Identification of a Biomarker Panel from Genome-Wide Methylation to Detect Early HPV-Associated Oropharyngeal Cancer

As oropharyngeal cancer (OPC) associated with human papillomavirus (HPV) increases in men, the need for a screening test to diagnose OPC early is crucial. This study agnostically identified differentially methylated CpG sites to identify additional biomarkers to improve screening for early OPC.DNA was extracted from oral gargles of 89 early cases and 108 frequency matched healthy controls, and processed for genome-wide methylation using the Illumina Infinium MethylationEPIC BeadChip. Selected sites were combined with our prior methylation data in the EPB41L3 gene (CpG sites 438, 427, and 425) and oral HPV16 and HPV18 status were considered as binary variables (positive/negative). Lasso regression identified CpG sites strongly associated with early OPC. ROC curves with AUC were generated. The panel was validated utilizing bootstrap resampling.Machine learning analyses identified 14 markers that are significantly associated with early OPC, including one EPB41L3 CpG site (438) and oral HPV16 status. A final model was trained on all available samples using the discovered panel and was able to predict early OPC compared with controls with an AUC of 0.970 on the training set. In the bootstrap validation sets, the average AUC was 0.935, indicating adequate internal validity.Our data suggest that this panel can detect OPC early, however external validation of this panel is needed. Further refinement of a panel of biomarkers to diagnose OPC earlier is urgently needed to prevent complex treatment of OPC and associated comorbidities, while reducing risk of recurrence.

PREVENTION RELEVANCE

This study identified biomarkers using genome-wide methylation to create a panel capable of discerning early oropharyngeal cancer (OPC) from those without OPC. Such a biomarker panel would be an effective tool to detect OPC early and prevent complications of treatment associated with later diagnosis.

CK1δ and CK1ε Signaling Sustains Mitochondrial Metabolism and Cell Survival in Multiple Myeloma

Multiple myeloma remains an incurable malignancy due to acquisition of intrinsic programs that drive therapy resistance. Here we report that casein kinase-1δ (CK1δ) and CK1ε are therapeutic targets in multiple myeloma that are necessary to sustain mitochondrial metabolism. Specifically, the dual CK1δ/CK1ε inhibitor SR-3029 had potent in vivo and ex vivo anti-multiple myeloma activity, including against primary multiple myeloma patient specimens. RNA sequencing (RNA-seq) and metabolic analyses revealed inhibiting CK1δ/CK1ε disables multiple myeloma metabolism by suppressing genes involved in oxidative phosphorylation (OxPhos), reducing citric acid cycle intermediates, and suppressing complexes I and IV of the electron transport chain. Finally, sensitivity of multiple myeloma patient specimens to SR-3029 correlated with elevated expression of mitochondrial genes, and RNA-seq from 687 multiple myeloma patient samples revealed that increased CSNK1D, CSNK1E, and OxPhos genes correlate with disease progression and inferior outcomes. Thus, increases in mitochondrial metabolism are a hallmark of multiple myeloma progression that can be disabled by targeting CK1δ/CK1ε.

SIGNIFICANCE

CK1δ and CK1ε are attractive therapeutic targets in multiple myeloma whose expression increases with disease progression and connote poor outcomes, and that are necessary to sustain expression of genes directing OxPhos.

Increased spatial coupling of integrin and collagen IV in the immunoresistant clear cell renal cell carcinoma tumor microenvironment

Background

Immunotherapy (IO) has improved survival for patients with advanced clear cell renal cell carcinoma (ccRCC), but resistance to therapy develops in most patients. We use cellular-resolution spatial transcriptomics in patients with IO naïve and IO exposed primary ccRCC tumors to better understand IO resistance. Spatial molecular imaging (SMI) was obtained for tumor and adjacent stroma samples. Spatial gene set enrichment analysis (GSEA) and autocorrelation (coupling with high expression) of ligand-receptor transcript pairs were assessed. Multiplex immunofluorescence (mIF) validation was used for significant autocorrelative findings and the cancer genome atlas (TCGA) and the clinical proteomic tumor analysis consortium (CPTAC) databases were queried to assess bulk RNA expression and proteomic correlates.

Results

21 patient samples underwent SMI. Viable tumors following IO harbored more stromal CD8+ T cells and neutrophils than IO naïve tumors. YES1 was significantly upregulated in IO exposed tumor cells. The epithelial-mesenchymal transition pathway was enriched on spatial GSEA and the associated transcript pair COL4A1 - ITGAV had significantly higher autocorrelation in the stroma. Fibroblasts, tumor cells, and endothelium had the relative highest expression. More integrin αV+ cells were seen in IO exposed stroma on mIF validation. Compared to other cancers in TCGA, ccRCC tumors have the highest expression of both COL4A1 and ITGAV . In CPTAC, collagen IV protein was more abundant in advanced stages of disease.

Conclusions

On spatial transcriptomics, COL4A1 and ITGAV were more autocorrelated in IO-exposed stroma compared to IO-naïve tumors, with high expression amongst fibroblasts, tumor cells, and endothelium. Integrin represents a potential therapeutic target in IO treated ccRCC.

Epigenetic Activation of TUSC3 Sensitizes Glioblastoma to Temozolomide Independent of MGMT Promoter Methylation Status

Temozolomide (TMZ) is an important first-line treatment for glioblastoma (GBM), but there are limitations to TMZ response in terms of durability and dependence on the promoter methylation status of the DNA repair gene O6-methylguanine DNA methyltransferase (MGMT). MGMT-promoter-hypermethylated (MGMT-M) GBMs are more sensitive to TMZ than MGMT-promoter-hypomethylated (MGMT-UM) GBMs. Moreover, TMZ resistance is inevitable even in TMZ-sensitive MGMT-M GBMs. Hence, epigenetic reprogramming strategies are desperately needed in order to enhance TMZ response in both MGMT-M and MGMT-UM GBMs. In this study, we present novel evidence that the epigenetic reactivation of Tumor Suppressor Candidate 3 (TUSC3) can reprogram sensitivity of GBM stem cells (GSCs) to TMZ irrespective of MGMT promoter methylation status. Interrogation of TCGA patient GBM datasets confirmed TUSC3 promoter regulation of TUSC3 expression and also revealed a strong positive correlation between TUSC3 expression and GBM patient survival. Using a combination of loss-of-function, gain-of-function and rescue studies, we demonstrate that TUSC3 reactivation is associated with enhanced TMZ response in both MGMT-M and MGMT-UM GSCs. Further, we provide novel evidence that the demethylating agent 5-Azacitidine (5-Aza) reactivates TUSC3 expression in MGMT-M GSCs, whereas the combination of 5-Aza and MGMT inhibitor Lomeguatrib is necessary for TUSC3 reactivation in MGMT-UM GSCs. Lastly, we propose a pharmacological epigenetic reactivation strategy involving TUSC3 that leads to significantly prolonged survival in MGMT-M and MGMT-UM orthotopic GSCs models. Collectively, our findings provide a framework and rationale to further explore TUSC3-mediated epigenetic reprogramming strategies that could enhance TMZ sensitivity and outcomes in GBM. Mechanistic and translational evidence gained from such studies could contribute towards optimal design of impactful trials for MGMT-UM GBMs that currently do not have good treatment options.

The Polyamine-Hypusine Circuit Controls an Oncogenic Translational Program Essential for Malignant Conversion in MYC-Driven Lymphoma

The MYC oncoprotein is activated in a broad spectrum of human malignancies and transcriptionally reprograms the genome to drive cancer cell growth. Given this, it is unclear if targeting a single effector of MYC will have therapeutic benefit. MYC activates the polyamine-hypusine circuit, which posttranslationally modifies the eukaryotic translation factor eIF5A. The roles of this circuit in cancer are unclear. Here we report essential intrinsic roles for hypusinated eIF5A in the development and maintenance of MYC-driven lymphoma, where the loss of eIF5A hypusination abolishes malignant transformation of MYC-overexpressing B cells. Mechanistically, integrating RNA sequencing, ribosome sequencing, and proteomic analyses revealed that efficient translation of select targets is dependent upon eIF5A hypusination, including regulators of G1-S phase cell-cycle progression and DNA replication. This circuit thus controls MYC's proliferative response, and it is also activated across multiple malignancies. These findings suggest the hypusine circuit as a therapeutic target for several human tumor types.

SIGNIFICANCE

Elevated EIF5A and the polyamine-hypusine circuit are manifest in many malignancies, including MYC-driven tumors, and eIF5A hypusination is necessary for MYC proliferative signaling. Not-ably, this circuit controls an oncogenic translational program essential for the development and maintenance of MYC-driven lymphoma, supporting this axis as a target for cancer prevention and treatment. See related commentary by Wilson and Klein, p. 248. This article is highlighted in the In This Issue feature, p. 247.

Complimentary genomic, pathologic, and artificial intelligence analysis on low-grade noninvasive bladder cancer to predict downstream recurrence

553

Background: Low-grade noninvasive (LGTa) bladder cancer is a relatively quiescent but heterogenous malignancy, characterized by downstream recurrences requiring repeated transurethral resections and frequent surveillance. Investigations to elucidate drivers of recurrence have been sparse, but will help risk-stratify patients with LGTa and allow augmentation of follow up protocols. Methods: Patients with LGTa index tumors were stratified by those with no downstream recurrences (nonrecurrent) vs. those with later recurrences (recurrent). RNA sequencing identified differentially expressed genes (DEGs), deconvoluted for cell-type using xCell. Pathologic analysis was performed by a genitourinary pathologist, then a deep-learning artificial intelligence (AI) platform was leveraged to correlate recurrence risk and recurrence-free survival (RFS) based on deep-learning algorithm of segmented nuclei. Results: Thirty index bladder tumors/patients were identified, 18 (60%) of which had later recurrence (Table). There were 238 DEGs recognized, with recurrent tumors expressing signatures for epithelial mesenchymal transition, myogenesis, TNFα signaling via NFκB, and angiogenesis. Recurrent tumors also demonstrated a higher tissue micoenvironment, stroma, and cancer-associated fibroblast score. Pathologic TME analysis validated these findings, with recurrent tumors demonstrating a higher frequency of inverted growth pattern and a higher median stroma percentage. Finally, the AI-derived signature was predictive of recurrence and risk-stratified the cohort (HR= 5.43 [95% CI 1.1-26.76]) for predicting high vs. low risk of recurrence. Patients in the high risk group had a 87.5% recurrence rate while those in the low risk group had a 28.5% recurrence rate (p<0.01). Conclusions: Using a multi-disciplinary approach, we identified key signatures in recurrent LGTa bladder cancer. Characterization of these factors is a critical first step in the risk-stratification of LGTa tumors, and may allow risk-stratification of surveillance protocols and identification of possible targets for chemoprevention trials. [Table: see text]

Deregulated expression of the 14q32 miRNA cluster in clear cell renal cancer cells

Clear cell renal cell carcinomas (ccRCC) are characterized by arm-wide chromosomal alterations. Loss at 14q is associated with disease aggressiveness in ccRCC, which responds poorly to chemotherapeutics. The 14q locus contains one of the largest miRNA clusters in the human genome; however, little is known about the contribution of these miRNAs to ccRCC pathogenesis. In this regard, we investigated the expression pattern of selected miRNAs at the 14q32 locus in TCGA kidney tumors and in ccRCC cell lines. We demonstrated that the miRNA cluster is downregulated in ccRCC (and cell lines) as well as in papillary kidney tumors relative to normal kidney tissues (and primary renal proximal tubule epithelial (RPTEC) cells). We demonstrated that agents modulating expression of DNMT1 (e.g., 5-Aza-deoxycytidine) could modulate 14q32 miRNA expression in ccRCC cell lines. Lysophosphatidic acid (LPA, a lysophospholipid mediator elevated in ccRCC) not only increased labile iron content but also modulated expression of a 14q32 miRNA. Through an overexpression approach targeting a subset of 14q32 miRNAs (specifically at subcluster A: miR-431-5p, miR-432-5p, miR-127-3p, and miR-433-3p) in 769-P cells, we uncovered changes in cellular viability and claudin-1, a tight junction marker. A global proteomic approach was implemented using these miRNA overexpressing cell lines which uncovered ATXN2 as a highly downregulated target. Collectively, these findings support a contribution of miRNAs at 14q32 in ccRCC pathogenesis.

