Search

Abstract Background Acute myeloid leukemia (AML) is a heterogeneous disease with a poor prognosis. Dysregulation of the epigenetic machinery is a significant contributor to disease development. Some AML patients benefit from treatment with hypomethylating agents (HMAs), but no predictive biomarkers for therapy response exist. Here, we investigated whether unbiased genome-wide assessment of pre-treatment DNA-methylation profiles in AML bone marrow blasts can help to identify patients who will achieve a remission after an azacytidine-containing induction regimen. Results A total of n = 155 patients with newly diagnosed AML treated in the AMLSG 12-09 trial were randomly assigned to a screening and a refinement and validation cohort. The cohorts were divided according to azacytidine-containing induction regimens and response status. Methylation status was assessed for 664,227 500-bp-regions using methyl-CpG immunoprecipitation-seq, resulting in 1755 differentially methylated regions (DMRs). Top regions were distilled and included genes such as WNT10A and GATA3. 80% of regions identified as a hit were represented on HumanMethlyation 450k Bead Chips. Quantitative methylation analysis confirmed 90% of these regions (36 of 40 DMRs). A classifier was trained using penalized logistic regression and fivefold cross validation containing 17 CpGs. Validation based on mass spectra generated by MALDI-TOF failed (AUC 0.59). However, discriminative ability was maintained by adding neighboring CpGs. A recomposed classifier with 12 CpGs resulted in an AUC of 0.77. When evaluated in the non-azacytidine containing group, the AUC was 0.76. Conclusions Our analysis evaluated the value of a whole genome methyl-CpG screening assay for the identification of informative methylation changes. We also compared the informative content and discriminatory power of regions and single CpGs for predicting response to therapy. The relevance of the identified DMRs is supported by their association with key regulatory processes of oncogenic transformation and support the idea of relevant DMRs being enriched at distinct loci rather than evenly distribution across the genome. Predictive response to therapy could be established but lacked specificity for treatment with azacytidine. Our results suggest that a predictive epigenotype carries its methylation information at a complex, genome-wide level, that is confined to regions, rather than to single CpGs. With increasing application of combinatorial regimens, response prediction may become even more complicated.

Abstract In this paper we propose PIICM, a probabilistic framework for dose–response prediction in high-throughput drug combination datasets. PIICM utilizes a permutation invariant version of the intrinsic co-regionalization model for multi-output Gaussian process regression, to predict dose–response surfaces in untested drug combination experiments. Coupled with an observation model that incorporates experimental uncertainty, PIICM is able to learn from noisily observed cell-viability measurements in settings where the underlying dose–response experiments are of varying quality, utilize different experimental designs, and the resulting training dataset is sparsely observed. We show that the model can accurately predict dose–response in held out experiments, and the resulting function captures relevant features indicating synergistic interaction between drugs.

The analysis of whole genomes of pan‐cancer data sets provides a challenge for researchers, and we contribute to the literature concerning the identification of robust subgroups with clear biological interpretation. Specifically, we tackle this unsupervised problem via a novel rank‐based Bayesian clustering method. The advantages of our method are the integration and quantification of all uncertainties related to both the input data and the model, the probabilistic interpretation of final results to allow straightforward assessment of the stability of clusters leading to reliable conclusions, and the transparent biological interpretation of the identified clusters since each cluster is characterized by its top‐ranked genomic features. We applied our method to RNA‐seq data from cancer samples from 12 tumor types from the Cancer Genome Atlas. We identified a robust clustering that mostly reflects tissue of origin but also includes pan‐cancer clusters. Importantly, we identified three pan‐squamous clusters composed of a mix of lung squamous cell carcinoma, head and neck squamous carcinoma, and bladder cancer, with different biological functions over‐represented in the top genes that characterize the three clusters. We also found two novel subtypes of kidney cancer that show different prognosis, and we reproduced known subtypes of breast cancer. Taken together, our method allows the identification of robust and biologically meaningful clusters of pan‐cancer samples.

Abstract Background Placenta-derived proteins in the systemic maternal circulation are suggested as potential biomarkers for placental function. However, the identity and longitudinal patterns of such proteins are largely unknown due to the inaccessibility of the human placenta and limitations in assay technologies. We aimed to identify proteins derived from and taken up by the placenta in the maternal circulation. Furthermore, we aimed to describe the longitudinal patterns across gestation of placenta-derived proteins as well as identify placenta-derived proteins that can serve as reference curves for placental function. Methods We analyzed proteins in plasma samples collected in two cohorts using the Somalogic 5000-plex platform. Antecubital vein samples were collected at three time points (gestational weeks 14–16, 22–24, and 30–32) across gestation in 70 healthy pregnancies in the longitudinal STORK cohort. In the cross sectional 4-vessel cohort, blood samples were collected simultaneously from the maternal antecubital vein (AV), radial artery (RA), and uterine vein (UV) during cesarean section in 75 healthy pregnancies. Placenta-derived proteins and proteins taken up by the placenta were identified using venoarterial differences (UV-RA). Placenta-derived proteins were defined as placenta-specific by comparison to the venoarterial difference in the antecubital vein-radial artery (AV-RA). These proteins were described longitudinally based on the STORK cohort samples using a linear mixed effects model per protein. Using a machine learning algorithm, we identified placenta-derived proteins that could predict gestational age, meaning that they closely tracked gestation, and were potential read-outs of placental function. Results Among the nearly 5000 measured proteins, we identified 256 placenta-derived proteins and 101 proteins taken up by the placenta (FDR

