Integrative genomic analyses of European intrahepatic cholangiocarcinoma: Novel ROS1 fusion gene and PBX1 as prognostic marker.

BACKGROUND: Cholangiocarcinoma (CCA) is a fatal cancer of the bile duct with a poor prognosis owing to limited therapeutic options. The incidence of intrahepatic CCA (iCCA) is increasing worldwide, and its molecular basis is emerging. Environmental factors may contribute to regional differences in the mutation spectrum of European patients with iCCA, which are underrepresented in systematic genomic and transcriptomic studies of the disease. METHODS: We describe an integrated whole-exome sequencing and transcriptomic study of 37 iCCAs patients in Germany. RESULTS: We observed as most frequently mutated genes ARID1A (14%), IDH1, BAP1, TP53, KRAS, and ATM in 8% of patients. We identified FGFR2::BICC1 fusions in two tumours, and FGFR2::KCTD1 and TMEM106B::ROS1 as novel fusions with potential therapeutic implications in iCCA and confirmed oncogenic properties of TMEM106B::ROS1 in vitro. Using a data integration framework, we identified PBX1 as a novel central regulatory gene in iCCA. We performed extended screening by targeted sequencing of an additional 40 CCAs. In the joint analysis, IDH1 (13%), BAP1 (10%), TP53 (9%), KRAS (7%), ARID1A (7%), NF1 (5%), and ATM (5%) were the most frequently mutated genes, and we found PBX1 to show copy gain in 20% of the tumours. According to other studies, amplifications of PBX1 tend to occur in European iCCAs in contrast to liver fluke-associated Asian iCCAs. CONCLUSIONS: By analyzing an additional European cohort of iCCA patients, we found that PBX1 protein expression was a marker of poor prognosis. Overall, our findings provide insight into key molecular alterations in iCCA, reveal new targetable fusion genes, and suggest that PBX1 is a novel modulator of this disease.

Aberrant JAK-STAT signaling-mediated chromatin remodeling impairs the sensitivity of NK/T-cell lymphoma to chidamide.

BACKGROUND: Natural killer/T-cell lymphoma (NKTL) is a rare type of aggressive and heterogeneous non-Hodgkin's lymphoma (NHL) with a poor prognosis and limited therapeutic options. Therefore, there is an urgent need to exploit potential novel therapeutic targets for the treatment of NKTL. Histone deacetylase (HDAC) inhibitor chidamide was recently approved for treating relapsed/refractory peripheral T-cell lymphoma (PTCL) patients. However, its therapeutic efficacy in NKTL remains unclear. METHODS: We performed a phase II clinical trial to evaluate the efficacy of chidamide in 28 relapsed/refractory NKTL patients. Integrative transcriptomic, chromatin profiling analysis and functional studies were performed to identify potential predictive biomarkers and unravel the mechanisms of resistance to chidamide. Immunohistochemistry (IHC) was used to validate the predictive biomarkers in tumors from the clinical trial. RESULTS: We demonstrated that chidamide is effective in treating relapsed/refractory NKTL patients, achieving an overall response and complete response rate of 39 and 18%, respectively. In vitro studies showed that hyperactivity of JAK-STAT signaling in NKTL cell lines was associated with the resistance to chidamide. Mechanistically, our results revealed that aberrant JAK-STAT signaling remodels the chromatin and confers resistance to chidamide. Subsequently, inhibition of JAK-STAT activity could overcome resistance to chidamide by reprogramming the chromatin from a resistant to sensitive state, leading to synergistic anti-tumor effect in vitro and in vivo. More importantly, our clinical data demonstrated that combinatorial therapy with chidamide and JAK inhibitor ruxolitinib is effective against chidamide-resistant NKTL. In addition, we identified TNFRSF8 (CD30), a downstream target of the JAK-STAT pathway, as a potential biomarker that could predict NKTL sensitivity to chidamide. CONCLUSIONS: Our study suggests that chidamide, in combination with JAK-STAT inhibitors, can be a novel targeted therapy in the standard of care for NKTL. TRIAL REGISTRATION: ClinicalTrials.gov, NCT02878278. Registered 25 August 2016, https://clinicaltrials.gov/ct2/show/NCT02878278.

