Slack, Graham W. MD; Gascoyne, Randy D. MD
In 1977, Sanger et al1 published their chain termination method for sequencing DNA, which achieved widespread use and popularity throughout the world and became the de facto gold standard for what we now call “first-generation” DNA sequencing. This advancement heralded a new era in molecular biology and paved the way for mapping of the human genome.
In the 1990s, 2 groups independently set out to map the haploid human genome, starting with the Human Genome Project in 1990. This massive scientific research project was conducted under the auspices of the publically funded International Human Genome Consortium (IHGC), an international collaboration consisting of 20 sequencing centers in 6 countries. Using Sanger sequencing, the project would ultimately cost 3 billion US dollars and take 14 years to complete.2,3 In 1998, Celera Genomics entered the race to become the first to map the human genome. In contrast to the IHGC efforts, this privately funded venture utilized newly developed “shotgun” whole-genome sequencing (WGS) technology. With this approach, their project would take only 3 years and cost a fraction of the IHGC project at 300 million US dollars. Both groups published incomplete draft genomes in 2001, and the IHGC published a near-complete (95%) draft in 2004.3,4 The result of this competitive scientific endeavor was the generation of a sequenced reference human genome. A byproduct was the refinement of “first-generation” Sanger sequencing technology and the introduction of “next-generation” sequencing technology, so-called “shotgun” or “massively parallel” sequencing.
In 2006, next-generation sequencing (NGS) technology was made commercially available and has since revolutionized the field of genomics. Today, several commercial platforms are available, including: Illumina Genome Analyzer, Roche 454, Applied Biosystems SOLiD, Pacific Biosystems, and Life Technologies Ion Torrent. A detailed description of each of these platforms is beyond the scope of this article and the reader is referred elsewhere for an in depth review.5–7 In general, NGS begins with the creation of a DNA library—a multitude of short DNA molecules generated from the fragmentation (“shotgun”) of much larger genetic material of interest (eg, genome). Each molecule within the library is isolated and sequenced in parallel using a “sequencing by synthesis” method. The sequenced reads are compared with a reference genome and the sequence of the large genetic material of interest is constructed. The latter step is data intensive and requires bioinformatics algorithms to complete. The platforms differ in their use of technology and chemistry and therefore differ in their throughput times, throughput volumes, read-lengths, errors rates, error types and cost. Since the introduction of NGS, the average cost per base has decreased steadily, whereas read length and overall throughput have increased. Despite this, a major drawback to this technology is the amount of time, cost, and infrastructure required, including data storage and computational resources, to analyze the massive amounts of sequencing data. As such, their use in experimental studies has to date been limited.
The haploid human genome is large and complex. It contains approximately 3 billion bases and houses an estimated 20,000 to 25,000 protein-encoding genes. Yet protein-encoding exons represent only 1% of the entire genome; introns and intergenic regions, including noncoding and repetitive DNA elements, account for the remaining 99% of the genome. Strategies for investigating the cancer genome include sequencing the whole genome or limiting the sequencing to exomes, transcriptomes (RNA-seq), or targeted resequencing of specific exons. WGS covers the entire genome, including all exons, introns, and intergenic regions. This approach reveals all classes of alterations and provides comprehensive characterization of the cancer genome. WGS is the ideal approach for mapping the genetic landscape of malignant tumors; however, its biggest limitation is its prohibitive cost, often leading investigators to use less expensive but more limited approaches. Exome sequencing covers the regions of the genome that encode for proteins. It can reveal single-nucleotide variants and insertion-deletion events (indels) within coding regions but will miss alterations in noncoding or regulatory regions and will fail to detect most chromosomal arrangements and copy number alterations.8 Part of its appeal is the relatively low cost to perform the sequencing. As the whole exome accounts for only 1% of the whole genome, this approach is significantly less expensive than WGS and is an financially attractive approach to examining the cancer genome. Whole-transcriptome sequencing, or RNA-seq, covers only the transcribed genes in a tumor. Messenger RNA is extracted from tumors and converted into cDNA, which is then sequenced. RNA-seq will reveal single-nucleotide variants in genes with sufficiently high levels of mRNA (ie, highly expressed genes) and can identify transcribed rearrangements (gene fusions). RNA-seq also provides very good quantitative gene expression data, superior in many respects to gene microchip arrays, such as Affymetrix U133. Limitations include the inability to detect abnormalities in noncoding regions of the genome. In addition, protein-truncating mutations can be missed if they result in nonsense-mediated RNA decay. All of these approaches vary significantly in their cost, the amount of information/depth of coverage provided by a specific amount of raw sequence data, and the ability of each approach to detect specific cancer-causing genetic alterations.9 One must be cognizant of these differences and limitations when interpreting data from studies using NGS platforms. One must also be aware that some significant genetic abnormalities occur only in a minority of tumors and their discovery by NGS is wholly dependent upon the inclusion of a sufficient number of cases within the discovery set to provide broad and representative coverage of the genetic landscape of a particular disease. This fact explains why some studies report abnormalities that others are not able to identify using NGS.10,11