Glioblastoma (GBM) is an aggressive and highly heterogeneous primary brain tumor. Glioma stem cells represent a subpopulation of tumor cells with stem cell traits that are presumed to be the cause of tumor relapse. There exists complex tumor heterogeneity in drug sensitivity patterns between glioma stem cell (GSC) cultures derived from different patients. Here, we describe that heterogeneity also exists between GSC cultures derived from multiple biopsies within a single tumor. From biopsies harvested within spatially distinct regions representing the entire tumor mass, we established seven GSC cultures and compared their stem cell properties, mutations, gene expression profiles, and drug sensitivity patterns against 115 different anticancer drugs. The results were compared to 14 GSC cultures derived from other patients. Between the multiregional-derived GSC cultures, we observed only minor differences in their phenotype, proliferative capacity, and global gene expression. Further, they displayed intratumoral heterogeneity in mutational profiles and sensitivity patterns to anticancer drugs. This heterogeneity, however, did not exceed the extensive heterogeneity found between GSC cultures derived from other GBM patients. Our results suggest that the use of GSC cultures from one single focal biopsy may underestimate the overall complexity of the GSC population and display the importance of including GSC cultures reflecting the entire tumor mass in drug screening strategies.

Abstract Background Foot disorders affect up to one quarter of the adult population. Plantar fasciopathy is a common cause of foot pain associated with decreased activity level and quality of life. Patient-reported outcome measures are important in assessing the burden of a condition as well as in research on the effects of interventions. The Foot Function Index revised short form (FFI-RS) is a region specific questionnaire frequently used in research. This study aimed to cross-culturally adapt the FFI-RS into Norwegian and to test its psychometric properties. Methods The FFI-RS was translated into Norwegian (FFI-RSN) following international guidelines. 139 patients with foot disorders (88% with plantar fasciopathy) were included at baseline to measure internal consistency, explorative factor analysis, construct validity and floor and ceiling effects. 54 patients were included after 1 week for test-retest reliability and smallest detectable change analyses. 100 patients were included for responsiveness and minimal important change at 3 months. Results Cronbach’s alpha for internal consistency was 0.97 and factor analysis supported the use of the total score of the FFI-RSN. Two out of three predefined hypotheses were confirmed by assessing the construct validity with Spearman’s correlation coefficient. Quadratic weighted Kappa for test-retest reliability showed 0.91 (95% CI 0.86–0.96) and the smallest detectable change was 6.5%. The minimal important change was 8.4% and the area under the receiver operating characteristic curve for responsiveness was 0.78 (95% CI 0.69–0.87). We found no floor or ceiling effects on the total score of the FFI-RSN. Conclusions The present study showed excellent reliability of the FFI-RSN and supports the use of the total score of the questionnaire. Furthermore, we found the FFI-RSN to have acceptable responsiveness in relation to change in general health. Smallest detectable change, minimal important change and responsiveness were presented as novel results of the total score of the FFI-RS. FFI-RSN can be used to evaluate global foot health in clinical or research settings with Norwegian patients suffering from plantar fasciopathy. Trial registration Clinical Trials.gov NCT04207164 . Initial release 01.11.19.

Given that both hearing and touch are ‘mechanical senses’ that respond to physical pressure or mechanical energy and that individuals appear to have a characteristic internal or spontaneous tempo, individual preferences in musical and touch rhythms might be related. We explored this in two experiments probing individual preferences for tempo in the tactile and auditory modalities. Study 1 collected ratings of received stroking on the forearm and measured the velocity the participants used for stroking a fur. Music tempo preferences were assessed as mean beats per minute of individually selected music pieces and via the adjustment of experimenter-selected music to a preferred tempo. Heart rate was recorded to measure levels of physiological arousal. We found that the preferred tempo of favorite (self-selected) music correlated positively with the velocity with which each individual liked to be touched. In Study 2, participants rated videos of repeated touch on someone else’s arm and videos of a drummer playing with brushes on a snare drum, both at a variety of tempos. We found that participants with similar rating patterns for the different stroking speeds did not show similar rating patterns for the different music beats. The results suggest that there may be a correspondence between preferences for favorite music and felt touch, but this is either weak or it cannot be evoked effectively with vicarious touch and/or mere drum beats. Thus, if preferences for touch and music are related, this is likely to be dependent on the specific type of stimulation.

Long-term antibiotics are prescribed for a variety of medical conditions, recently including low back pain with Modic changes. The molecular impact of such treatment is unknown. We conducted longitudinal transcriptome and epigenome analyses in patients (n = 100) receiving amoxicillin treatment or placebo for 100 days in the Antibiotics in Modic Changes (AIM) study. Gene expression and DNA methylation were investigated at a genome-wide level at screening, after 100 days of treatment, and at one-year follow-up. We identified intra-individual longitudinal changes in gene expression and DNA methylation in patients receiving amoxicillin, while few changes were observed in patients receiving placebo. After 100 days of amoxicillin treatment, 28 genes were significantly differentially expressed, including the downregulation of 19 immunoglobulin genes. At one-year follow-up, the expression levels were still not completely restored. The significant changes in DNA methylation (n = 4548 CpGs) were mainly increased methylation levels between 100 days and one-year follow-up. Hence, the effects on gene expression occurred predominantly during treatment, while the effects on DNA methylation occurred after treatment. In conclusion, unrecognized side effects of long-term amoxicillin treatment were revealed, as alterations were observed in both gene expression and DNA methylation that lasted long after the end of treatment.

Abstract Background Important objectives in cancer research are the prediction of a patient’s risk based on molecular measurements such as gene expression data and the identification of new prognostic biomarkers (e.g. genes). In clinical practice, this is often challenging because patient cohorts are typically small and can be heterogeneous. In classical subgroup analysis, a separate prediction model is fitted using only the data of one specific cohort. However, this can lead to a loss of power when the sample size is small. Simple pooling of all cohorts, on the other hand, can lead to biased results, especially when the cohorts are heterogeneous. Results We propose a new Bayesian approach suitable for continuous molecular measurements and survival outcome that identifies the important predictors and provides a separate risk prediction model for each cohort. It allows sharing information between cohorts to increase power by assuming a graph linking predictors within and across different cohorts. The graph helps to identify pathways of functionally related genes and genes that are simultaneously prognostic in different cohorts. Conclusions Results demonstrate that our proposed approach is superior to the standard approaches in terms of prediction performance and increased power in variable selection when the sample size is small.