A genomic-augmented multivariate prognostic model for the survival of natural-killer/T-cell lymphoma patients from an international cohort.

With lowering costs of sequencing and genetic profiling techniques, genetic drivers can now be detected readily in tumors but current prognostic models for Natural-killer/T cell lymphoma (NKTCL) have yet to fully leverage on them for prognosticating patients. Here, we used next-generation sequencing to sequence 260 NKTCL tumors, and trained a genomic prognostic model (GPM) with the genomic mutations and survival data from this retrospective cohort of patients using LASSO Cox regression. The GPM is defined by the mutational status of 13 prognostic genes and is weakly correlated with the risk-features in International Prognostic Index (IPI), Prognostic Index for Natural-Killer cell lymphoma (PINK), and PINK-Epstein-Barr virus (PINK-E). Cox-proportional hazard multivariate regression also showed that the new GPM is independent and significant for both progression-free survival (PFS, HR: 3.73, 95% CI 2.07-6.73; p < .001) and overall survival (OS, HR: 5.23, 95% CI 2.57-10.65; p = .001) with known risk-features of these indices. When we assign an additional risk-score to samples, which are mutant for the GPM, the Harrell's C-indices of GPM-augmented IPI, PINK, and PINK-E improved significantly (p < .001, χ2 test) for both PFS and OS. Thus, we report on how genomic mutational information could steer toward better prognostication of NKTCL patients.

CREBBP cooperates with the cell cycle machinery to attenuate chidamide sensitivity in relapsed/refractory diffuse large B-cell lymphoma.

Diffuse large B-cell lymphoma (DLBCL) exhibits frequent inactivating mutations of the histone acetyltransferase CREBBP, highlighting the attractiveness of targeting CREBBP deficiency as a therapeutic strategy. In this study, we demonstrate that chidamide, a novel histone deacetylase (HDAC) inhibitor, is effective in treating a subgroup of relapsed/refractory DLBCL patients, achieving an overall response rate (ORR) of 25.0% and a complete response (CR) rate of 15.0%. However, the clinical response to chidamide remains poor, as most patients exhibit resistance, hampering the clinical utility of the drug. Functional in vitro and in vivo studies have shown that CREBBP loss of function is correlated with chidamide sensitivity, which is associated with modulation of the cell cycle machinery. A combinatorial drug screening of 130 kinase inhibitors targeting cell cycle regulators identified AURKA inhibitors, which inhibit the G2/M transition during the cell cycle, as top candidates that synergistically enhanced the antitumor effects of chidamide in CREBBP-proficient DLBCL cells. Our study demonstrates that CREBBP inactivation can serve as a potential biomarker to predict chidamide sensitivity, while combination of an AURKA inhibitor and chidamide is a novel therapeutic strategy for the treatment of relapsed/refractory DLBCL.

Evaluation of the PIK3 pathway in peripheral T-cell lymphoma and NK/T-cell lymphoma.

Peripheral T-cell lymphomas (PTCL) and natural killer (NK)/T-cell lymphomas (NKTCL) are a heterogeneous group of aggressive malignancies with dismal outcomes and limited treatment options. While the phosphatidylinositol 3-kinase (PIK3) pathway has been shown to be highly activated in many B-cell lymphomas, its therapeutic relevance in PTCL and NKTCL remains unclear. The aim of this study is to investigate the expression of PIK3 and phosphatase and tensin homolog (PTEN) in these subtypes of lymphoma and to identify potential therapeutic targets for clinical testing. Therefore, the expression of PIK3α, PIK3ß, PIK3γ, PIK3δ and PTEN was analyzed in 88 cases of PTCL and NKTCL samples by immunohistochemistry. All PTCL and NKTCL samples demonstrated high expression of PIK3 isoforms. In particular, high PIK3α expression was significantly associated with poor survival, even after adjustment for age, International Prognostic Index (IPI) score and anthracycline-based chemotherapy in first line. Notably, copanlisib, a pan-class I inhibitor with predominant activities towards PIK3α and PIK3δ isoforms, effectively inhibited phosphorylation of AKT, 4E-BP-1 and STAT3, causing G0 /G1 cell cycle arrest and resulting in suppression of tumour cell growth in vitro and in vivo. This study provides evidence that targeting the PIK3 pathway, particularly simultaneous inhibition of PIK3α and δ, could be a promising approach for the treatment of PTCL and NKTCL.