In recent years, NGS technology has been applied to lymphoid neoplasms and has provided some early insight into the mutational landscape of several lymphoid cancers. Lymphoma is a genetic disease, and somatically acquired mutations are believed to provide lymphoma cells a growth or survival advantage by disrupting key pathways in a cell cycle control, proliferation, differentiation, and apoptosis, or by altering homeostatic interactions within the microenvironment. Somatic alterations to the genome can stratify into 2 general classes: driver events and passenger events. Driver events occur in key “cancer” genes and confer a growth or survival advantage. They are essential for converting a normal lymphocyte into a malignant lymphocyte. Typically these events affect one of a number of predefined consensus cancer genes. A list of consensus cancer genes can be found in the COSMIC database on the Sanger site (www.sanger.ac.uk/genetics/CGP/cosmic/). Driver events can also be inferred using other parameters, such as their frequency within a specific cancer subtype or their tendency to target very specific regions of genes (so-called “hot-spot” mutations).8,10 Passenger events, which make up the majority of genetic alterations in a cancer cell do not confer a growth or survival advantage; they only travel passively within the expanding tumor clone. Identification of recurrent genetic alterations and distinguishing between driver and passenger events is a critical step in understanding tumor biology and moving forward with the investigation of candidate genes in functional studies.
The following is a review of the major discoveries in the genetic landscape of lymphoma using NGS technology. A summary of these discoveries is presented in Table 1.
NEXT-GENERATION SEQUENCING DISCOVERIES IN LYMPHOMA
Diffuse Large B-Cell Lymphoma
Diffuse large B-cell lymphoma (DLBCL) is the most common non-Hodgkin lymphoma in the western world. It is neoplasm of large B cells growing in a diffuse pattern that does not otherwise meet the criteria for a more specific diagnosis.12 DLBCL is a biologically heterogenous disease that has an equally heterogenous response to standard therapeutic regimens, including rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP). Gene expression profiling studies have shown that DLBCL can be segregated in molecularly distinct subgroups based on a cell of origin gene signature: germinal center B-cell (GCB) like and activated B-cell (ABC) like. GCB-type DLBCL has a more favorable prognosis and better response to R-CHOP therapy than the more aggressive ABC-type DLBCL.13–15
Morin et al16 were the first to examine DBLCL with NGS technology. Using a combination of WGS, exome sequencing, and RNA-seq, they identified a recurrent and very targeted somatic mutation affecting the polycomb repressor-2 complex gene EZH2 in 22% of DLBCL, all of which were confined to the GCB subtype. EZH2 encodes a histone methyltransferase that is responsible for trimethylating Lys27 of histone H3 (H3K27) and plays an important role in gene regulation.17 The observed mutations all targeted the SET domain of the EZH2 protein (Tyr641) resulting in several nonsynonymous amino acid substitutions of the tyrosine moiety within the protein’s catalytic domain. Initial in vitro functional data suggested these were loss-of-function mutations; however, the heterozygous nature of the alterations and the highly localized mutations suggested otherwise. Subsequent studies have shown that these mutations are in fact a gain-of-function alteration.18,19 The novel discovery of a recurrent EZH2 mutation in DLBCL has provided investigators a new target for novel therapies in this disease, and recent studies have shown that selective EZH2 inhibitors can prevent the growth of tumor cells in DLBCL cell lines and murine xenograft models.20–22 These recent findings highlight the importance of NGS discoveries and their potential for rapid translation of targeted therapies into clinical practice.
Ngo et al23 examined DLBCL using a combination of RNA interference studies and RNA-seq and identified a recurrent somatic mutation involving MYD88 in 29% of ABC-DLBCL cases, a mutation that was virtually absent in the GCB subtype. Furthermore, their study showed that the MYD88 mutation is a gain-of-function mutation that is highly targeted as all cases harbored the same amino acid substitution, L265P, within the MYD88 Toll/IL-1 receptor domain. Interestingly, MYD88 is a known adaptor protein that activates the NF-kB pathway after stimulation of toll-like receptors. This feature is consistent with observations that ABC-type DLBCL is associated with constitutive activation of the NF-kB pathway.24
Using a combination of whole-exome sequencing and genome-wide single-nucleotide polymorphism copy number array analysis, Pasqualucci et al25 also examined DLBCL and identified additional recurrent somatic mutations of genes involved in gene regulation: CREBBP and E300. CREBBP and E300 both function as histone and nonhistone acetyltransferases and serve as transcriptional coactivators in a number of signaling pathways.26 CREBBP mutations were observed in 22% of all DLBCL, with enrichment in the GCB subtype (32% vs. 13% in ABC, P<0.01), whereas E300 mutations were observed in 10% of all DLBCL. CREBBP deletions were also identified in an additional 10% of cases and were mutually exclusive of CREBBP mutations. The functional significance of these mutations is also highlighted. Tumor cells that harbored the mutant genes were deficient in acetylating BCL6 and p53, leading to constitutive activation of the BLC6 oncoprotein and to decreased p53 tumor suppressor activity.