A Targetable Myeloid Inflammatory State Governs Disease Recurrence in Clear-Cell Renal Cell Carcinoma

It is poorly understood how the tumor immune microenvironment influences disease recurrence in localized clear-cell renal cell carcinoma (ccRCC). Here we performed whole-transcriptomic profiling of 236 tumors from patients assigned to the placebo-only arm of a randomized, adjuvant clinical trial for high-risk localized ccRCC. Unbiased pathway analysis identified myeloid-derived IL6 as a key mediator. Furthermore, a novel myeloid gene signature strongly correlated with disease recurrence and overall survival on uni- and multivariate analyses and is linked to TP53 inactivation across multiple data sets. Strikingly, effector T-cell gene signatures, infiltration patterns, and exhaustion markers were not associated with disease recurrence. Targeting immunosuppressive myeloid inflammation with an adenosine A2A receptor antagonist in a novel, immunocompetent, Tp53-inactivated mouse model significantly reduced metastatic development. Our findings suggest that myeloid inflammation promotes disease recurrence in ccRCC and is targetable as well as provide a potential biomarker-based framework for the design of future immuno-oncology trials in ccRCC.

SIGNIFICANCE

Improved understanding of factors that influence metastatic development in localized ccRCC is greatly needed to aid accurate prediction of disease recurrence, clinical decision-making, and future adjuvant clinical trial design. Our analysis implicates intratumoral myeloid inflammation as a key driver of metastasis in patients and a novel immunocompetent mouse model. This article is highlighted in the In This Issue feature, p. 2221.

TCR-Independent Metabolic Reprogramming Precedes Lymphoma-Driven Changes in T-cell Fate

Chronic T-cell receptor (TCR) signaling in the tumor microenvironment is known to promote T-cell dysfunction. However, we reasoned that poorly immunogenic tumors may also compromise T cells by impairing their metabolism. To address this, we assessed temporal changes in T-cell metabolism, fate, and function in models of B-cell lymphoma driven by Myc, a promoter of energetics and repressor of immunogenicity. Increases in lymphoma burden most significantly impaired CD4+ T-cell function and promoted regulatory T cell (Treg) and Th1-cell differentiation. Metabolomic analyses revealed early reprogramming of CD4+ T-cell metabolism, reduced glucose uptake, and impaired mitochondrial function, which preceded changes in T-cell fate. In contrast, B-cell lymphoma metabolism remained robust during tumor progression. Finally, mitochondrial functions were impaired in CD4+ and CD8+ T cells in lymphoma-transplanted OT-II and OT-I transgenic mice, respectively. These findings support a model, whereby early, TCR-independent, metabolic interactions with developing lymphomas limits T cell-mediated immune surveillance.

Proteogenomic, Epigenetic, and Clinical Implications of Recurrent Aberrant Splice Variants in Clear Cell Renal Cell Carcinoma

BACKGROUND

Alternative mRNA splicing can be dysregulated in cancer, resulting in the generation of aberrant splice variants (SVs). Given the paucity of actionable genomic mutations in clear cell renal cell carcinoma (ccRCC), aberrant SVs may be an avenue to novel mechanisms of pathogenesis.

OBJECTIVE

To identify and characterize aberrant SVs enriched in ccRCC.

DESIGN, SETTING, AND PARTICIPANTS

Using RNA-seq data from the Cancer Cell Line Encyclopedia, we identified neojunctions uniquely expressed in ccRCC. Candidate SVs were then checked for expression across normal tissue in the Genotype-Tissue Expression Project and primary tumor tissue from The Cancer Genome Atlas (TCGA), Clinical Proteomic Tumor Analysis Consortium (CPTAC), and our institutional Total Cancer Care database.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Clinicopathologic, genomic, and survival data were available for all cohorts. Epigenetic data were available for the TCGA and CPTAC cohorts. Proteomic data were available for the CPTAC cohort. The association of aberrant SV expression with these variables was examined using the Kruskal-Wallis test, pairwise t test, Spearman correlation test, and Cox regression analysis.

RESULTS AND LIMITATIONS

Our pipeline identified 16 ccRCC-enriched SVs. EGFR, HPCAL1-SV and RNASET2-SV expression was negatively correlated with gene-specific CpG methylation. We derived a survival risk score based primarily on the expression of five SVs (RNASET2, FGD1, PDZD2, COBLL1, and PTPN14), which was consistent and applicable across multiple cohorts on multivariate analysis. The splicing factor RBM4, which modulates splicing of HIF-1α, exhibited significantly lower expression at the protein level in the high-risk group, as defined by our SV-based score.

CONCLUSIONS

We describe 16 aberrant SVs enriched in ccRCC, many of which are associated with disease biology and/or clinical outcomes. This study provides a novel strategy for identifying and characterizing disease-specific aberrant SVs.

PATIENT SUMMARY

We describe a method to identify disease targets and biomarkers using transcriptomic analysis beyond somatic mutations or gene expression. Kidney tumors express unique splice variants that may provide additional prognostic information following surgery.

Ovarian cancer: Age, prognosis, and biologic underpinnings

e17555

Background: The prognosis of ovarian cancer worsens markedly with age. Yet few data are available to explain this phenomenon. Two hypotheses can be made. Changes in pharmacokinetics (PK) and treatment tolerance limit the dosing of chemotherapy, and/or changes in tumor and host-tumor biology limit treatment efficacy. Objective:To identify PK, body composition, and biologic changes with age that may impact prognosis. We reported PK findings in a previous abstract (Extermann et al., ASCO 2021) In this abstract we focus on the biologic findings. Methods: We conducted a prospective cohort study of women with stage III/IV high-grade serous ovarian cancers treated with neoadjuvant intravenous carboplatin (C) AUC 5 every 3 weeks and paclitaxel (P) dosed either weekly at 80 mg/m2 or every 3 weeks at 175 mg/m2. Body composition was assessed using baseline CT scans at the L3 level. Pre- and post-chemotherapy plasma samples were assayed by ELISA for CRP, IL-6, TNF alpha, TNF-receptor 1 and 2, d-dimers, vitamins B12 and D, methylmalonic acid, and cortisol levels. Tissue samples pre- and post-chemotherapy were analyzed for gene expression, gene methylation, and immunohistochemistry for inflammatory markers and senescence. Toxicity was assessed as first cycle nadir ANC and G4 hematologic (H) or grade 3-4 non-hematologic (NH) toxicity by CTCAE 4.0 criteria over a 3 cycles follow-up. The correlation between continuous measures was assessed by Spearman correlation coefficient, and the association with toxicity was evaluated by Wilcoxon rank sum test. The association with relapse-free survival (RFS) and overall survival (OS) was calculated by Log-rank and Cox regression analyzes. Results: Seventeen patients, ages 38 to 86 (median 61) were eligible, 14 underwent surgery. Treatment delivery was high with a median relative dose intensity of 100%. The mean percentage of senescent cells (CAV-1 +) was 49.5% in the stroma, vs 12.8% in the tumor (p < 0.001) prechemotherapy, and 15.7% in the stroma vs 21.4% in the tumor post-chemotherapy (pre-post stroma p < 0.01). The percentage of PD-L1 positive cells and CTLA-4 positive cells was very low: mean < 1% both pre-and post-chemotherapy. The results from gene expression and methylation showed an increase in the inflammatory pathways and a decrease in the apoptosis pathways over time. Vitamin D levels decreased with age. Prechemotherapy CD8 percentage in the stroma and post chemotherapy d-dimers were associated with RFS. Conclusions: Neoadjuvant treatment delivery was high in an academic setting. Our results provide insights in the underexplored area of biologic response to chemotherapy in patients with ovarian cancer. Although our results did not provide clear candidates to explain the worsening of prognosis with age, they offer clues to develop future research on this issue.

Proteogenomic and clinical implications of unique recurrent splice variants in clear cell renal cell carcinoma

380

Background: Alternative mRNA splicing is recognized as a key driver of proteomic diversity. In cancer, this splicing process can be altered resulting in generation of aberrant splice variants (SvPs) that can contribute to tumor pathogenesis. However, our understanding of the significance of splice variants in clear cell renal cell carcinoma (ccRCC) is currently limited. Given the lack of actionable genomic mutations in ccRCC, aberrant SpVs may be the avenue to new pathogenic mechanisms and biomarkers. Methods: We implemented a novel pipeline to screen for and select SpVs frequent in and relatively specific to ccRCC. We started with RNA-seq data from the Cancer Cell Line Encyclopedia to identify SpVs specific to ccRCC cell lines. These were then screened across normal tissue in the Genotype-Tissue Expression Project (GTEx) and excluded if expressed. We analyzed bulk RNA-seq data of ccRCC primary tumors obtained from our institutional Total Cancer Care cohort (TCC; n = 111), The Cancer Genome Atlas (TCGA; n = 484) and the Clinical Proteomic Tumor Analysis Consortium (CPTAC; n = 110) to analyze SpV expression in these samples. Using raw proteomics files from the CPTAC portal, proteins were identified and quantified using MaxQuant. Associations of SvP with protein expression were filtered by a Spearman correlation cutoff of +/-0.3. The Enrichr R library was used for pathway enrichment. Finally, we correlated splice variant expression with overall (OS) and cancer-specific survival (CSS). Using LASSO Cox regression analysis, we derived a SpV-based risk score trained on OS from the TCGA cohort and validated on the TCC and CPTAC cohorts. Results: Our pipeline selected 16 previously uncharacterized SpVs, including variants of suspected oncogenes and tumor suppressors. Proteogenomic analysis identified interesting biological associations. Among patients with high levels of EGFR SpV, we found significantly higher expression of the protein regulatory T cell marker CD70 (padj = 0.03). MVK SvP was highly correlated with 25 proteins enriched for the mTOR pathway (padj = 0.002). We derived a survival risk score based on expression of 5 SpVs ( PDZD2, COBLL1, PTPN14, RNASET2, FGD1) in the TCGA cohort. This risk score remained significant on multivariate analysis (HR 1.4, p = 0.002) adjusting for covariates including AJCC stage. This was validated on multivariate analysis in the TCC (HR 3.56, p < 0.001) and CPTAC (HR 3.18, p = 0.019) cohorts. Conclusions: Our novel pipeline selected 16 unique SpVs frequent in and relatively specific for ccRCC. Some are associated with proteins expressed in oncogenic pathways, suggesting a potential role in disease pathogenesis. Additionally, our SpV-based risk score is strongly associated with OS and CSS across multiple cohorts. This study provides a template for identifying and characterizing disease-specific aberrant SpVs to aid discovery of new mechanisms and biomarkers.

