SOX11/PRDX2 axis modulates redox homeostasis and chemoresistance in aggressive mantle cell lymphoma.

Mantle cell lymphoma (MCL) is an incurable B-cell neoplasm characterized by an aggressive behavior, short responses to conventional therapies and SOX11 overexpression, which is associated with aggressive disease features and inferior clinical outcome of patients. Oxidative stress is known to induce tumorigenesis and tumor progression, whereas high expression levels of antioxidant genes have been associated with chemoresistance in different cancers. However, the role of oxidative stress in MCL pathogenesis and the involvement of SOX11 regulating redox homeostasis in MCL cells are largely unknown. Here, by integrating gene set enrichment analysis of two independent series of MCL, we observed that SOX11+ MCL had higher reactive oxygen species (ROS) levels compared to SOX11- MCL primary tumors and increased expression of Peredoxine2 (PRDX2), which upregulation significantly correlated with SOX11 overexpression, higher ROS production and worse overall survival of patients. SOX11 knockout (SOX11KO) significantly reduced PRDX2 expression, and SOX11KO and PRDX2 knockdown (PRDX2KD) had increased ROS levels and ROS-mediated tumor cell death upon treatment with drugs, compared to control MCL cell lines. Our results suggest an aberrant redox homeostasis associated with chemoresistance in aggressive MCL through SOX11-mediated PRDX2 upregulation, highlighting PRDX2 as promising target for new therapeutic strategies to overcome chemoresistance in aggressive MCLs.

BCL3 rearrangements in B-cell lymphoid neoplasms occur in two breakpoint clusters associated with different diseases.

The t(14;19)(q32;q13) often juxtaposes BCL3 with immunoglobulin heavy chain (IGH) resulting in overexpression of the gene. In contrast to other oncogenic translocations, BCL3 rearrangement (BCL3-R) has been associated with a broad spectrum of lymphoid neoplasms. Here we report an integrative whole-genome sequence, transcriptomic, and DNA methylation analysis of 13 lymphoid neoplasms with BCL3-R. The resolution of the breakpoints at single base-pair revealed that they occur in two clusters at 5' (n=9) and 3' (n=4) regions of BCL3 associated with two different biological and clinical entities. Both breakpoints were mediated by aberrant class switch recombination of the IGH locus. However, the 5' breakpoints (upstream) juxtaposed BCL3 next to an IGH enhancer leading to overexpression of the gene whereas the 3' breakpoints (downstream) positioned BCL3 outside the influence of the IGH and were not associated with its expression. Upstream BCL3-R tumors had unmutated IGHV, trisomy 12, and mutated genes frequently seen in chronic lymphocytic leukemia (CLL) but had an atypical CLL morphology, immunophenotype, DNA methylome, and expression profile that differ from conventional CLL. In contrast, downstream BCL3-R neoplasms were atypical splenic or nodal marginal zone lymphomas (MZL) with mutated IGHV, complex karyotypes and mutated genes typical of MZL. Two of the latter four tumors transformed to a large B-cell lymphoma. We designed a novel fluorescence in situ hybridization assay that recognizes the two different breakpoints and validated these findings in 17 independent tumors. Overall, upstream or downstream breakpoints of BCL3-R are mainly associated with two subtypes of lymphoid neoplasms with different (epi)genomic, expression, and clinicopathological features resembling atypical CLL and MZL, respectively.

Chronic lymphocytic leukemia patient-derived xenografts recapitulate clonal evolution to Richter transformation.

Chronic lymphocytic leukemia (CLL) is a B-cell neoplasm with a heterogeneous clinical behavior. In 5-10% of patients the disease transforms into a diffuse large-B cell lymphoma known as Richter transformation (RT), which is associated with dismal prognosis. Here, we aimed to establish patient-derived xenograft (PDX) models to study the molecular features and evolution of CLL and RT. We generated two PDXs by injecting CLL (PDX12) and RT (PDX19) cells into immunocompromised NSG mice. Both PDXs were morphologically and phenotypically similar to RT. Whole-genome sequencing analysis at different time points of the PDX evolution revealed a genomic landscape similar to RT tumors from both patients and uncovered an unprecedented RT subclonal heterogeneity and clonal evolution during PDX generation. In PDX12, the transformed cells expanded from a very small subclone already present at the CLL stage. Transcriptomic analysis of PDXs showed a high oxidative phosphorylation (OXPHOS) and low B-cell receptor (BCR) signaling similar to the RT in the patients. IACS-010759, an OXPHOS inhibitor, reduced proliferation, and circumvented resistance to venetoclax. In summary, we have generated new RT-PDX models, one of them from CLL cells that mimicked the evolution of CLL to RT uncovering intrinsic features of RT cells of therapeutical value.