Morin et al10 further described the mutational landscape of DLBCL. Utilizing 96 patient samples and 10 DLBCL cell lines in combination with WGS, exome sequencing, and RNA-seq, they identified recurrent mutations in several genes affecting histone modification, highlighting the importance of this cellular process in the pathobiology of DLBCL. Notably, MLL2 was identified as a highly recurrent target for mutation with inactivating mutations found in 32% of DLBCL genomes. MLL2 is one of 6 human H3K4-specific methyltransferases and is known to regulate the transcription of a diverse set of genes.27,28 Mutations in MEF2B, a gene that cooperates with CREBBP and EP300 to acetylate histones, were also newly described. Pasqualucci et al11 subsequently combined exome sequencing in 6 cases of DLBCL with targeted resequencing in 115 cases of DLBCL and largely confirmed these findings, identifying MLL2 mutations in 24% of DLBCL. The same group also published a follow-up study describing the presence of recurrent mutations in CD58 and B2M, 2 genes directly impacting the microenvironment in DLBCL.29 B2M is closely linked to HLA class I expression on malignant B cells and CD58 is required for NK cell killing. The identification of these mutations establishes escape from immune cell recognition as an important oncogenic event in DLBCL lymphomagenesis.
Recently, Lohr et al30 examined the mutational landscape of DLBCL, performing whole-exome sequencing on 49 patient samples of DLBCL. Their results confirmed the findings of Morin, Ngo, and Pasqualucci, identifying EZH2, MYD88, CREBBP, MLL2, MEF2B, and CD58, in addition to several other genes, as targets of recurrent mutation in DLBCL. They also identified KRAS, BRAF, and NOTCH1 as targets of rare but potential driver mutations in a subset of DLBCL cases.
Finally, Scott et al31 recently used NGS technology to identify a novel recurrent somatic gene fusion in DLBCL. Using RNA-seq in 96 cases of DLBCL, they discovered a novel fusion involving TBL1XR1 and TP63. This fusion was exclusive to GCB-type DLBCL and was present in 5% of GCB-type cases. This discovery was confirmed by fluorescence in situ hybridization of 187 cases of DLBCL in a tissue microarray (4/82 GCB vs. 0/105 ABC harbored the gene fusion) as well as by positive RT-PCR in 2 cases. The recurrent nature of this gene fusion and its apparent restriction to the GCB subtype suggest that it plays a role in lymphomagenesis, the precise nature of which waits to be elucidated. In addition, these preliminary results also suggest that the fusion maybe associated with primary refractory disease, a finding with potential clinical relevance.
Follicular lymphoma (FL) is the second most common non-Hodgkin lymphoma in the western world and the most common low-grade lymphoma. Like a subset of DLBCL, it is a neoplasm of germinal center B-cells; however, it grows in at least a partially follicular pattern.12 In contrast to DLBCL, it is generally an indolent disease, and although it is easily treated, it is considered universally incurable.
To date, all published studies examining the mutational landscape of FL using NGS technology have been conducted in parallel with DLBCL, and these studies have been described above. In brief, Morin et al16 identified the EZH2 mutation in 7% of FL. EZH2 mutations are less frequently observed than in DLBC; however, their presence in both FL and GCB-type DLBCL supports the notion that the polycomb repressor complex-2 plays a significant role in the pathogenesis of at least a subset of lymphomas of germinal center origin. Pasqualucci et al25 also identified CREBBP mutations in 33% of FL, whereas deletions were rare and an additional 9% of cases exhibited mutations in E300. Deletions of E300 were not identified. Morin et al10 identified the MLL2 and MEF2B mutations in 89% and 13% of FL, respectively. Finally, Scott et al31 also showed the novel TBL1XR1/TP63 fusion gene occurs at very low incidence in FL, having been identified in only one of 81 cases.