Disrupting the MYC-TFEB Circuit Impairs Amino Acid Homeostasis and Provokes Metabolic Anergy

MYC family oncoproteins are regulators of metabolic reprogramming that sustains cancer cell anabolism. Normal cells adapt to nutrient-limiting conditions by activating autophagy, which is required for amino acid (AA) homeostasis. Here we report that the autophagy pathway is suppressed by Myc in normal B cells, in premalignant and neoplastic B cells of Eμ-Myc transgenic mice, and in human MYC-driven Burkitt lymphoma. Myc suppresses autophagy by antagonizing the expression and function of transcription factor EB (TFEB), a master regulator of autophagy. Mechanisms that sustained AA pools in MYC-expressing B cells include coordinated induction of the proteasome and increases in AA transport. Reactivation of the autophagy-lysosomal pathway by TFEB disabled the malignant state by disrupting mitochondrial functions, proteasome activity, amino acid transport, and amino acid and nucleotide metabolism, leading to metabolic anergy, growth arrest and apoptosis. This phenotype provides therapeutic opportunities to disable MYC-driven malignancies, including AA restriction and treatment with proteasome inhibitors.

A Novel Role of BIRC3 in Stemness Reprogramming of Glioblastoma

Stemness reprogramming remains a largely unaddressed principal cause of lethality in glioblastoma (GBM). It is therefore of utmost importance to identify and target mechanisms that are essential for GBM stemness and self-renewal. Previously, we implicated BIRC3 as an essential mediator of therapeutic resistance and survival adaptation in GBM. In this study, we present novel evidence that BIRC3 has an essential noncanonical role in GBM self-renewal and stemness reprogramming. We demonstrate that BIRC3 drives stemness reprogramming of human GBM cell lines, mouse GBM cell lines and patient-derived GBM stem cells (GSCs) through regulation of BMP4 signaling axis. Specifically, BIRC3 induces stemness reprogramming in GBM through downstream inactivation of BMP4 signaling. RNA-Seq interrogation of the stemness reprogramming hypoxic (pseudopalisading necrosis and perinecrosis) niche in GBM patient tissues further validated the high BIRC3/low BMP4 expression correlation. BIRC3 knockout upregulated BMP4 expression and prevented stemness reprogramming of GBM models. Furthermore, siRNA silencing of BMP4 restored stemness reprogramming of BIRC3 knockout in GBM models. In vivo silencing of BIRC3 suppressed tumor initiation and progression in GBM orthotopic intracranial xenografts. The stemness reprograming of both GSCs and non-GSCs populations highlights the impact of BIRC3 on intra-tumoral cellular heterogeneity GBM. Our study has identified a novel function of BIRC3 that can be targeted to reverse stemness programming of GBM.

Abstract PO-090: TGF-β induced EMT gene expression is associated with promoter demethylation in pancreatic cancer

INTRODUCTION: Transforming growth factor-β (TGF-β), a ubiquitous molecule in pancreatic ductal adenocarcinoma (PDA) tumors, acts as a potent inducer of tumor invasion by regulating proteins involved in metastasis and driving epithelial-mesenchymal transition (EMT). We hypothesized that TGF-β signaling-induced EMT is regulated by DNA methylation. To evaluate this, we investigated the association between EMT gene promoter methylation status and gene expression in the TCGA PDA dataset. We also investigated vimentin gene (a mesenchymal marker)-specific changes in DNA methylation due to TGF-β treatment in PDA cells. METHODS: The methylation status of EMT-related genes (ZEB2, CDH2, Vimentin) was interrogated in the TCGA PDA data set. β-values (proportion of methylated gene probes in a location) were compared to gene expression levels. Next, a PDA cell line (Panc-1) was treated with TGF-β (10ng/ml), vehicle control, DNA methyltransferase inhibitor (5-azacytidine, AZA, 10mM), or TGF-β plus AZA to evaluate the effects on the DNA methylation status of the vimentin gene with methylation-specific PCR (MSP) against the promoter region of vimentin. Unmethylation and methylation levels at the vimentin promoter region were then compared. RESULTS: We found that β values of promotor regions of ZEB2, CDH2, and Vimentin were moderately negatively correlated with gene expression (Pearson r varies from 0.45 to 0.63) in the PDA TCGA data set. In Panc-1 cells, vimentin gene expression was increased following treatment with AZA compared to controls suggesting that DNA demethylation increases vimentin gene expression. MSP demonstrated that TGF-β treatment increased unmethylated vimentin gene levels compared to vehicle control treatment. TGF-β treatment increased unmethylated vimentin gene levels more so than methylated gene levels. CONCLUSIONS: These results demonstrate that gene expression of EMT-related genes may be at least partially regulated by DNA methylation. We demonstrate that TGF-β treatment leads to increased Vimentin promotor demethylation (lower β value) and increased gene expression in Panc-1 cells. This observation is consistent with our findings in the TCGA data set that lower β-values are associated with increased expression of mesenchymal genes, indicative of EMT. Further studies are underway to investigate the relationship between DNA methylation and TGF-β-induced EMT in general.

Citation Format: Manjul Rana, Abul Elahi, Abidemi O. Ajidahun, Rita G. Kansal, Anders E. Berglund, David Shibata, Evan S. Glazer. TGF-β induced EMT gene expression is associated with promoter demethylation in pancreatic cancer [abstract]. In: Proceedings of the AACR Virtual Special Conference on Pancreatic Cancer; 2021 Sep 29-30. Philadelphia (PA): AACR; Cancer Res 2021;81(22 Suppl):Abstract nr PO-090.

The Impact of Epigenetic Modifications on Adaptive Resistance Evolution in Glioblastoma

Glioblastoma (GBM) is a highly lethal cancer that is universally refractory to the standard multimodal therapies of surgical resection, radiation, and chemotherapy treatment. Temozolomide (TMZ) is currently the best chemotherapy agent for GBM, but the durability of response is epigenetically dependent and often short-lived secondary to tumor resistance. Therapies that can provide synergy to chemoradiation are desperately needed in GBM. There is accumulating evidence that adaptive resistance evolution in GBM is facilitated through treatment-induced epigenetic modifications. Epigenetic alterations of DNA methylation, histone modifications, and chromatin remodeling have all been implicated as mechanisms that enhance accessibility for transcriptional activation of genes that play critical roles in GBM resistance and lethality. Hence, understanding and targeting epigenetic modifications associated with GBM resistance is of utmost priority. In this review, we summarize the latest updates on the impact of epigenetic modifications on adaptive resistance evolution in GBM to therapy.

Abstract 1635: Sirt2 blockade promotes T cell metabolism and restores the anti-tumor immunity

Background: The majority of cancer patients remain refractory to existing cancer immunotherapies. Despite the growing evidence that dysregulated metabolism contributes to the exhaustion of tumor-infiltrating T lymphocytes (TILs) and the loss of their effector functions within the metabolically restricted tumor microenvironment (TME), actionable targets to rescue metabolic fitness and anti-tumor activity of TILs remain elusive. Sirt2 is an NAD+ dependent histone deacetylase and conflicting evidences suggest its tumor-suppressor and oncogenic roles. In inflammatory response, Sirt2 suppresses inflammation via negative regulation of NF-κB subunit; however, the role of Sirt2 in tumor immunity has not been described.

Method: Human TILs from non-small cell lung cancer (NSCLC) tumor samples and matched peripheral blood T cells were analyzed for Sirt2 expression by flow cytometry. The role of Sirt2 in anti-tumor immunity was studied by in vivo B16F10 tumor challenge in Wild-type (Wt) and Sirt2 knockout (Sirt2KO) mice. The role of Sirt2 in T cell effector functions was investigated ex vivo by CFSE proliferation assay, IFN-γ ELISpot assay, intracellular staining of effector markers and LDH cytotoxicity assay on Wt versus Sirt2KO T cells. Sirt2 targets were identified using mass spectrometry (MS) and Co-immunoprecipitation analyses. T cells metabolic changes were investigated using seahorse bioanalyzer and LC-MS/MS Metabolomic profiling. Sirt2 blockade in human T cells was performed using AGK2, a Sirt2 selective inhibitor.

Result: We show that Sirt2 expression is upregulated within the TME, and its upregulation in human TILs is associated with a poor clinical response to immunotherapy in a phase I clinical trial in advanced NSCLC. We also show that, Sirt2 deficiency in mice boosts T cell effector functions and tumor rejection in vivo. Our molecular and metabolomic studies revealed multiple metabolic pathways as Sirt2 targets including glycolysis, TCA-cycle, FAO and glutaminolysis. We found that Sirt2 deacetylase deficiency increased acetylation and enzymatic activity of key metabolic enzymes leading to a hyper metabolic status of T cells. Finally pharmacologic inhibition of Sirt2 in human TILs isolated from NSCLC patients enhances their metabolic fitness and effector functions.

Conclusion Our findings indicate Sirt2 as a master suppressor of T cell metabolism amenable to therapeutic targeting and Sirt2 inhibition reprograms T cell metabolic fitness to optimally sustain their effector function within the metabolically challenging TME, thus, leading to an effective anti-tumor immune response.

Citation Format: Imene Hamaidi, Lin Zhang, Nayoung Kim, Min Hsuan Wang, Cristina Iclozan, Bin Fang, Min Liu, John M. Koomen, Anders E. Berglund, Sean J. Yoder, Jiqiang Yao, Robert W. Engelman, Ben C. Creelan, Jose R. Conejo-Garcia, James J. Mulé, Scott J. Antonia, Sungjune Kim. Sirt2 blockade promotes T cell metabolism and restores the anti-tumor immunity [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 1635.

KLK3 SNP-SNP interactions for prediction of prostate cancer aggressiveness

Risk classification for prostate cancer (PCa) aggressiveness and underlying mechanisms remain inadequate. Interactions between single nucleotide polymorphisms (SNPs) may provide a solution to fill these gaps. To identify SNP-SNP interactions in the four pathways (the angiogenesis-, mitochondria-, miRNA-, and androgen metabolism-related pathways) associated with PCa aggressiveness, we tested 8587 SNPs for 20,729 cases from the PCa consortium. We identified 3 KLK3 SNPs, and 1083 (P < 3.5 × 10-9) and 3145 (P < 1 × 10-5) SNP-SNP interaction pairs significantly associated with PCa aggressiveness. These SNP pairs associated with PCa aggressiveness were more significant than each of their constituent SNP individual effects. The majority (98.6%) of the 3145 pairs involved KLK3. The 3 most common gene-gene interactions were KLK3-COL4A1:COL4A2, KLK3-CDH13, and KLK3-TGFBR3. Predictions from the SNP interaction-based polygenic risk score based on 24 SNP pairs are promising. The prevalence of PCa aggressiveness was 49.8%, 21.9%, and 7.0% for the PCa cases from our cohort with the top 1%, middle 50%, and bottom 1% risk profiles. Potential biological functions of the identified KLK3 SNP-SNP interactions were supported by gene expression and protein-protein interaction results. Our findings suggest KLK3 SNP interactions may play an important role in PCa aggressiveness.

