Improving high-resolution copy number variation analysis from next generation sequencing using unique molecular identifiers.

BACKGROUND: Recently, copy number variations (CNV) impacting genes involved in oncogenic pathways have attracted an increasing attention to manage disease susceptibility. CNV is one of the most important somatic aberrations in the genome of tumor cells. Oncogene activation and tumor suppressor gene inactivation are often attributed to copy number gain/amplification or deletion, respectively, in many cancer types and stages. Recent advances in next generation sequencing protocols allow for the addition of unique molecular identifiers (UMI) to each read. Each targeted DNA fragment is labeled with a unique random nucleotide sequence added to sequencing primers. UMI are especially useful for CNV detection by making each DNA molecule in a population of reads distinct. RESULTS: Here, we present molecular Copy Number Alteration (mCNA), a new methodology allowing the detection of copy number changes using UMI. The algorithm is composed of four main steps: the construction of UMI count matrices, the use of control samples to construct a pseudo-reference, the computation of log-ratios, the segmentation and finally the statistical inference of abnormal segmented breaks. We demonstrate the success of mCNA on a dataset of patients suffering from Diffuse Large B-cell Lymphoma and we highlight that mCNA results have a strong correlation with comparative genomic hybridization. CONCLUSION: We provide mCNA, a new approach for CNV detection, freely available at https://gitlab.com/pierrejulien.viailly/mcna/ under MIT license. mCNA can significantly improve detection accuracy of CNV changes by using UMI.

Application of the cghRA framework to the genomic characterization of Diffuse Large B-Cell Lymphoma.

MOTIVATION: Although sequencing-based technologies are becoming the new reference in genome analysis, comparative genomic hybridization arrays (aCGH) still constitute a simple and reliable approach for copy number analysis. The most powerful algorithms to analyze such data have been freely provided by the scientific community for many years, but combining them is a complex scripting task. RESULTS: The cghRA framework combines a user-friendly graphical interface and a powerful object-oriented command-line interface to handle a full aCGH analysis, as is illustrated in an original series of 107 Diffuse Large B-Cell Lymphomas. New algorithms for copy-number calling, polymorphism detection and minimal common region prioritization were also developed and validated. While their performances will only be demonstrated with aCGH, these algorithms could actually prove useful to any copy-number analysis, whatever the technique used. AVAILABILITY AND IMPLEMENTATION: R package and source for Linux, MS Windows and MacOS are freely available at http://bioinformatics.ovsa.fr/cghRA. CONTACT: mareschal@ovsa.fr or fabrice.jardin@chb.unicancer.fr. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Oncogenic events rather than antigen selection pressure may be the main driving forces for relapse in diffuse large B-cell lymphomas.

Little is known on the phylogenetic relationship between diagnostic and relapse clones of diffuse large B-cell lymphoma (DLBCL). We applied high throughput sequencing (HTS) of the VDJ locus of Immunoglobulin heavy chain (IGHV) on 14 DLBCL patients with serial samples, including tumor biopsies and/or peripheral blood mononuclear cells (PBMC). Phylogenetic data were consolidated with targeted sequencing and cytogenetics. Phylogeny clearly showed that DLBCL relapse could occur according either an early or a late divergent mode. These two modes of divergence were independent from the elapsed time between diagnosis and relapse. We found no significant features for antigen selection pressure in complementary determining region both at diagnosis and relapse for 9/12 pairs and a conserved negative selection pressure for the three remaining cases. Targeted HTS and conventional cytogenetics revealed a branched vs. linear evolution for 5/5 IGHV early divergent cases, but unexpected such "oncogenetic" branched evolution could be found in at least 2/7 IGHV late divergent cases. Thus, if BCR signaling is mandatory for DLBCL emergence, oncogenetic events under chemotherapy selection pressure may be the main driving forces at relapse. Finally, circulating subclones with divergent IGHV somatic hypermutations patterns from initial biopsy could be detected in PBMC at diagnosis for 4/6 patients and, for two of them, at least one was similar to the ones found at relapse. This study highlights that oncogenetic intraclonal diversity of DLBCL should be evaluated beyond the scope a single biopsy and represents a rationale for future investigations using peripheral blood for lymphoid malignancies genotyping. Am. J. Hematol. 92:68-76, 2017. © 2016 Wiley Periodicals, Inc.