Cytogenomics of B-cell non-Hodgkin lymphomas: The "old" meets the "new".

For the routine diagnosis of haematological neoplasms an integrative approach is used considering the morphology, and the immunophenotypic, and molecular features of the tumor sample, along with clinical information. The identification and characterization of recurrent chromosomal aberrations mainly detected by conventional and molecular cytogenetics in the tumor cells has a major impact on the classification of lymphoid neoplasms. Some of the B-cell non-Hodgkin lymphomas are characterized by particular chromosomal aberrations, highlighting the relevance of conventional and molecular cytogenetic studies in their diagnosis and prognosis. In the current genomics era, next generation sequencing provides relevant information as the mutational profiles of haematological malignancies, improving their classification and also the clinical management of the patients. In addition, other new technologies have emerged recently, such as the optical genome mapping, which can overcome some of the limitations of conventional and molecular cytogenetics and may become more widely used in the cytogenetic laboratories in the upcoming years. Moreover, epigenetic alterations may complement genetic changes for a deeper understanding of the pathogenesis underlying B-cell neoplasms and a more precise risk-based patient stratification. Overall, here we describe the current state of the genomic data integrating chromosomal rearrangements, copy number alterations, and somatic variants, as well as a succinct overview of epigenomic changes, which altogether constitute a comprehensive diagnostic approach in B-cell non-Hodgkin lymphomas.

MALAT1 expression is associated with aggressive behavior in indolent B-cell neoplasms.

MALAT1 long non-coding RNA has oncogenic roles but has been poorly studied in indolent B-cell neoplasms. Here, MALAT1 expression was analyzed using RNA-seq, microarrays or qRT-PCR in primary samples from clinico-biological subtypes of chronic lymphocytic leukemia (CLL, n = 266), paired Richter transformation (RT, n = 6) and follicular lymphoma (FL, n = 61). In peripheral blood (PB) CLL samples, high MALAT1 expression was associated with a significantly shorter time to treatment independently from other known prognostic factors. Coding genes expressed in association with MALAT1 in CLL were predominantly related to oncogenic pathways stimulated in the lymph node (LN) microenvironment. In RT paired samples, MALAT1 levels were lower, concordant with their acquired increased independency of external signals. Moreover, MALAT1 levels in paired PB/LN CLLs were similar, suggesting that the prognostic value of MALAT1 expression in PB is mirroring expression differences already present in LN. Similarly, high MALAT1 expression in FL predicted for a shorter progression-free survival, in association with expression pathways promoting FL pathogenesis. In summary, MALAT1 expression is related to pathophysiology and more aggressive clinical behavior of indolent B-cell neoplasms. Particularly in CLL, its levels could be a surrogate marker of the microenvironment stimulation and may contribute to refine the clinical management of these patients.

RFcaller: a machine learning approach combined with read-level features to detect somatic mutations.

The cost reduction in sequencing and the extensive genomic characterization of a wide variety of cancers are expanding tumor sequencing to a wide number of research groups and the clinical practice. Although specific pipelines have been generated for the identification of somatic mutations, their results usually differ considerably, and a common approach is to use several callers to achieve a more reliable set of mutations. This procedure is computationally expensive and time-consuming, and it suffers from the same limitations in sensitivity and specificity as other approaches. Expert revision of mutant calls is therefore required to verify calls that might be used for clinical diagnosis. This step could take advantage of machine learning techniques, as they provide a useful approach to incorporate expert-reviewed information for the identification of somatic mutations. Here we present RFcaller, a pipeline based on machine learning algorithms, for the detection of somatic mutations in tumor-normal paired samples that does not require large computing resources. RFcaller shows high accuracy for the detection of substitutions and insertions/deletions from whole genome or exome data. It allows the detection of mutations in driver genes missed by other approaches, and has been validated by comparison to deep and Sanger sequencing.

Unraveling the genetics of transformed splenic marginal zone lymphoma.

