Tumour genomic and microenvironmental heterogeneity for integrated prediction of 5-year biochemical recurrence of prostate cancer: a retrospective cohort study.

BACKGROUND: Clinical prognostic groupings for localised prostate cancers are imprecise, with 30-50% of patients recurring after image-guided radiotherapy or radical prostatectomy. We aimed to test combined genomic and microenvironmental indices in prostate cancer to improve risk stratification and complement clinical prognostic factors. METHODS: We used DNA-based indices alone or in combination with intra-prostatic hypoxia measurements to develop four prognostic indices in 126 low-risk to intermediate-risk patients (Toronto cohort) who will receive image-guided radiotherapy. We validated these indices in two independent cohorts of 154 (Memorial Sloan Kettering Cancer Center cohort [MSKCC] cohort) and 117 (Cambridge cohort) radical prostatectomy specimens from low-risk to high-risk patients. We applied unsupervised and supervised machine learning techniques to the copy-number profiles of 126 pre-image-guided radiotherapy diagnostic biopsies to develop prognostic signatures. Our primary endpoint was the development of a set of prognostic measures capable of stratifying patients for risk of biochemical relapse 5 years after primary treatment. FINDINGS: Biochemical relapse was associated with indices of tumour hypoxia, genomic instability, and genomic subtypes based on multivariate analyses. We identified four genomic subtypes for prostate cancer, which had different 5-year biochemical relapse-free survival. Genomic instability is prognostic for relapse in both image-guided radiotherapy (multivariate analysis hazard ratio [HR] 4·5 [95% CI 2·1-9·8]; p=0·00013; area under the receiver operator curve [AUC] 0·70 [95% CI 0·65-0·76]) and radical prostatectomy (4·0 [1·6-9·7]; p=0·0024; AUC 0·57 [0·52-0·61]) patients with prostate cancer, and its effect is magnified by intratumoral hypoxia (3·8 [1·2-12]; p=0·019; AUC 0·67 [0·61-0·73]). A novel 100-loci DNA signature accurately classified treatment outcome in the MSKCC low-risk to intermediate-risk cohort (multivariate analysis HR 6·1 [95% CI 2·0-19]; p=0·0015; AUC 0·74 [95% CI 0·65-0·83]). In the independent MSKCC and Cambridge cohorts, this signature identified low-risk to high-risk patients who were most likely to fail treatment within 18 months (combined cohorts multivariate analysis HR 2·9 [95% CI 1·4-6·0]; p=0·0039; AUC 0·68 [95% CI 0·63-0·73]), and was better at predicting biochemical relapse than 23 previously published RNA signatures. INTERPRETATION: This is the first study of cancer outcome to integrate DNA-based and microenvironment-based failure indices to predict patient outcome. Patients exhibiting these aggressive features after biopsy should be entered into treatment intensification trials. FUNDING: Movember Foundation, Prostate Cancer Canada, Ontario Institute for Cancer Research, Canadian Institute for Health Research, NIHR Cambridge Biomedical Research Centre, The University of Cambridge, Cancer Research UK, Cambridge Cancer Charity, Prostate Cancer UK, Hutchison Whampoa Limited, Terry Fox Research Institute, Princess Margaret Cancer Centre Foundation, PMH-Radiation Medicine Program Academic Enrichment Fund, Motorcycle Ride for Dad (Durham), Canadian Cancer Society.

A tiling resolution DNA microarray with complete coverage of the human genome.

We constructed a tiling resolution array consisting of 32,433 overlapping BAC clones covering the entire human genome. This increases our ability to identify genetic alterations and their boundaries throughout the genome in a single comparative genomic hybridization (CGH) experiment. At this tiling resolution, we identified minute DNA alterations not previously reported. These alterations include microamplifications and deletions containing oncogenes, tumor-suppressor genes and new genes that may be associated with multiple tumor types. Our findings show the need to move beyond conventional marker-based genome comparison approaches, that rely on inference of continuity between interval markers. Our submegabase resolution tiling set for array CGH (SMRT array) allows comprehensive assessment of genomic integrity and thereby the identification of new genes associated with disease.

Allelic loss of the loci containing the androgen synthesis gene, StAR, is prognostic for relapse in intermediate-risk prostate cancer.