Whole exome sequencing of relapsed/refractory patients expands the repertoire of somatic mutations in diffuse large B-cell lymphoma.

Despite the many efforts already spent to enumerate somatic mutations in diffuse large B-cell lymphoma (DLBCL), previous whole-genome and whole-exome studies conducted on patients of mixed outcomes failed at characterizing the 30% of patients who will relapse or resist current immunochemotherapies. To address this issue, we performed whole-exome sequencing of normal/tumoral DNA pairs in 14 relapsed/refractory (R/R) patients subclassified by full-transcriptome arrays (six activated B-cell like, three germinal center B-cell like, and five primary mediastinal B-cell lymphomas), from the LNH-03 LYSA clinical trial program. Aside from well-known DLBCL features, gene and pathway level recurrence analyses proposed several interesting leads including TBL1XR1 and activating mutations in IRF4 or in the insulin regulation pathway. Sequencing-based copy number analysis defined 23 short recurrently altered regions involving genes such as REL, CDKN2A, HYAL2, and TP53. Moreover, it highlighted mutations in genes such as GNA13, CARD11, MFHAS1, and PCLO as associated with secondary variant allele amplification events. The five primary mediastinal B-cell lymphomas (PMBL), while unexpected in a R/R cohort, showed a significantly higher mutation rate (P = 0.003) and provided many insights on this classical Hodgkin lymphoma related subtype. Novel genes such as XPO1, MFHAS1, and ITPKB were found particularly mutated, along with various cytokine-based signaling pathways. Among these analyses, somatic events in the NF-κB pathway were found preponderant in the three DLBCL subtypes, confirming its major implication in DLBCL aggressiveness and pinpointing several new candidate genes.

Detection and prognostic value of recurrent exportin 1 mutations in tumor and cell-free circulating DNA of patients with classical Hodgkin lymphoma.

Classical Hodgkin lymphoma is one of the most common lymphomas and shares clinical and genetic features with primary mediastinal B-cell lymphoma. In this retrospective study, we analyzed the recurrent hotspot mutation of the exportin 1 (XPO1, p.E571K) gene, previously identified in primary mediastinal B-cell lymphoma, in biopsies and plasma circulating cell-free DNA from patients with classical Hodgkin lymphoma using a highly sensitive digital PCR technique. A total of 94 patients were included in the present study. This widely expressed XPO1 E571K mutation is present in one quarter of classical Hodgkin lymphoma patients (24.2%). Mutated and wild-type classical Hodgkin lymphomas were similar regarding the main clinical features. Patients with a detectable XPO1 mutation at the end of treatment displayed a tendency toward shorter progression-free survival, as compared to patients with undetectable mutation in plasma cell-free DNA (2-year progression-free survival: 57.1%, 95% confidence interval: 30.1-100% versus 2-year progression-free survival: 90.5%, 95% confidence interval: 78.8-100%, respectively, P=0.0601). To conclude, the detection of the XPO1 E571K mutation in biopsy and plasma cell-free DNA by digital PCR may be used as a novel biomarker in classical Hodgkin lymphoma for both diagnosis and minimal residual disease, and pinpoints a crucial role of XPO1 in classical Hodgkin lymphoma pathogenesis. The detection of somatic mutation in the plasma cell-free DNA of patients represents a major technological advance in the context of liquid biopsies and noninvasive management of classical Hodgkin lymphoma.