The genetic mechanisms associated with splenic marginal zone lymphoma (SMZL) transformation are not well defined. We studied 41 patients with SMZL that eventually underwent large B-cell lymphoma transformation. Tumor material was obtained either only at diagnosis (9 patients), at diagnosis and transformation (18 patients), and only at transformation (14 patients). Samples were categorized in 2 groups: (1) at diagnosis (SMZL, n = 27 samples), and (2) at transformation (SMZL-T, n = 32 samples). Using copy number arrays and a next-generation sequencing custom panel, we identified that the main genomic alterations in SMZL-T involved TNFAIP3, KMT2D, TP53, ARID1A, KLF2, 1q gains, and losses of 9p21.3 (CDKN2A/B) and 7q31-q32. Compared with SMZL, SMZL-T had higher genomic complexity, and higher incidence of TNFAIP3 and TP53 alterations, 9p21.3 (CDKN2A/B) losses, and 6p gains. SMZL and SMZL-T clones arose by divergent evolution from a common altered precursor cell that acquired different genetic alterations in virtually all evaluable cases (92%, 12 of 13 cases). Using whole-genome sequencing of diagnostic and transformation samples in 1 patient, we observed that the SMZL-T sample carried more genomic aberrations than the diagnostic sample, identified a translocation t(14;19)(q32;q13) present in both samples, and detected a focal B2M deletion due to chromothripsis acquired at transformation. Survival analysis showed that KLF2 mutations, complex karyotype, and International Prognostic Index score at transformation were predictive of a shorter survival from transformation (P = .001; P = .042; and P = .007; respectively). In summary, SMZL-T are characterized by higher genomic complexity than SMZL, and characteristic genomic alterations that could represent key players in the transformation event.

Genomic landscape of follicular lymphoma across a wide spectrum of clinical behaviors.

While some follicular lymphoma (FL) patients do not require treatment or experience prolonged responses, others relapse early, and little is known about genetic alterations specific to patients with a particular clinical behavior. We selected 56 grade 1-3A FL patients according to their need of treatment or timing of relapse: never treated (n = 7), non-relapsed (19), late relapse (14), early relapse or POD24 (11), and primary refractory (5). We analyzed 56 diagnostic and 12 paired relapse lymphoid tissue biopsies and performed copy number alteration (CNA) analysis and next generation sequencing (NGS). We identified six focal driver losses (1p36.32, 6p21.32, 6q14.1, 6q23.3, 9p21.3, 10q23.33) and 1p36.33 copy-neutral loss of heterozygosity (CN-LOH). By integrating CNA and NGS results, the most frequently altered genes/regions were KMT2D (79%), CREBBP (67%), TNFRSF14 (46%) and BCL2 (40%). Although we found that mutations in PIM1, FOXO1 and TMEM30A were associated with an adverse clinical behavior, definitive conclusions cannot be drawn, due to the small sample size. We identified common precursor cells harboring early oncogenic alterations of the KMT2D, CREBBP, TNFRSF14 and EP300 genes and 16p13.3-p13.2 CN-LOH. Finally, we established the functional consequences of mutations by means of protein modeling (CD79B, PLCG2, PIM1, MCL1 and IRF8). These data expand the knowledge on the genomics behind the heterogeneous FL population and, upon replication in larger cohorts, could contribute to risk stratification and the development of targeted therapies.

Tumorigenic role of Musashi-2 in aggressive mantle cell lymphoma.

SOX11 overexpression has been associated with aggressive behavior of mantle cell lymphomas (MCL). SOX11 is overexpressed in embryonic and cancer stem cells (CSC) of some tumors. Although CSC have been isolated from primary MCL, their relationship to SOX11 expression and contribution to MCL pathogenesis and clinical evolution remain unknown. Here, we observed enrichment in leukemic and hematopoietic stem cells gene signatures in SOX11+ compared to SOX11- MCL primary cases. Musashi-2 (MSI2) emerged as one of the most significant upregulated stem cell-related genes in SOX11+ MCLs. SOX11 is directly bound to the MSI2 promoter upregulating its expression in vitro. MSI2 intronic enhancers were strongly activated in SOX11+ MCL cell lines and primary cases. MSI2 upregulation was significantly associated with poor overall survival independently of other high-risk features of MCL. MSI2 knockdown decreased the expression of genes related to apoptosis and stem cell features and significantly reduced clonogenic growth, tumor cell survival and chemoresistance in MCL cells. MSI2-knockdown cells had reduced tumorigenic engraftment into mice bone marrow and spleen compared to control cells in xenotransplanted mouse models. Our results suggest that MSI2 might play a key role in sustaining stemness and tumor cell survival, representing a possible novel target for therapeutic interventions in MCL.