In molecular biology, advances in high-throughput technologies have made it possible to study complex multivariate phenotypes and their simultaneous associations with high-dimensional genomic and other omics data, a problem that can be studied with high-dimensional multi-response regression, where the response variables are potentially highly correlated. To this purpose, we recently introduced several multivariate Bayesian variable and covariance selection models, e.g., Bayesian estimation methods for sparse seemingly unrelated regression for variable and covariance selection. Several variable selection priors have been implemented in this context, in particular the hotspot detection prior for latent variable inclusion indicators, which results in sparse variable selection for associations between predictors and multiple phenotypes. We also propose an alternative, which uses a Markov random field (MRF) prior for incorporating prior knowledge about the dependence structure of the inclusion indicators. Inference of Bayesian seemingly unrelated regression (SUR) by Markov chain Monte Carlo methods is made computationally feasible by factorization of the covariance matrix amongst the response variables. In this paper we present BayesSUR, an R package, which allows the user to easily specify and run a range of different Bayesian SUR models, which have been implemented in C++ for computational efficiency. The R package allows the specification of the models in a modular way, where the user chooses the priors for variable selection and for covariance selection separately. We demonstrate the performance of sparse SUR models with the hotspot prior and spike-and-slab MRF prior on synthetic and real data sets representing eQTL or mQTL studies and in vitro anti-cancer drug screening studies as examples for typical applications.

Methotrexate is one of the cornerstones of rheumatoid arthritis (RA) therapy. Genetic factors or single nucleotide polymorphisms (SNPs) are responsible for 15%–30% of the variation in drug response. Identification of clinically effective SNP biomarkers for predicting methotrexate (MTX) sensitivity has been a challenge. The aim of this study was to explore the association between the disease related outcome of MTX treatment and 23 SNPs in 8 genes of the MTX pathway, as well as one pro-inflammatory related gene in RA patients naïve to MTX. Categorical outcomes such as Disease Activity Score (DAS)-based European Alliance of Associations for Rheumatology (EULAR) non-response at 4 months, The American College of Rheumatology and EULAR (ACR/EULAR) non-remission at 6 months, and failure to sustain MTX monotherapy from 12 to 24 months were assessed, together with continuous outcomes of disease activity, joint pain and fatigue. We found that the SNPs rs1801394 in the MTRR gene, rs408626 in DHFR gene, and rs2259571 in AIF-1 gene were significantly associated with disease activity relevant continuous outcomes. Additionally, SNP rs1801133 in the MTHFR gene was identified to be associated with improved fatigue. Moreover, associations with p values at uncorrected significance level were found in SNPs and different categorical outcomes: 1) rs1476413 in the MTHFR gene and rs3784864 in ABCC1 gene are associated with ACR/EULAR non-remission; 2) rs1801133 in the MTHFR gene is associated with EULAR response; 3) rs246240 in the ABCC1 gene, rs2259571 in the AIF-1 gene, rs2274808 in the SLC19A1 gene and rs1476413 in the MTHFR gene are associated with failure to MTX monotherapy after 12–24 months. The results suggest that SNPs in genes associated with MTX activity may be used to predict MTX relevant-clinical outcomes in patients with RA.

Summary: Current statistical models for drug response prediction and biomarker identification fall short in leveraging the shared and unique information from various cancer tissues and multi-omics profiles. We developed mix-lasso model that introduces an additional sample group penalty term to capture tissue-specific effects of features on pan-cancer response prediction. The mix-lasso model takes into account both the similarity between drug responses (i.e., multi-task learning), and the heterogeneity between multi-omics data (multi-modal learning). When applied to large-scale pharmacogenomics dataset from Cancer Therapeutics Response Portal, mix-lasso enabled accurate drug response predictions and identification of tissue-specific predictive features in the presence of various degrees of missing data, drug-drug correlations, and high-dimensional and correlated genomic and molecular features that often hinder the use of statistical approaches in drug response modeling. Compared to tree lasso model, mix-lasso identified a smaller number of tissue-specific features, hence making the model more interpretable and stable for drug discovery applications.

Abstract Background Chronic brain pathology and pre-stroke cognitive impairment (PCI) is predictive of post-stroke dementia. The aim of the current study was to measure pre-stroke neurodegenerative and vascular disease burden found on brain MRI and to assess the association between pre-stroke imaging pathology and PCI, whilst also looking for potential sex differences. Methods This prospective brain MRI cohort is part of the multicentre Norwegian cognitive impairment after stroke (Nor-COAST) study. Patients hospitalized with acute ischemic or hemorrhagic stroke were included from five participating stroke units. Visual rating scales were used to categorize baseline MRIs (N = 410) as vascular, neurodegenerative, mixed, or normal, based on the presence of pathological imaging findings. Pre-stroke cognition was assessed by interviews of patients or caregivers using the Global Deterioration Scale (GDS). Stroke severity was assessed with the National Institute of Health Stroke Scale (NIHSS). Univariate and multiple logistic regression analyses were performed to investigate the association between imaging markers, PCI, and sex. Results Patients’ (N = 410) mean (SD) age was 73.6 (±11) years; 182 (44%) participants were female, the mean (SD) NIHSS at admittance was 4.1 (±5). In 68% of the participants, at least one pathological imaging marker was found. Medial temporal lobe atrophy (MTA) was present in 30% of patients, white matter hyperintensities (WMH) in 38% of patients and lacunes in 35% of patients. PCI was found in 30% of the patients. PCI was associated with cerebrovascular pathology (OR 2.5; CI = 1.4 to 4.5, p = 0.001) and mixed pathology (OR 3.4; CI = 1.9 to 6.1, p = 0.001) but was not associated with neurodegeneration (OR 1.0; CI = 0.5 to 2.2; p = 0.973). Pathological MRI markers, including MTA and lacunes, were more prevalent among men, as was a history of clinical stroke prior to the index stroke. The OR of PCI for women was not significantly increased (OR 1.2; CI = 0.8 to 1.9; p = 0.3). Conclusions Pre-stroke chronic brain pathology is common in stroke patients, with a higher prevalence in men. Vascular pathology and mixed pathology are associated with PCI. There were no significant sex differences for the risk of PCI. Trial registration NCT02650531 , date of registration: 08.01.2016.