A clinicohaematological prognostic model for nasal-type natural killer/T-cell lymphoma: A multicenter study.

Extranodal NK/T-cell lymphoma, nasal type (NKTL) is an aggressive type of non-Hodgkin lymphoma closely associated with Epstein-Barr virus and characterized by varying degrees of systemic inflammation. We aim to examine the prognostic significance of peripheral blood neutrophil-lymphocyte ratio (NLR) in patients with NKTL. Therefore, we conducted a retrospective review of 178 patients with biopsy-proven NKTL from the National Cancer Centre Singapore and Samsung Medical Center, South Korea. Using receiver operating curve analysis, an optimal cut-off for high NLR (>3.5) in predicting overall survival (OS) was derived. Survival analysis was performed using the Kaplan-Meier method and multivariable Cox proportional regression. In patients with high NLR, estimated 5-year OS was 25% compared to 53% in those with low NLR. In multivariable analysis, high NLR, in addition to age ≥60 years, presence of B-symptoms and stage III/IV at diagnosis, was independently correlated with worse OS (HR 2.08; 95% CI 1.36 to 3.18; p = 0.0008) and progression-free survival (HR 1.66; 95% CI 1.11 to 2.46; p = 0.0128). A new prognostic index (NABS score) derived from these factors stratified patients into low (0), low-intermediate (1), high-intermediate (2) and high (3-4) risk subgroups, which were associated with 5-year OS of 76.5%, 55.7%, 29.2% and 0% respectively. In conclusion, high NLR is an independent prognostic marker and the NABS model can be used to risk-stratify NKTL patients.

Genomic and transcriptomic landscapes of Epstein-Barr virus in extranodal natural killer T-cell lymphoma.

Extranodal natural killer T-cell lymphoma (nasal type; NKTCL) is an aggressive malignancy strongly associated with Epstein-Barr virus (EBV) infection. However, the role of EBV in NKTCL development is unclear, largely due to the lack of information about EBV genome and transcriptome in NKTCL. Here, using high-throughput sequencing, we obtained whole genome (n = 27) and transcriptome datasets (n = 18) of EBV derived from NKTCL tumor biopsies. We assembled 27 EBV genomes and detected an average of 1,152 single nucleotide variants and 44.8 indels (<50 bp) of EBV per sample. We also identified frequent focal EBV genome deletions and integrated EBV fragments in the host genome. Moreover, Phylogenetic analysis revealed that NKTCL-derived EBVs are closely clustered; transcriptome analysis revealed less activation of both latent and lytic genes and larger amount of T-cell epitope alterations in NKTCL, as compared with other EBV-associated cancers. Furthermore, we observed transcriptional defects of the BARTs miRNA by deletion, and the disruption of host NHEJ1 by integrated EBV fragment, implying novel pathogenic mechanisms of EBV. Taken together, we reported for the first time global mutational and transcriptional profiles of EBV in NKTCL clinical samples, revealing important somatic events of EBV and providing insights to better understanding of EBV's contribution in tumorigenesis.

Oncogenic activation of the STAT3 pathway drives PD-L1 expression in natural killer/T-cell lymphoma.

