Charlotte Soneson

The consensus molecular subtypes of colorectal cancer

Colorectal cancer (CRC) is a frequently lethal disease with heterogeneous outcomes and drug responses. To resolve inconsistencies among the reported gene expression–based CRC classifications and facilitate clinical translation, we formed an international consortium dedicated to large-scale data sharing and analytics across expert groups. We show marked interconnectivity between six independent classification systems coalescing into four consensus molecular subtypes (CMSs) with distinguishing features: CMS1 (microsatellite instability immune, 14%), hypermutated, microsatellite unstable and strong immune activation; CMS2 (canonical, 37%), epithelial, marked WNT and MYC signaling activation; CMS3 (metabolic, 13%), epithelial and evident metabolic dysregulation; and CMS4 (mesenchymal, 23%), prominent transforming growth factor–β activation, stromal invasion and angiogenesis. Samples with mixed features (13%) possibly represent a transition phenotype or intratumoral heterogeneity. We consider the CMS groups the most robust classification system currently available for CRC—with clear biological interpretability—and the basis for future clinical stratification and subtype-based targeted interventions.

A comparison of methods for differential expression analysis of RNA-seq data

BackgroundFinding genes that are differentially expressed between conditions is an integral part of understanding the molecular basis of phenotypic variation. In the past decades, DNA microarrays have been used extensively to quantify the abundance of mRNA corresponding to different genes, and more recently high-throughput sequencing of cDNA (RNA-seq) has emerged as a powerful competitor. As the cost of sequencing decreases, it is conceivable that the use of RNA-seq for differential expression analysis will increase rapidly. To exploit the possibilities and address the challenges posed by this relatively new type of data, a number of software packages have been developed especially for differential expression analysis of RNA-seq data.ResultsWe conducted an extensive comparison of eleven methods for differential expression analysis of RNA-seq data. All methods are freely available within the R framework and take as input a matrix of counts, i.e. the number of reads mapping to each genomic feature of interest in each of a number of samples. We evaluate the methods based on both simulated data and real RNA-seq data.ConclusionsVery small sample sizes, which are still common in RNA-seq experiments, impose problems for all evaluated methods and any results obtained under such conditions should be interpreted with caution. For larger sample sizes, the methods combining a variance-stabilizing transformation with the ‘limma’ method for differential expression analysis perform well under many different conditions, as does the nonparametric SAMseq method.

Distal and proximal colon cancers differ in terms of molecular, pathological and clinical features

BACKGROUND Differences exist between the proximal and distal colon in terms of developmental origin, exposure to patterning genes, environmental mutagens, and gut flora. Little is known on how these differences may affect mechanisms of tumorigenesis, side-specific therapy response or prognosis. We explored systematic differences in pathway activation and their clinical implications. MATERIALS AND METHODS Detailed clinicopathological data for 3045 colon carcinoma patients enrolled in the PETACC3 adjuvant chemotherapy trial were available for analysis. A subset of 1404 samples had molecular data, including gene expression and DNA copy number profiles for 589 and 199 samples, respectively. In addition, 413 colon adenocarcinoma from TCGA collection were also analyzed. Tumor side-effect on anti-epidermal growth factor receptor (EGFR) therapy was assessed in a cohort of 325 metastatic patients. Outcome variables considered were relapse-free survival and survival after relapse (SAR). RESULTS Proximal carcinomas were more often mucinous, microsatellite instable (MSI)-high, mutated in key tumorigenic pathways, expressed a B-Raf proto-oncogene, serine/threonine kinase (BRAF)-like and a serrated pathway signature, regardless of histological type. Distal carcinomas were more often chromosome instable and EGFR or human epidermal growth factor receptor 2 (HER2) amplified, and more frequently overexpressed epiregulin. While risk of relapse was not different per side, SAR was much poorer for proximal than for distal stage III carcinomas in a multivariable model including BRAF mutation status [N = 285; HR 1.95, 95% CI (1.6-2.4), P < 0.001]. Only patients with metastases from a distal carcinoma responded to anti-EGFR therapy, in line with the predictions of our pathway enrichment analysis. CONCLUSIONS Colorectal carcinoma side is associated with differences in key molecular features, some immediately druggable, with important prognostic effects which are maintained in metastatic lesions. Although within side significant molecular heterogeneity remains, our findings justify stratification of patients by side for retrospective and prospective analyses of drug efficacy and prognosis.

Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences

High-throughput sequencing of cDNA (RNA-seq) is used extensively to characterize the transcriptome of cells. Many transcriptomic studies aim at comparing either abundance levels or the transcriptome composition between given conditions, and as a first step, the sequencing reads must be used as the basis for abundance quantification of transcriptomic features of interest, such as genes or transcripts. Various quantification approaches have been proposed, ranging from simple counting of reads that overlap given genomic regions to more complex estimation of underlying transcript abundances. In this paper, we show that gene-level abundance estimates and statistical inference offer advantages over transcript-level analyses, in terms of performance and interpretability. We also illustrate that the presence of differential isoform usage can lead to inflated false discovery rates in differential gene expression analyses on simple count matrices but that this can be addressed by incorporating offsets derived from transcript-level abundance estimates. We also show that the problem is relatively minor in several real data sets. Finally, we provide an R package ( tximport) to help users integrate transcript-level abundance estimates from common quantification pipelines into count-based statistical inference engines.

Long-lasting stem cell-like memory CD8 + T cells with a naïve-like profile upon yellow fever vaccination

The yellow fever vaccine induces a CD8+ T stem cell–like memory subset with a naïve-like profile that persists long term. Yellow fever vaccine induces long-term naïve-like memory In the ongoing quest to find better models of human disease, humans themselves are frequently overlooked. New vaccines for viral infections have hit barriers in translating attempts to induce protective immunity by producing long-lasting memory T cell responses. Now, Fuertes Marraco et al. report that individuals who receive the current yellow fever vaccine develop just that. They found that yellow fever–specific CD8+ T cells with a naïve-like phenotype persisted in vaccinated individuals for more than 25 years. These cells were capable of self-renewal and resembled the stem cell–like memory subset. Thus, by studying vaccinated individuals and building on their own success, researcher may learn—in people—what exactly makes long-term memory T cells tick. Efficient and persisting immune memory is essential for long-term protection from infectious and malignant diseases. The yellow fever (YF) vaccine is a live attenuated virus that mediates lifelong protection, with recent studies showing that the CD8+ T cell response is particularly robust. Yet, limited data exist regarding the long-term CD8+ T cell response, with no studies beyond 5 years after vaccination. We investigated 41 vaccinees, spanning 0.27 to 35 years after vaccination. YF-specific CD8+ T cells were readily detected in almost all donors (38 of 41), with frequencies decreasing with time. As previously described, effector cells dominated the response early after vaccination. We detected a population of naïve-like YF-specific CD8+ T cells that was stably maintained for more than 25 years and was capable of self-renewal ex vivo. In-depth analyses of markers and genome-wide mRNA profiling showed that naïve-like YF-specific CD8+ T cells in vaccinees (i) were distinct from genuine naïve cells in unvaccinated donors, (ii) resembled the recently described stem cell–like memory subset (Tscm), and (iii) among all differentiated subsets, had profiles closest to naïve cells. Our findings reveal that CD8+ Tscm are efficiently induced by a vaccine in humans, persist for decades, and preserve a naïveness-like profile. These data support YF vaccination as an optimal mechanistic model for the study of long-lasting memory CD8+ T cells in humans.

Early changes in the hypothalamic region in prodromal Huntington disease revealed by MRI analysis.

