Abstract 1343: Mesenchymal cell populations associated with different stages of prostate cancer progression in mice and human

Mesenchymal cells in the prostate cancer (PCa) tumor microenvironment (TME) contribute to the biological and clinical history of PCa. Indeed, mesenchymal cells heavily interact with cancer cells, immune cells, and the other cellular and non-cellular components of the TME to favor or hinder carcinogenesis and tumor progression. Using a comprehensive array of genetically engineered mouse models (GEMMs) of prostate cancer, 8 mesenchymal populations with different transcriptional programs are preferentially enriched in specific GEMMs at different stages of PCa. Here, we determine the transferability of this mesenchymal cluster designation from mice PCa models to human PCa cases. To this end, we compared: a) Tmprss2-ERG (T-ERG) mouse and ERG+ human cases; b) Pb4-Cre+/-;Ptenf/f;LSL-MYCN+/+;Rb1f/f (PRN) mouse and PCa bone metastasis. We generated scRNA-seq data for > 8000 mesenchymal cells from ERG+ (n=6) and ERG- (n=3) PCa patients, and we retrieved data for bone metastasis mesenchymal cells (osteoblasts, osteoclasts, endothelial cells, pericytes; 1,872 total cells) from GSE143791. To transfer the stromal mouse clusters’ labels to human data, human gene symbols were converted to their mouse counterparts, then both datasets were restricted to overlapping genes. For the human PCa cases, label transfer was performed through ‘ingest’ using the scRNA-seq data from the mouse T-ERG model as reference for the human ERG+ cases and data from the remaining GEMMs as reference for the human ERG- cases. For bone metastases cases, mouse stromal data from all GEMMs were used to project the 8 stromal clusters to the mesenchymal cells in the bone metastases microenvironment. Not surprisingly, ERG+ human samples were enriched (> 60% of total stromal cells) in mouse stroma clusters predominantly present in T-ERG mouse model, characterized by the expression of Wnt regulators and AR. Common populations to all murine models, representing myofibroblasts and immunomodulatory fibroblasts (expressing Gpx3, C3, C7, Cfh), were also commonly present in patients, irrespectively to the ERG status. In the PCa bone metastases, mesenchymal clusters enriched in the PRN model were strongly represented in human bone metastases, comprising > 60% of total stromal cells. These cells were characterized by high expression of POSTN and MKI67, as well as bone-specific genes like BGN. Altogether, these findings suggest that our mesenchymal cluster designation developed using GEMMs can be meaningfully applied to human PCa, and that the different transcriptional programs we identified in distinct mesenchymal population are conserved across species. This lays the foundation for the utilization of defined genetically-engineered models in defining the interactions and cross-talks between different mesenchymal populations in relation to cancer and immune cells and other components of the TME in human prostate cancer.

Citation Format: Mohamed Omar, Hubert Pakula, Filippo Pederzoli, Giuseppe N. Fanelli, Tania Panellinni, Ryan Carelli, Silvia Rodrigues, Caroline Fidalgo-Ribeiro, Pier V. Nuzzo, Lucie V. Emmenis, Mohammad Mohammad, Madhavi Jere, Caitlin Unkenholz, David Rickman, Christopher Barbieri, Brian Robinson, Luigi Marchionni, Massimo Loda. Mesenchymal cell populations associated with different stages of prostate cancer progression in mice and human [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1343.

Abstract LB036: Molecular insight into ERG transcriptional activity to unravel novel therapeutic options for ERG-positive castration resistant prostate cancer

Gene fusions of the AR-regulated gene TMPRSS2 with the ETS family transcription factor ERG characterize about 40% of castration resistant prostate cancer (CRPC) - an advanced and lethal stage of prostate cancer. AR-mediated overexpression of ERG hijacks AR inhibiting lineage-specific differentiation and increases cell invasion and migration. In vivo work revealed a causal role of ERG in initiating prostate cancer. Notably, in vitro and in vivo models of CRPC endogenously bearing T2ERG depend on ERG activity for proliferation and survival. Thus, pursuing the development of ERG-based therapeutic strategies represents a promising route for CRPC treatment, although major challenges in targeting transcription factors such as ERG have been reported. Yet, synthetic lethality approaches including genome-wide loss-of-function screens can be employed to avoid direct ERG targeting. Here, we used metastatic PCa cell lines (LNCaP, 22Rv1 and DU145) engineered with inducible systems to overexpress ERG to show that high levels of ERG expression lead to an apparent reduction in the fitness of AR-positive and negative cells. Prolonged overexpression of a mutant form of ERG, which is not able to bind DNA, does not affect cell viability indicating that ERG DNA binding and transcriptional activity are necessary for the induction of this phenotype. The ERG-mediated reduction in cellular fitness is influenced by cell adhesion and is characterized by cell cycle arrest mediated by several cell cycle inhibitors induced by ERG. As revealed by gene expression analyses and functional assays, ERG induces a less proliferative, more dedifferentiated and invasive phenotype, which lead to an apparent reduction of cellular fitness in vitro, but is characteristic of more aggressive tumor cells in vivo. Altogether, these results underscore the reliability of the developed in vitro models for ERG activity, which will be used for screening to identify therapeutic vulnerabilities unlocked by ERG in CRPC. Importantly, we also uncovered confounding factors (i.e. loss of ERG-positive cells not due to screen selection) that could impinge on the results arising from in vitro screens.