Classical Hodgkin Lymphoma and Primary Mediastinal Large B-Cell Lymphoma
Classic Hodgkin lymphoma (CHL) is a tumor composed of malignant Hodgkin and Reed-Sternberg cells within a variably cellular inflammatory background.12 Hodgkin and Reed-Sternberg cells are thought to arise from germinal center B-cells that have lost the normal repertoire of B-cell antigen expression. They usually weakly express the B-cell antigen PAX5 but lack other B-cell antigens including CD45, CD20, and CD79a. They are always positive for CD30 and most cases also coexpress CD15. Conversely, primary mediastinal large B-cell lymphoma (PMBCL) is composed of a diffuse proliferation of large B cells, arising from putative thymic B cells.12 They express all B-cell antigens, including CD20, CD79a, and PAX5. In addition, CD30 is expressed by a variable number of cells the majority of tumors, but CD15 is usually absent. Despite these differences, there is clinicopathologic overlap between these 2 tumors suggesting a shared underlying biology. This is highlighted by the fact that the current World Health Organization classification now includes a gray-zone diagnosis for cases with overlapping features: B-cell lymphoma, unclassifiable, with features intermediate between DLBCL and classic Hodgkin lymphoma.
Steidl et al32 used RNA-seq to study 2 Hodgkin lymphoma cell lines as a discovery platform. They found several interesting alterations but focused their work on the highest read abnormality—a recurrent gene fusion involving the transcription factor CIITA located on (chromosome 16p) and promiscuous translocation partners. They extended their study to examine 55 cases of CHL and 77 cases of PMBCL in tissue microarray using fluorescence in situ hybridization for CIITA rearrangement. They identified the rearrangement in 15% of CHL and 38% of PMBCL. The fact that this novel rearrangement was identified in both tumor types further supports the shared clinicopathologic features between these 2 entities. In addition, half of the fusion partners in PMBCL involved the ligands of PD-1. PD-1 is a cell membrane protein that negatively regulates T-cell function; thus, this discovery strongly supports the hypothesis that cultivating immune privilege in PMCBL is a critical oncogenic event.33
Chronic Lymphocytic Leukemia
Chronic lymphocytic leukemia (CLL) is a neoplasm of small mature B-cells that express CD5 and CD23 and infiltrate the bone marrow and peripheral blood with variable involvement of the spleen, liver, and lymph nodes.12 This incurable disease is the most common adult leukemia in the western world and is clinically heterogenous, ranging from patients with highly stable disease to patients with rapidly progressive disease and high mortality. Several markers of prognosis have been identified, including the expression of antigens CD38 and ZAP70; cytogenetic alterations, such as deletions of chromosomes 11q, 13q, 17p and trisomy 12; and the mutation status of the variable region of the IGH gene. CLL that harbors a mutated IGH gene has a better prognosis than unmutated cases.
As part of the collaboration on behalf of the International Cancer Genome Consortium (ICGC), Puente et al34 performed WGS of 4 CLL cases, including 2 with unmutated IGH and 2 with mutated IGH. They found recurrent mutations in several genes, with the 4 most common being NOTCH1, MYD88, XPO1, and KLHL6. A large extension set was used to more firmly establish the frequency of mutations (NOTCH1 12.2%, MYD88 2.9%, XPO1 2.4%, KLHL6 1.8%) as well as their clinical relevance. The authors showed that MYD88 and KLHL6 mutations were associated with CLL with mutated IGH, whereas NOTCH1 and XPO1 mutations were associated with CLL unmutated IGH.
Fabbri et al35 used WGS to examine 5 cases of CLL and found a similar array of mutations in the CLL genome as Puente and colleagues. In an extension set, Fabbri and colleagues examined the NOTCH1 mutation and showed that it was associated with disease progression and poor prognosis, with NOTCH1 mutations being present in 8.3% of CLL at diagnosis, 31% of cases in transformation, and 20.8% of cases with chemorefractoriness.
Three studies have also identified SF3B1 as a recurrently mutated gene in a subset of CLL cases using NGS approaches.36–38 SF3B1 is a critical member of the RNA splicing machinery and is mutated in roughly 10% to 20% of CLL cases. This mutation is associated with treatment refractoriness and aggressive clinical behavior. It is mutually exclusive of the del17p and is associated with the presence of a del11q.37 Of note, SF3B1 was not identified as a mutated gene in the 2 investigations discussed above, likely owing to the frequency of SF3B1 mutation in CLL and the low numbers of cases examined by these initial studies.34,35 This serves to highlight the value of performing WGS a cohort of samples large enough detect low-frequency recurrent genetic events with confidence.
Splenic B-Cell Marginal Zone Lymphoma
Splenic marginal zone B-cell lymphoma (SMZL) is the most common primary lymphoma of the spleen. It is composed a small B cells that proliferate within the marginal zone of the splenic white pulp, but it can also infiltrate splenic hilar lymph nodes, liver, bone marrow, and peripheral blood.12 Within the peripheral blood characteristic “villous lymphocytes” are seen. SMZL has recently been shown to be associated with the chromosomal abnormality, del7q(q22-24), which is seen in nearly 40% of cases.39 The frequent and specific nature of this abnormality in SMZL suggests this region of the genome harbors a tumor suppressor gene, but little else is known about the genetic landscape of this disease.