Background: More accurate risk assessments are needed to improve prostate cancer management. Objective: To identify blood-based protein biomarkers that provided prognostic information for risk stratification. Design, setting, and participants: Mass spectrometry was used to identify biomarker candidates from blood, and validation studies were performed in four independent cohorts retrospectively collected between 1988 and 2015. Outcome measurements and statistical analysis: The primary outcome objectives were progression-free survival, prostate cancer–specific survival (PCSS), and overall survival. Statistical analyses to assess survival and model performance were performed. Results and limitation: Serum leucine-rich α-2-glycoprotein 1 (LRG1) was found to be elevated in fatal prostate cancer. LRG1 provided prognostic information independent of metastasis and increased the accuracy in predicting PCSS, particularly in the first 3 yr. A high LRG1 level is associated with an average of two-fold higher risk of disease-progression and mortality in both high-risk and metastatic patients. However, our study design, with a retrospective analysis of samples spanning several decades back, limits the assessment of the clinical utility of LRG1 in today’s clinical practice. Thus, independent prospective studies are needed to establish LRG1 as a clinically useful biomarker for patient management. Conclusions: High blood levels of LRG1 are unfavourable in newly diagnosed high-risk and metastatic prostate cancer, and LRG1 increased the accuracy of risk stratification of prostate cancer patients. Patient summary: High blood levels of leucine-rich α-2-glycoprotein 1 are unfavourable in newly diagnosed high-risk and metastatic prostate cancer.

Rheumatoid arthritis (RA) is a complex disease with a wide range of underlying susceptibility factors. Recently, dysregulation of microRNAs (miRNAs) in RA have been reported in several immune cell types from blood. However, B cells have not been studied in detail yet. Given the autoimmune nature of RA with the presence of autoantibodies, CD19+ B cells are a key cell type in RA pathogenesis and alterations in CD19+ B cell subpopulations have been observed in patient blood. Therefore, we aimed to reveal the global miRNA repertoire and to analyze miRNA expression profile differences in homogenous RA patient phenotypes in blood-derived CD19+ B cells. Small RNA sequencing was performed on CD19+ B cells of newly diagnosed untreated RA patients (n=10), successfully methotrexate (MTX) treated RA patients in remission (MTX treated RA patients, n=18) and healthy controls (n=9). The majority of miRNAs was detected across all phenotypes. However, significant expression differences between MTX treated RA patients and controls were observed for 27 miRNAs, while no significant differences were seen between the newly diagnosed patients and controls. Several of the differentially expressed miRNAs were previously found to be dysregulated in RA including miR-223-3p, miR-486-3p and miR-23a-3p. MiRNA target enrichment analysis, using the differentially expressed miRNAs and miRNA-target interactions from miRTarBase as input, revealed enriched target genes known to play important roles in B cell activation, differentiation and B cell receptor signaling, such as STAT3, PRDM1 and PTEN. Interestingly, many of those genes showed a high degree of correlated expression in CD19+ B cells in contrast to other immune cell types. Our results suggest important regulatory functions of miRNAs in blood-derived CD19+ B cells of MTX treated RA patients and motivate for future studies investigating the interactive mechanisms between miRNA and gene targets, as well as the possible predictive power of miRNAs for RA treatment response.

Abstract Background Several studies have documented the variety of post-stroke psychosocial challenges, which are complex, multifaceted, and affect a patient’s rehabilitation and recovery. Due to the consequences of these challenges, psychosocial well-being should be considered an important outcome of the stroke rehabilitation. Thus, a valid and reliable instrument that is appropriate for the stroke population is required. The factor structure of the Norwegian version of GHQ-28 has not previously been examined when applied to a stroke population. The purpose of this study was to explore the psychometric properties of the GHQ-28 when applied in the stroke population included in the randomized controlled trial; “Psychosocial well-being following stroke”, by evaluating the internal consistency, exploring the factor structure, construct validity and measurement invariance. Methods Data were obtained from 322 individuals with a stroke onset within the past month. The Kaiser-Meyer-Olkin (KMO) test was used to test the sampling adequacy for exploratory factor analysis, and the Bartlett’s test of sphericity was used to test equal variances. Internal consistency was analysed using Cronbach’s alpha. The factor structure of the GHQ-28 was evaluated by exploratory factor analysis (EFA), and a confirmatory factor analysis (CFA) was used to determine the goodness of fit to the original structure of the outcome measurement. Measurement invariance for two time points was evaluated by configural, metric and scalar invariance. Results The results from the EFA supported the four-factor dimensionality, but some of the items were loaded on different factors compared to those of the original structure. The differences resulted in a reduced goodness of fit in the CFA. Measurement invariance at two time points was confirmed. Conclusions The change in mean score from one to six months on the GHQ-28 and the factor composition are assumed to be affected by characteristics in the stroke population. The results, when applying the GHQ-28 in a stroke population, and sub-factor analysis based on the original factor structure should be interpreted with caution. Trial registration ClinicalTrials.gov, NCT02338869, registered 10/04/2014.

