primary papillary serous carcinomas, normal ovary and distal fallopian tube using Next Generation Sequencing and a custom expression array. Chromosomal gains and losses were also examined by CGH. SYBR Green reagents were used to measure relative expression of target gene expression by quantitative real-time PCR. Functional impact of altered miRNA expression was tested using standard MTT and Caspase 3/7 assays to measure proliferation and apoptosis (Promega). Key outcome demographics were coded and correlated with miRNA and gene expression by Kaplan-Meier analysis. Results: A total of 140 miRNAs were differentially expressed when papillary serous ovarian cancers were compared to either fimbrae of normal fallopian or normal ovary. Of these, 36 miRNAs were found to correlate with either overall survival, disease free interval (DFI) or platinum sensitivity. Nineteen (19) of these clinically significant miRNAs mapped to single primate-specific genomic locus located at 19q13.41. This locus spanned 125 Kb of non-coding DNA and encoded a total of 44 miRNAs, most all of which showed significant copy number variation in papillary serous ovarian cancers (n = 178) and showed copy losses in the majority of tumors. Using established algorithms for target prediction, we found that this miRNA cluster collectively targeted more than 2800 distinct genes. Key loci included gene products implicated in the epithelial-to-mesenchymal transition (Snail, Slug) as well as both the G1-S and G2-M cell cycle checkpoints (MYCN and Wee1). Transfection of established ovarian cancer cell lines with individual 19q13.41 miRNAs significantly reduced expression of Snail, Slug, Wee1, resulting in altered proliferation and apoptosis. Conclusions: Altered expression of 19q13.41 cluster miRNAs correlate with significant clinical outcomes for women with papillary serous ovarian cancers. These miRNAs appear to play a key role in regulating the expression of gene products critical for the ongoing proliferation and metastasis of ovarian cancer. Future work will focus on dissecting the role of individual 19q13.41 miRNAs in ovarian and other cancers as well as validating the novel nanoparticle-based strategies we have developed for therapeutic miRNA delivery. Supported by NIH TCGA and the Ovarian Cancer Research Foundation (OCRF) Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2029.