A five-CpG DNA methylation score to predict metastatic-lethal outcomes in men treated with radical prostatectomy for localized prostate cancer.

BACKGROUND: Prognostic biomarkers for localized prostate cancer (PCa) could improve personalized medicine. Our group previously identified a panel of differentially methylated CpGs in primary tumor tissue that predict disease aggressiveness, and here we further validate these biomarkers. METHODS: Pyrosequencing was used to assess CpG methylation of eight biomarkers previously identified using the HumanMethylation450 array; CpGs with strongly correlated (r >0.70) results were considered technically validated. Logistic regression incorporating the validated CpGs and Gleason sum was used to define and lock a final model to stratify men with metastatic-lethal versus non-recurrent PCa in a training dataset. Coefficients from the final model were then used to construct a DNA methylation score, which was evaluated by logistic regression and Receiver Operating Characteristic (ROC) curve analyses in an independent testing dataset. RESULTS: Five CpGs were technically validated and all were retained (P < 0.05) in the final model. The 5-CpG and Gleason sum coefficients were used to calculate a methylation score, which was higher in men with metastatic-lethal progression (P = 6.8 × 10-6 ) in the testing dataset. For each unit increase in the score there was a four-fold increase in risk of metastatic-lethal events (odds ratio, OR = 4.0, 95%CI = 1.8-14.3). At 95% specificity, sensitivity was 74% for the score compared to 53% for Gleason sum alone. The score demonstrated better prediction performance (AUC = 0.91; pAUC = 0.037) compared to Gleason sum alone (AUC = 0.87; pAUC = 0.025). CONCLUSIONS: The DNA methylation score improved upon Gleason sum for predicting metastatic-lethal progression and holds promise for risk stratification of men with aggressive tumors. This prognostic score warrants further evaluation as a tool for improving patient outcomes.

ESTIMATING AND COMPARING CANCER PROGRESSION RISKS UNDER VARYING SURVEILLANCE PROTOCOLS.

Outcomes after cancer diagnosis and treatment are often observed at discrete times via doctor-patient encounters or specialized diagnostic examinations. Despite their ubiquity as endpoints in cancer studies, such outcomes pose challenges for analysis. In particular, comparisons between studies or patient populations with different surveillance schema may be confounded by differences in visit frequencies. We present a statistical framework based on multistate and hidden Markov models that represents events on a continuous time scale given data with discrete observation times. To demonstrate this framework, we consider the problem of comparing risks of prostate cancer progression across multiple active surveillance cohorts with different surveillance frequencies. We show that the different surveillance schedules partially explain observed differences in the progression risks between cohorts. Our application permits the conclusion that differences in underlying cancer progression risks across cohorts persist after accounting for different surveillance frequencies.

Comparative Analysis of Biopsy Upgrading in Four Prostate Cancer Active Surveillance Cohorts.

Background: Active surveillance (AS) is increasingly accepted for managing low-risk prostate cancer, yet there is no consensus about implementation. This lack of consensus is due in part to uncertainty about risks for disease progression, which have not been systematically compared or integrated across AS studies with variable surveillance protocols and dropout to active treatment. Objective: To compare risks for upgrading from a Gleason score (GS) of 6 or less to 7 or more across AS studies after accounting for differences in surveillance intervals and competing treatments and to evaluate tradeoffs of more versus less frequent biopsies. Design: Joint statistical model of longitudinal prostate-specific antigen (PSA) levels and risks for biopsy upgrading. Setting: Johns Hopkins University (JHU); Canary Prostate Active Surveillance Study (PASS); University of California, San Francisco (UCSF); and University of Toronto (UT) AS studies. Patients: 2576 men aged 40 to 80 years with a GS between 2 and 6 and clinical stage T1 or T2 prostate cancer enrolled between 1995 and 2014. Measurements: PSA levels and biopsy GSs. Results: After variable surveillance intervals and competing treatments were accounted for, estimated risks for biopsy upgrading were similar in the PASS and UT studies but higher in UCSF and lower in JHU studies. All cohorts had a delay of 3 to 5 months in detecting upgrading with biennial biopsies starting after a first confirmatory biopsy versus annual biopsies. Limitation: The model does not account for possible misclassification of biopsy GS. Conclusion: Men in different AS studies have different risks for biopsy upgrading after variable surveillance protocols and competing treatments are accounted for. Despite these differences, the consequences of more versus less frequent biopsies seem to be similar across cohorts. Biennial biopsies seem to be an acceptable alternative to annual biopsies. Primary Funding Source: National Cancer Institute.