Mature T-cell lymphomas, including peripheral T-cell lymphoma (PTCL) and extranodal NK/T-cell lymphoma (NKTL), represent a heterogeneous group of non-Hodgkin lymphomas with dismal outcomes and limited treatment options. To determine the extent of involvement of the JAK/STAT pathway in this malignancy, we performed targeted capture sequencing of 188 genes in this pathway in 171 PTCL and NKTL cases. A total of 272 nonsynonymous somatic mutations in 101 genes were identified in 73% of the samples, including 258 single-nucleotide variants and 14 insertions or deletions. Recurrent mutations were most frequently located in STAT3 and TP53 (15%), followed by JAK3 and JAK1 (6%) and SOCS1 (4%). A high prevalence of STAT3 mutation (21%) was observed specifically in NKTL. Novel STAT3 mutations (p.D427H, E616G, p.E616K, and p.E696K) were shown to increase STAT3 phosphorylation and transcriptional activity of STAT3 in the absence of cytokine, in which p.E616K induced programmed cell death-ligand 1 (PD-L1) expression by robust binding of activated STAT3 to the PD-L1 gene promoter. Consistent with these findings, PD-L1 was overexpressed in NKTL cell lines harboring hotspot STAT3 mutations, and similar findings were observed by the overexpression of p.E616K and p.E616G in the STAT3 wild-type NKTL cell line. Conversely, STAT3 silencing and inhibition decreased PD-L1 expression in STAT3 mutant NKTL cell lines. In NKTL tumors, STAT3 activation correlated significantly with PD-L1 expression. We demonstrated that STAT3 activation confers high PD-L1 expression, which may promote tumor immune evasion. The combination of PD-1/PD-L1 antibodies and STAT3 inhibitors might be a promising therapeutic approach for NKTL, and possibly PTCL.

ConsensusDriver Improves upon Individual Algorithms for Predicting Driver Alterations in Different Cancer Types and Individual Patients.

Existing cancer driver prediction methods are based on very different assumptions and each of them can detect only a particular subset of driver genes. Here we perform a comprehensive assessment of 18 driver prediction methods on more than 3,400 tumor samples from 15 cancer types, all to determine their suitability in guiding precision medicine efforts. We categorized these methods into five groups: functional impact on proteins in general (FI) or specific to cancer (FIC), cohort-based analysis for recurrent mutations (CBA), mutations with expression correlation (MEC), and methods that use gene interaction network-based analysis (INA). The performance of driver prediction methods varied considerably, with concordance with a gold standard varying from 9% to 68%. FI methods showed relatively poor performance (concordance <22%), while CBA methods provided conservative results but required large sample sizes for high sensitivity. INA methods, through the integration of genomic and transcriptomic data, and FIC methods, by training cancer-specific models, provided the best trade-off between sensitivity and specificity. As the methods were found to predict different subsets of driver genes, we propose a novel consensus-based approach, ConsensusDriver, which significantly improves the quality of predictions (20% increase in sensitivity) in patient subgroups or even individual patients. Consensus-based methods like ConsensusDriver promise to harness the strengths of different driver prediction paradigms.Significance: These findings assess state-of-the-art cancer driver prediction methods and develop a new and improved consensus-based approach for use in precision oncology. Cancer Res; 78(1); 290-301. ©2017 AACR.

Critical Assessment of Metagenome Interpretation-a benchmark of metagenomics software.

Methods for assembly, taxonomic profiling and binning are key to interpreting metagenome data, but a lack of consensus about benchmarking complicates performance assessment. The Critical Assessment of Metagenome Interpretation (CAMI) challenge has engaged the global developer community to benchmark their programs on highly complex and realistic data sets, generated from ∼700 newly sequenced microorganisms and ∼600 novel viruses and plasmids and representing common experimental setups. Assembly and genome binning programs performed well for species represented by individual genomes but were substantially affected by the presence of related strains. Taxonomic profiling and binning programs were proficient at high taxonomic ranks, with a notable performance decrease below family level. Parameter settings markedly affected performance, underscoring their importance for program reproducibility. The CAMI results highlight current challenges but also provide a roadmap for software selection to answer specific research questions.