BACKGROUND: Androgen deprivation therapy (ADT) and novel agents targeting the androgen synthesis axis (e.g., abiraterone acetate) are adjuvant therapies that are currently, or may in the future be, combined with radiotherapy to reduce the chance of disease relapse. Little is known about allelic loss or gain pertaining to genes associated with the androgen synthesis axis and whether this is prognostic in patients who receive localized radiotherapy. In this hypothesis generating study, we conducted an array comparative genomic hybridization (aCGH) analysis of 33 androgen synthesis genes to identify potential prognostic factors for radiotherapy outcome. METHODS: aCGH analysis of tumor DNA prospectively derived from frozen needle biopsies of 126 men with intermediate-risk disease who underwent image-guided radiotherapy (IGRT) to a mean dose of 76.4 Gy was conducted. Statistical analyses were conducted for allelic loss or gain in genes as potential prognostic factors relative to prostate specific antigen, Gleason-score, and T-category. RESULTS: We observed that allelic losses of loci containing the genes StAR and HSD17B2 were associated with increased genetic instability (as determined by percentage genome alteration). On multivariate analyses these loci were prognostic for biochemical disease-free relapse (StAR: HR = 2.84, 95% CI: 1.44-5.61, P = 0.00269; HSD17B2: HR = 1.97, 95% CI: 1.06-3.64, P = 0.031). The results were validated in a surgical cohort of 131 intermediate-risk patients. CONCLUSIONS: Allelic losses of the loci containing StAR and HSD17B2 have significant prognostic value for intermediate-risk prostate cancer. With this hypothesis generating information future studies should test StAR and HSD17B2 losses as biomarkers of androgen response in combined modality protocols.

Copy number alterations of c-MYC and PTEN are prognostic factors for relapse after prostate cancer radiotherapy.

Despite the use of PSA, Gleason score, and T-category as prognosticators in intermediate-risk prostate cancer, 20-40% of patients will fail local therapy. In order to optimize treatment approaches for intermediate-risk patients, additional genetic prognosticators are needed. Previous reports using array comparative genomic hybridization (aCGH) in radical prostatectomy cohorts suggested a combination of allelic loss of the PTEN gene on 10q and allelic gain of the c-MYC gene on 8q were associated with metastatic disease. We tested whether copy number alterations (CNAs) in PTEN (allelic loss) and c-MYC (allelic gain) were associated with biochemical relapse following modern-era, image-guided radiotherapy (mean dose 76.4 Gy). We used aCGH analyses validated by fluorescence in-situ hybridization (FISH) of DNA was derived from frozen, pre-treatment biopsies in 126 intermediate-risk prostate cancer patients. Patients whose tumors had CNAs in both PTEN and c-MYC had significantly increased genetic instability (percent genome alteration; PGA) compared to tumors with normal PTEN and c-MYC status (p < 0.0001). We demonstrate that c-MYC gain alone, or combined c-MYC gain and PTEN loss, were increasingly prognostic for relapse on multivariable analyses (hazard ratios (HR) of 2.58/p = 0.005 and 3.21/p = 0.0004; respectively). Triaging patients by the use of CNAs within pre-treatment biopsies may allow for better use of systemic therapies to target sub-clinical metastases or locally recurrent disease and improve clinical outcomes.

Integrating the multiple dimensions of genomic and epigenomic landscapes of cancer.

Advances in high-throughput, genome-wide profiling technologies have allowed for an unprecedented view of the cancer genome landscape. Specifically, high-density microarrays and sequencing-based strategies have been widely utilized to identify genetic (such as gene dosage, allelic status, and mutations in gene sequence) and epigenetic (such as DNA methylation, histone modification, and microRNA) aberrations in cancer. Although the application of these profiling technologies in unidimensional analyses has been instrumental in cancer gene discovery, genes affected by low-frequency events are often overlooked. The integrative approach of analyzing parallel dimensions has enabled the identification of (a) genes that are often disrupted by multiple mechanisms but at low frequencies by any one mechanism and (b) pathways that are often disrupted at multiple components but at low frequencies at individual components. These benefits of using an integrative approach illustrate the concept that the whole is greater than the sum of its parts. As efforts have now turned toward parallel and integrative multidimensional approaches for studying the cancer genome landscape in hopes of obtaining a more insightful understanding of the key genes and pathways driving cancer cells, this review describes key findings disseminating from such high-throughput, integrative analyses, including contributions to our understanding of causative genetic events in cancer cell biology.