Gene expression panel predicts metastatic-lethal prostate cancer outcomes in men diagnosed with clinically localized prostate cancer.

Prognostic biomarkers are needed to distinguish patients with clinically localized prostate cancer (PCa) who are at high risk of metastatic progression. The tumor transcriptome may reveal its aggressiveness potential and have utility for predicting adverse patient outcomes. Genomewide gene expression levels were measured in primary tumor samples of 383 patients in a population-based discovery cohort, and from an independent clinical validation dataset of 78 patients. Patients were followed for ≥ 5 years after radical prostatectomy to ascertain outcomes. Area under the receiver-operating characteristic curve (AUC), partial AUC (pAUC, 95% specificity), and P-value criteria were used to detect and validate the differentially expressed transcripts. Twenty-three differentially expressed transcripts in patients with metastatic-lethal compared with nonrecurrent PCa were validated (P < 0.05; false discovery rate < 0.20) in the independent dataset. The addition of each validated transcript to a model with Gleason score showed that 17 transcripts significantly improved the AUC (range: 0.83-0.88; all P-values < 0.05). These differentially expressed mRNAs represent genes with diverse cellular functions related to tumor aggressiveness. This study validated 23 gene transcripts for predicting metastatic-lethal PCa in patients surgically treated for clinically localized disease. Several of these mRNA biomarkers have clinical potential for identifying the subset of PCa patients with more aggressive tumors who would benefit from closer monitoring and adjuvant therapy.

Epigenome-Wide Tumor DNA Methylation Profiling Identifies Novel Prognostic Biomarkers of Metastatic-Lethal Progression in Men Diagnosed with Clinically Localized Prostate Cancer.

PURPOSE: Aside from Gleason sum, few factors accurately identify the subset of prostate cancer patients at high risk for metastatic progression. We hypothesized that epigenetic alterations could distinguish prostate tumors with life-threatening potential. EXPERIMENTAL DESIGN: Epigenome-wide DNA methylation profiling was performed in surgically resected primary tumor tissues from a population-based (n = 430) and a replication (n = 80) cohort of prostate cancer patients followed prospectively for at least 5 years. Metastasis was confirmed by positive bone scan, MRI, CT, or biopsy, and death certificates confirmed cause of death. AUC, partial AUC (pAUC, 95% specificity), and P value criteria were used to select differentially methylated CpG sites that robustly stratify patients with metastatic-lethal from nonrecurrent tumors, and which were complementary to Gleason sum. RESULTS: Forty-two CpG biomarkers stratified patients with metastatic-lethal versus nonrecurrent prostate cancer in the discovery cohort, and eight of these CpGs replicated in the validation cohort based on a significant (P < 0.05) AUC (range, 0.66-0.75) or pAUC (range, 0.007-0.009). The biomarkers that improved discrimination of patients with metastatic-lethal prostate cancer include CpGs in five genes (ALKBH5, ATP11A, FHAD1, KLHL8, and PI15) and three intergenic regions. In the validation dataset, the AUC for Gleason sum alone (0.82) significantly increased with the addition of four individual CpGs (range, 0.86-0.89; all P <0.05). CONCLUSIONS: Eight differentially methylated CpGs that distinguish patients with metastatic-lethal from nonrecurrent tumors were validated. These novel epigenetic biomarkers warrant further investigation as they may improve prognostic classification of patients with clinically localized prostate cancer and provide new insights on tumor aggressiveness. Clin Cancer Res; 23(1); 311-9. ©2016 AACR.

Characterizing dynamic changes in the human blood transcriptional network.