Recurrent mutations of the exportin 1 gene (XPO1) and their impact on selective inhibitor of nuclear export compounds sensitivity in primary mediastinal B-cell lymphoma.

Primary mediastinal B-cell lymphoma (PMBL) is an entity of B-cell lymphoma distinct from the other molecular subtypes of diffuse large B-cell lymphoma (DLBCL). We investigated the prevalence, specificity, and clinical relevance of mutations of XPO1, which encodes a member of the karyopherin-ß nuclear transporters, in a large cohort of PMBL. PMBL cases defined histologically or by gene expression profiling (GEP) were sequenced and the XPO1 mutational status was correlated to genetic and clinical characteristics. The XPO1 mutational status was also assessed in DLBCL, Hodgkin lymphoma (HL) and mediastinal gray-zone lymphoma (MGZL).The biological impact of the mutation on Selective Inhibitor of Nuclear Export (SINE) compounds (KPT-185/330) sensitivity was investigated in vitro. XPO1 mutations were present in 28/117 (24%) PMBL cases and in 5/19 (26%) HL cases but absent/rare in MGZL (0/20) or DLBCL (3/197). A higher prevalence (50%) of the recurrent codon 571 variant (p.E571K) was observed in GEP-defined PMBL and was associated with shorter PFS. Age, International Prognostic Index and bulky mass were similar in XPO1 mutant and wild-type cases. KPT-185 induced a dose-dependent decrease in cell proliferation and increased cell-death in PMBL cell lines harboring wild type or XPO1 E571K mutant alleles. Experiments in transfected U2OS cells further confirmed that the XPO1 E571K mutation does not have a drastic impact on KPT-330 binding. To conclude the XPO1 E571K mutation represents a genetic hallmark of the PMBL subtype and serves as a new relevant PMBL biomarker. SINE compounds appear active for both mutated and wild-type protein. Am. J. Hematol. 91:923-930, 2016. © 2016 Wiley Periodicals, Inc.

Outcome of patients older than 60 years with classical Hodgkin lymphoma treated with front line ABVD chemotherapy: frequent pulmonary events suggest limiting the use of bleomycin in the elderly.

There is no standard of care in elderly classical Hodgkin lymphoma (cHL) patients. ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine), the standard chemotherapy for younger patients, is also used in elderly patients but little is known about toxicity and efficacy. We retrospectively analysed 147 patients aged 60 years and over treated with ABVD in three French haematological centres. Treatment regimen modification was applied in 56 patients for toxicity or HL progression. Bleomycin was removed or reduced in 53 patients, mainly for pulmonary toxicity. Neither initial characteristics nor treatment characteristics were found to correlate with lung toxicity. One hundred and seventeen patients achieved a complete remission, 6 a partial remission, 16 had refractory disease and 8 were non-evaluable. Five-year overall survival was estimated at 67%. With a median follow-up of 58 months, 51 patients died and 14% of deaths were related to lung toxicity. Our study confirms the efficacy of ABVD in elderly patients even if results are inferior to those obtained in younger patients with the same regimen. ABVD can be proposed as front-line chemotherapy in selected elderly cHL patients. The frequency of pulmonary events leads us to propose to either reduce the dose of bleomycin or to remove it from the regimen.

Rgb: a scriptable genome browser for R.

SUMMARY: Thanks to its free licensing and the development of initiatives like Bioconductor, R has become an essential part of the bioinformatics toolbox in the past years and is more and more confronted with genomically located data. While separate solutions are available to manipulate and visualize such data, no R package currently offers the efficiency required for computationally intensive tasks such as interactive genome browsing. The package proposed here fulfills this specific need, providing a multilevel interface suitable for most needs, from a completely interfaced genome browser to low-level classes and methods. Its time and memory efficiency have been challenged in a human dataset, where it outperformed existing solutions by several orders of magnitude. AVAILABILITY AND IMPLEMENTATION: R sources and packages are freely available at the CRAN repository and dedicated Web site: http://bioinformatics.ovsa.fr/Rgb. Distributed under the GPL 3 license, compatible with most operating systems (Windows, Linux, Mac OS) and architectures. CONTACT: maressyl@gmail.com or fabrice.jardin@chb.unicancer.fr SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PET/CT before autologous stem cell transplantation predicts outcome in refractory/relapsed follicular lymphoma.