Citation Format: Francesca Lorenzin, Giulia Fracassi, Yari Ciani, Paola Gasperini, David Rickman, Francesca Demichelis. Molecular insight into ERG transcriptional activity to unravel novel therapeutic options for ERG-positive castration resistant prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr LB036.

Delta-like protein 3 expression and therapeutic targeting in neuroendocrine prostate cancer

Histologic transformation to small cell neuroendocrine prostate cancer occurs in a subset of patients with advanced prostate cancer as a mechanism of treatment resistance. Rovalpituzumab tesirine (SC16LD6.5) is an antibody-drug conjugate that targets delta-like protein 3 (DLL3) and was initially developed for small cell lung cancer. We found that DLL3 is expressed in most of the castration-resistant neuroendocrine prostate cancer (CRPC-NE) (36 of 47, 76.6%) and in a subset of castration-resistant prostate adenocarcinomas (7 of 56, 12.5%). It shows minimal to no expression in localized prostate cancer (1 of 194) and benign prostate (0 of 103). DLL3 expression correlates with neuroendocrine marker expression, RB1 loss, and aggressive clinical features. DLL3 in circulating tumor cells was concordant with matched metastatic biopsy (87%). Treatment of DLL3-expressing prostate cancer xenografts with a single dose of SC16LD6.5 resulted in complete and durable responses, whereas DLL3-negative models were insensitive. We highlight a patient with neuroendocrine prostate cancer with a meaningful clinical and radiologic response to SC16LD6.5 when treated on a phase 1 trial. Overall, our findings indicate that DLL3 is preferentially expressed in CRPC-NE and provide rationale for targeting DLL3 in patients with DLL3-positive metastatic prostate cancer.

Abstract CN06-01: Lineage plasticity and the neuroendocrine phenotype as a resistance mechanism in prostate cancer

Loss of androgen receptor (AR) signaling dependence occurs in approximately 15-20% of treatment resistant prostate cancers, and this may manifest clinically as transformation from a prostate adenocarcinoma histology to a castration resistant neuroendocrine prostate cancer (CRPC-NE). The diagnosis of CRPC-NE currently relies on a metastatic tumor biopsy, which is invasive for patients and often challenging to diagnose due to tumor heterogeneity. By studying whole exome sequencing and whole genome bisulfite sequencing of metastatic tumor biopsies and matched cell free DNA (cfDNA), we identified distinct genomic and epigenomic features of CRPC-NE and patterns of tumor evolution that occur during clinical progression and treatment resistance. Loss of RB1 and TP53 are enriched in CRPC-NE compared with castration resistant prostate adenocarcinoma, as are significant changes in DNA methylation, which are detectable by cfDNA. There was significantly higher concordance between cfDNA and biopsy tissue genomic alterations in CRPC-NE patients compared to castration resistant adenocarcinoma, supporting greater intra-individual genomic consistency across metastases. cfDNA and serial tumor biopsies allowed for the tracking of dynamic clonal and subclonal tumor cell populations as patients progressed and identified CRPC-NE alterations sometimes prior to the development of clinical features. CRPC-NE appears to arise clonally from a prostate adenocarcinoma precursor likely through a dynamic process of clonal selection and trans-differentiation that occurs during resistance to AR-directed therapies. In addition to loss of AR expression and canonical AR signaling, we identified a dysregulation of key pathways in CRPC-NE including loss of Notch signaling, reactivation of developmental programs, and gain of neuronal/neuroendocrine programs including critical lineage determining transcription factors (LDTFs) such as ASCL1, MYCN, BRN2, pointing to novel biomarkers and potential targets for CRPC-NE. Patient-derived organoids and xenografts were used to interrogate the timing by which genomic and epigenomic changes and LDTFs contribute to lineage plasticity and the development of CRPC-NE as a resistance mechanism in prostate cancer.

Citation Format: Himisha Beltran, Sheng-Yu Ku, Vincenza Conteduca, Alessandro Romanel, Loredana Puca, Michael Sigouros, Juan Miguel Mosquera, Scott T. Tagawa, Andrea Sboner, Olivier Elemento, David Goodrich, David Rickman, Amina Zoubeidi, Francesca Demichelis. Lineage plasticity and the neuroendocrine phenotype as a resistance mechanism in prostate cancer [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2019 Oct 26-30; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2019;18(12 Suppl):Abstract nr CN06-01. doi:10.1158/1535-7163.TARG-19-CN06-01

Integrative Molecular Analysis of Patients With Advanced and Metastatic Cancer

PURPOSE

We developed a precision medicine program for patients with advanced cancer using integrative whole-exome sequencing and transcriptome analysis.

PATIENTS AND METHODS

Five hundred fifteen patients with locally advanced/metastatic solid tumors were prospectively enrolled, and paired tumor/normal sequencing was performed. Seven hundred fifty-nine tumors from 515 patients were evaluated.