Gene expression data generated systematically in a given system over multiple time points provides a source of perturbation that can be leveraged to infer causal relationships among genes explaining network changes. Previously, we showed that food intake has a large impact on blood gene expression patterns and that these responses, either in terms of gene expression level or gene-gene connectivity, are strongly associated with metabolic diseases. In this study, we explored which genes drive the changes of gene expression patterns in response to time and food intake. We applied the Granger causality test and the dynamic Bayesian network to gene expression data generated from blood samples collected at multiple time points during the course of a day. The simulation result shows that combining many short time series together is as powerful to infer Granger causality as using a single long time series. Using the Granger causality test, we identified genes that were supported as the most likely causal candidates for the coordinated temporal changes in the network. These results show that PER1 is a key regulator of the blood transcriptional network, in which multiple biological processes are under circadian rhythm regulation. The fasted and fed dynamic Bayesian networks showed that over 72% of dynamic connections are self links. Finally, we show that different processes such as inflammation and lipid metabolism, which are disconnected in the static network, become dynamically linked in response to food intake, which would suggest that increasing nutritional load leads to coordinate regulation of these biological processes. In conclusion, our results suggest that food intake has a profound impact on the dynamic co-regulation of multiple biological processes, such as metabolism, immune response, apoptosis and circadian rhythm. The results could have broader implications for the design of studies of disease association and drug response in clinical trials.

The effect of food intake on gene expression in human peripheral blood.

Human gene expression traits have been shown to be dependent on gender, age and time of day in blood and other tissues. However, other factors that may impact gene expression have not been systematically explored. For example, in studies linking blood gene expression to obesity related traits, whether the fasted or fed state will be the most informative is an open question. Here, we employed a two-arm cross-over design to perform a genome-wide survey of gene expression in human peripheral blood to address explicitly this type of question. We were able to distinguish expression changes due to individual and time-specific effects from those due to food intake. We demonstrate that the transcriptional response to food intake is robust by constructing a classifier from the gene expression traits with >90% accuracy classifying individuals as being in the fasted or fed state. Gene expression traits that were best able to discriminate the fasted and fed states were more heritable and achieved greater coherence with respect to pathways associated with metabolic traits. The connectivity structure among gene expression traits was explored in the context of coexpression networks. Changes in the connectivity structure were observed between the fasted and fed states. We demonstrate that differential expression and differential connectivity are two complementary ways to characterize changes between fasted and fed states. Both gene sets were significantly enriched for genes associated with obesity related traits. Our results suggest that the pair of fasted/fed blood expression profiles provide more comprehensive information about an individual's metabolic states.

Diurnal variation of the human adipose transcriptome and the link to metabolic disease.

BACKGROUND: Circadian (diurnal) rhythm is an integral part of the physiology of the body; specifically, sleep, feeding behavior and metabolism are tightly linked to the light-dark cycle dictated by earth's rotation. METHODS: The present study examines the effect of diurnal rhythm on gene expression in the subcutaneous adipose tissue of overweight to mildly obese, healthy individuals. In this well-controlled clinical study, adipose biopsies were taken in the morning, afternoon and evening from individuals in three study arms: treatment with the weight loss drug sibutramine/fasted, placebo/fed and placebo/fasted. RESULTS: The results indicated that diurnal rhythm was the most significant driver of gene expression variation in the human adipose tissue, with at least 25% of the genes having had significant changes in their expression levels during the course of the day. The mRNA expression levels of core clock genes at a specific time of day were consistent across multiple subjects on different days in all three arms, indicating robust diurnal regulation irrespective of potential confounding factors. The genes essential for energy metabolism and tissue physiology were part of the diurnal signature. We hypothesize that the diurnal transition of the expression of energy metabolism genes reflects the shift in the adipose tissue from an energy-expending state in the morning to an energy-storing state in the evening. Consistent with this hypothesis, the diurnal transition was delayed by fasting and treatment with sibutramine. Finally, an in silico comparison of the diurnal signature with data from the publicly-available Connectivity Map demonstrated a significant association with transcripts that were repressed by mTOR inhibitors, suggesting a possible link between mTOR signaling, diurnal gene expression and metabolic regulation. CONCLUSION: Diurnal rhythm plays an important role in the physiology and regulation of energy metabolism in the adipose tissue and should be considered in the selection of novel targets for the treatment of obesity and other metabolic disorders.

Genetics of gene expression and its effect on disease.

Common human diseases result from the interplay of many genes and environmental factors. Therefore, a more integrative biology approach is needed to unravel the complexity and causes of such diseases. To elucidate the complexity of common human diseases such as obesity, we have analysed the expression of 23,720 transcripts in large population-based blood and adipose tissue cohorts comprehensively assessed for various phenotypes, including traits related to clinical obesity. In contrast to the blood expression profiles, we observed a marked correlation between gene expression in adipose tissue and obesity-related traits. Genome-wide linkage and association mapping revealed a highly significant genetic component to gene expression traits, including a strong genetic effect of proximal (cis) signals, with 50% of the cis signals overlapping between the two tissues profiled. Here we demonstrate an extensive transcriptional network constructed from the human adipose data that exhibits significant overlap with similar network modules constructed from mouse adipose data. A core network module in humans and mice was identified that is enriched for genes involved in the inflammatory and immune response and has been found to be causally associated to obesity-related traits.