Integrative genomic analyses identify BRF2 as a novel lineage-specific oncogene in lung squamous cell carcinoma.

BACKGROUND: Traditionally, non-small cell lung cancer is treated as a single disease entity in terms of systemic therapy. Emerging evidence suggests the major subtypes--adenocarcinoma (AC) and squamous cell carcinoma (SqCC)--respond differently to therapy. Identification of the molecular differences between these tumor types will have a significant impact in designing novel therapies that can improve the treatment outcome. METHODS AND FINDINGS: We used an integrative genomics approach, combing high-resolution comparative genomic hybridization and gene expression microarray profiles, to compare AC and SqCC tumors in order to uncover alterations at the DNA level, with corresponding gene transcription changes, which are selected for during development of lung cancer subtypes. Through the analysis of multiple independent cohorts of clinical tumor samples (>330), normal lung tissues and bronchial epithelial cells obtained by bronchial brushing in smokers without lung cancer, we identified the overexpression of BRF2, a gene on Chromosome 8p12, which is specific for development of SqCC of lung. Genetic activation of BRF2, which encodes a RNA polymerase III (Pol III) transcription initiation factor, was found to be associated with increased expression of small nuclear RNAs (snRNAs) that are involved in processes essential for cell growth, such as RNA splicing. Ectopic expression of BRF2 in human bronchial epithelial cells induced a transformed phenotype and demonstrates downstream oncogenic effects, whereas RNA interference (RNAi)-mediated knockdown suppressed growth and colony formation of SqCC cells overexpressing BRF2, but not AC cells. Frequent activation of BRF2 in >35% preinvasive bronchial carcinoma in situ, as well as in dysplastic lesions, provides evidence that BRF2 expression is an early event in cancer development of this cell lineage. CONCLUSIONS: This is the first study, to our knowledge, to show that the focal amplification of a gene in Chromosome 8p12, plays a key role in squamous cell lineage specificity of the disease. Our data suggest that genetic activation of BRF2 represents a unique mechanism of SqCC lung tumorigenesis through the increase of Pol III-mediated transcription. It can serve as a marker for lung SqCC and may provide a novel target for therapy. Please see later in the article for the Editors' Summary.

Methods for high throughput validation of amplified fragment pools of BAC DNA for constructing high resolution CGH arrays.

BACKGROUND: The recent development of array based comparative genomic hybridization (CGH) technology provides improved resolution for detection of genomic DNA copy number alterations. In array CGH, generating spotting solution is a multi-step process where bacterial artificial chromosome (BAC) clones are converted to replenishable PCR amplified fragments pools (AFP) for use as spotting solution in a microarray format on glass substrate. With completion of the human and mouse genome sequencing, large BAC clone sets providing complete genome coverage are available for construction of whole genome BAC arrays. Currently, Southern hybridization, fluorescent in-situ hybridization (FISH), and BAC end sequencing methods are commonly used to identify the initial BAC clone but not the end product used for spotting arrays. The AFP sequencing technique described in this study is a novel method designed to verify the identity of array spotting solution in a high throughput manner. RESULTS: We show here that Southern hybridization, FISH, and AFP sequencing can be used to verify the identity of final spotting solutions using less than 10% of the AFP product. Single pass AFP sequencing identified over half of the 960 AFPs analyzed. Moreover, using two vector primers approximately 90% of the AFP spotting solutions can be identified. CONCLUSIONS: In this feasibility study we demonstrate that current methods for identifying initial BAC clones can be adapted to verify the identity of AFP spotting solutions used in printing arrays. Of these methods, AFP sequencing proves to be the most efficient for large scale identification of spotting solution in a high throughput manner.

Two-dimensional display and whole genome comparison of bacterial pathogen genomes of high G+C DNA content.

High-resolution comparison of bacterial genomes facilitates the identification of the genetic changes responsible for clinically relevant phenotypes. For this purpose we have established a method for the display and comparison of high G+C bacterial genomes in two dimensions. Here we describe the application of two-dimensional bacterial genomic display to resolve the genomes of Bordetella pertussis, Mycobacterium avium and Mycobacterium tuberculosis, and its utility in strain comparison and detection of insertion and substitution mutations.