Huntington disease (HD) is a fatal neurodegenerative disorder caused by an expanded CAG repeat. Its length can be used to estimate the time of clinical diagnosis, which is defined by overt motor symptoms. Non-motor symptoms begin before motor onset, and involve changes in hypothalamus-regulated functions such as sleep, emotion and metabolism. Therefore we hypothesized that hypothalamic changes occur already prior to the clinical diagnosis. We performed voxel-based morphometry and logistic regression analyses of cross-sectional MR images from 220 HD gene carriers and 75 controls in the Predict-HD study. We show that changes in the hypothalamic region are detectable before clinical diagnosis and that its grey matter contents alone are sufficient to distinguish HD gene carriers from control cases. In conclusion, our study shows, for the first time, that alterations in grey matter contents in the hypothalamic region occur at least a decade before clinical diagnosis in HD using MRI.

Integrative analysis of gene expression and copy number alterations using canonical correlation analysis

BackgroundWith the rapid development of new genetic measurement methods, several types of genetic alterations can be quantified in a high-throughput manner. While the initial focus has been on investigating each data set separately, there is an increasing interest in studying the correlation structure between two or more data sets. Multivariate methods based on Canonical Correlation Analysis (CCA) have been proposed for integrating paired genetic data sets. The high dimensionality of microarray data imposes computational difficulties, which have been addressed for instance by studying the covariance structure of the data, or by reducing the number of variables prior to applying the CCA. In this work, we propose a new method for analyzing high-dimensional paired genetic data sets, which mainly emphasizes the correlation structure and still permits efficient application to very large data sets. The method is implemented by translating a regularized CCA to its dual form, where the computational complexity depends mainly on the number of samples instead of the number of variables. The optimal regularization parameters are chosen by cross-validation. We apply the regularized dual CCA, as well as a classical CCA preceded by a dimension-reducing Principal Components Analysis (PCA), to a paired data set of gene expression changes and copy number alterations in leukemia.ResultsUsing the correlation-maximizing methods, regularized dual CCA and PCA+CCA, we show that without pre-selection of known disease-relevant genes, and without using information about clinical class membership, an exploratory analysis singles out two patient groups, corresponding to well-known leukemia subtypes. Furthermore, the variables showing the highest relevance to the extracted features agree with previous biological knowledge concerning copy number alterations and gene expression changes in these subtypes. Finally, the correlation-maximizing methods are shown to yield results which are more biologically interpretable than those resulting from a covariance-maximizing method, and provide different insight compared to when each variable set is studied separately using PCA.ConclusionsWe conclude that regularized dual CCA as well as PCA+CCA are useful methods for exploratory analysis of paired genetic data sets, and can be efficiently implemented also when the number of variables is very large.

Serum neurofilament light chain in early relapsing remitting MS is increased and correlates with CSF levels and with MRI measures of disease severity.

Background/objectives: Neurofilament light chain (NfL) levels in the cerebrospinal fluid (CSF) of multiple sclerosis (MS) patients correlate with the degree of neuronal injury. To date, little is known about NfL concentrations in the serum of relapsing remitting multiple sclerosis (RRMS) patients and their relationship with CSF levels and magnetic resonance imaging (MRI) measures of disease severity. We aimed to validate the quantification of NfL in serum samples of RRMS, as a biofluid source easily accessible for longitudinal studies. Methods: A total of 31 RRMS patients underwent CSF and serum sampling. After a median time of 3.6 years, 19 of these RRMS patients, 10 newly recruited RRMS patients and 18 healthy controls had a 3T MRI and serum sampling. NfL concentrations were determined by electrochemiluminescence immunoassay. Results: NfL levels in serum were highly correlated to levels in CSF (r = 0.62, p = 0.0002). Concentrations in serum were higher in patients than in controls at baseline (p = 0.004) and follow-up (p = 0.0009) and did not change over time (p = 0.56). Serum NfL levels correlated with white matter (WM) lesion volume (r = 0.68, p < 0.0001), mean T1 (r = 0.40, p = 0.034) and T2* relaxation time (r = 0.49, p = 0.007) and with magnetization transfer ratio in normal appearing WM (r = −0.41, p = 0.029). Conclusion: CSF and serum NfL levels were highly correlated, and serum concentrations were increased in RRMS. Serum NfL levels correlated with MRI markers of WM disease severity. Our findings further support longitudinal studies of serum NfL as a potential biomarker of on-going disease progression and as a potential surrogate to quantify effects of neuroprotective drugs in clinical trials.