TFEB links MYC signaling to epigenetic control of myeloid differentiation and acute myeloid leukemia

MYC oncoproteins regulate transcription of genes directing cell proliferation, metabolism and tumorigenesis. A variety of alterations drive MYC expression in acute myeloid leukemia (AML) and enforced MYC expression in hematopoietic progenitors is sufficient to induce AML. Here we report that AML and myeloid progenitor cell growth and survival rely on MYC-directed suppression of Transcription Factor EB (TFEB), a master regulator of the autophagy-lysosome pathway. Notably, although originally identified as an oncogene, TFEB functions as a tumor suppressor in AML, where it provokes AML cell differentiation and death. These responses reflect TFEB control of myeloid epigenetic programs, by inducing expression of isocitrate dehydrogenase-1 (IDH1) and IDH2, resulting in global hydroxylation of 5-methycytosine. Finally, activating the TFEB-IDH1/IDH2-TET2 axis is revealed as a targetable vulnerability in AML. Thus, epigenetic control by a MYC-TFEB circuit dictates myeloid cell fate and is essential for maintenance of AML.

The prognostic and predictive implications of the 12-chemokine score in muscle invasive bladder cancer

466

Background: Adaptive anti-tumor immunity can be orchestrated by lymph node-like immune cell aggregates within the tumor microenvironment (TME) called tertiary lymphoid structures (TLSs). TLSs are postulated to be the gateway of lymphocyte infiltration into the TME, and are privileged sites for coordinated tumor antigen presentation and lymphocyte priming, differentiation, and proliferation, leading to a robust tumor-specific immune response. A 12-chemokine metagene grouping (12-CK score) has previously been described that correlates with the presence of TLSs in other solid tumor types. In this study, we explored the prognostic implication of the 12-CK score in bladder cancer and its correlation with the presence of TLSs. Methods: Cystectomy specimens from 132 patients with bladder cancer were arrayed on Affymetrix microarrays. 12-CK scores were normalized with > 1 denoting high scores (12-CKHi). Immunohistochemistry (IHC) antibody staining was performed for DC-LAMP, CD20, CD4, and CD8. A GU pathologist scored TLSs into Types I-III, with type III representing fully developed TLSs. The Fisher’s exact test was used to test the associations between the 12-CK scores and the type of lymphoid aggregate. Overall survival was estimated using the Kaplan Meier method. Findings were validated using 12-CK scores extracted from TCGA transcriptome sequencing data and the IMvigor210CoreBiologies package. Results: Twenty-five (n = 25) patients had 12CK scores > 1 and were classified as 12CK-High. Pathologic review of 43 bladder tumor specimens confirmed higher levels of Type III TLS patients (33% vs. 9%, p = 0.03), B cells (p = 0.002), CD8 T cells (p = 0.01), and activated DC (p = 0.01) in 12-CKHi compared to 12-CKLo. 12-CKHi was found to have a progression-free survival (PFS, HR 0.29, p = 0.003, Fig1a), disease specific survival (DSS, HR 0.29, p = 0.004, Fig1b), and overall survival (OS, HR 0.55, p = 0.03, fig1c) advantage compared to 12-CKLo in the Moffitt patient cohort. These results were validated using the publically available RNA expression data from TCGA. TCGA patients with 12-CKHi (18%,n = 72) had improved PFS ( HR 0.55, p = 0.007, fig1d), DSS (HR = 0.40, p = 0.002, fig1e), and 0S (HR = 0.59, p = 0.01, fig1f). From the IMVIGOR-210 patient who were 12-CKHi were more likely to have a complete response (p < 0.05, fig1g) and have a 11.2mo OS benefit (fig1h) after treatment using atezolizumab. Conclusions: Three important findings emerged from the current study: 12CK-High scores corresponded with formation of TLS in the TME; favorable prognosis in surgically treated MIBC patients; and CR in atezolizumab-treated patients. The findings herein suggest the 12CK gene signature to be a clinically actable biomarker for predicting response to immune checkpoint blockade. We believe the 12CK signature may serve as an important tool to refine patient selection for immune checkpoint blockade treatment.

Alcohol Intake and Alcohol-SNP Interactions Associated with Prostate Cancer Aggressiveness

Excessive alcohol intake is a well-known modifiable risk factor for many cancers. It is still unclear whether genetic variants or single nucleotide polymorphisms (SNPs) can modify alcohol intake's impact on prostate cancer (PCa) aggressiveness. The objective is to test the alcohol-SNP interactions of the 7501 SNPs in the four pathways (angiogenesis, mitochondria, miRNA, and androgen metabolism-related pathways) associated with PCa aggressiveness. We evaluated the impacts of three excessive alcohol intake behaviors in 3306 PCa patients with European ancestry from the PCa Consortium. We tested the alcohol-SNP interactions using logistic models with the discovery-validation study design. All three excessive alcohol intake behaviors were not significantly associated with PCa aggressiveness. However, the interactions of excessive alcohol intake and three SNPs (rs13107662 [CAMK2D, p = 6.2 × 10-6], rs9907521 [PRKCA, p = 7.1 × 10-5], and rs11925452 [ROBO1, p = 8.2 × 10-4]) were significantly associated with PCa aggressiveness. These alcohol-SNP interactions revealed contrasting effects of excessive alcohol intake on PCa aggressiveness according to the genotypes in the identified SNPs. We identified PCa patients with the rs13107662 (CAMK2D) AA genotype, the rs11925452 (ROBO1) AA genotype, and the rs9907521 (PRKCA) AG genotype were more vulnerable to excessive alcohol intake for developing aggressive PCa. Our findings support that the impact of excessive alcohol intake on PCa aggressiveness was varied by the selected genetic profiles.

TMPRSS2-ERG fusion impacts anterior tumor location in men with prostate cancer

BACKGROUND

In prostate cancer (PCa), lack of androgen receptor (AR) regulated TMPRSS2-ETS-related gene (ERG) gene fusion (ERG^negative ) status has been associated with African American race; however, the implications of ERG status for the location of dominant tumors within the prostate remains understudied.

METHODS

An African American-enriched multiinstitutional cohort of 726 PCa patients consisting of both African American men (AAM; n = 254) and European American men (EAM; n = 472) was used in the analyses. Methods of categorical analysis were used. Messenger RNA (mRNA) expression differences between anterior and posterior tumor lesions were analyzed using Wilcoxon rank-sum tests with multiple comparison corrections.

RESULTS

Anti-ERG immunohistochemistry staining showed that the association between ERG status and anterior tumors is independent of race and is consistently robust for both AAM (ERG^negative 81.4% vs. ERG^positive 18.6%; p = .005) and EAM (ERG^negative 60.4% vs. ERG^positive 39.6%; p < .001). In a multivariable model, anterior tumors were more likely to be IHC-ERG^negative (odds ratio [OR]: 3.20; 95% confidence interval [CI]: 2.14-4.78; p < .001). IHC-ERG^negative were also more likely to have high-grade tumors (OR: 1.73; 95% CI: 1.06-2.82; p = .02). In the exploratory genomic analysis, mRNA expression of location-dependent genes is highly influenced by ERG status and African American race. However, tumor location did not impact the expression of AR or the major canonical AR-target genes (KLK3, AMACR, and MYC).

CONCLUSIONS

ERG^negative tumor status is the strongest predictor of anterior prostate tumors, regardless of race. Furthermore, AR expression and canonical AR signaling do not impact tumor location.

Accessible Pipeline for Translational Research Using TCGA: Examples of Relating Gene Mechanism to Disease-Specific Outcomes

Bioinformatic scientists are often asked to do widespread analyses of publicly available datasets in order to identify genetic alterations in cancer for genes of interest; therefore, we sought to create a set of tools to conduct common statistical analyses of The Cancer Genome Atlas (TCGA) data. These tools have been developed in response to requests from our collaborators to ask questions, validate findings, and better understand the function of their gene of interest. We describe here what data we have used, how to obtain it, and what figures we have found useful.

Anti-tumor efficacy of plasmid encoding emm55 in a murine melanoma model

Emm55 is a bacterial gene derived from Streptococcus pyogenes (S. pyogenes) that was cloned into a plasmid DNA vaccine (pAc/emm55). In this study, we investigated the anti-tumor efficacy of pAc/emm55 in a B16 murine melanoma model. Intralesional (IL) injections of pAc/emm55 significantly delayed tumor growth compared to the pAc/Empty group. There was a significant increase in the CD8+ T cells infiltrating into the tumors after pAc/emm55 treatment compared to the control group. In addition, we observed that IL injection of pAc/emm55 increased antigen-specific T cell infiltration into tumors. Depletion of CD4+ or CD8+ T cells abrogated the anti-tumor effect of pAc/emm55. Combination treatment of IL injection of pAc/emm55 with anti-PD-1 antibody significantly delayed tumor growth compared to either monotherapy. pAc/emm55 treatment combined with PD-1 blockade enhanced anti-tumor immune response and improved systemic anti-tumor immunity. Together, these strategies may lead to improvements in the treatment of patients with melanoma.

Sirt2 Inhibition Enhances Metabolic Fitness and Effector Functions of Tumor-Reactive T Cells

Dysregulated metabolism is a key driver of maladaptive tumor-reactive T lymphocytes within the tumor microenvironment. Actionable targets that rescue the effector activity of antitumor T cells remain elusive. Here, we report that the Sirtuin-2 (Sirt2) NAD⁺-dependent deacetylase inhibits T cell metabolism and impairs T cell effector functions. Remarkably, upregulation of Sirt2 in human tumor-infiltrating lymphocytes (TILs) negatively correlates with response to TIL therapy in advanced non-small-cell lung cancer. Mechanistically, Sirt2 suppresses T cell metabolism by targeting key enzymes involved in glycolysis, tricarboxylic acid-cycle, fatty acid oxidation, and glutaminolysis. Accordingly, Sirt2-deficient murine T cells exhibit increased glycolysis and oxidative phosphorylation, resulting in enhanced proliferation and effector functions and subsequently exhibiting superior antitumor activity. Importantly, pharmacologic inhibition of Sirt2 endows human TILs with these superior metabolic fitness and effector functions. Our findings unveil Sirt2 as an unexpected actionable target for reprogramming T cell metabolism to augment a broad spectrum of cancer immunotherapies.

Abstract 3520: Immunogenomic landscape of grade group 5 prostate cancer predicts risk of lethal outcomes and may inform treatment response

Introduction

Prognostic and therapeutic implications of the immunogenomic landscape in poorly differentiated Grade Group 5 (GG5) prostate cancer (PCa) remains largely unknown. While novel treatment strategies including immuno-radiotherapy have been hypothesized to attenuate disease progression in advanced PCa, limited data exist to show underlying mechanisms related to immunogenomcis and its role in the management of GG5 PCa.

Method

This retrospective analysis includes 822 PCa patients who underwent radical prostatectomy (RP) with pathologic GG5 disease. Immunogenomic interaction between the immune content score (ICS) and Decipher score was used to predict the metastasis and Prostate cancer specific mortality (PCSM). Immunogenomic interaction categories were used to assess patient response to radiation therapy, measured by the PORTOS score. Additionally, 135 biopsy based tumor samples were also used to validate the immunogenomic correlation between ICS and Decipher as well as the association with PORTOS score. A Kruskal-Wallis test was used to compare median scores whereas methods of survival analysis were used for time dependent endpoints. Principal component analysis (PCA) was used to show sample clustering based on PORTOS score.