Copy Number Alteration and Mutational Profile of High-Grade B-Cell Lymphoma with MYC and BCL2 and/or BCL6 Rearrangements, Diffuse Large B-Cell Lymphoma with MYC-Rearrangement, and Diffuse Large B-Cell Lymphoma with MYC-Cluster Amplification.

Diffuse large B-cell lymphoma (DLBCL) with MYC alteration is classified as high-grade B-cell lymphoma with MYC and BCL2 and/or BCL6 rearrangements (double/triple-hit lymphoma; DHL/THL), DLBCL with MYC rearrangement (single-hit lymphoma; SHL) and DLBCL with MYC-cluster amplification (MCAD). To elucidate the genetic features of DHL/THL, SHL, and MCAD, 23 lymphoma cases from Tokai University Hospital were analyzed. The series included 10 cases of DHL/THL, 10 cases of SHL and 3 cases of MCAD. The analysis used whole-genome copy number microarray analysis (OncoScan) and a custom-made next-generation sequencing (NGS) panel of 115 genes associated with aggressive B-cell lymphomas. The copy number alteration (CNA) profiles were similar between DHL/THL and SHL. MCAD had fewer CNAs than those of DHL/THL and SHL, except for +8q24. The NGS profile characterized DHL/THL with a higher "mutation burden" than SHL (17 vs. 10, p = 0.010), and the most relevant genes for DHL/THL were BCL2 and SOCS1, and for SHL was DTX1. MCAD was characterized by mutations of DDX3X, TCF3, HLA-A, and TP53, whereas MYC was unmutated. In conclusion, DHL/THL, SHL, and MCAD have different profiles.

Real-World Data on Chronic Myelomonocytic Leukemia: Clinical and Molecular Characteristics, Treatment, Emerging Drugs, and Patient Outcomes.

Despite emerging molecular information on chronic myelomonocytic leukemia (CMML), patient outcome remains unsatisfactory and little is known about the transformation to acute myeloid leukemia (AML). In a single-center cohort of 219 CMML patients, we explored the potential correlation between clinical features, gene mutations, and treatment regimens with overall survival (OS) and clonal evolution into AML. The most commonly detected mutations were TET2, SRSF2, ASXL1, and RUNX1. Median OS was 34 months and varied according to age, cytogenetic risk, FAB, CPSS and CPSS-Mol categories, and number of gene mutations. Hypomethylating agents were administered to 37 patients, 18 of whom responded. Allogeneic stem cell transplantation (alloSCT) was performed in 22 patients. Two-year OS after alloSCT was 60.6%. Six patients received targeted therapy with IDH or FLT3 inhibitors, three of whom attained a long-lasting response. AML transformation occurred in 53 patients and the analysis of paired samples showed changes in gene mutation status. Our real-world data emphasize that the outcome of CMML patients is still unsatisfactory and alloSCT remains the only potentially curative treatment. However, targeted therapies show promise in patients with specific gene mutations. Complete molecular characterization can help to improve risk stratification, understand transformation, and personalize therapy.

Detection of early seeding of Richter transformation in chronic lymphocytic leukemia.

Richter transformation (RT) is a paradigmatic evolution of chronic lymphocytic leukemia (CLL) into a very aggressive large B cell lymphoma conferring a dismal prognosis. The mechanisms driving RT remain largely unknown. We characterized the whole genome, epigenome and transcriptome, combined with single-cell DNA/RNA-sequencing analyses and functional experiments, of 19 cases of CLL developing RT. Studying 54 longitudinal samples covering up to 19 years of disease course, we uncovered minute subclones carrying genomic, immunogenetic and transcriptomic features of RT cells already at CLL diagnosis, which were dormant for up to 19 years before transformation. We also identified new driver alterations, discovered a new mutational signature (SBS-RT), recognized an oxidative phosphorylation (OXPHOS)high-B cell receptor (BCR)low-signaling transcriptional axis in RT and showed that OXPHOS inhibition reduces the proliferation of RT cells. These findings demonstrate the early seeding of subclones driving advanced stages of cancer evolution and uncover potential therapeutic targets for RT.