PURPOSE: Salvage of young patients with follicular lymphoma (FL) after R-CHOP includes salvage immunochemotherapy followed by autologous stem cell transplantation (ASCT). Previous studies dealing with relapsed Hodgkin lymphoma have shown the prognostic value of PET/CT prior to ASCT. METHODS: We retrospectively analysed 59 patients with refractory/relapsed FL after first-line R-CHOP who were chemosensitive (as evaluated by CT) to the salvage treatment and who proceeded to ASCT. The role of PET/CT in this setting to define chemosensitivity is not definitely established. So we focused on the prognostic value of PET/CT performed after salvage treatment, before ASCT. RESULTS: The estimated 3-year progression-free survival (PFS) and overall survival were 63.1% (50.9-78.3%) and 90.5% (82.8 - 98.8%), respectively, and did not differ significantly according to their Follicular Lymphoma International Prognostic Index at relapse, conditioning regimen, or type of salvage. PFS was significantly lower in PET/CT-positive patients, according to the International Harmonization Project revised response criteria, with a 3-year PFS of 45.5% (26.6 - 77.8%) versus 72.6% (58.5 - 90.0%; p = 0.039). To better refine prognosis, we applied two types of throsholds: a Deauville five-point scale positive threshold of ≥3 (3-year PFS of 74.9%, range 61.0 - 92.1% %, versus 42.8%, range 24.7 - 74.4%; p = 0.02), and a ≥70% ∆SUVmax threshold between presalvage and pre-ASCT PET/CT (3-year PFS of 72.4%, range 57.5 - 91.3% versus 13.3%, 2.2 - 81.7%; p < 10(-3)). The PET/CT findings before ASCT were independently correlated with PFS in our series. CONCLUSION: PET/CT negativity before ASCT is a desirable and achievable goal in the management of chemosensitive FL relapsing after first-line R-CHOP.

Targetable activating mutations are very frequent in GCB and ABC diffuse large B-cell lymphoma.

Diffuse large B cell lymphoma (DLBCL) is an aggressive and heterogeneous malignancy that can be divided in two major subgroups, germinal center B-cell-like (GCB) and activated B-cell-like (ABC). Activating mutations of genes involved in the BCR and NF-κB pathways (CD79A, CD79B, MYD88, and CARD11) or in epigenetic regulation (EZH2) have been recently reported, preferentially in one of the two DLBCL subtypes. We analyzed the mutational status of these five recurrently mutated genes in a cohort of 161 untreated de novo DLBCL. Overall, 93 mutations were detected, in 61 (38%) of the patients. The L265P MYD88 mutation was the most frequent MYD88 variant (n = 18), observed exclusively in the ABC subtype. CD79A/CD79B ITAM domains were targeted in ABC DLBCL (12/77; 16%), whereas CARD11 mutations were equally distributed in the two subtypes. The EZH2 Y641 substitution was found almost exclusively in the GCB subgroup (15/62; 24%). Twenty cases (12%) displayed two activating mutations, including the most frequent CD79/MYD88 variants combination (n = 8) which is observed exclusively in the ABC subtype. When considering only ABC DLBCL patients treated by rituximab plus chemotherapy, the presence of an activating NF-κB mutation was associated with an unfavorable outcome (3-years OS 26% for mutated cases versus 67% for the cases without mutations, P = 0.0337). Our study demonstrates that activating and targetable mutations are observed at a very high frequency in DLBCL at the time of diagnosis, indicating that sequencing of a limited number of genes could help tailor an optimal treatment strategy in DLBCL.