ObjectiveThe role of neutrophil extracellular traps (NETs) in acute heart failure is unknown. We recently showed that interleukin 8, a putative NETs stimulator, was associated with myocardial recovery in acute heart failure complicating ST-elevation myocardial infarction (STEMI). In this exploratory post-hoc study, we aimed to investigate the role of NETs components in relation to myocardial function and interleukin 8 in STEMI patients with symptomatic acute heart failure.MethodsIn 61 STEMI patients developing acute heart failure within 48 hours of successful revascularization, wall motion score index (WMSI), global longitudinal strain (GLS) and left ventricular ejection fraction (LVEF) were assessed by echocardiography at baseline and on day 5. Blood drawn at baseline and days 1, 2 and 5 was used to quantify double-stranded DNA (dsDNA), myeloperoxidase-DNA complexes (MPO-DNA) and citrullinated histone 3 (CitH3). The area under the curve (AUC) of each NETs marker and interleukin 8 was approximated for the first 5 days.ResultsdsDNAAUC and MPO-DNAAUC correlated significantly with change in WMSI from baseline to day 5 (rs = 0.28 for both, p≤0.05), whereas NETs AUCs did not correlate with changes in GLS and LVEF. dsDNAAUC was significantly correlated with interleukin 8AUC (r = 0.40, p = 0.003). However, mixed model regression could not identify a significant effect of the NETs components on myocardial function parameters.ConclusionsIn this cohort with acute heart failure complicating STEMI, NETs components were partly correlated with myocardial function and interleukin 8 levels, yet no causal relationship between NETs components and myocardial recovery could be established.Clinical trial registrationClinicalTrials.gov, identifier: NCT00324766.

Abstract Background A Bayesian mixed model approach using integrated nested Laplace approximations (INLA) allows us to construct flexible models that can account for pedigree structure. Using these models, we estimate genome-wide patterns of DNA methylation heritability (h 2 ), which are currently not well understood, as well as h 2 of blood lipid measurements. Methods We included individuals from the Genetics of Lipid Lowering Drugs and Diet Network (GOLDN) study with Infinium 450 K cytosine-phosphate-guanine (CpG) methylation and blood lipid data pre- and posttreatment with fenofibrate in families with up to three-generation pedigrees. For genome-wide patterns, we constructed 1 model per CpG with methylation as the response variable, with a random effect to model kinship, and age and gender as fixed effects. Results In total, 425,791 CpG sites pre-, but only 199,027 CpG sites posttreatment were found to have nonzero heritability. Across these CpG sites, the distributions of h 2 estimates are similar in pre- and posttreatment (pre: median = 0.31, interquartile range [IQR] = 0.16; post: median = 0.34, IQR = 0.20). Blood lipid h 2 estimates were similar pre- and posttreatment with overlapping 95% credibility intervals. Heritability was nonzero for treatment effect, that is, the difference between pre- and posttreatment blood lipids. Estimates for triglycerides h 2 are 0.48 (pre), 0.42 (post), and 0.21 (difference); likewise for high-density lipoprotein cholesterol h 2 the estimates are 0.61, 0.68, and 0.10. Conclusions We show that with INLA, a fully Bayesian approach to estimate DNA methylation h 2 is possible on a genome-wide scale. This provides uncertainty assessment of the estimates, and allows us to perform model selection via deviance information criterion (DIC) to identify CpGs with strong evidence for nonzero heritability.

Abstract Background An important feature in many genomic studies is quality control and normalization. This is particularly important when analyzing epigenetic data, where the process of obtaining measurements can be bias prone. The GAW20 data was from the Genetics of Lipid Lowering Drugs and Diet Network (GOLDN), a study with multigeneration families, where DNA cytosine-phosphate-guanine (CpG) methylation was measured pre- and posttreatment with fenofibrate. We performed quality control assessment of the GAW20 DNA methylation data, including normalization, assessment of batch effects and detection of sample swaps. Results We show that even after normalization, the GOLDN methylation data has systematic differences pre- and posttreatment. Through investigation of (a) CpGs sites containing a single nucleotide polymorphism, (b) the stability of breeding values for methylation across time points, and (c) autosomal gender-associated CpGs, 13 sample swaps were detected, 11 of which were posttreatment. Conclusions This paper demonstrates several ways to perform quality control of methylation data in the absence of raw data files and highlights the importance of normalization and quality control of the GAW20 methylation data from the GOLDN study.

Abstract Background In order to investigate the impact of adherence to recommendations of physical activity and sedentary time on health outcomes in clinical trials, there is a need for feasible tools such as questionnaires that can give representative estimates of these measures. The primary aim of the present study was to validate two such questionnaires and their ability to estimate adherence to the recommendations of physical activity defined as moderate-to- vigorous physical activity or moderate physical activity of at least 150 min/week in colorectal cancer patients. Secondarily, self-reported sedentary time from the HUNT-PAQ was also evaluated. Methods Participants from 'The Norwegian dietary guidelines and colorectal cancer survival-study’ (CRC-NORDIET study) completed two short questionnaires; the NORDIET-FFQ (n = 78) and the HUNT-PAQ (n = 77). The physical activity monitor SenseWear Armband Mini was used as the reference method during seven consecutive days. Results The NORDIET-FFQ provided better estimates of time in moderate-to- vigorous physical activity and moderate physical activity than the HUNT-PAQ. The NORDIET-FFQ was unable to rank individual time in moderate-to- vigorous physical activity and moderate physical activity (Spearman’s rho = 0.08, p = 0.509 and Spearman’s rho rho = 0.01, p = 0.402, respectively). All intensities were under-reported by the HUNT-PAQ, but ranking of individual time in moderate physical activity and sedentary time were acceptable among women only (Spearman’s rho = 0.37, p = 0.027 and Spearman’s rho = 0.36, p = 0.035, respectively). The HUNT-PAQ correctly classified 71% of those not meeting the recommendations (sensitivity), and the NORDIET-FFQ correctly classified 63% of those who met the recommendations (specificity). About 67% and 33% reported to meet the recommendation of moderate-to- vigorous physical activity with the NORDIET-FFQ and HUNT-PAQ, respectively, whereas 55% actually met the moderate-to- vigorous physical activity according to the SenseWear Armband Mini. Conclusions The NORDIET-FFQ provided better specificity and better estimates of PA than the HUNT-PAQ. The HUNT-PAQ provided better sensitivity, and provided better ranking of PA and sedentary time among women than NORDIET-FFQ. It is important to be aware of the limitations documented in the present study. Trial registration The study is registered on the National Institutes of Health Clinical Trials (Identifier: NCT01570010). Registered 4 April 2012.