A glioma classification scheme based on coexpression modules of EGFR and PDGFRA

Significance Classification of cancer provides crucial guidance for clinical treatment and mechanistic studies. Our work extends previous glioma classification studies in that we established EGFR module (EM)/PDGFRA module (PM) glioma classification scheme based on gene coexpression modules around key signaling pathways conserved in neural development and gliomagenesis. We identified coexpressed EM and PM genes as classifiers. Based on the EM and PM signatures, our classification scheme robustly assigns adult low-grade and high-grade diffuse gliomas into three major subtypes that are distinct in patient survival, and in transcriptomic and genomic patterns. Our work suggests that EM and PM genes may play currently unrecognized roles in gliomagenesis. EM/PM glioma classification scheme forms a framework toward establishing molecular diagnostic tools and identifying new therapeutic targets to combat gliomas. We hypothesized that key signaling pathways of glioma genesis might enable the molecular classification of gliomas. Gene coexpression modules around epidermal growth factor receptor (EGFR) (EM, 29 genes) or platelet derived growth factor receptor A (PDGFRA) (PM, 40 genes) in gliomas were identified. Based on EM and PM expression signatures, nonnegative matrix factorization reproducibly clustered 1,369 adult diffuse gliomas WHO grades II-IV from four independent databases generated in three continents, into the subtypes (EM, PM and EMlowPMlow gliomas) in a morphology-independent manner. Besides their distinct patterns of genomic alterations, EM gliomas were associated with higher age at diagnosis, poorer prognosis, and stronger expression of neural stem cell and astrogenesis genes. Both PM and EMlowPMlow gliomas were associated with younger age at diagnosis and better prognosis. PM gliomas were enriched in the expression of oligodendrogenesis genes, whereas EMlowPMlow gliomas were enriched in the signatures of mature neurons and oligodendrocytes. The EM/PM-based molecular classification scheme is applicable to adult low-grade and high-grade diffuse gliomas, and outperforms existing classification schemes in assigning diffuse gliomas to subtypes with distinct transcriptomic and genomic profiles. The majority of the EM/PM classifiers, including regulators of glial fate decisions, have not been extensively studied in glioma biology. Subsets of these classifiers were coexpressed in mouse glial precursor cells, and frequently amplified or lost in an EM/PM glioma subtype-specific manner, resulting in somatic copy number alteration-dependent gene expression that contributes to EM/PM signatures in glioma samples. EM/PM-based molecular classification provides a molecular diagnostic framework to expedite the search for new glioma therapeutic targets.

The correlation pattern of acquired copy number changes in 164 ETV6/RUNX1-positive childhood acute lymphoblastic leukemias

The ETV6/RUNX1 fusion gene, present in 25% of B-lineage childhood acute lymphoblastic leukemia (ALL), is thought to represent an initiating event, which requires additional genetic changes for leukemia development. To identify additional genetic alterations, 24 ETV6/RUNX1-positive ALLs were analyzed using 500K single nucleotide polymorphism arrays. The results were combined with previously published data sets, allowing us to ascertain genomic copy number aberrations (CNAs) in 164 cases. In total, 45 recurrent CNAs were identified with an average number of 3.5 recurrent changes per case (range 0-13). Twenty-six percent of cases displayed a set of recurrent CNAs identical to that of other cases in the data set. The majority (74%), however, displayed a unique pattern of recurrent CNAs, indicating a large heterogeneity within this ALL subtype. As previously demonstrated, alterations targeting genes involved in B-cell development were common (present in 28% of cases). However, the combined analysis also identified alterations affecting nuclear hormone response (24%) to be a characteristic feature of ETV6/RUNX1-positive ALL. Studying the correlation pattern of the CNAs allowed us to highlight significant positive and negative correlations between specific aberrations. Furthermore, oncogenetic tree models identified ETV6, CDKN2A/B, PAX5, del(6q) and +16 as possible early events in the leukemogenic process.