Kiel et al40 performed WGS on 6 discovery cases of SMZL identifying recurrent NOTCH2 mutations in 3 cases. Targeted resequencing of the NOTCH2 gene in an additional 93 cases identified NOTCH2 mutations in 25.3% of SMZL. The mutation seems to be relatively specific for SMZL as it was only found in 5% of nonsplenic marginal zone lymphomas and was not identified in other B-cell lymphomas, including: CLL, mantle cell lymphoma (MCL), hairy cell leukemia (HCL), and FL. They showed that the mutations clustered near the C-terminal proline/glutamate/serine/threonine (PEST)-rich domain of the NOTCH2 protein, resulting in protein truncation and gain-of-function. They also found that a NOTCH2 mutation was associated with a worse clinical outcome. Using whole-exome sequencing in 8 discovery cases and targeted resequencing in an additional 109 extension cases, Rossi et al41 also identified a recurrent mutation of NOTCH2 in 21.3% of cases and further confirmed its high specificity for SMZL.
Hairy Cell Leukemia
HCL is a neoplasm of small mature B cells that infiltrate the spleen, peripheral blood, and bone marrow, resulting in splenomegaly, lymphocytosis, and cytopenias. The malignant cells exhibit characteristic features, including: round to oval nuclei with spongy reticular chromatin; hair-like cytoplasmic projections; and a CD11c+, CD25+, CD103+ phenotype.42 HCL is sensitive to purine analogs, such as cladribine, and prolonged remissions can be achieved. Until recently, HCL had not been associated with any recurrent genetic abnormalities, and little was known of its underlying genetic deficiencies.
Tiacci et al42 performed WGS in 1 case of HCL and found a heterozygous BRAF mutation (V600E) similar to the mutation previously identified in melanoma. They subsequently performed targeted resequencing of the BRAF gene in an additional 47 HCL cases and confirmed the presence of a BRAF (V600E) mutation in 100% of cases. Importantly, this mutation was not identified in other small B-cell neoplasms, including SMZL, HCL-variant, and splenic diffuse red pulp small B-cell lymphoma, establishing it as a disease-defining alternation. Treatments are available that target this mutation and are currently being tested in HCL in addition to other cancers.
Lymphoplasmacytic Lymphoma/Waldenström Macroglobulinemia
Lymphoplasmacytic lymphoma (LPL) is a neoplasm of post-GCBs, consisting of small mature lymphocytes, plasmacytoid lymphocytes, and plasma cells.12 Waldenström macroglobulinemia (WM) is defined as any case of LPL that involves the bone marrow and is associated with an IgM paraprotein. Patients with WM exhibit broad symptomatology, including: cytopenias and their associated sequelae, autoimmune phenomena, neuropathy, coagulopathies, and cryglobulinemia. The t(9;14) was initially described in association with LPL/WM but subsequent studies failed to show the association and it is not currently believed to be a disease-specific abnormality. No other specific chromosomal or genetic abnormalities had been previously described in LPL/WM.
Treon et al43 performed WGS in 30 cases of LPL and discovered a highly recurrent point mutation in MYD88, occurring in 90% of cases. Targeted resequencing of MYD88 in an additional 54 cases of WM and 3 cases of non-IgM LPL confirmed the results, identify the mutation in 91% of WM and 100% of non-IgM LPL. The mutation is the same one described in ABC-type DLBCL (see above), which results in a L265P amino acid substitution, with gain-of-function activity and constitutive downstream activation of the NF-kB pathway, consistent with the postgerminal center origin of this tumor. The mutation was only infrequently identified in other postgerminal center B-cell neoplasms with plasmacytic differentiation, including: marginal zone lymphoma, plasma cell myeloma, and IgM monoclonal gammopathy of undetermined significance, suggesting it is relatively specific for LPL/WM and may aid in establishing a definitive diagnosis in cases with overlapping features. Interestingly, in 2 of 3 cases of WM with wild-type MYD88 an MLL2 mutation was present, another mutation also described in DLBCL and FL. The authors also identified another recurrent mutation in LPL/WM: ARID1A was mutated in 17% of cases resulting in protein truncation, frameshift changes, or amino acid insertions. Patients with both MYD88 and ARID1A mutations had a heavier disease burden, with greater involvement of the bone marrow, lower hematocrit, and lower platelet count observed.