RESULTS

Most frequent tumor types were prostate (19.4%), brain (16.5%), bladder (15.4%), and kidney cancer (9.2%). Most frequently altered genes were TP53 (33%), CDKN2A (11%), APC (10%), KTM2D (8%), PTEN (8%), and BRCA2 (8%). Pathogenic germline alterations were present in 10.7% of patients, most frequently CHEK2 (1.9%), BRCA1 (1.5%), BRCA2 (1.5%), and MSH6 (1.4%). Novel gene fusions were identified, including a RBM47-CDK12 fusion in a metastatic prostate cancer sample. The rate of clinically relevant alterations was 39% by whole-exome sequencing, which was improved by 16% by adding RNA sequencing. In patients with more than one sequenced tumor sample (n = 146), 84.62% of actionable mutations were concordant.

CONCLUSION

Integrative analysis may uncover informative alterations for an advanced pan-cancer patient population. These alterations are consistent in spatially and temporally heterogeneous samples.

Abstract 134: Immunogenomic landscape of neuroendocrine prostate cancer (NEPC)

Background: NEPC is a histological subtype of advanced prostate cancer, predominantly arising clonally from castrate resistant prostate adenocarcinoma (CRPC) as a mechanism of resistance. Given current immunotherapeutic strategies only showing modest clinical benefit in prostate cancer patients and NEPC having clinical features aligned with small cell lung carcinoma (SCLC) we investigated the immune landscape of NEPC in relation to prostate cancer subtypes and SCLC to identify potential targets.

Methods: We evaluated RNA-seq from a 190 patient cohort including benign prostate (n=29; 25 PCa matched), localized prostate adenocarcinoma (PCa; n=68), hormone-naïve metastatic prostate adenocarcinoma (mPCa; n=11), CRPC (n=54), NEPC (n=25; 11 de novo) with follow-up data, and SCLC (n=28) (Rudin et al., Nat Gen 2012). Additionally, 234 prostate cancer patients had tumor mutational burden (TMB) determined by WES. Unsupervised clustering of FPKMs was performed to identify a 232 gene, immune-rich cluster, used to categorize immune status and prioritize validation of targets by IHC.

Results: Prostate cancer is known to have a relatively low TMB. Similarly, the median TMB of NEPC is akin to CRPC (38.0 vs 37.0 p = 0.44) while significantly lower than SCLC (38.0 vs 142.5, p <0.001). Unsupervised assessment identified a predominantly ‘cold’ immune status across subtypes, with ‘hot’ tumors (n=8) associated with metastatic tumors of the LN and bone and ‘intermediate’ NEPC tumors (n=9) associated with de novo cases. Worse overall survival was associated with intermediate vs cold T-cell immune status (66.5 mo vs 101.5 mo; p <0.001). Further analysis of NEPC showed lower expression of cytokines (p<0.01) as well as variation in checkpoint markers. Specifically, NEPC tumors had significantly lower expression of PD1 in relation to CRPC (p = 0.0001) and SCLC (p = <0.0001), higher PDL1 expression than CRPC (p = 0.05) but comparable with SCLC (p = 0.93) and lower PDL2 expression than PCa and SCLC (p = 0.03; <0.001, respectively).

Conclusion: NEPC is characterized by a relatively ‘cold’ tumor immune microenvironment similar to other metastatic prostate cancer subtypes but higher PDL1 expression comparable to SCLC. Association of colder tumors with treatment-induced disease, inverse correlation between survival outcome and immune infiltration, as well as novel expression changes in cytokines and checkpoint markers support further investigation into the immune landscape and potential targets for NEPC.

Citation Format: Alison M. Ferguson, Bhavneet Bhinder, Vincenza Conteduca, Michael Sigouros, Andrea Sboner, David Nanus, Scott Tagawa, David Rickman, Olivier Elemento, Himisha Beltran. Immunogenomic landscape of neuroendocrine prostate cancer (NEPC) [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 134.

Immunogenomic landscape of neuroendocrine prostate cancer (NEPC)

224

Background: Prostate cancer (PCa) shows limited clinical benefit from current immunotherapy strategies. NEPC is a histologic subtype of advanced PCa that most often arises clonally from castrate resistant prostate adenocarcinoma (CRPC) as a mechanism of resistance, but shares pathologic, clinical, and molecular features with small cell lung carcinoma (SCLC). We investigated the immune landscape of NEPC in relation to other PCa subtypes and SCLC to identify potential immunological targets. Methods: We evaluated RNA-seq from 190 patients comprising benign prostate (n = 29; 25 PCa matched), localized PCa (n = 68), hormone-naïve metastatic prostate adenocarcinoma (mPCa; n = 11), CRPC (n = 54), NEPC (n = 25) with follow-up data, and SCLC (n = 28) (Rudin et al., Nat Gen 2012). Additionally, 290 patients had WES data available for tumor mutational burden (TMB). Unsupervised clustering of FPKMs was performed to identify an immune rich cluster of 232 genes, which was used to categorize immune status and prioritize validation of select targets by IHC. Results: Median TMB of NEPC was similar to CRPC (38.0 vs 37.0 p= 0.44) but significantly lower than SCLC (38.0 vs 142.5, p< 0.001). Unsupervised assessment of T-cell related gene expression identified a predominantly cold immune status across subtypes, with hot (n = 8) tumors associated with metastatic tumors of the LN and bone. Worse overall survival was seen with intermediate vs cold T-cell immune status (66.5 mo vs 101.5 mo; p < 0.001). Further analysis of NEPC showed lower expression of cytokines ( p< 0.01) and variation in checkpoint markers. Specifically, NEPC had significantly lower expression of PD1 in relation to CRPC ( p= 0.0001) and SCLC ( p = < 0.0001), higher PDL1 than CRPC ( p= 0.05) but comparable with SCLC ( p= 0.93) and lower PDL2 than PCa and SCLC ( p= 0.03; < 0.001, respectively). Conclusions: NEPC is characterized by a relatively ‘cold’ tumor immune microenvironment similar to other metastatic prostate cancer subtypes but higher PDL1 expression comparable to SCLC. The inverse correlation between survival outcome and immune infiltration, as well as the novel expression changes in cytokines and checkpoint markers support further investigation into the immune landscape and potential targets for NEPC.