Variations in DNA elucidate molecular networks that cause disease.

Identifying variations in DNA that increase susceptibility to disease is one of the primary aims of genetic studies using a forward genetics approach. However, identification of disease-susceptibility genes by means of such studies provides limited functional information on how genes lead to disease. In fact, in most cases there is an absence of functional information altogether, preventing a definitive identification of the susceptibility gene or genes. Here we develop an alternative to the classic forward genetics approach for dissecting complex disease traits where, instead of identifying susceptibility genes directly affected by variations in DNA, we identify gene networks that are perturbed by susceptibility loci and that in turn lead to disease. Application of this method to liver and adipose gene expression data generated from a segregating mouse population results in the identification of a macrophage-enriched network supported as having a causal relationship with disease traits associated with metabolic syndrome. Three genes in this network, lipoprotein lipase (Lpl), lactamase beta (Lactb) and protein phosphatase 1-like (Ppm1l), are validated as previously unknown obesity genes, strengthening the association between this network and metabolic disease traits. Our analysis provides direct experimental support that complex traits such as obesity are emergent properties of molecular networks that are modulated by complex genetic loci and environmental factors.

An integrative genomics approach to infer causal associations between gene expression and disease.

A key goal of biomedical research is to elucidate the complex network of gene interactions underlying complex traits such as common human diseases. Here we detail a multistep procedure for identifying potential key drivers of complex traits that integrates DNA-variation and gene-expression data with other complex trait data in segregating mouse populations. Ordering gene expression traits relative to one another and relative to other complex traits is achieved by systematically testing whether variations in DNA that lead to variations in relative transcript abundances statistically support an independent, causative or reactive function relative to the complex traits under consideration. We show that this approach can predict transcriptional responses to single gene-perturbation experiments using gene-expression data in the context of a segregating mouse population. We also demonstrate the utility of this approach by identifying and experimentally validating the involvement of three new genes in susceptibility to obesity.

A comprehensive transcript index of the human genome generated using microarrays and computational approaches.

BACKGROUND: Computational and microarray-based experimental approaches were used to generate a comprehensive transcript index for the human genome. Oligonucleotide probes designed from approximately 50,000 known and predicted transcript sequences from the human genome were used to survey transcription from a diverse set of 60 tissues and cell lines using ink-jet microarrays. Further, expression activity over at least six conditions was more generally assessed using genomic tiling arrays consisting of probes tiled through a repeat-masked version of the genomic sequence making up chromosomes 20 and 22. RESULTS: The combination of microarray data with extensive genome annotations resulted in a set of 28,456 experimentally supported transcripts. This set of high-confidence transcripts represents the first experimentally driven annotation of the human genome. In addition, the results from genomic tiling suggest that a large amount of transcription exists outside of annotated regions of the genome and serves as an example of how this activity could be measured on a genome-wide scale. CONCLUSIONS: These data represent one of the most comprehensive assessments of transcriptional activity in the human genome and provide an atlas of human gene expression over a unique set of gene predictions. Before the annotation of the human genome is considered complete, however, the previously unannotated transcriptional activity throughout the genome must be fully characterized.

Discovering modes of action for therapeutic compounds using a genome-wide screen of yeast heterozygotes.

Modern medicine faces the challenge of developing safer and more effective therapies to treat human diseases. Many drugs currently in use were discovered without knowledge of their underlying molecular mechanisms. Understanding their biological targets and modes of action will be essential to design improved second-generation compounds. Here, we describe the use of a genome-wide pool of tagged heterozygotes to assess the cellular effects of 78 compounds in Saccharomyces cerevisiae. Specifically, lanosterol synthase in the sterol biosynthetic pathway was identified as a target of the antianginal drug molsidomine, which may explain its cholesterol-lowering effects. Further, the rRNA processing exosome was identified as a potential target of the cell growth inhibitor 5-fluorouracil. This genome-wide screen validated previously characterized targets or helped identify potentially new modes of action for over half of the compounds tested, providing proof of this principle for analyzing the modes of action of clinically relevant compounds.