Mantle Cell Lymphoma
MCL is a moderately aggressive neoplasm of naïve pregerminal center B cells derived from the inner mantle zone with a median survival of 4 to 5 years. The disease typically affects lymph nodes and can infiltrate the liver, spleen, gastrointestinal tract, bone marrow, and peripheral blood.12 Tumor cells are usually small and express CD5, CD43, and FMC7, but variants with blastoid or pleomorphic features that are associated with more aggressive behavior exist. The genetic hallmark of this disease is the t(11;14)(q13;q32), which juxtaposes CCND1 on chromosome 11q13 with IGH on chromosome 14q32, resulting in constitutive overexpression of the cyclin D1 protein and cell cycle deregulation. This genetic alteration is considered the primary genetic event in MCL biology, but it is not sufficient to cause the disease and secondary genetic events are required to induce lymphomagenesis. Recurrent mutation of genes such as TP53, ATM, and CCND1 have been described; however, these findings insufficiently explain the biological underpinnings of MCL.
Kridel et al44 performed RNA-seq in 18 patient samples of MCL and 2 MCL cell lines and identified a novel and recurrent mutation involving NOTCH1. They confirmed this finding by performing targeted resequencing in a larger cohort of patient samples and cell cultures, identifying NOTCH1 mutations in a total of 12% of patient samples and 20% of MCL cell lines. Interestingly, 86% of the mutations consisted of small frameshift indels or nonsense mutations occurring in exon 34 and were predicted to cause truncations of the C-terminus of the PEST domain, similar to the mutations observed in CLL. They also found those tumors that harbored a NOTCH1 mutation behaved more aggressively and were associated with inferior overall survival.44 This finding has potential importance as inhibitors of NOTCH signaling are available.
Peripheral T-Cell Lymphoma
Peripheral T-cell lymphomas are a heterogenous group of malignant neoplasms derived from mature T cells. They are relatively rare, account for only 12% of all non-Hodgkin lymphomas and are generally aggressive with poor response to therapy. The most common disease within this group is peripheral T-cell lymphoma, not otherwise specified (PTCL, NOS). PTCL, NOS is itself heterogenous, consisting of those T-cell lymphomas that do not otherwise meet the diagnostic criteria for more specific disease entities. Anaplastic T-cell lymphoma, ALK-negative is a newly recognized provisional disease entity by the World Health Organization. It shares histopathologic features with anaplastic large cell lymphoma (ALCL), ALK-positive but does not harbor the characteristic translocation involving the ALK gene seen in ALCL, ALK-positive and thus the malignant cells do not express ALK. Whether ALCL, ALK-negative is distinct from PTCL, NOS and warrants its own diagnostic category is not clear and more studies examining tumor biology are required.
Feldman et al45 used NGS as a discovery tool in ALCL, ALK-negative. They examined a single case with a known rearrangement at chromosome 6p25.3. They found the DUSP22 phosphatase on chromosome 6p25.3 formed a translocation with the fragile site FRA7H on chromosome 7q32.3. They extended their investigation to 29 additional cases of ALK-negative ALCL that harbored a 6p25.3 rearrangement. Using fluorescence in situ hybridization, they found that 45% of these cases harbored the t(6;7)(p25.3;q32.3) showing that it is a recurrent event in ALK-negative ALCLs. They also showed the t(6;7)(p25.3;q32.3) was associated with downregulation of DUSP22 and upregulation of MIR29 microRNAs on 7q32.3, suggesting DUSP22 may act as a tumor suppressor, whereas MIR29 may have oncogenic function.
Vasmatzis et al46 examined 16 patient samples of peripheral T-cell lymphomas and 6 T-cell lymphoma cell lines with NGS and identified 13 gene rearrangements in 21 cases. Nine of the rearrangements were intrachromosomal and involved p53-related genes in 38% of cases. Two cases harbored the inv(3)(q26q28) leading to a TBL1XR1/TP63 gene fusion. The remaining 4 rearrangements were interchromosomal. One case harbored the t(3;6)(q28;p22.3) corresponding to a TP63/ATXN1 gene fusion. The previously described t(6;7)(p25.3;q32.3) was also observed. The authors also showed the TP63 gene fusions encoded for a fusion protein homologous to ΔNp63, a dominant-negative p63 isoform that inhibits the p53 pathway. These findings are virtually identical to those originally described by Scott et al31 in the GCB subtype of DLBCL.
Noting the recurrent nature of rearrangements involving the TP63 gene, Vasmatzis and colleagues expanded their study to 190 cases of various peripheral T-cell lymphomas, including: PTCL, NOS; ALCL, ALK-negative and ALK-positive; angioimmunoblastic T-cell lymphoma; mycosis fungoides; enteropathy associated T-cell lymphoma; hepatosplenic T-cell lymphoma; subcutaneous panniculitis-like T-cell lymphoma; T-cell large granular lymphocytic leukemia; and extranodal NK/T-cell lymphoma, nasal type. Using fluorescence in situ hybridization, they identified recurrent rearrangements involving TP63 in 9.2% of PTCL, NOS and 11.8% of ALCL, ALK-negative. In the majority of cases, the partner gene was TBL1XR1. Rearrangements involving TP63 were not identified in the other examined lymphoma types. These findings suggest rearrangements of TP63 may represent a mechanism for deregulation of the p53 pathway in a subset of peripheral T-cell lymphomas.