Abstract PR08: The N-Myc transcriptional program driving the neuroendocrine prostate cancer phenotype

Emerging observations from clinical trials suggest that a subset of castration resistant prostate adenocarcinomas (CRPC) eventually evolve or progress to a predominantly small cell carcinoma or neuroendocrine phenotype [1]. Neuroendocrine prostate cancer (NEPC) does not typically express the androgen receptor (AR) or secrete prostate specific antigen (PSA), and often expresses markers of neuroendocrine lineage [2]. NEPC is clinically aggressive and carries a poor prognosis with an average survival of less than one year [3, 4]. Identification of effective treatment strategies for this lethal subtype of prostate cancer represents a significant unmet need in the clinic. We have previously discovered significant over-expression and gene amplification of AURKA (encoding Aurora-A) and MYCN (encoding N-Myc) in NEPC as compared to prostate adenocarcinoma [5, 6]. As in neuroblastoma [7], N-Myc interacts with Aurora-A in NEPC which leads to a co-stabilization of both proteins and that ectopic expression of N-Myc or Aurora-A induces neuroendocrine transformation of prostate adenocarcinoma cells [5]. However, the molecular mechanisms that underlie N-Myc driven NEPC have yet to be characterized. We performed RNA-sequencing (RNAseq) and ChIP-sequencing from multiple stable prostate adenocarcinoma cells with and without N-Myc over-expression. RNAseq reads were aligned to the hg19 reference genome using TopHat. Based on GSEA , pathway analysis and further validation of our N-Myc signature, we found that N-Myc is recruited to AR-bound enhancers and AR target genes, dramatically reduces AR signaling and induces a profile enriched in pro-metastatic, dedifferentiation and Polycomb Repressive Complex deregulated genes as well as other genes encoding targetable proteins. RNAseq data from 128 clinical samples (17 NEPC, 10 castrate resistant prostate cancer, 68 prostate adenocarcinoma patient tumors and 33 matched benign prostate samples) were used to assess clinical relevance and a novel Nanostring assay, targeted RT-PCR and ChIP-PCR to validate our findings. We used DESeq and MISO to identify differential expressed genes and differential exon expression. We have identified a signature of deregulated genes including specific alternatively spliced mRNA variants and splicing factors associated with N-Myc over-expression that are both biologically and clinically relevant. This included several variant transcripts from genes that are implicated in cancer-related signaling (e.g.PIK3C2A,ATM,HUWE1) and an up-regulation of the splicing factor NOVA1 (neuro-oncological ventral antigen 1). In conclusion, our findings have the potential to ultimately lead to the identification of a new class of disease specific biomarkers and therapeutic alternatives for this aggressive subgroup of prostate cancer.

References:

1. Beltran, H., S. Tomlins, et al., Clin Cancer Res, 2014. 20(11): p. 2846-50.

2. Wang, W. and J.I. Epstein, Am J Surg Pathol, 2008. 32(1): p. 65-71.

3. Palmgren, J.S., S.S. Karavadia, et al., Semin Oncol, 2007. 34(1): p. 22-9.

4. Wang, H.T., Y.H. Yao, et al., J Clin Oncol, 2014.

5. Beltran, H., D.S. Rickman, et al., Cancer Discovery, 2011. 1(6): p. 487-495.

6. Mosquera, J.M., H. Beltran, et al., Neoplasia, 2013. 15(1): p. 1-10.

7. Otto, T., S. Horn, et al., Cancer Cell, 2009. 15(1): p. 67-78.

Citation Format: Etienne Dardenne, Kaitlyn Gayert, Andrea Sboner, Cynthia Cheung, Martin Eilers, Mark Rubin, Himisha Beltran, Olivier Elemento, David Rickman. The N-Myc transcriptional program driving the neuroendocrine prostate cancer phenotype. [abstract]. In: Proceedings of the AACR Special Conference on Myc: From Biology to Therapy; Jan 7-10, 2015; La Jolla, CA. Philadelphia (PA): AACR; Mol Cancer Res 2015;13(10 Suppl):Abstract nr PR08.

Abstract 3688: Target identification for anticancer molecules using a Big Data approach

Identifying the targets of bioactive compounds has been a major challenge, with efforts being driven by case specific experimentation - a slow and failure-prone process. Various methods have attempted to facilitate discovery, but frequently rely on unavailable cohorts of known binding ligands or complex 3D structures. On the other hand, there has recently been an increase of large-scale genomic, chemical, and structural datasets. By integrating these, molecules can be represented as over 100,00 unique data points, and we hypothesized that combining these various sources into a comprehensive prediction engine could radically improve our ability to predict the drug targets and identify novel anticancer compounds.