Several mechanisms lead to the inactivation of the CDKN2A (P16), P14ARF, or CDKN2B (P15) genes in the GCB and ABC molecular DLBCL subtypes.

Diffuse large B-cell lymphoma (DLBCL) represents the most frequent type of aggressive lymphoma. Deletions of the CDKN2A locus, encoding the proteins CDKN2A (P16), P14ARF, and of the CDKN2B locus, encoding the protein CDKN2B (P15), affect one-third of DLBCL patients. Although other mechanisms that decrease gene expression have been reported, such as promoter methylation, the prognostic value of these mechanisms is still unclear. We studied the deletion and methylation status of these genes in 171 patients and correlated the genomic results with their mRNA expression level and clinical outcome. CDKN2A, P14ARF, and CDKN2B deletions were significantly correlated with decreased mRNA expression (P<0.0001, P<0.0001, and P=0.0148, respectively). P14ARF was methylated in only two patients (1.3%), whereas CDKN2A and CDKN2B were methylated in 36.7 and 31.4% of patients, respectively. Methylation levels greater than 25% were associated with decreased expression of CDKN2A (P=0.0169). CDKN2A and CDKN2B inactivation by deletion or methylation was observed in 42.7 and 37.4% of cases, respectively. Including P14ARF deletions, we identified an inactivating mechanism for at least one of these genes in 47% of patients. Although gene inactivation was not correlated with the international prognostic index, P14ARF and CDKN2B inactivation was significantly associated with shorter survival (P=0.0048 and P=0.0413, respectively), whereas CDKN2A was not (P=0.085). Low mRNA expression levels of these genes were correlated with the ABC phenotype. Furthermore, our results show that an inactivating methylation was more frequent in the GCB phenotype.

Performances of Targeted RNA Sequencing for the Analysis of Fusion Transcripts, Gene Mutation, and Expression in Hematological Malignancies.

RNA sequencing holds great promise to improve the diagnostic of hematological malignancies, because this technique enables to detect fusion transcripts, to look for somatic mutations in oncogenes, and to capture transcriptomic signatures of nosological entities. However, the analytical performances of targeted RNA sequencing have not been extensively described in diagnostic samples. Using a targeted panel of 1385 cancer-related genes in a series of 100 diagnosis samples and 8 controls, we detected all the already known fusion transcripts and also discovered unknown and/or unsuspected fusion transcripts in 12 samples. Regarding the analysis of transcriptomic profiles, we show that targeted RNA sequencing is performant to discriminate acute lymphoblastic leukemia entities driven by different oncogenic translocations. Additionally, we show that 86% of the mutations identified at the DNA level are also detectable at the messenger RNA (mRNA) level, except for nonsense mutations that are subjected to mRNA decay. We conclude that targeted RNA sequencing might improve the diagnosis of hematological malignancies. Standardization of the preanalytical steps and further refinements of the panel design and of the bioinformatical pipelines will be an important step towards its use in standard diagnostic procedures.

Challenging conventional karyotyping by next-generation karyotyping in 281 intensively treated patients with AML.

Although copy number alterations (CNAs) and translocations constitute the backbone of the diagnosis and prognostication of acute myeloid leukemia (AML), techniques used for their assessment in routine diagnostics have not been reconsidered for decades. We used a combination of 2 next-generation sequencing-based techniques to challenge the currently recommended conventional cytogenetic analysis (CCA), comparing the approaches in a series of 281 intensively treated patients with AML. Shallow whole-genome sequencing (sWGS) outperformed CCA in detecting European Leukemia Net (ELN)-defining CNAs and showed that CCA overestimated monosomies and suboptimally reported karyotype complexity. Still, the concordance between CCA and sWGS for all ELN CNA-related criteria was 94%. Moreover, using in silico dilution, we showed that 1 million reads per patient would be enough to accurately assess ELN-defining CNAs. Total genomic loss, defined as a total loss ≥200 Mb by sWGS, was found to be a better marker for genetic complexity and poor prognosis compared with the CCA-based definition of complex karyotype. For fusion detection, the concordance between CCA and whole-transcriptome sequencing (WTS) was 99%. WTS had better sensitivity in identifying inv(16) and KMT2A rearrangements while showing limitations in detecting lowly expressed PML-RARA fusions. Ligation-dependent reverse transcription polymerase chain reaction was used for validation and was shown to be a fast and reliable method for fusion detection. We conclude that a next-generation sequencing-based approach can replace conventional CCA for karyotyping, provided that efforts are made to cover lowly expressed fusion transcripts.