OPERA-LG: efficient and exact scaffolding of large, repeat-rich eukaryotic genomes with performance guarantees.

The assembly of large, repeat-rich eukaryotic genomes represents a significant challenge in genomics. While long-read technologies have made the high-quality assembly of small, microbial genomes increasingly feasible, data generation can be expensive for larger genomes. OPERA-LG is a scalable, exact algorithm for the scaffold assembly of large, repeat-rich genomes, out-performing state-of-the-art programs for scaffold correctness and contiguity. It provides a rigorous framework for scaffolding of repetitive sequences and a systematic approach for combining data from different second-generation and third-generation sequencing technologies. OPERA-LG provides an avenue for systematic augmentation and improvement of thousands of existing draft eukaryotic genome assemblies.

Patient-specific driver gene prediction and risk assessment through integrated network analysis of cancer omics profiles.

Extensive and multi-dimensional data sets generated from recent cancer omics profiling projects have presented new challenges and opportunities for unraveling the complexity of cancer genome landscapes. In particular, distinguishing the unique complement of genes that drive tumorigenesis in each patient from a sea of passenger mutations is necessary for translating the full benefit of cancer genome sequencing into the clinic. We address this need by presenting a data integration framework (OncoIMPACT) to nominate patient-specific driver genes based on their phenotypic impact. Extensive in silico and in vitro validation helped establish OncoIMPACT's robustness, improved precision over competing approaches and verifiable patient and cell line specific predictions (2/2 and 6/7 true positives and negatives, respectively). In particular, we computationally predicted and experimentally validated the gene TRIM24 as a putative novel amplified driver in a melanoma patient. Applying OncoIMPACT to more than 1000 tumor samples, we generated patient-specific driver gene lists in five different cancer types to identify modes of synergistic action. We also provide the first demonstration that computationally derived driver mutation signatures can be overall superior to single gene and gene expression based signatures in enabling patient stratification and prognostication. Source code and executables for OncoIMPACT are freely available from http://sourceforge.net/projects/oncoimpact.

The importance of study design for detecting differentially abundant features in high-throughput experiments.

High-throughput assays, such as RNA-seq, to detect differential abundance are widely used. Variable performance across statistical tests, normalizations, and conditions leads to resource wastage and reduced sensitivity. EDDA represents a first, general design tool for RNA-seq, Nanostring, and metagenomic analysis, that rationally selects tests, predicts performance, and plans experiments to minimize resource wastage. Case studies highlight EDDA's ability to model single-cell RNA-seq, suggesting ways to reduce sequencing costs up to five-fold and improving metagenomic biomarker detection through improved test selection. EDDA's novel mode-based normalization for detecting differential abundance improves robustness by 10% to 20% and precision by up to 140%.

Differential co-expression framework to quantify goodness of biclusters and compare biclustering algorithms.

BACKGROUND: Biclustering is an important analysis procedure to understand the biological mechanisms from microarray gene expression data. Several algorithms have been proposed to identify biclusters, but very little effort was made to compare the performance of different algorithms on real datasets and combine the resultant biclusters into one unified ranking. RESULTS: In this paper we propose differential co-expression framework and a differential co-expression scoring function to objectively quantify quality or goodness of a bicluster of genes based on the observation that genes in a bicluster are co-expressed in the conditions belonged to the bicluster and not co-expressed in the other conditions. Furthermore, we propose a scoring function to stratify biclusters into three types of co-expression. We used the proposed scoring functions to understand the performance and behavior of the four well established biclustering algorithms on six real datasets from different domains by combining their output into one unified ranking. CONCLUSIONS: Differential co-expression framework is useful to provide quantitative and objective assessment of the goodness of biclusters of co-expressed genes and performance of biclustering algorithms in identifying co-expression biclusters. It also helps to combine the biclusters output by different algorithms into one unified ranking i.e. meta-biclustering.