To test this hypothesis, we developed BANDIT - a Bayesian Approach to find Novel Drug Interaction Targets. BANDIT integrates datasets on drug efficacies, post-treatment transcriptional responses, drug structures, known adverse effects, and bioassay sensitivities, in order to compute an overall Bayesian likelihood and predict drugs that may share a biological target. When applied to a test set of known drugs, BANDIT achieved a predictive power of 89% at identifying drug pairs known to share a target (AUROC = 0.89). Across these datasets we observed that the most predictive were structural similarity and similar survival responses across various cell lines. Moreover the power of the classifier steadily increased from 60% to 89% as the number of included data types was increased- indicating the strength of BANDIT's Big Data approach.

We then used BANDIT to determine drugs that could be used for cancer therapy. Specifically, we looked for non-cancer drugs predicted to share a target with known anticancer drugs and for novel targets of known anticancer drugs. We predicted two common cancer drugs, Resveratrol and Genistein, to share a target (Likelihood ratio (LR) = 79.8; Top .05% of predictions). Studies have shown that resveratrol enhances the apoptotic effect of Genistein and our prediction reveals that a possible mechanism for the additive effect could be the dual inhibition of a single target. We also predicted the unreported potential of Vismodegib - used to inhibit the Hedgehog signaling pathway in basal cell carcinoma - to act as a tyrosine kinase inhibitor (LR = 348.1; Top .01% of predictions). Additionally, microtubule-targeting drugs have been important chemotherapy agents, and we predicted Mebendazole and Romidepsin to both inhibit tubulin formation (Likelihood ratios = 178.3 & 210.1 respectively; Top .02% of predictions), thus presenting the possibility for them to be used in an anti-tubulin chemotherapeutic context.

Altogether, BANDIT provides a novel, broadly applicable way to identify novel targets for new and established drugs and could greatly expedite pharmaceutical research. Moreover, it could significantly impact therapeutic decisions by determining drugs that could be repositioned to target important cancer regulatory proteins.

Citation Format: Neel S. Madhukar, Linda Huang, Kaitlyn Gayvert, David Rickman, Olivier Elemento. Target identification for anticancer molecules using a Big Data approach. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3688. doi:10.1158/1538-7445.AM2015-3688

Abstract 4745: Precision cancer medicine program for whole-exome sequencing of metastatic tumors reveals biomarkers of response

Background: Understanding molecular mechanisms of response and resistance to anticancer therapies requires prospective patient follow-up and clinical and functional validation of both common and low frequency mutations. We describe an evidence-based precision medicine program designed to bring whole exome sequencing into clinical practice and illustrate how this can be utilized to identify novel predictive biomarkers associated with exceptional systemic response.

Methods: Metastatic and treatment-resistant cancer patients are prospectively enrolled under an IRB approved protocol for paired metastatic tumor and normal WES through our Precision Medicine Clinic. Using a comprehensive computational pipeline, point mutations, indels and copy number alterations are detected. Mutations are categorized as Category I-III based on actionability and a report is generated and discussed in tumor board. Patients are followed prospectively for correlation of molecular information with clinical response and patient outcomes. A rapid response functional team follows on key findings to expand actionability based on the WES results.

Results: In the first year, 154 tumor-normal pairs from 97 patients were sequenced, with an average coverage of 95X of over 21,000 genes and 16 somatic alterations detected on average per patient. In total, 16 mutations were Category I, 98 were Category II, and 1474 were Category III. Tumor purity ranged from 14 to near 100%. Among unexpected findings, a prostate cancer exceptional responder was identified that harbored a somatic hemizygous deletion of the DNA repair gene FANCA and likely loss of function of the second allele through germline missense variant. Follow-up experiments established that loss of FANCA function was associated with platinum hypersensitivity both in vitro and in patient-derived xenografts, thus providing biologic rationale and functional evidence for his extreme clinical response.

Conclusions: The establishment of a clinical program for whole exome sequencing of metastatic tumors with prospective follow-up of patients can help identify candidate predictive biomarkers of response.

Citation Format: Himisha Beltran, Kenneth Eng, Juan Miguel Mosquera, Alexandros Sigaras, Alessandro Romanel, Hanna Rennart, Myriam Kossai, Chantal Pauli, Bishoy Faltas, Jacqueline Fontugne, Brian Robinson, David M. Nanus, Scott T. Tagawa, Jenny Z. Xiang, Francesca Demichelis, David Rickman, Andrea Sboner, Olivier Elemento, Mark A. Rubin. Precision cancer medicine program for whole-exome sequencing of metastatic tumors reveals biomarkers of response. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4745. doi:10.1158/1538-7445.AM2015-4745

Precision medicine program for whole-exome sequencing (WES) provides new insight on platinum sensitivity in advanced prostate cancer (PCa)