Cancers elicit an immune response by modifying the microenvironment. The immune system plays a pivotal role in cancer recognition and eradication. While the potential clinical value of infiltrating lymphocytes at the tumor site has been assessed in breast cancer, circulating cytokines – the molecules coordinating and fine-tuning immune response – are still poorly characterized. Using two breast cancer cohorts (MicMa, n = 131, DCTB, n = 28) and the multiplex Luminex platform, we measured the levels of 27 cytokines in the serum of breast cancer patients prior to treatment. We investigated the cytokine levels in relation to clinicopathological characteristics and in perspective of the tumor infiltrating immune cells predicted from the bulk mRNA expression data. Unsupervised clustering analysis of the serum cytokine levels in the MicMa cohort identified a cluster of pro-inflammatory, pro-angiogenic, and Th2-related cytokines which was associated with poor prognosis. Notably high levels of platelet derived growth factor BB (PDGF) reflected a more aggressive tumor phenotype and larger tumor size. A significant positive correlation between serum levels of interferon gamma-induced protein 10 (IP10) and its mRNA expression at the tumor site suggested that tumor-IP10-production may outflow to the bloodstream. High IP10 serum levels were associated with a worse prognosis. Finally, we found serum levels of both PDGF and IP10 associated with enrichment scores of specific tumor infiltrating immune cells. Our study suggests that monitoring cytokine circulating levels in breast cancer could be used to characterize breast cancers and the immune composition of their microenvironment through readily available biological material.

A high concentration of circulating vascular endothelial growth factor (VEGF) in cancer patients is associated with an aggressive tumor phenotype. Here, serum levels of 27 cytokines and blood cell counts were assessed in breast cancer patients receiving neoadjuvant chemotherapy with or without bevacizumab (Bev) in a randomized cohort of 132 patients with non-metastatic HER2-negative tumors. Cytokine levels were determined prior to treatment and at various time-points. The cytotoxic chemotherapy regimen of fluorouracil, epirubicin, and cyclophosphamide (FEC) had a profound impact on both circulating white blood cells and circulating cytokine levels. At the end of FEC treatment, the global decrease in cytokine levels correlated with the drop in white blood cell counts and was significantly greater in the patients of the Bev arm for cytokines, such as VEGF-A, IL-12, IP-10 and IL-10. Among patients who received Bev, those with pathological complete response (pCR) exhibited significantly lower levels of VEGF-A, IFN-γ, TNF-α and IL-4 than patients without pCR. This effect was not observed in the chemotherapy-only arm. Certain circulating cytokine profiles were found to correlate with different immune cell types at the tumor site. For the Bev arm patients, the serum cytokine levels correlated with higher levels of cytotoxic T cells at the end of the therapy regimen, which was indicative of treatment response. The higher response rate for Bev-treated patients and stronger correlations between serum cytokine levels and infiltrating CD8T cells merits further investigation.

Chronic lymphocytic leukemia is a malignancy of mature B cells that strongly depend on microenvironmental factors, and their deprivation has been identified as a promising treatment approach for this incurable disease. Cytokine array screening of 247 chronic lymphocytic leukemia serum samples revealed elevated levels of tumor necrosis factor (TNF) receptor-1 which were associated with poor clinical outcome. We detected a microenvironment-induced expression of TNF receptor-1 in chronic lymphocytic leukemia cells in vitro, and an aberrantly high expression of this receptor in the proliferation centers of patients’ lymph nodes. Stimulation of TNF receptor-1 with TNF-α enhanced nuclear factor κ-light-chain-enhancer of activated B cells (NFκB) activity and viability of chronic lymphocytic leukemia cells, which was inhibited by wogonin. The therapeutic effects of wogonin were analyzed in mice after adoptive transfer of Eμ-T-cell leukemia 1 (TCL1) leukemic cells. Wogonin treatment prevented leukemia development when given early after transplantation. The treatment of full-blown leukemia resulted in the loss of the TNF receptor-1 on chronic lymphocytic leukemia cells and their mobilization to blood. Targeting TNF receptor-1 signaling is therefore proposed for the treatment of chronic lymphocytic leukemia.

IntroductionMiddle cerebral artery (MCA) and umbilical artery (UA) Doppler blood flow pulsatility indices (PIs) and MCA peak systolic velocity (PSV) are essential variables for clinically evaluating fetal well-being. Here we examined how a maternal meal influenced these Doppler blood flow velocity variables.MethodsThis prospective cohort study included 89 healthy Caucasian women with normal singleton pregnancies (median age, 32 years). Measurements were performed at gestational weeks 30 and 36, representing the start and near the end of the energy-depositing period. Measured variables included the MCA-PI, UA-PI, fetal heart rate (FHR) and MCA-PSV. The cerebroplacental ratio (CPR) was calculated as the ratio of MCA-PI to UA-PI. The first examination was performed in the fasting state at 08:30 a.m. Then participants ate a standard breakfast (approximate caloric intake, 400kcal), and the examination was repeated ~105 min after the meal.ResultsWithout adjustment for FHR, fetal MCA-PI decreased after the meal at week 30 (‒0.115; p = 0.012) and week 36 (‒0.255; p < 0.001). All PI values were negatively correlated with FHR. After adjustment for FHR, MCA-PI still decreased after the meal at week 30 (‒0.087; p = 0.044) and week 36 (‒0.194; p < 0.001). The difference between the two gestational weeks was non-significant (p = 0.075). UA-PI values did not significantly change at week 30 (p = 0.253) or week 36 (p = 0.920). CPR revealed significant postprandial decreases of -0.17 at week 30 (p = 0.006) and -0.22 at week 36 (p = 0.001). Compared to fasting values, MCA-PSV was significantly higher after food intake: +3.9 cm/s at week 30 (p < 0.001) and +5.9 cm/s at week 36 (p < 0.001).ConclusionIn gestational weeks 30 and 36, we observed a postprandial influence that was apparently specific to fetal cerebral blood flow.