These 2 studies and the work of Scott and colleagues highlight the power of NGS as a discovery tool for novel translocations in lymphoid cancer.
In the coming years, more lymphoid cancers will be sequenced and our understanding of the driver events that lead to lymphomagenesis will be expanded. We will also see gains and further insight into those events that lead to disease progression and lymphoma transformation. The knowledge gained form NGS will hopefully lead to the creation and implementation of novel and targeted therapeutic strategies resulting in a reduction in overall lymphoma morbidity and mortality: the recent development of specific EZH2 inhibitors that followed the discovery in 2010 of EZH2 mutations in FL and DLBCL shows we are well on our way to achieving this. It may also come to pass that with further refinements in technology, NGS will become cost permissive, and it will become part of the routine diagnostic workup of patient specimens. Analyzing and describing individual patient tumors on a genome-wide scale will allow for truly personalized medicine.
1. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA. 1977;74:5463–5467
2. Lander ES, Linton LM, Birren B, et al. Initial sequencing and analysis of the human genome. Nature. 2001;409:860–921
3. . Finishing the euchromatic sequence of the human genome. Nature. 2004;431:931–945
4. Venter JC, Adams MD, Myers EW, et al. The sequence of the human genome. Science. 2001;291:1304–1351
5. Chin L, Andersen JN, Futreal PA. Cancer genomics: from discovery science to personalized medicine. Nat Med. 2011;17:297–303
6. Metzker ML. Sequencing technologies—the next generation. Nat Rev Genet. 2010;11:31–46
7. Shendure J, Ji H. Next-generation DNA sequencing. Nat Biotechnol. 2008;26:1135–1145
8. Stratton MR. Exploring the genomes of cancer cells: progress and promise. Science. 2011;331:1553–1558
9. Ross JS, Cronin M. Whole cancer genome sequencing by next-generation methods. Am J Clin Pathol. 2011;136:527–539
10. Morin RD, Mendez-Lago M, Mungall AJ, et al. Frequent mutation of histone-modifying genes in non-Hodgkin lymphoma. Nature. 2011;476:298–303
11. Pasqualucci L, Trifonov V, Fabbri G, et al. Analysis of the coding genome of diffuse large B-cell lymphoma. Nat Genet. 2011;43:830–837
12. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 20084th ed Lyon IARC
13. Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature. 2000;403:503–511
14. Rosenwald A, Wright G, Chan WC, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med. 2002;346:1937–1947
15. Lenz G, Wright G, Dave SS, et al. Stromal gene signatures in large-B-cell lymphomas. N Engl J Med. 2008;359:2313–2323
16. Morin RD, Johnson NA, Severson TM, et al. Somatic mutations altering EZH2 (Tyr641) in follicular and diffuse large B-cell lymphomas of germinal-center origin. Nat Genet. 2010;42:181–185
17. Kirmizis A, Bartley SM, Kuzmichev A, et al. Silencing of human polycomb target genes is associated with methylation of histone H3 Lys 27. Genes Dev. 2004;18:1592–1605
18. Sneeringer CJ, Scott MP, Kuntz KW, et al. Coordinated activities of wild-type plus mutant EZH2 drive tumor-associated hypertrimethylation of lysine 27 on histone H3 (H3K27) in human B-cell lymphomas. Proc Natl Acad Sci USA. 2010;107:20980–20985
19. Yap DB, Chu J, Berg T, et al. Somatic mutations at EZH2 Y641 act dominantly through a mechanism of selectively altered PRC2 catalytic activity, to increase H3K27 trimethylation. Blood. 2011;117:2451–2459
20. Qi W, Chan H, Teng L, et al. Selective inhibition of Ezh2 by a small molecule inhibitor blocks tumor cells proliferation. Proc Natl Acad Sci USA. 2012;109:21360–21365
21. McCabe MT, Ott HM, Ganji G, et al. EZH2 inhibition as a therapeutic strategy for lymphoma with EZH2-activating mutations. Nature. 2012;492:108–112
22. Knutson SK, Wigle TJ, Warholic NM, et al. A selective inhibitor of EZH2 blocks H3K27 methylation and kills mutant lymphoma cells. Nat Chem Biol. 2012;8:890–896