158

Background: WES has provided insight into the genomic landscape of PCa. The next step for precision medicine requires prospective patient (pt) follow-up and clinical and functional validation of both common and low frequency mutations. We describe a precision medicine WES program and illustrate how this can be used to identify novel biomarkers associated with response. Methods: Metastatic PCa pts were prospectively enrolled. WES of metastatic biopsies or rapid autopsies and normal DNA were sequenced by Illumina HiSeq 2500. Results: Tumor-normal pairs from 71 pts with metastatic PCa (11 hormone naïve, 38 CRPC, 23 NEPC) were sequenced including 3 rapid autopsies, with a biopsy success rate of >95%, avg tumor purity 10-95%, avg coverage 85X. The spectrum of mutations across metastatic PCa and how serial biopsies and rapid autopsies were used to assess heterogeneity and clonal evolution will be presented. WES from a metastatic PCa pt (PM12) with exceptional response to platinum with complete remission of liver and lung metastases at 2 years follow-up, demonstrated a hypermutated genotype and hemizygous deletion of the DNA repair gene FANCA in both his primary and metastatic PCa. A loss of function germline variant was detected within the second allele, and only the mutated allele was expressed in his tumor. Genome editing of FANCA using CRISPR in PCa cell lines resulted in cisplatin hypersensitivity and a significant decrease in FANC complex formation. In patient derived xenograft (PDX) models, PM12’s PDX was significantly more sensitive to cisplatin compared to a control Pca PDX of similar morphology but lacking FANCA deletion. By screening larger cohorts, FANCA loss was detected by FISH in 16% of localized PCa (n= 69), 14% metastatic Pca (n= 29), and not detected in any benign prostate (n=69). Conclusions: Our study provides a proof of principle for developing a precision medicine approach to cancer care with the capability to identify potential biomarkers of response. Our findings suggest that a subset of PCa with FANCA loss may be particularly vulnerable to cytotoxic therapy and provides biologic rationale to help explain the exceptional response of pt PM12 to platinum chemotherapy.

Aggressive variants of castration-resistant prostate cancer

A subset of patients with advanced castration-resistant prostate cancer may eventually evolve into an androgen receptor (AR)-independent phenotype, with a clinical picture associated with the development of rapidly progressive disease involving visceral sites and hormone refractoriness, often in the setting of a low or modestly rising serum prostate-specific antigen level. Biopsies performed in such patients may vary, ranging from poorly differentiated carcinomas to mixed adenocarcinoma-small cell carcinomas to pure small cell carcinomas. These aggressive tumors often demonstrate low or absent AR protein expression and, in some cases, express markers of neuroendocrine differentiation. Because tumor morphology is not always predicted by clinical behavior, the terms "anaplastic prostate cancer" or "neuroendocrine prostate cancer" have been used descriptively to describe these rapidly growing clinical features. Patients meeting clinical criteria of anaplastic prostate cancer have been shown to predict for poor prognosis, and these patients may be considered for platinum-based chemotherapy treatment regimens. Therefore, understanding variants within the spectrum of advanced prostate cancer has important diagnostic and treatment implications.

Biobanking after robotic-assisted radical prostatectomy: a quality assessment of providing prostate tissue for RNA studies

BACKGROUND

RNA quality is believed to decrease with ischaemia time, and therefore open radical prostatectomy has been advantageous in allowing the retrieval of the prostate immediately after its devascularization. In contrast, robotic-assisted laparoscopic radical prostatectomies (RALP) require the completion of several operative steps before the devascularized prostate can be extirpated, casting doubt on the validity of this technique as a source for obtaining prostatic tissue. We seek to establish the integrity of our biobanking process by measuring the RNA quality of specimens derived from robotic-assisted laparoscopic radical prostatectomy.

METHODS

We describe our biobanking process and report the RNA quality of prostate specimens using advanced electrophoretic techniques (RNA Integrity Numbers, RIN). Using multivariate regression analysis we consider the impact of various clinicopathological correlates on RNA integrity.

RESULTS

Our biobanking process has been used to acquire 1709 prostates, and allows us to retain approximately 40% of the prostate specimen, without compromising the histopathological evaluation of patients. We collected 186 samples from 142 biobanked prostates, and demonstrated a mean RIN of 7.25 (standard deviation 1.64) in 139 non-stromal samples, 73% of which had a RIN ≥ 7. Multivariate regression analysis revealed cell type--stromal/epithelial and benign/malignant--and prostate volume to be significant predictors of RIN, with unstandardized coefficients of 0.867(p = 0.001), 1.738(p < 0.001) and -0.690(p = 0.009) respectively. A mean warm ischaemia time of 120 min (standard deviation 30 min) was recorded, but multivariate regression analysis did not demonstrate a relationship with RIN within the timeframe of the RALP procedure.

CONCLUSIONS

We demonstrate the robustness of our protocol--representing the concerted efforts of dedicated urology and pathology departments--in generating RNA of sufficient concentration and quality, without compromising the histopathological evaluation and diagnosis of patients. The ischaemia time associated with our prostatectomy technique using a robotic platform does not negatively impact on biobanking for RNA studies.

Abstract 957: Aurora kinase and N-myc are involved in neuroendocrine differentiation of prostate cancer and are new drug targets

Background: Neuroendocrine prostate cancer (NEPC) is an aggressive variant of prostate cancer that can arise de novo or from existing prostate adenocarcinoma (PCA), and is promoted by the use of androgen deprivation therapy. We sought to better understand the molecular transformation of NEPC and identify new drug targets.