Chronic T cell receptor stimulation unmasks NK receptor signaling in peripheral T cell lymphomas via epigenetic reprogramming.

Peripheral T cell lymphomas (PTCLs) represent a significant unmet medical need with dismal clinical outcomes. The T cell receptor (TCR) is emerging as a key driver of T lymphocyte transformation. However, the role of chronic TCR activation in lymphomagenesis and in lymphoma cell survival is still poorly understood. Using a mouse model, we report that chronic TCR stimulation drove T cell lymphomagenesis, whereas TCR signaling did not contribute to PTCL survival. The combination of kinome, transcriptome, and epigenome analyses of mouse PTCLs revealed a NK cell-like reprogramming of PTCL cells with expression of NK receptors (NKRs) and downstream signaling molecules such as Tyrobp and SYK. Activating NKRs were functional in PTCLs and dependent on SYK activity. In vivo blockade of NKR signaling prolonged mouse survival, demonstrating the addiction of PTCLs to NKRs and downstream SYK/mTOR activity for their survival. We studied a large collection of human primary samples and identified several PTCLs recapitulating the phenotype described in this model by their expression of SYK and the NKR, suggesting a similar mechanism of lymphomagenesis and establishing a rationale for clinical studies targeting such molecules.

Integrated transcriptomic and genomic analysis improves prediction of complete remission and survival in elderly patients with acute myeloid leukemia.

Relevant molecular tools for treatment stratification of patients ≥65 years with acute myeloid leukemia (AML) are lacking. We combined clinical data with targeted DNA- and full RNA-sequencing of 182 intensively and palliatively treated patients to predict complete remission (CR) and survival in AML patients ≥65 years. Intensively treated patients with NPM1 and IDH2R172 mutations had longer overall survival (OS), whereas mutated TP53 conferred lower CR rates and shorter OS. FLT3-ITD and TP53 mutations predicted worse OS in palliatively treated patients. Gene expression levels most predictive of CR were combined with somatic mutations for an integrated risk stratification that we externally validated using the beatAML cohort. We defined a high-risk group with a CR rate of 20% in patients with mutated TP53, compared to 97% CR in low-risk patients defined by high expression of ZBTB7A and EEPD1 without TP53 mutations. Patients without these criteria had a CR rate of 54% (intermediate risk). The difference in CR rates translated into significant OS differences that outperformed ELN stratification for OS prediction. The results suggest that an integrated molecular risk stratification can improve prediction of CR and OS and could be used to guide treatment in elderly AML patients.

RNA Splicing Alterations Induce a Cellular Stress Response Associated with Poor Prognosis in Acute Myeloid Leukemia.