Results

Among GG5 patients 64.8% had high Decipher risk score (&gt; 0.60) whereas 35.5% had low-average risk score (≤ 0.60). When comparing Decipher risk categories with ICS, a linear association was observed with the highest ICS quartile showing the highest Decipher score (Spearman's correlation 0.23, p &lt; 2.4e−11). In a multivariable model, high Decipher score was associated with risk metastasis (HR = 1.63, 95% CI: 1.12 - 2.37, p = 0.01) and PCSM (HR = 1.61, 95% CI: 0.93 – 2.78, p = 0.08) whereas high ICS did not show significant effect on lethal outcomes. However, in the interaction model, effect of decipher on lethal outcomes was augmented in relation with higher ICS. Patients with ICSHigh and DecipherHigh showed significantly higher risk of metastasis (HR = 3.29, 95% CI: 1.53 - 7.07, p = 0.002) and PCSM (HR = 4.15, 95% CI: 1.55 - 11.0, p = 0.004). Additionally, ICSHigh and DecipherHigh also showed higher PORTOS score compared to those with ICSLow and DecipherLow (p &lt; 2.2e−16). Similar to RP cases, in the validation biopsy data Decipher score and ICS were associated (p = 0.01). Likewise, the association of immunogenomic categories with PORTOS score was also consistent in the validation dataset (p = 0.0002). PCA in both RP and validation data using other T cells (i.e. CD4+, CD8+ and Treg) along with ICS, showed separation of high and low PORTOS score categories in relation to immune cells.

Conclusions

The immunogenomic landscape of GG5 disease has major prognostic and therapeutic implications in aggressive PCa. Immunogenomic interaction between immune content and genomic risk classifiers can be used to identify subsets of patients who are at increased risk of metastasis and PCSM, and may have enhanced response to radiotherapy.

Citation Format: Kosj Yamoah, Shivanshu Awasthi, Brandon Mahal, Shuang G. Zhao, Julieta Abraham-Miranda, Travis A. Gerke, Elai Davicioni, Robert J. Rounbehler, Anders E. Berglund, Daniel G. Grass, Jong Y. Park, Julio M. Pow-Sang, Daniel Fernandez, Rohit K. Jain, Edward M. Schaeffer, Stephen Freedland, Daniel E. Spratt, Robert B. Den, Manish Kohli, Timothy R. Rebbeck, Felix Y. Fang. Immunogenomic landscape of grade group 5 prostate cancer predicts risk of lethal outcomes and may inform treatment response [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 3520.

Multifaceted Function of MicroRNA-299-3p Fosters an Antitumor Environment Through Modulation of Androgen Receptor and VEGFA Signaling Pathways in Prostate Cancer

Prostate cancer (PCa) is one of the most common cancers to affect men worldwide. Androgen receptor (AR) signaling is central to PCa and PCa therapy. MicroRNAs (miRNAs) play crucial roles in the regulation of prostate cancer through modulation of signaling pathways. In the present study, we illustrate the functional significance and therapeutic benefit of miR-299-3p, an AR targeting microRNA, in PCa progression. We noted loss of expression of miR-299-3p in prostate tumors compared to noncancerous prostate tissues. Replenishment of miR-299-3p in C4-2B, 22Rv-1 and PC-3 cells contributed to cell cycle arrest, reduced proliferation, migration and increased expression of apoptotic markers. Additionally, overexpression of miR-299-3p induced a reduction of AR, PSA and VEGFA expression. AGO-RNA pulldown experiment showed enrichment of AR, VEGFA and miR-299-3p in C4-2B cells overexpressing miR-299-3p. miR-299-3p overexpression also inhibited epithelial mesenchymal transition, expression of Slug, TGF-β3, phospho-AKT and phospho-PRAS40, but increased expression of E-cadherin. Furthermore, miR-299 overexpression resulted in reduced tumor growth in xenograft models and increased drug sensitivity. Overall, this study has identified novel mechanisms of antitumor and antimigration function of miR-299-3p through modulation of AR and VEGFA signaling pathways which lead to improved drug sensitivity of PCa.

Multi-Dimensional Flow Cytometry Analyses Reveal a Dichotomous Role for Nitric Oxide in Melanoma Patients Receiving Immunotherapy

Phenotyping of immune cell subsets in clinical trials is limited to well-defined phenotypes, due to technological limitations of reporting flow cytometry multi-dimensional phenotyping data. We developed a multi-dimensional phenotyping analysis tool and applied it to detect nitric oxide (NO) levels in peripheral blood immune cells before and after adjuvant ipilimumab co-administration with a peptide vaccine in melanoma patients. We analyzed inhibitory and stimulatory markers for immune cell phenotypes that were felt to be important in the NO analysis. The pipeline allows visualization of immune cell phenotypes without knowledge of clustering techniques and to categorize cells by association with relapse-free survival (RFS). Using this analysis, we uncovered the potential for a dichotomous role of NO as a pro- and anti-melanoma factor. NO was found in subsets of immune-suppressor cells associated with shorter-term (≤ 1 year) RFS, whereas NO was also present in immune-stimulatory effector cells obtained from patients with significant longer-term (> 1 year) RFS. These studies provide insights into the cell-specific immunomodulatory role of NO. The methods presented herein can be applied to monitor the pro- and anti-tumor effects of a variety of immune-based therapeutics in cancer patients. Clinical Trial Registration Number: NCT00084656 (https://clinicaltrials.gov/ct2/show/NCT00084656).

Expression of the BAD pathway is a marker of triple-negative status and poor outcome

Triple-negative breast cancer (TNBC) has few therapeutic targets, making nonspecific chemotherapy the main treatment. Therapies enhancing cancer cell sensitivity to cytotoxic agents could significantly improve patient outcomes. A BCL2-associated agonist of cell death (BAD) pathway gene expression signature (BPGES) was derived using principal component analysis (PCA) and evaluated for associations with the TNBC phenotype and clinical outcomes. Immunohistochemistry was used to determine the relative expression levels of phospho-BAD isoforms in tumour samples. Cell survival assays evaluated the effects of BAD pathway inhibition on chemo-sensitivity. BPGES score was associated with TNBC status and overall survival (OS) in breast cancer samples of the Moffitt Total Cancer Care dataset and The Cancer Genome Atlas (TCGA). TNBC tumours were enriched for the expression of phospho-BAD isoforms. Further, the BPGES was associated with TNBC status in breast cancer cell lines of the Cancer Cell Line Encyclopedia (CCLE). Targeted inhibition of kinases known to phosphorylate BAD protein resulted in increased sensitivity to platinum agents in TNBC cell lines compared to non-TNBC cell lines. The BAD pathway is associated with triple-negative status and OS. TNBC tumours were enriched for the expression of phosphorylated BAD protein compared to non-TNBC tumours. These findings suggest that the BAD pathway it is an important determinant of TNBC clinical outcomes.

Commercial Gene Expression Tests for Prostate Cancer Prognosis Provide Paradoxical Estimates of Race-Specific Risk

BACKGROUND

Commercial gene expression signatures of prostate cancer prognosis were developed and validated in cohorts of predominantly European American men (EAM). Limited research exists on the value of such signatures in African American men (AAM), who have poor prostate cancer outcomes. We explored differences in gene expression between EAM and AAM for three commercially available panels recommended by the National Comprehensive Cancer Network for prostate cancer prognosis.

METHODS

A total of 232 EAM and 95 AAM patients provided radical prostatectomy specimens. Gene expression was quantified using NanoString for 60 genes spanning the Oncotype DX Prostate, Prolaris, and Decipher panels. A continuous expression-based risk score was approximated for each. Differential expression, intrapanel coexpression, and risk by race were assessed.

RESULTS

Clinical and pathologic features were similar between AAM and EAM. Differential expression by race was observed for 48% of genes measured, although the magnitudes of expression differences were small. Coexpression patterns were more strongly preserved by race group for Oncotype DX and Decipher than Prolaris. Poorer prognosis was estimated in EAM versus AAM for Oncotype DX (P < 0.001), whereas negligible prognostic differences were predicted between AAM and EAM using Prolaris or Decipher (P > 0.05).

CONCLUSIONS

Because of observed racial differences across three commercial gene expression panels for prostate cancer prognosis, caution is warranted when applying these panels in clinical decision-making in AAM.

IMPACT

Differences in gene expression by race for three commercial panels for prostate cancer prognosis indicate that further study of their effectiveness in AAM with long-term follow-up is warranted.

Paradoxical epigenetic regulation of XAF1 mediates plasticity towards adaptive resistance evolution in MGMT-methylated glioblastoma

Epigenetic regulation of O⁶-alkylguanine DNA alkyltransferase (MGMT) is surrogate of intrinsic resistance to temozolomide (TMZ). However, mechanisms associated with adaptive resistance evolution of glioblastoma (GBM) relative to MGMT methylation remain unclear. We hereby report a paradoxical yet translational epigenetic regulation of plasticity towards adaptive resistance in GBM. Based on an adaptive resistance model of GBM cells with differential MGMT methylation profiles, MGMT-hypermethylation enhanced genetic and phenotypic plasticity towards adaptive resistance to TMZ while MGMT hypomethylation limited plasticity. The resulting model-associated adaptive resistance gene signature negatively correlated with GBM patient survival. XAF1, a tumor suppressor protein, paradoxically emerged as a mediator of differential plasticities towards adaptive resistance to TMZ through epigenetic regulation. XAF1 promoted resistance both in-vitro and in-vivo. Furthermore, XAF1 expression negatively correlated with XAF1 promoter methylation status, and negatively correlate with GBM patient survival. Collectively, XAF1 appears to have a pradoxical yet translational role in GBM.

Abstract 4471: Commercial gene expression tests for prostate cancer prognosis provide paradoxical estimates of race-specific risk

Background: Current National Comprehensive Cancer Network (NCCN) guidelines recommend three gene expression-based tests for prostate cancer (PCa) prognosis in men with low or favorable intermediate risk disease: Decipher, Oncotype DX Prostate, and Prolaris. The three tests feature varying numbers of genes with minimal overlap. Importantly, development and validation efforts for all three panels were undertaken in predominantly European American men (EAM) cohorts, and limited research exists on the value of such signatures in African American men (AAM), who have poorer PCa outcomes. Here, we explored differences in gene expression between EAM and AAM for the three panels recommended by the NCCN for PCa prognosis.

Methods: 232 EAM and 95 AAM patients provided radical prostatectomy specimens. Gene expression was quantified using Nanostring for 60 genes spanning the Oncotype DX Prostate, Prolaris, and Decipher panels. Differential expression and intrapanel co-expression by race were assessed using Mann-Whitney tests and Spearman’s correlations, respectively. A continuous expression-based risk score was approximated for each panel with higher scores indicating worse outcomes. Race-specific risks were compared using Mann-Whitney tests and Spearman’s correlations.

Results: Clinical and pathologic features were similar between AAM and EAM. Differential expression by race was observed for 48% of genes measured, though the magnitudes of expression differences were small. Co-expression patterns were more strongly preserved by race group for Oncotype DX and Decipher versus Prolaris (integrative correlations of 0.87, 0.73, and 0.62, respectively). Paradoxically, poorer prognosis was estimated in EAM versus AAM for Prolaris and Oncotype DX (p = 0.01 for both), whereas worse prognosis was predicted for AAM versus EAM using Decipher (p &lt; 0.001).

Discussion: Due to observed racial differences across three commercial gene expression panels for PCa prognosis, caution is warranted when applying these panels in clinical decision-making in AAM. Replication of our findings directly on the commercial panels with long-term follow-up is warranted.

Citation Format: Jordan H. Creed, Anders E. Berglund, Robert J. Rounbehler, Shivanshu Awasthi, John L. Cleveland, Jong Y. Park, Kosj Yamoah, Travis A. Gerke. Commercial gene expression tests for prostate cancer prognosis provide paradoxical estimates of race-specific risk [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 4471.