Methods: Using Next Generation RNA sequencing and oligonucleotide arrays, we profiled NEPC, PCA, and benign prostate (BEN), and validated findings on tumors from a large cohort of patients using IHC and FISH. Functional studies were performed using NCI-H660 (NEPC), VCaP and LnCaP (PCA), RWPE (BEN) cell lines and xenografts.

Results: There was a spectrum of NEPC, ranging from pure NEPC to those with mixed features of PCA and NEPC. ERG rearrangement was detected by FISH break-apart in 47% of NEPC, but ERG protein expression by IHC was absent and corresponded directly with lack of androgen receptor expression. We sequenced 7 NEPC, 30 PCA, 5 BEN. There were significant molecular differences between NEPC and PCA, with 936/25932 genes differentially expressed (P&lt;0.001). The cell cycle kinases Aurora kinase A and B (AURKA, AURKB) were overexpressed in NEPC, and notably this was independent of cell proliferation (Ki67 expression). N-myc (MYCN), a gene frequently amplified in neuroblastoma, was overexpressed in NEPC (P&lt;0.001), and both AURKA and MYCN were amplified. Using IHC and FISH, we validated these findings on a large cohort (30 BEN, 118 PCA, 30 NEPC) and found majority (&gt;80%) of NEPC showed Aurora kinase A and B overexpression, and 35% harbored AURKA and co-existent MYCN amplification. A small subset of PCA (5%) and no BEN were positive. Transfection of MYCN induced AURKA expression as well as Aurora kinase activity in RWPE in vitro, and AURKA could also induce MYCN expression. MYCN and AURKA independently induced expression of the neuroendocrine (NE) markers, SYP and NSE. After validating NCI-H660 as model of NEPC, we observed enhanced in vitro and in vivo sensitivity to the Aurora kinase inhibitor PHA-739358 compared to LnCaP and VCaP, with &gt; 50% tumor shrinkage in NEPC xenografts and minimal to no effect in PCA. Phosphorylated histone 3 expression, a downstream marker of Aurora kinase activity, was significantly inhibited in the treated NCI-H660 but not in treated PCA xenografts. Notably, NE marker expression was completely suppressed in the treated NCI-H660 xenografts, again supporting a role of Aurora kinase in modulating the NE phenotype.

Conclusions: There is likely clonal origin of NEPC from PCA (with ERG fusion positivity seen in both), but ERG expression is limited to PCA and driven by AR signaling. We discovered significant overexpression and gene amplification of Aurora kinases and N-myc in NEPC and a small subset of PCA, and evidence that that they cooperate and induce a NE phenotype in prostate cells. In vitro and in vivo data confirms that these are novel drug targets for NEPC.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 957. doi:10.1158/1538-7445.AM2011-957

Molecular characterization of neuroendocrine prostate cancer (NEPC) and identification of new drug targets

19

Background: NEPC is an aggressive variant of prostate cancer that can arise de novo or from existing prostate adenocarcinoma (PCA). We sought to better understand the molecular transformation of NEPC and identify new drug targets.

Methods: We used Next Generation RNA sequencing and oligonucleotide arrays to profile 7 NEPC, 30 PCA, 5 benign prostate (BEN), and validated findings on tumors from a large cohort of patients (30 NEPC, 118 PCA, 30 BEN) using IHC and FISH. Functional studies were performed using NCI-H660 (NEPC), VCaP and LnCaP (PCA), RWPE (BEN).

Results: ERG rearrangement was present in 47% of NEPC, but ERG protein expression was absent and corresponded directly with lack of AR expression. 936/25932 genes were differentially expressed in NEPC versus PCA (P<0.001). Aurora kinases (AURKA, AURKB) and N-myc (MYCN) were overexpressed in NEPC (P<0.001) and AURKA and MYCN amplified. Using IHC and FISH, we validated these findings on a large cohort and found majority (>80%) of NEPC showed Aurora overexpression, 35% had AURKA and MYCN amplification. A small subset of PCA (5%) and no BEN were positive. Transfection of MYCN induced AURKA expression and kinase activity in vitro, and MYCN or AURKA could induce expression of neuroendocrine (NE) markers (SYP, NSE). After validating NCI-H660 as model of NEPC, we observed dramatic and enhanced in vitro and in vivo sensitivity to the Aurora kinase inhibitor PHA-739358 in NCI-H660 compared to minimal to no effect in LnCaP and VCaP. Phospho-H3 expression, a downstream marker of Aurora kinase activity, was inhibited in the treated NCI-H660 and not in PCA. Notably, NE marker expression was also suppressed in the treated NCI-H660 xenografts, again supporting a role of Aurora kinase in modulating the NE phenotype.

Conclusions: There is likely clonal origin of NEPC from PCA (with ERG fusion positivity seen in both), but ERG expression is limited to PCA and driven by AR signaling. We discovered significant overexpression and gene amplification of Aurora kinases and N-myc in NEPC and a small subset of PCA, and evidence that that they cooperate and induce a NE phenotype in prostate cells. In vitro and in vivo data confirms that these are novel drug targets for NEPC.