TP53 Abnormalities Are Underlying the Poor Outcome Associated with Chromothripsis in Chronic Lymphocytic Leukemia Patients with Complex Karyotype.

Chromothripsis (cth) has been associated with a dismal outcome and poor prognosis factors in patients with chronic lymphocytic leukemia (CLL). Despite being correlated with high genome instability, previous studies have not assessed the role of cth in the context of genomic complexity. Herein, we analyzed a cohort of 33 CLL patients with cth and compared them against a cohort of 129 non-cth cases with complex karyotypes. Nine cth cases were analyzed using optical genome mapping (OGM). Patterns detected by genomic microarrays were compared and the prognostic value of cth was analyzed. Cth was distributed throughout the genome, with chromosomes 3, 6 and 13 being those most frequently affected. OGM detected 88.1% of the previously known copy number alterations and several additional cth-related rearrangements (median: 9, range: 3-26). Two patterns were identified: one with rearrangements clustered in the region with cth (3/9) and the other involving both chromothriptic and non-chromothriptic chromosomes (6/9). Cases with cth showed a shorter time to first treatment (TTFT) than non-cth patients (median TTFT: 2 m vs. 15 m; p = 0.013). However, when stratifying patients based on TP53 status, cth did not affect TTFT. Only TP53 maintained its significance in the multivariate analysis for TTFT, including cth and genome complexity defined by genomic microarrays (HR: 1.60; p = 0.029). Our findings suggest that TP53 abnormalities, rather than cth itself, underlie the poor prognosis observed in this subset.

Cell-Free DNA for Genomic Analysis in Primary Mediastinal Large B-Cell Lymphoma.

High-throughput sequencing of cell-free DNA (cfDNA) has emerged as a promising noninvasive approach in lymphomas, being particularly useful when a biopsy specimen is not available for molecular analysis, as it frequently occurs in primary mediastinal large B-cell lymphoma (PMBL). We used cfDNA for genomic characterization in 20 PMBL patients by means of a custom NGS panel for gene mutations and low-pass whole-genome sequencing (WGS) for copy number analysis (CNA) in a real-life setting. Appropriate cfDNA to perform the analyses was obtained in 18/20 cases. The sensitivity of cfDNA to detect the mutations present in paired FFPE samples was 69% (95% CI: 60-78%). The mutational landscape found in cfDNA samples was highly consistent with that of the tissue, with the most frequently mutated genes being B2M (61%), SOCS1 (61%), GNA13 (44%), STAT6 (44%), NFKBIA (39%), ITPKB (33%), and NFKBIE (33%). Overall, we observed a 75% concordance to detect CNA gains/losses between DNA microarray and low-pass WGS. The sensitivity of low-pass WGS was remarkably higher for clonal CNA (18/20, 90%) compared to subclonal alterations identified by DNA microarray. No significant associations between cfDNA amount and tumor burden or outcome were found. cfDNA is an excellent alternative source for the accurate genetic characterization of PMBL cases.

Serum soluble CD23 levels are an independent predictor of time to first treatment in chronic lymphocytic leukemia.

Serum soluble CD23 (sCD23) levels have been acknowledged as a prognostic factor in patients with chronic lymphocytic leukemia (CLL), but their potential relevance has not been analyzed in recent times. We retrospectively studied 338 CLL, small lymphocytic lymphoma, or CLL-type monoclonal B-cell lymphocytosis patients from a single institution, with available sCD23 levels at diagnosis. Baseline features and outcomes were compared between patients with sCD23 ≤/>1000 UI/L. The 140 patients (41%) who had sCD23 > 1000 UI/L showed adverse-risk clinical and biological characteristics. High sCD23 levels were predictive of a shorter time to first treatment (5-year probability of requiring treatment: 60 vs. 20%, p < 0.0001; hazard ratio (HR) = 1.72, p = 0.003 in a multivariable model also including the CLL International Prognostic Index and the absolute lymphocyte count), and a poorer 5-year overall survival (70 vs. 82%, p = 0.0009). These data suggest the potential of sCD23 to predict treatment-free survival and to shed light on mechanisms of activity and resistance to CD23-directed therapies.

PanCancer analysis of somatic mutations in repetitive regions reveals recurrent mutations in snRNA U2.