Distinct transcriptional repertoire of the androgen receptor in ETS fusion-negative prostate cancer

Background

Prostate cancer (PCa) tumors harboring translocations of ETS family genes with the androgen responsive TMPRSS2 gene (ETS+ tumors) provide a robust biomarker for detecting PCa in approximately 70% of patients. ETS+ PCa express high levels of the androgen receptor (AR), yet PCa tumors lacking ETS fusions (ETS−) also express AR and demonstrate androgen-regulated growth. In this study, we evaluate the differences in the AR-regulated transcriptomes between ETS+ and ETS− PCa tumors.

Methods

10,608 patient tumors from three independent PCa datasets classified as ETS+ (samples overexpressing ERG or other ETS family members) or ETS− (all other PCa) were analyzed for differential gene expression using false-discovery-rate adjusted methods and gene-set enrichment analysis (GSEA).

Results

Based on the expression of AR-dependent genes and an unsupervised Principal Component Analysis (PCA) model, AR-regulated gene expression alone was able to separate PCa samples into groups based on ETS status in all PCa databases. ETS status distinguished several differentially expressed genes in both TCGA (6.9%) and GRID (6.6%) databases, with 413 genes overlapping in both databases. Importantly, GSEA showed enrichment of distinct androgen-responsive genes in both ETS− and ETS+ tumors, and AR ChIP-seq data identified 131 direct AR-target genes that are regulated in an ETS-specific fashion. Notably, dysregulation of ETS-dependent AR-target genes within the metabolic and non-canonical WNT pathways was associated with clinical outcomes.

Conclusions

ETS status influences the transcriptional repertoire of the AR, and ETS− PCa tumors appear to rely on distinctly different AR-dependent transcriptional programs to drive and sustain tumorigenesis.

Abstract A045: All happy families are alike: Transcriptomic homogeneity in indolent prostate tumors is a useful prognostic biomarker

Purpose: To derive a gene expression score of transcriptomic homogeneity in localized prostate tumor specimens, and assess its prognostic utility for metastatic or lethal disease.

Background: At the time of prostate cancer diagnosis, improved biomarkers are needed to distinguish tumors with metastatic potential versus those likely to follow an indolent course. Previous evidence suggests that gene expression aberrations are homogeneous in nonaggressive tumors, while transcriptomic profiles are disregulated in widely varying patterns across metastasis-prone tumors. Here, we investigate whether a measure of transcriptomic homogeneity has prognostic utility.

Study Population: Gene expression profiles of noncancerous prostate tissue from the Genotype-Tissue Expression (GTEx) project (n = 106) were used to develop the homogeneity measure. Prognostic utility of the score was assessed in a study nested in the Health Professionals Follow-up Study (HPFS) and Physicians’ Health Study (PHS), in which whole-transcriptome gene expression was quantified from archival surgical tumor tissue. Cases (n = 113) were men who died of prostate cancer or developed metastatic disease, and controls (n = 291) had at least 8 years of metastasis-free survival. Men were diagnosed with prostate cancer between 1982 and 2005, and median follow-up time was 14.0 years. TCGA data (n = 333) were further used to investigate associations of the expression score with Gleason.

Methods: RNA-seq profiles from GTEx were filtered to retain genes persistently expressed across noncancerous prostate tissue specimens. Remaining genes were placed into decile bins on the basis of median expression across samples. The 25 lowest variance genes within each bin were selected to generate a 250-gene panel. Transcriptome homogeneity scores for patients in HPFS and PHS were calculated by summing the standard deviations over the decile bins, where standard deviations were calculated using patient-specific data across 25 genes in each bin. Associations between the score and lethal disease were assessed through logistic regression and AUC analyses. Associations between the homogeneity score and Gleason score in TCGA samples were evaluated via Pearson correlations.

Results: High transcriptomic heterogeneity was strongly associated with lethal or metastatic outcomes. A crude odds ratio of 3.36 (95% CI 2.12-5.41; p&lt;0.001) was estimated for patients with above median heterogeneity scores compared to those with homogenous expression (those with less than or equal to median heterogeneity). This association persisted with adjustment for Gleason (&lt;7, 3+4, 4+3, 8, 9-10) with an odds ratio of 2.04 (95% CI 1.19-3.51; p=0.01). Addition of heterogeneity scores to Gleason improved the AUC of the Gleason-only model from 0.82 (95% CI 0.77-0.86) to 0.84 (95% CI 0.80-0.88; p=0.01 for improvement). Higher heterogeneity also tracked with higher Gleason score in TCGA with a correlation of 0.33 (95% CI 0.23-0.42; p&lt;0.001).

Conclusions: Transcriptomic homogeneity, as measured by variation across gene sets that are typically stable in normal prostate tissue, is able to distinguish indolent from lethal prostate cancers. Further investigation and discovery of gene sets using this novel approach is likely to deliver new biologic insights and clinically actionable biomarker sets.

Citation Format: Travis Gerke, Svitlana Tyekucheva, Jordan H. Creed, Kathryn L. Penney, Jennifer A. Sinnott, Ericka Ebot, Anders E. Berglund, Massimo Loda, Meir J. Stampfer, Peter Kraft, Giovanni Parmigiani, Lorelei A. Mucci. All happy families are alike: Transcriptomic homogeneity in indolent prostate tumors is a useful prognostic biomarker [abstract]. In: Proceedings of the AACR Special Conference: Prostate Cancer: Advances in Basic, Translational, and Clinical Research; 2017 Dec 2-5; Orlando, Florida. Philadelphia (PA): AACR; Cancer Res 2018;78(16 Suppl):Abstract nr A045.

Tristetraprolin Is a Prognostic Biomarker for Poor Outcomes among Patients with Low-Grade Prostate Cancer

Background: We studied the utility of the tumor suppressor Tristetraprolin (TTP, ZFP36) as a clinically relevant biomarker of aggressive disease in prostate cancer patients after radical prostatectomy (RP).Methods: TTP RNA expression was measured in an RP cohort of patients treated at Moffitt Cancer Center (MCC) and obtained from six publically available RP datasets with biochemical recurrence (BCR; total n = 1,394) and/or metastatic outcome data (total n = 1,222). TTP protein expression was measured by immunohistochemistry in a tissue microarray of 153 MCC RP samples. The time to BCR or metastasis based on TTP RNA or protein levels was calculated using the Kaplan-Meier analysis. Univariable and multivariable Cox proportional hazard models were performed on multiple cohorts to evaluate if TTP is a clinically relevant biomarker and to assess if TTP improves upon the Cancer of the Prostate Risk Assessment postsurgical (CAPRA-S) score for predicting clinical outcomes.Results: In all of the RP patient cohorts, prostate cancer with low TTP RNA or protein levels had decreased time to BCR or metastasis versus TTP-high tumors. Further, the decreased time to BCR in TTP-low prostate cancer was more pronounced in low-grade tumors. Finally, pooled survival analysis suggests that TTP RNA expression provides independent information beyond CAPRA-S to predict BCR.Conclusions: TTP is a promising prostate cancer biomarker for predicting which RP patients will have poor outcomes, especially for low-grade prostate cancer patients.Impact: This study suggests that TTP RNA expression can be used to enhance the accuracy of CAPRA-S to predict outcomes in patients treated with RP. Cancer Epidemiol Biomarkers Prev; 27(11); 1376-83. ©2018 AACR.

Induction of Tertiary Lymphoid Structures With Antitumor Function by a Lymph Node-Derived Stromal Cell Line

Tertiary lymphoid structures (TLSs) associate with better prognosis in certain cancer types, but their underlying formation and immunological benefit remain to be determined. We established a mouse model of TLSs to study their contribution to antitumor immunity. Because the stroma in lymph nodes (sLN) participates in architectural support, lymphogenesis, and lymphocyte recruitment, we hypothesized that TLSs can be created by sLN. We selected a sLN line with fibroblast morphology that expressed sLN surface markers and lymphoid chemokines. The subcutaneous injection of the sLN line successfully induced TLSs that attracted infiltration of host immune cell subsets. Injection of MC38 tumor lysate-pulsed dendritic cells activated TLS-residing lymphocytes to demonstrate specific cytotoxicity. The presence of TLSs suppressed MC38 tumor growth in vivo by improving antitumor activity of tumor-infiltrating lymphocytes with downregulated immune checkpoint proteins (PD-1 and Tim-3). Future engineering of sLN lines may allow for further enhancements of TLS functions and immune cell compositions.

Radiosensitivity of Lung Metastases by Primary Histology and Implications for Stereotactic Body Radiation Therapy Using the Genomically Adjusted Radiation Dose

INTRODUCTION

We assessed the radiosensitivity of lung metastases on the basis of primary histologic type by using a validated gene signature and model lung metastases for the gnomically adjusted radiation dose (GARD).

METHODS

Tissue samples were identified from our prospective observational protocol. The radiosensitivity index (RSI) 10-gene assay was run on samples and calculated alongside the GARD by using the previously published algorithms. A cohort of 105 patients with 137 lung metastases treated with stereotactic body radiation therapy (SBRT) at our institution was used for clinical correlation.

RESULTS

A total of 138 unique metastatic lung lesions from our institution's tissue biorepository were identified for inclusion. There were significant differences in the RSI of lung metastases on the basis of histology. In order of decreasing radioresistance, the median RSIs for the various histologic types of cancer were endometrial adenocarcinoma (0.49), soft-tissue sarcoma (0.47), melanoma (0.44), rectal adenocarcinoma (0.43), renal cell carcinoma (0.33), head and neck squamous cell cancer (0.33), colon adenocarcinoma (0.32), and breast adenocarcinoma (0.29) (p = 0.002). We modeled the GARD for these samples and identified the biologically effective dose necessary to optimize local control. The 12- and 24-month Kaplan-Meier rates of local control for radioresistant versus radiosensitive histologic types from our clinical correlation cohort after lung SBRT were 92%/87% and 100%, respectively (p = 0.02).

CONCLUSIONS

In this analysis, we have noted significant differences in radiosensitivity on the basis of primary histologic type of lung metastases and have modeled the biologically effective dose necessary to optimize local control. This study suggests that primary histologic type may be an additional factor to consider in selection of SBRT dose to the lung and that dose personalization may be feasible.

Tristetraprolin disables prostate cancer maintenance by impairing proliferation and metabolic function

Tristetraprolin (TTP) is an RNA-binding protein that post-transcriptionally suppresses gene expression by delivering mRNA cargo to processing bodies (P-bodies) where the mRNA is degraded. TTP functions as a tumor suppressor in a mouse model of B cell lymphoma, and in some human malignancies low TTP expression correlates with reduced survival. Here we report important prognostic and functional roles for TTP in human prostate cancer. First, gene expression analysis of prostate tumors revealed low TTP expression correlates with patients having high-risk Gleason scores and increased biochemical recurrence. Second, in prostate cancer cells with low levels of endogenous TTP, inducible TTP expression inhibits their growth and proliferation, as well as their clonogenic growth. Third, TTP functions as a tumor suppressor in prostate cancer, as forced TTP expression markedly impairs the tumorigenic potential of prostate cancer cells in a mouse xenograft model. Finally, pathway analysis of gene expression data suggested metabolism is altered by TTP expression in prostate tumor cells, and metabolic analyses revealed that such processes are impaired by TTP, including mitochondrial respiration. Collectively, these findings suggest that TTP is an important prognostic indicator for prostate cancer, and augmenting TTP function would effectively disable the metabolism and proliferation of aggressive prostate tumors.