No significant financial relationships to disclose.

An ANOCEF genomic and transcriptomic microarray study of the response to radiotherapy or to alkylating first-line chemotherapy in glioblastoma patients

BACKGROUND

The molecular characteristics associated with the response to treatment in glioblastomas (GBMs) remain largely unknown. We performed a retrospective study to assess the genomic characteristics associated with the response of GBMs to either first-line chemotherapy or radiation therapy. The gene expression (n = 56) and genomic profiles (n = 67) of responders and non-responders to first-line chemotherapy or radiation therapy alone were compared on Affymetrix Plus 2 gene expression arrays and BAC CGH arrays.

RESULTS

According to Verhaak et al.'s classification system, mesenchymal GBMs were more likely to respond to radiotherapy than to first-line chemotherapy, whereas classical GBMs were more likely to respond to first-line chemotherapy than to radiotherapy. In patients treated with radiation therapy alone, the response was associated with differential expression of microenvironment-associated genes; the expression of hypoxia-related genes was associated with short-term progression-free survival (< 5 months), whereas the expression of immune genes was associated with prolonged progression-free survival (> 10 months). Consistently, infiltration of the tumor by both CD3 and CD68 cells was significantly more frequent in responders to radiotherapy than in non-responders. In patients treated with first-line chemotherapy, the expression of stem-cell genes was associated with resistance to chemotherapy, and there was a significant association between response to treatment and p16 locus deletions. Consistently, in an independent data set of patients treated with either radiotherapy alone or with both radiotherapy and adjuvant chemotherapy, we found that patients with the p16 deletion benefited from adjuvant chemotherapy regardless of their MGMT promoter methylation status, whereas in patients without the p16 deletion, this benefit was only observed in patients with a methylated MGMT promoter.

CONCLUSION

Differential expression of microenvironment genes and p16 locus deletion are associated with responses to radiation therapy and to first-line chemotherapy, respectively, in GBM. Recently identified transcriptomic subgroups of GBMs seem to respond differently to radiotherapy and to first-line chemotherapy.

Biological and clinical relevance of transcriptionally active human papillomavirus (HPV) infection in oropharynx squamous cell carcinoma

Human papillomaviruses (HPV) are associated with a subset of head and neck squamous cell carcinoma (HNSCC), particularly HPV16. This study analyzed the presence and genotype of high risk HPVs, viral DNA load and transcription of the E6/E7 mRNAs, in 231 consecutive HNSCC. Twelve out of 30 HPV16 DNA-positive tumors displayed high E6/E7 mRNAs levels and were localized in the oropharyngeal region. While HPV-free and non-transcriptionally active HPV-related patients showed similar 5-years survival rates, E6/E7 expression was associated with a better prognosis. This emphasizes the importance of considering the transcriptional status of HPV-positive tumors for patient stratification. A gene expression profiling analysis of these different types of tumors was carried out. The most significant differentially expressed gene was CDKN2A, a known biomarker for HPV-related cancer. Assessing both the expression level of the E6/E7 mRNAs and of CDKN2A in HNSCC is required to detect active HPV infection. Chromosomic alterations were investigated by Comparative Genomic Hybridation (CGH) analysis of tumors with transcriptionally active HPV and HPV-negative tumors. The loss of the chromosomal region 16q was found to be a major genetic event in HPV-positive lesions. A cluster of genes located in 16q21-24 displayed decreased expression levels, notably APP-BP1 that is involved in the modulation of the transcriptional activity of p53. In conclusion, this study highlights important criteria required to predict clinically active HPV infection, identifies new biological pathways implicated in HPV tumorigenesis and increases the understanding of HPV-HNSCC physiopathology that is required to develop new targets for therapy.

Hepatic stem-like phenotype and interplay of Wnt/beta-catenin and Myc signaling in aggressive childhood liver cancer

Hepatoblastoma, the most common pediatric liver cancer, is tightly linked to excessive Wnt/beta-catenin signaling. Here, we used microarray analysis to identify two tumor subclasses resembling distinct phases of liver development and a discriminating 16-gene signature. beta-catenin activated different transcriptional programs in the two tumor types, with distinctive expression of hepatic stem/progenitor markers in immature tumors. This highly proliferating subclass was typified by gains of chromosomes 8q and 2p and upregulated Myc signaling. Myc-induced hepatoblastoma-like tumors in mice strikingly resembled the human immature subtype, and Myc downregulation in hepatoblastoma cells impaired tumorigenesis in vivo. Remarkably, the 16-gene signature discriminated invasive and metastatic hepatoblastomas and predicted prognosis with high accuracy.

Microarray-based method for combinatorial library sequence mapping and characterization

Here we describe a DNA-chip-based method for high-throughput sequence mapping. This involves competitive hybridization between short and differentially labeled fluorescent oligonucleotide probes and glass-supported PCR products. Competition between an excess of oligonucleotide probes targeting the same sequence segment improves sequence discrimination and reduces sensitivity to experimental conditions such as probe concentrations, hybridization, and washing temperatures and durations. The method was found to be particularly adapted to sequence mapping of combinatorial libraries obtained by DNA shuffling between members of a gene family. We present an application of this technique for the characterization of recombination biases in combinatorial libraries used in directed evolution.