Current somatic mutation callers are biased against repetitive regions, preventing the identification of potential driver alterations in these loci. We developed a mutation caller for repetitive regions, and applied it to study repetitive non protein-coding genes in more than 2200 whole-genome cases. We identified a recurrent mutation at position c.28 in the gene encoding the snRNA U2. This mutation is present in B-cell derived tumors, as well as in prostate and pancreatic cancer, suggesting U2 c.28 constitutes a driver candidate associated with worse prognosis. We showed that the GRCh37 reference genome is incomplete, lacking the U2 cluster in chromosome 17, preventing the identification of mutations in this gene. Furthermore, the 5'-flanking region of WDR74, previously described as frequently mutated in cancer, constitutes a functional copy of U2. These data reinforce the relevance of non-coding mutations in cancer, and highlight current challenges of cancer genomic research in characterizing mutations affecting repetitive genes.

Ibrutinib in Combination With Rituximab for Indolent Clinical Forms of Mantle Cell Lymphoma (IMCL-2015): A Multicenter, Open-Label, Single-Arm, Phase II Trial.

PURPOSE: The need for an individualized management of indolent clinical forms in mantle cell lymphoma (MCL) is increasingly recognized. We hypothesized that a tailored treatment with ibrutinib in combination with rituximab (IR) could obtain significant responses in these patients. METHODS: This is a multicenter single-arm, open-label, phase II study with a two-stage design conducted in 12 Spanish GELTAMO sites (ClinicalTrials.gov identifier: NCT02682641). Previously untreated MCL patients with indolent clinical forms defined by the following criteria were eligible: no disease-related symptoms, nonblastoid variants, Ki-67 < 30%, and largest tumor diameter ≤ 3 cm. Both leukemic non-nodal and nodal subtypes were recruited. Patients received ibrutinib 560 mg once daily and a total of eight doses of rituximab 375 mg/m2. Ibrutinib could be discontinued after 2 years in the case of sustained undetectable minimal residual disease (MRD). The primary end point was the complete response (CR) rate achieved after 12 cycles according to Lugano criteria. RESULTS: Fifty patients with MCL (male 66%; median age 65 years) were enrolled. After 12 cycles of treatment, 42 (84%; 95% CI, 74 to 94) patients had an overall response, including 40 (80%; 95% CI, 69 to 91) with CR. Moreover, undetectable MRD in peripheral blood was achieved in 87% (95% CI, 77 to 97) of cases. At 2 years, 24 of 35 evaluable patients (69%) could discontinue ibrutinib because of undetectable MRD. Four patients had disease progression; three were non-nodal MCL and carried high genomic complexity and TP53 mutations at enrollment. No unexpected toxicity was seen except one patient with severe aplastic anemia. CONCLUSION: Frontline IR combination achieves a high rate of CRs and undetectable MRD in indolent clinical forms of MCL. Discontinuation seems appropriate in cases with undetectable MRD, except for TP53-mutated cases.

Genomic and transcriptomic profiling reveals distinct molecular subsets associated with outcomes in mantle cell lymphoma.

Mantle cell lymphoma (MCL) is a phenotypically and genetically heterogeneous malignancy in which the genetic alterations determining clinical indications are not fully understood. Here, we performed a comprehensive whole-exome sequencing analysis of 152 primary samples derived from 134 MCL patients, including longitudinal samples from 16 patients and matched RNA-Seq data from 48 samples. We classified MCL into 4 robust clusters (C1-C4). C1 featured mutated immunoglobulin heavy variable (IGHV), CCND1 mutation, amp(11q13), and active B cell receptor (BCR) signaling. C2 was enriched with del(11q)/ATM mutations and upregulation of NF-κB and DNA repair pathways. C3 was characterized by mutations in SP140, NOTCH1, and NSD2, with downregulation of BCR signaling and MYC targets. C4 harbored del(17p)/TP53 mutations, del(13q), and del(9p), and active MYC pathway and hyperproliferation signatures. Patients in these 4 clusters had distinct outcomes (5-year overall survival [OS] rates for C1-C4 were 100%, 56.7%, 48.7%, and 14.2%, respectively). We also inferred the temporal order of genetic events and studied clonal evolution of 16 patients before treatment and at progression/relapse. Eleven of these samples showed drastic clonal evolution that was associated with inferior survival, while the other samples showed modest or no evolution. Our study thus identifies genetic subsets that clinically define this malignancy and delineates clonal evolution patterns and their impact on clinical outcomes.