Evaluating somatic tumor mutation detection without matched normal samples

Background

Observations of recurrent somatic mutations in tumors have led to identification and definition of signaling and other pathways that are important for cancer progression and therapeutic targeting. As tumor cells contain both an individual’s inherited genetic variants and somatic mutations, challenges arise in distinguishing these events in massively parallel sequencing datasets. Typically, both a tumor sample and a “normal” sample from the same individual are sequenced and compared; variants observed only in the tumor are considered to be somatic mutations. However, this approach requires two samples for each individual.

Results

We evaluate a method of detecting somatic mutations in tumor samples for which only a subset of normal samples are available. We describe tuning of the method for detection of mutations in tumors, filtering to remove inherited variants, and comparison of detected mutations to several matched tumor/normal analysis methods. Filtering steps include the use of population variation datasets to remove inherited variants as well a subset of normal samples to remove technical artifacts. We then directly compare mutation detection with tumor-only and tumor-normal approaches using the same sets of samples. Comparisons are performed using an internal targeted gene sequencing dataset (n = 3380) as well as whole exome sequencing data from The Cancer Genome Atlas project (n = 250). Tumor-only mutation detection shows similar recall (43–60%) but lesser precision (20–21%) to current matched tumor/normal approaches (recall 43–73%, precision 30–82%) when compared to a “gold-standard” tumor/normal approach. The inclusion of a small pool of normal samples improves precision, although many variants are still uniquely detected in the tumor-only analysis.

Conclusions

A detailed method for somatic mutation detection without matched normal samples enables study of larger numbers of tumor samples, as well as tumor samples for which a matched normal is not available. As sensitivity/recall is similar to tumor/normal mutation detection but precision is lower, tumor-only detection is more appropriate for classification of samples based on known mutations. Although matched tumor-normal analysis is preferred due to higher precision, we demonstrate that mutation detection without matched normal samples is possible for certain applications.

Electronic supplementary material

The online version of this article (10.1186/s40246-017-0118-2) contains supplementary material, which is available to authorized users.

BIRC3 is a biomarker of mesenchymal habitat of glioblastoma, and a mediator of survival adaptation in hypoxia-driven glioblastoma habitats

Tumor hypoxia is an established facilitator of survival adaptation and mesenchymal transformation in glioblastoma (GBM). The underlying mechanisms that direct hypoxia-mediated survival in GBM habitats are unclear. We previously identified BIRC3 as a mediator of therapeutic resistance in GBM to standard temozolomide (TMZ) chemotherapy and radiotherapy (RT). Here we report that BIRC3 is a biomarker of the hypoxia-mediated adaptive mesenchymal phenotype of GBM. Specifically, in the TCGA dataset elevated BIRC3 gene expression was identified as a superior and selective biomarker of mesenchymal GBM versus neural, proneural and classical subtypes. Further, BIRC3 protein was highly expressed in the tumor cell niches compared to the perivascular niche across multiple regions in GBM patient tissue microarrays. Tumor hypoxia was found to mechanistically induce BIRC3 expression through HIF1-alpha signaling in GBM cells. Moreover, in human GBM xenografts robust BIRC3 expression was noted within hypoxic regions of the tumor. Importantly, selective inhibition of BIRC3 reversed therapeutic resistance of GBM cells to RT in hypoxic microenvironments through enhanced activation of caspases. Collectively, we have uncovered a novel role for BIRC3 as a targetable biomarker and mediator of hypoxia-driven habitats in GBM.

Abstract 5716: Tristetraprolin is a prognostic biomarker for biochemical recurrence in low Gleason score patients

PURPOSE: Most prostate cancer patients currently classified as having a low-risk of disease-specific death are overtreated in the United States, resulting in hundreds of thousands of men to have debilitating side effects from unnecessary treatment. However, ~30% of low-risk patients on active surveillance are later reclassified and require therapeutic intervention. Thus, there is a dire need for biomarkers that can discriminate which prostate tumors currently classified as low-risk will progress into aggressive disease.

EXPERIMENTAL DESIGN: Retrospective analyses of two independent prostate cancer expression datasets were performed to determine the relationship between the expression levels of the tumor suppressor Tristetraprolin (TTP, ZFP36) and biochemical recurrence (BCR). Further, TTP expression levels were measured by quantitative real-time PCR (qRT-PCR) in a cohort of prostate tumors from Moffitt Cancer Center patients having long-term follow-up (n=382), and will be correlated to associated clinical data, including BCR. Finally, the Cancer of the Prostate Risk Assessment Postsurgical (CAPRA-S) scores, a clinical tool used to predict prostate cancer outcomes following radical prostatectomy, were calculated along with BCR for the patients in one of the prostate cancer datasets.

RESULTS: Analysis of The Cancer Genome Atlas (TCGA) prostate adenocarcinoma dataset revealed that tumors with low TTP expression have an increased rate of BCR compared to those with high TTP levels (p=0.0096). Further, assessment of the GEO dataset GSE21034 (Taylor et al., Cancer Cell, 2010) established that prostate cancer patients with low Gleason scores (6 and 3+4) have an increased chance of BCR if their tumors express low levels of TTP compared to men with high TTP expressing tumors (p=0.0001). To validate these findings, patients with low Gleason score tumors and decreased levels of TTP in the Moffitt Cancer Center cohort are being analyzed to determine their rate of BCR. In contrast, CAPRA-S scores for low Gleason score tumors in the GSE20134 dataset failed to distinguish if a patient would have BCR (p=0.0511).

CONCLUSIONS: Prostate cancer patients with Gleason score 6 or 3+4, which are currently classified as having a low-risk of dying from prostate cancer, have an increased risk of BCR if their tumor expresses low levels of the tumor suppressor TTP. Our data suggests that TTP may be a biomarker that clearly discriminates which low Gleason score tumors will progress into aggressive disease.

Citation Format: Anders E. Berglund, Travis Gerke, Shivanshu Awasthi, G. Daniel Grass, Hyun Y. Park, John L. Cleveland, Jong Y. Park, Kosj Yamoah, Robert J. Rounbehler. Tristetraprolin is a prognostic biomarker for biochemical recurrence in low Gleason score patients [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 5716. doi:10.1158/1538-7445.AM2017-5716

Abstract 206: Proteogenomic classifications and outcome in squamous cell carcinoma of the lung

Introduction: Genomic analyses have yielded a tremendous amount of knowledge on the genetic changes in lung cancers, but translating this information into actionable data benefitting patients has proven difficult. The integration of proteomic analyses with genomics and gene expression profiling allows a more detailed description of the biological processes, thus improving our understanding of cancer phenotypes. These insights can potentially be used for better classification and help to guide patient selection for targeted therapies.

Experimental: We analyzed 116 surgically resected squamous cell lung carcinoma samples for copy-number alterations, gene expression profiling, targeted exome-sequencing and global proteomic profiling. The cohort consisted of mostly early stage tumors (83% Stage I or II) with complete follow-up (median 58 months). Copy number status was analyzed using the Affymetrix CytoScan array, DNA mutation status was assessed using a customized version of the Agilent Comprehensive Cancer Panel for targeted sequencing, and gene expression profiling was carried out by RNA-sequencing. Proteomic analysis was performed using TMT labeling, 12- fraction bRPLC separation and LC-MS/MS analysis with a Thermo Q-Exactive mass spectrometer. Database searches were performed using multiple search engines against RefSeq version 78, and summarized using IDPicker 3.

Results: The non-redundant protein inventory consisted of more than 6,000 protein groups with a protein FDR &lt;5%. Tumors were classified according to well established gene expression criteria into 4 classes: Classical, Basal, Primitive and Secretory. At the protein level, the Classical subtype was associated with xenobiotic and energy metabolism; the Basal subtype with defense responses and extracellular matrix changes; the Primitive subtype with nucleic acid metabolism; and the Secretory subtype with p38 signaling. These findings confirm and expand on previous mRNA expression studies of squamous cell lung carcinoma. Proteomics-based classification identified two sub-categories within the Classical subtype, which were characterized by inflammatory and stress response signaling. Within this group, patients with high expression of inflammation-associated proteins had better cancer-specific survival than those with low expression (p=0.04, Log-rank test). Targeted exome sequencing of 154 cancer-associated genes revealed frequent mutations in TP53, CDKN2A, NFE2L2, and other genes. Proteomic expression of genes located in amplified chromosomal regions was used to identify driver genes in squamous cell lung carcinoma.

Conclusion: Our results provide new biological insights from the addition of protein measurements to genomic datasets that have the potential to improve classification. The data suggest that proteins involved in immune responses are important for the biological behavior and outcome of the Classical subtype in squamous cell lung carcinoma.

Citation Format: Robbert J.C. Slebos, Paul A. Stewart, Eric A. Welsh, Ling Cen, Yonghong Zhang, Zhihua Chen, Chia-Ho Cheng, Fredrik Pettersson, Anders E. Berglund, Guolin Zhang, Bin Fang, Victoria Izumi, Sean J. Yoder, Katherine M. Fellows, Y Ann Chen, Jamie Teer, Steven Eschrich, John Koomen, Eric Haura. Proteogenomic classifications and outcome in squamous cell carcinoma of the lung [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 206. doi:10.1158/1538-7445.AM2017-206

Evaluation of invasive breast cancer samples using a 12-chemokine gene expression score: correlation with clinical outcomes

Background

A unique 12-chemokine gene expression score (CS) accurately predicted the presence of tumor-localized, ectopic lymph node-like structures (TL-ELNs) and improved overall survival (OS) in primary colorectal cancer and metastatic melanoma. We analyzed the correlation between CS, clinicopathological variables, molecular data, and 366 survival in Moffitt Cancer Center’s Total Cancer Care (TCC) patients with non-metastatic breast cancer.

Methods

Affymetrix gene expression profiles were used to interrogate the CS by the principal component method. Breast tumors were classified as high or low score based on median split, and correlations between clinicopathologic variables, PAM50 molecular subtype, and ELN formation were analyzed using the TCC dataset. Differences in overall survival (OS) and recurrence-free survival (RFS) in the larger KM Plot breast cancer public datasets were compared using Kaplan-Meier curves.

Results

We divided the Total Cancer Care (TCC) breast cancer patients into two groups of high or low CS. Mean CS was 0.24 (range, 2.2–2.1). Patients with higher CS were more likely to be white (172 vs. 159; p = 0.03), had poorly differentiated tumors (112 vs. 59; p <0.0001), ER/PR negative (41 vs. 26) and HER2 positive (36 vs. 19; p = 0.001), and contain TL-ELNs. Higher CS scores were also seen in the basal and HER2+ molecular subtypes. In the KM Plot breast cancer datasets higher CS patients demonstrated superior OS (HR = 0.73, p = 0.008) and RFS (HR 0.76, p = <0.0001), especially in basal and HER2+ patients.

Conclusions

High CS breast tumors tend to be higher grade, basal or HER2+, and present more frequently in Caucasians. However, this group of patients also shows the presence of TL-ELNs within the tumor microenvironment and has better survival outcomes. The CS is a novel tool that can identify breast cancer patients with tumors of a unique intratumoral immune composition and better prognosis. Whether or not the CS is a predictive response marker in breast cancer patients undergoing immunotherapy remains to be determined.

Electronic supplementary material

The online version of this article (doi:10.1186/s13058-017-0864-z) contains supplementary material, which is available to authorized users.