PURPOSE: RNA splicing is a fundamental biological process that generates protein diversity from a finite set of genes. Recurrent somatic mutations of splicing factor genes are common in some hematologic cancers but are relatively uncommon in acute myeloid leukemia (AML, < 20% of patients). We examined whether RNA splicing differences exist in AML, even in the absence of splicing factor mutations. EXPERIMENTAL DESIGN: We developed a bioinformatics pipeline to study alternative RNA splicing in RNA-sequencing data from large cohorts of patients with AML. RESULTS: We have identified recurrent differential alternative splicing between patients with poor and good prognosis. These splicing events occurred even in patients without any discernible splicing factor mutations. Alternative splicing recurrently occurred in genes with specific molecular functions, primarily related to protein translation. Developing tools to predict the functional impact of alternative splicing on the translated protein, we discovered that approximately 45% of the splicing events directly affected highly conserved protein domains. Several splicing factors were themselves misspliced and the splicing of their target transcripts were altered. Studying differential gene expression in the same patients, we identified that alternative splicing of protein translation genes in ELNAdv patients resulted in the induction of an integrated stress response and upregulation of inflammation-related genes. Finally, using machine learning techniques, we identified a splicing signature of four genes which refine the accuracy of existing risk prognosis schemes and validated it in a completely independent cohort. CONCLUSIONS: Our discoveries therefore identify aberrant alternative splicing as a molecular feature of adverse AML with clinical relevance.See related commentary by Bowman, p. 3503.

Refining diffuse large B-cell lymphoma subgroups using integrated analysis of molecular profiles.

BACKGROUND: Gene expression profiling (GEP), next-generation sequencing (NGS) and copy number variation (CNV) analysis have led to an increasingly detailed characterization of the genomic profiles of DLBCL. The aim of this study was to perform a fully integrated analysis of mutational, genomic, and expression profiles to refine DLBCL subtypes. A comparison of our model with two recently published integrative DLBCL classifiers was carried out, in order to best reflect the current state of genomic subtypes. METHODS: 223 patients with de novo DLBCL from the prospective, multicenter and randomized LNH-03B LYSA clinical trials were included. GEP data was obtained using Affymetrix GeneChip arrays, mutational profiles were established by Lymphopanel NGS targeting 34 key genes, CNV analysis was obtained by array CGH, and FISH and IHC were performed. Unsupervised independent component analysis (ICA) was applied to GEP data and integrated analysis of multi-level molecular data associated with each component (gene signature) was performed. FINDINGS: ICA identified 38 components reflecting transcriptomic variability across our DLBCL cohort. Many of the components were closely related to well-known DLBCL features such as cell-of-origin, stromal and MYC signatures. A component linked to gain of 19q13 locus, among other genomic alterations, was significantly correlated with poor OS and PFS. Through this integrated analysis, a high degree of heterogeneity was highlighted among previously described DLBCL subtypes. INTERPRETATION: The results of this integrated analysis enable a global and multi-level view of DLBCL, as well as improve our understanding of DLBCL subgroups.

Mediation analysis reveals common mechanisms of RUNX1 point mutations and RUNX1/RUNX1T1 fusions influencing survival of patients with acute myeloid leukemia.

Alterations of RUNX1 in acute myeloid leukemia (AML) are associated with either a more favorable outcome in the case of the RUNX1/RUNX1T1 fusion or unfavorable prognosis in the case of point mutations. In this project we aimed to identify genes responsible for the observed differences in outcome that are common to both RUNX1 alterations. Analyzing four AML gene expression data sets (n = 1514), a total of 80 patients with RUNX1/RUNX1T1 and 156 patients with point mutations in RUNX1 were compared. Using the statistical tool of mediation analysis we identified the genes CD109, HOPX, and KIAA0125 as candidates for mediator genes. In an analysis of an independent validation cohort, KIAA0125 again showed a significant influence with respect to the impact of the RUNX1/RUNX1T1 fusion. While there were no significant results for the other two genes in this smaller validation cohort, the observed relations linked with mediation effects (i.e., those between alterations, gene expression and survival) were almost without exception as strong as in the main analysis. Our analysis demonstrates that mediation analysis is a powerful tool in the identification of regulative networks in AML subgroups and could be further used to characterize the influence of genetic alterations.

Authors' Reply.

Authors' Reply to the Letter to the Editor by Y. Lynn Wang (MYD88 mutations and sensitivity to ibrutinib therapy).