We have developed the R package c060 with the aim of improving R software func- tionality for high-dimensional risk prediction modeling, e.g., for prognostic modeling of survival data using high-throughput genomic data. Penalized regression models provide a statistically appealing way of building risk prediction models from high-dimensional data. The popular CRAN package glmnet implements an efficient algorithm for fitting penalized Cox and generalized linear models. However, in practical applications the data analysis will typically not stop at the point where the model has been fitted. One is for example often interested in the stability of selected features and in assessing the prediction performance of a model and we provide functions to deal with both of these tasks. Our R functions are computationally efficient and offer the possibility of speeding up computing time through parallel computing. Another feature which can drastically reduce computing time is an efficient interval-search algorithm, which we have implemented for selecting the optimal parameter combination for elastic net penalties. These functions have been useful in our daily work at the Biostatistics department (C060) of the German Cancer Research Center where prognostic modeling of patient survival data is of particular interest. Although we focus on a survival data application of penalized Cox models in this article, the functions in our R package are in general applicable to all types of regression models implemented in the glmnet package, with the exception of prediction error curves, which are specific to time-to-event data.

OBJECTIVES:The current concepts of human fetal-placental amino acid exchange and metabolism are mainly based on animal-, in vitro- and ex vivo models. We aimed to determine and assess the paired relationships between concentrations and arteriovenous differences of 19 amino acids on the maternal and fetal sides of the human placenta in a large study sample. METHODS:This cross-sectional in vivo study included 179 healthy women with uncomplicated term pregnancies. During planned cesarean section, we sampled blood from incoming and outgoing vessels on the maternal (radial artery and uterine vein) and fetal (umbilical vein and artery) sides of the placenta. Amino acid concentrations were measured by liquid chromatography-tandem mass spectrometry. We calculated paired arteriovenous differences and performed Wilcoxon signed-rank tests and Spearman's correlations. RESULTS:In the umbilical circulation, we observed a positive venoarterial difference (fetal uptake) for 14 amino acids and a negative venoarterial difference (fetal release) for glutamic acid (p

Women pregnant following kidney transplantation are at high risk of preeclampsia. Identifying the effects of preeclampsia on pregnancy outcome and allograft function in kidney transplanted women, and predicting which women will require more targeted follow-up and possible therapeutic intervention, could improve both maternal and neonatal outcome. In this retrospective cohort study of all pregnancies following kidney transplantation in Norway between 1969 and 2013, we used medical records to identify clinical characteristics predictive of preeclampsia. 175 pregnancies were included, in which preeclampsia was diagnosed in 65. Pregnancies with preeclampsia had significantly higher postpartum serum creatinine levels, higher risks of preterm delivery, caesarean delivery, and small for gestational age infants. In the final multivariate model chronic hypertension (aOR = 5.02 [95% CI, 2.47-10.18]), previous preeclampsia (aOR = 3.26 [95% CI, 1.43-7.43]), and elevated serum creatinine (≥125 μmol/L) at the start of pregnancy (aOR = 5.79 [95% CI, 1.91-17.59]) were prognostic factors for preeclampsia. Based on this model the risk was 19% when none of these factors were present, 45-59% risk when one was present, 80-87% risk when two were present, and 96% risk when all three were present. We suggest that the risk of preeclampsia in pregnancies in kidney transplanted women can be predicted with these variables, which are easily available at the start of pregnancy.

Despite much evidence on epigenetic abnormalities in cancer, it is currently unclear to what extent epigenetic alterations can be associated with tumors’ clonal genetic origins. Here, we show that the prostate intratumor heterogeneity in DNA methylation and copy-number patterns can be explained by a unified evolutionary process. By assaying multiple topographically distinct tumor sites, premalignant lesions, and lymph node metastases within five cases of prostate cancer, we demonstrate that both DNA methylation and copy-number heterogeneity consistently reflect the life history of the tumors. Furthermore, we show cases of genetic or epigenetic convergent evolution and highlight the diversity in the evolutionary origins and aberration spectrum between tumor and metastatic subclones. Importantly, DNA methylation can complement genetic data by serving as a proxy for activity at regulatory domains, as we show through identification of high epigenetic heterogeneity at androgen-receptor-bound enhancers. Epigenome variation thereby expands on the current genome-centric view on tumor heterogeneity.