NKX3.1 haploinsufficiency is prognostic for prostate cancer relapse following surgery or image-guided radiotherapy.

BACKGROUND: Despite the use of prostate specific antigen (PSA), Gleason-score, and T-category as prognostic factors, up to 40% of patients with intermediate-risk prostate cancer will fail radical prostatectomy or precision image-guided radiotherapy (IGRT). Additional genetic prognosticators are needed to triage these patients toward intensified combination therapy with novel targeted therapeutics. We tested the role of the NKX3.1 gene as a determinant of treatment outcome given its reported roles in tumor initiating cell (TIC) renewal, the DNA damage response, and cooperation with c-MYC during prostate cancer progression. METHODS: Using high-resolution array comparative genomic hybridization (aCGH), we profiled the copy number alterations in TIC genes using tumor DNA from frozen needle biopsies derived from 126 intermediate-risk patients who underwent IGRT. These data were correlated to biochemical relapse-free rate (bRFR) by the Kaplan-Meier method and Cox proportional hazards models. RESULTS: A screen of the aCGH-IGRT data for TIC genes showed frequent copy number alterations for NKX3.1, PSCA, and c-MYC. NKX3.1 haploinsufficiency was associated with increased genomic instability independent of PSA, T-category, and Gleason-score. After adjusting for clinical factors in a multivariate model, NKX3.1 haploinsufficiency was associated with bRFR when tested alone (HR = 3.05, 95% CI: 1.46-6.39, P = 0.0030) or when combined with c-MYC gain (HR = 3.88, 95% CI: 1.78-8.49, P = 0.00067). A similar association was observed for patients following radical prostatectomy with a public aCGH database. NKX3.1 status was associated with positive biopsies post-IGRT and increased clonogen radioresistance in vitro. CONCLUSIONS: Our results support the use of genomic predictors, such as NKX3.1 status, in needle biopsies for personalized approaches to prostate cancer management.

Contrasting transcriptional responses of a virulent and an attenuated strain of Mycobacterium tuberculosis infecting macrophages.

BACKGROUND: H37Rv and H37Ra are well-described laboratory strains of Mycobacterium tuberculosis derived from the same parental strain, H37, that show dramatically different pathogenic phenotypes. METHODOLOGY/PRINCIPAL FINDINGS: In this study, the transcriptomes of the two strains during axenic growth in broth and during intracellular growth within murine bone-marrow macrophages were compared by whole genome expression profiling. We identified and compared adaptations of either strain upon encountering an intracellular environment, and also contrasted the transcriptomes of the two strains while inside macrophages. In the former comparison, both strains induced genes that would facilitate intracellular survival including those involved in mycobactin synthesis and fatty acid metabolism. However, this response was stronger and more extensive for H37Rv than for H37Ra. This was manifested as the differential expression of a greater number of genes and an increased magnitude of expression for these genes in H37Rv. In comparing intracellular transcriptional signatures, fifty genes were found to be differentially expressed between the strains. Of these fifty, twelve were under control of the PhoPR regulon. Further differences between strains included genes whose products were members of the ESAT-6 family of proteins, or were associated with their secretion. CONCLUSIONS/SIGNIFICANCE: Along with the recent identification of single nucleotide polymorphisms in H37Ra when compared to H37Rv, our demonstration of differential expression of PhoP-regulated and ESX-1 region-related genes during macrophage infection further highlights the significance of these genes in the attenuation of H37Ra.

Two-dimensional bacterial genome display: a method for the genomic analysis of mycobacteria.

Annually, Mycobacterium tuberculosis is the cause of approximately three million deaths worldwide. It would appear that currently available therapies for this disease are inadequate. The identification of genes involved in mycobacterial virulence will facilitate the design of new prophylactic and therapeutic interventions. A method for high-resolution comparison of bacterial genomes has been developed to facilitate the identification of genes possibly involved in the virulence of clinically relevant mycobacteria. This 'two-dimensional bacterial genome display' (2DBGD) method utilizes two-dimensional DNA electrophoresis to separate, on the basis of size and G+C content, genomic fragments generated with different restriction endonucleases. The use of this method to identify genomic differences between species, strains and, most importantly, isogenic mutants of mycobacteria is reported. That 2DBGD can be used to identify differences resulting from either insertional mutagenesis using a gentamicin-resistance gene or from a frameshift mutation is demonstrated.