Abstract TMEM-037: GENE EXPRESSION ANALYSIS IDENTIFIES IMMUNOLOGICAL MRNA UP REGULATION IN CELLS IN THE ASCITES OF SEROUS OVARIAN CANCER AS COMPARED TO MATCHED PRIMARY TUMOR

BACKGROUND: Ovarian cancer patients with advanced disease frequently develop ascites. It is currently unclear to what extent the molecular phenotype of the cells in ascites differs from that of the primary ovarian tumor. There is also speculation that the tumor cells in ascites may represent a more treatment resistant cell population. An improved understanding of the biology of the cellular environment in ovarian cancer-related ascites may reveal opportunities to develop new targeted therapies that may not be apparent through analysis of the primary tumor alone. The aim of the study was to evaluate potential biological differences between the cellular environment of primary ovarian tumors and that of the ascites fluid.

METHODS: We performed mRNA expression in n=12 matched pairs of primary serous OVCA tumor and the cellular component of ascites using the HTG Edge System and the EdgeSeq Oncology Biomarker Panel assay. After normalization, log expression values for 2650 genes were compared between tumor and ascites samples using principle component analysis (PCA) modeling. The molecular profiles of primary tumor and ascites were compared using a t-test. Additionally, the CiberSort algorithm was applied to estimate the relative composition of tumor infiltrating lymphocytes in tumor and ascites samples.

RESULTS: PCA modeling demonstrated a clear difference between the primary tumor and ascites samples using all genes. A group comparison (q&lt;0.01, FDR corrected p-value, fold change &gt;2) indicated that 49 genes were differentially expressed between primary tumor and ascites samples. Ascites samples showed significantly higher expression of T-cell markers, CD2, CD3D, CD4, and CD8A, as well as the monocyte/macrophage markers CD14, CCL4, CD163, CCR1, IL10, and ITGAM. CiberSort analysis revealed a high proportion of eosinophils (p &lt;2 x 10^-7) and monocytes (p&lt;0.0002) in ascites as compared to tumor while resting mast cells were proportionately greater in the tumor samples (p &lt; 4 x 10^-5).

CONCLUSIONS: In this study, we report a significant difference in mRNA expression between primary tumors from serous OVCA and ascites. Ascites samples were found to have higher expression of T cell, monocyte and macrophage mRNA. Interestingly, ascites samples also expressed an increased proportion of eosinophils as compared to primary tumor. Eosinophils have been suggested to be essential for CD8+ T-cell mediated tumor rejection and have been associated with improved outcomes in prognosis. We also report an increased proportion of resting mast cells in primary tumor. Mast cells are increasingly thought to play a role in the regulation of tumor growth. In summary, this study demonstrates that molecular profile of cells in ascites is different than that of primary tumor and that this difference is potentially due to immune modulators.

Citation Format: Sharon E. Robertson, MD, MPH, Douglas C. Marchion, PhD, Yin Xiong, PhD, Anders E. Berglund, PhD, Anthony M. Magliocco, MD. GENE EXPRESSION ANALYSIS IDENTIFIES IMMUNOLOGICAL MRNA UP REGULATION IN CELLS IN THE ASCITES OF SEROUS OVARIAN CANCER AS COMPARED TO MATCHED PRIMARY TUMOR [abstract]. In: Proceedings of the 11th Biennial Ovarian Cancer Research Symposium; Sep 12-13, 2016; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(11 Suppl):Abstract nr TMEM-037.

Association of changes in T regulatory cells (Treg) during nivolumab treatment with melanoma outcome

3031

Background: PD-1 blocking antibodies have significant efficacy in the treatment of melanoma; however, many patients fail to respond and resistance mechanisms remain unknown. We addressed the role of Tregs, an immunosuppressive T-cell population, in patient outcome after treatment with nivolumab. Methods: Peripheral blood mononuclear cells (PBMC) were obtained from patients on trials with nivolumab as adjuvant therapy for resected disease or as treatment for metastatic melanoma. To measure suppression, Tregs were flow-sorted from PBMC and evaluated in allogeneic mixed lymphocyte reactions. Tregs and conventional CD4 T-cells were evaluated for gene expression changes by RNA-sequencing. Treg percentages and phosphorylated STAT3 (pSTAT3) expression were evaluated by flow cytometry. The effects of PD-1 blockade with nivolumab were evaluated in vitro using T-cells from baseline patient PBMC samples. Results: Tregs from responding patients or adjuvant patients without evidence of disease (NED) had reduced suppressive function post-nivolumab (p < 0.05), but no changes were observed in relapsing/non-responding patients; their Tregs were more suppressive than NED/responding Tregs (p < 0.001). NED Tregs had unique gene expression changes and associated pathways post-nivolumab compared to relapsing patient Tregs and conventional CD4 T-cells, including up-regulation of proliferation pathways (q < 8e-19) and downregulation of oxidative phosphorylation (q < 7e-5). NED Tregs had upregulation of pSTAT3 expression post-nivolumab (p < 0.05), which was not observed in relapsing patients. Evaluation of Tregs from patients with active disease also showed upregulation of pSTAT3 in responders (p < 0.05) but not non-responders. The relative increase in Treg pSTAT3 was associated with increased overall survival (R2= 0.49, p < 0.05). In vitro assays using PD-1 blocking antibodies recapitulated the increase in pSTAT3 (p < 0.05) and Treg percentages (p < 0.001), which were diminished with the addition of a STAT3 inhibitor (p < 0.01). Conclusions: These results demonstrate previously unknown roles of decreased Treg suppressive function and induction of STAT3 as biomarkers of patient’s outcome to nivolumab therapy.

The pan-cancer, pan-biomarker landscape for precision immuno-oncology

43

Background: Precision medicine has the opportunity to also change medical practice. However, there are many cancer treatments for which there are few tools for individualized selection based on patient-specific attributes. This is especially the case for the immune check point inhibitors, where a subset of patients receive almost unimaginable benefit, there is a significant incidence of severe, life threatening, and persistent autoimmune oriented toxicities, and the financial burden is real for patients. At least 3 distinct predictive immunooncology biomarkers (IOMark) have been proposed. This includes tumor mutation analysis, expression of PD-L1, and tumor-specific t cell subset (Weber Predictor). However, there is little data on the intrapatient relationship of these putative response markers. Methods: Analysis of individual patient IOMark status was determined in TCGA patients that had cancers for which immunotherapy is FDA approved (n = 2039; bladder, colorectal, kidney, lung, melanoma, stomach). Patients were categorized as putative high/low predicted responders for each biomarker [neoantigen burden (whole exome cutpoint = 225), PD-L1 (median), Weber predictor (upper 50% for each of CD8/Ki67/EOMES)]. Results: A high neoantigen burden was detected in 1%-60% of the tumors (median 31%; range kidney to melanoma). There was less variance for high PD-L1 (26%-78%, median 49%; colorectal to lung). The Weber predictor ranged from 2% to 41% (kidney to stomach). A subset of each cancer type were positive for each of the response markers (0% to 12%; median 6%; renal to stomach). There is not a high association between the IOMarks. Conclusions: Immuno-oncology is in need of robust IOMark, to allow more informed patient treatment. Analysis of the currently most promising IOMarks show high variation across tumor types and within tumor types. Only a subset of patients has high response predicted for all 3 IOMarks. The IOMarks are independent and identify distinct patient features.

Methylomic classifiers of anal cancer outcomes: An NRG Oncology / RTOG 98-11 tissue study

588

Background: Genome-wide epigenetic events appear to play a role in the development and behavior of HPV+ cancers. The value of adjuvant therapy following chemoradiation for localized anal cancer (AC) remains unclear. Molecular prognostication to identify patients (pts) who may be at higher risk for recurrence would be valuable. The goal was to define methylomic profiles predictive of disease-free (DFS) and overall (OS) survival in pts with AC. Methods: Genomic DNA was extracted, processed and methylation status at ~450,000 CpG loci examined (Illumina HumanMethylation450 Array). A multistep bioinformatics methodology was applied to develop a prognostic methylomic classifier for OS and DFS: (1) feature selection for methylated regions (β-value interquartile range ≥ 0.2, ≥ 2 adjacent significant probes within a CpG Island and p < 0.05 by univariate Cox proportional hazards) (2) selected features were entered into a supervised principal component analysis (PCA) and 3 components (PC1, PC2, PC3) were derived (3) classifier was built using forward selection multivariate regression models [PC1, PC2, PC3 alone and in combination with clinical features (size: > T2 vs. ≤ T2, nodal status: N0 vs N+)] using a 10-fold cross-validation (4) final model prediction risk score was generated, dichotomized and evaluated for prognostic values in Cox regression analysis. Results: 121 AC specimens from RTOG 98-11 were examined. The methylomic-only classifier model trended towards statistical significance (log-rank p = 0.05; HR = 1.96; 95% CI 0.99-3.88) in DFS (PC1, PC3 selected). In the combined model with clinical features, the final classifier included T status and epigenetic features (PC1, PC3) and was strongly predictive for DFS (p < 0.0001, HR = 4.45; 2.02-9.76). Final OS classifier models [methylomic-only (p = 0.28 HR = 1.55; 0.70-3.44) or combined (p = 0.013 HR = 2.88; 1.20-6.89)] were not as accurate. Conclusions: Methylomic and clinical features synergize to predict DFS in AC. Multivariate modeling reveal independent contributions from clinical and methylomic variables. Epigenomic profiling may contribute to identification of high-risk pts who may benefit from adjuvant strategies. Support: U10CA180822, U10CA180868, U24CA196067

Radiosensitivity Differences Between Liver Metastases Based on Primary Histology Suggest Implications for Clinical Outcomes After Stereotactic Body Radiation Therapy

PURPOSE/OBJECTIVES

Evidence from the management of oligometastases with stereotactic body radiation therapy (SBRT) reveals differences in outcomes based on primary histology. We have previously identified a multigene expression index for tumor radiosensitivity (RSI) with validation in multiple independent cohorts. In this study, we assessed RSI in liver metastases and assessed our clinical outcomes after SBRT based on primary histology.

METHODS AND MATERIALS

Patients were identified from our prospective, observational protocol. The previously tested RSI 10 gene assay was run on samples and calculated using the published algorithm. An independent cohort of 33 patients with 38 liver metastases treated with SBRT was used for clinical correlation.

RESULTS

A total of 372 unique metastatic liver lesions were identified for inclusion from our prospective, institutional metadata pool. The most common primary histologies for liver metastases were colorectal adenocarcinoma (n=314, 84.4%), breast adenocarcinoma (n=12, 3.2%), and pancreas neuroendocrine (n=11, 3%). There were significant differences in RSI of liver metastases based on histology. The median RSIs for liver metastases in descending order of radioresistance were gastrointestinal stromal tumor (0.57), melanoma (0.53), colorectal neuroendocrine (0.46), pancreas neuroendocrine (0.44), colorectal adenocarcinoma (0.43), breast adenocarcinoma (0.35), lung adenocarcinoma (0.31), pancreas adenocarcinoma (0.27), anal squamous cell cancer (0.22), and small intestine neuroendocrine (0.21) (P<.0001). The 12-month and 24-month Kaplan-Meier rates of local control (LC) for colorectal lesions from the independent clinical cohort were 79% and 59%, compared with 100% for noncolorectal lesions (P=.019), respectively.

CONCLUSIONS

In this analysis, we found significant differences based on primary histology. This study suggests that primary histology may be an important factor to consider in SBRT radiation dose selection.