Chronic lymphocytic leukemia (CLL) is known for its strong dependency on the tumor microenvironment. We found progranulin (GRN), a protein that has been linked to inflammation and cancer, to be upregulated in the serum of CLL patients compared to healthy controls, and increased GRN levels to be associated with an increased hazard for disease progression and death. This raised the question of whether GRN is a functional driver of CLL. We observed that recombinant GRN did not directly affect viability, activation, or proliferation of primary CLL cells in vitro. However, GRN secretion was induced in co-cultures of CLL cells with stromal cells that enhanced CLL cell survival. Gene expression profiling and protein analyses revealed that primary mesenchymal stromal cells (MSCs) in co-culture with CLL cells acquire a cancer-associated fibroblast-like phenotype. Despite its upregulation in the co-cultures, GRN treatment of MSCs did not mimic this effect. To test the relevance of GRN for CLL in vivo, we made use of the Eμ-TCL1 CLL mouse model. As we detected strong GRN expression in myeloid cells, we performed adoptive transfer of Eμ-TCL1 leukemia cells to bone marrow chimeric Grn−/− mice that lack GRN in hematopoietic cells. Thereby, we observed that CLL-like disease developed comparable in Grn−/− chimeras and respective control mice. In conclusion, serum GRN is found to be strongly upregulated in CLL, which indicates potential use as a prognostic marker, but there is no evidence that elevated GRN functionally drives the disease.

Non-coding RNAs are much more common than previously thought. However, for the vast majority of non-coding RNAs, the cellular function remains enigmatic. The two long non-coding RNA (lncRNA) genes DLEU1 and DLEU2 map to a critical region at chromosomal band 13q14.3 that is recurrently deleted in solid tumors and hematopoietic malignancies like chronic lymphocytic leukemia (CLL). While no point mutations have been found in the protein coding candidate genes at 13q14.3, they are deregulated in malignant cells, suggesting an epigenetic tumor suppressor mechanism. We therefore characterized the epigenetic makeup of 13q14.3 in CLL cells and found histone modifications by chromatin-immunoprecipitation (ChIP) that are associated with activated transcription and significant DNA-demethylation at the transcriptional start sites of DLEU1 and DLEU2 using 5 different semi-quantitative and quantitative methods (aPRIMES, BioCOBRA, MCIp, MassARRAY, and bisulfite sequencing). These epigenetic aberrations were correlated with transcriptional deregulation of the neighboring candidate tumor suppressor genes, suggesting a coregulation in cis of this gene cluster. We found that the 13q14.3 genes in addition to their previously known functions regulate NF-kB activity, which we could show after overexpression, siRNA-mediated knockdown, and dominant-negative mutant genes by using Western blots with previously undescribed antibodies, by a customized ELISA as well as by reporter assays. In addition, we performed an unbiased screen of 810 human miRNAs and identified the miR-15/16 family of genes at 13q14.3 as the strongest inducers of NF-kB activity. In summary, the tumor suppressor mechanism at 13q14.3 is a cluster of genes controlled by two lncRNA genes that are regulated by DNA-methylation and histone modifications and whose members all regulate NF-kB. Therefore, the tumor suppressor mechanism in 13q14.3 underlines the role both of epigenetic aberrations and of lncRNA genes in human tumorigenesis and is an example of colocalization of a functionally related gene cluster.

Over the last few years, circulating microRNAs (miRNAs) have emerged as promising novel and minimally invasive markers for various diseases, including cancer. We already showed that certain miRNAs are deregulated in the plasma of breast cancer patients when compared to healthy women. Herein we have further explored their potential to serve as breast cancer early detection markers in blood plasma. Circulating miR-127-3p, miR-376a and miR-652, selected as candidates from a miRNA array-based screening, were found to be associated with breast cancer for the first time (n = 417). Further we validated our previously reported circulating miRNAs (miR-148b, miR-376c, miR-409-3p and miR-801) in an independent cohort (n = 210) as elevated in the plasma of breast cancer patients compared to healthy women. We described, for the first time in breast cancer, an over-representation of deregulated miRNAs (miR-127-3p, miR-376a, miR-376c and miR-409-3p) originating from the chromosome 14q32 region. The inclusion of patients with benign breast tumors enabled the observation that miR-148b, miR-652 and miR-801 levels are even elevated in the plasma of women with benign tumors when compared to healthy controls. Furthermore, an analysis of samples stratified by cancer stage demonstrated that miR-127-3p, miR-148b, miR-409-3p, miR-652 and miR-801 can detect also stage I or stage II breast cancer thus making them attractive candidates for early detection. Finally, ROC curve analysis showed that a panel of these seven circulating miRNAs has substantial diagnostic potential with an AUC of 0.81 for the detection of benign and malignant breast tumors, which further increased to 0.86 in younger women (up to 50 years of age).

Background Chronic lymphocytic leukemia has a variable clinical course. Genomic aberrations identify prognostic subgroups, pointing towards distinct underlying biological mechanisms that are poorly understood. In particular it remains unclear whether the prognostic subgroups of chronic lymphocytic leukemia are characterized by different levels of leukemogenic proteins.Design and Methods Expression of 23 proteins involved in apoptosis, proliferation, DNA damage, and signaling or whose genes map to chromosomal regions known to be critical in chronic lymphocytic leukemia was quantified in 185 cytogenetically well characterized cases of chronic lymphocytic leukemia using immunoblotting. Cases were categorized hierarchically into deletion(17p), deletion(11q), trisomy 12, deletion(13q) as sole abnormality or normal karyotype. Statistical analysis was performed for expression differences between these subgroups. In addition, the expression levels of CDK4, P27 and P53 were quantified over the clinical course and compared to levels in immunopurified B cells from healthy individuals.Results In subgroups with a good prognosis, differential expression was mainly seen for proteins that regulate apoptosis. In contrast, in cytogenetic subgroups with a worse prognosis, differential expression was mostly detected for proteins that control DNA damage and proliferation. Expression levels of CDK4, P27 and P53 were higher compared to those in B cells from healthy individuals and significantly correlated with increasing hierarchical risk. In addition, no significant longitudinal changes of expression levels of CDK4, P27 and P53 could be detected in chronic lymphocytic leukemia patients.Conclusions Differences in expression levels of apoptosis- and proliferation-controlling proteins define distinct prognostic subgroups of chronic lymphocytic leukemia and uncover a correlation of levels of CDK4, P27 and P53 proteins with higher hierarchical risk.