23. Ngo VN, Young RM, Schmitz R, et al. Oncogenically active MYD88 mutations in human lymphoma. Nature. 2011;470:115–119
24. Davis RE, Brown KD, Siebenlist U, et al. Constitutive nuclear factor kappaB activity is required for survival of activated B cell-like diffuse large B cell lymphoma cells. J Exp Med. 2001;194:1861–1874
25. Pasqualucci L, Dominguez-Sola D, Chiarenza A, et al. Inactivating mutations of acetyltransferase genes in B-cell lymphoma. Nature. 2011;471:189–195
26. Ogryzko VV, Schiltz RL, Russanova V, et al. The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell. 1996;87:953–959
27. Shilatifard A. Molecular implementation and physiological roles for histone H3 lysine 4 (H3K4) methylation. Curr Opin Cell Biol. 2008;20:341–348
28. Issaeva I, Zonis Y, Rozovskaia T, et al. Knockdown of ALR (MLL2) reveals ALR target genes and leads to alterations in cell adhesion and growth. Mol Cell Biol. 2007;27:1889–1903
29. Challa-Malladi M, Lieu YK, Califano O, et al. Combined genetic inactivation of beta2-Microglobulin and CD58 reveals frequent escape from immune recognition in diffuse large B cell lymphoma. Cancer Cell. 2011;20:728–740
30. Lohr JG, Stojanov P, Lawrence MS, et al. Discovery and prioritization of somatic mutations in diffuse large B-cell lymphoma (DLBCL) by whole-exome sequencing. Proc Natl Acad Sci USA. 2012;109:3879–3884
31. Scott DW, Mungall KL, Ben-Neriah S, et al. TBL1XR1/TP63: a novel recurrent gene fusion in B-cell non-Hodgkin lymphoma. Blood. 2012;119:4949–4952
32. Steidl C, Shah SP, Woolcock BW, et al. MHC class II transactivator CIITA is a recurrent gene fusion partner in lymphoid cancers. Nature. 2011;471:377–381
33. Steidl C, Gascoyne RD. The molecular pathogenesis of primary mediastinal large B-cell lymphoma. Blood. 2011;118:2659–2669
34. Puente XS, Pinyol M, Quesada V, et al. Whole-genome sequencing identifies recurrent mutations in chronic lymphocytic leukaemia. Nature. 2011;475:101–105
35. Fabbri G, Rasi S, Rossi D, et al. Analysis of the chronic lymphocytic leukemia coding genome: role of NOTCH1 mutational activation. J Exp Med. 2011;208:1389–1401
36. Rossi D, Bruscaggin A, Spina V, et al. Mutations of the SF3B1 splicing factor in chronic lymphocytic leukemia: association with progression and fludarabine-refractoriness. Blood. 2011;118:6904–6908
37. Wang L, Lawrence MS, Wan Y, et al. SF3B1 and other novel cancer genes in chronic lymphocytic leukemia. N Engl J Med. 2011;365:2497–2506
38. Quesada V, Conde L, Villamor N, et al. Exome sequencing identifies recurrent mutations of the splicing factor SF3B1 gene in chronic lymphocytic leukemia. Nat Genet. 2012;44:47–52
39. Baro C, Salido M, Espinet B, et al. New chromosomal alterations in a series of 23 splenic marginal zone lymphoma patients revealed by Spectral Karyotyping (SKY). Leuk Res. 2008;32:727–736
40. Kiel MJ, Velusamy T, Betz BL, et al. Whole-genome sequencing identifies recurrent somatic NOTCH2 mutations in splenic marginal zone lymphoma. J Exp Med. 2012;209:1553–1565
41. Rossi D, Trifonov V, Fangazio M, et al. The coding genome of splenic marginal zone lymphoma: activation of NOTCH2 and other pathways regulating marginal zone development. J Exp Med. 2012;209:1537–1551
42. Tiacci E, Trifonov V, Schiavoni G, et al. BRAF mutations in hairy-cell leukemia. N Engl J Med. 2011;364:2305–2315
43. Treon SP, Xu L, Yang G, et al. MYD88 L265P somatic mutation in Waldenstrom’s macroglobulinemia. N Engl J Med. 2012;367:826–833
44. Kridel R, Meissner B, Rogic S, et al. Whole transcriptome sequencing reveals recurrent NOTCH1 mutations in mantle cell lymphoma. Blood. 2012;119:1963–1971
45. Feldman AL, Dogan A, Smith DI, et al. Discovery of recurrent t(6;7)(p25.3;q32.3) translocations in ALK-negative anaplastic large cell lymphomas by massively parallel genomic sequencing. Blood. 2011;117:915–919
46. Vasmatzis G, Johnson SH, Knudson RA, et al. Genome-wide analysis reveals recurrent structural abnormalities of TP63 and other p53-related genes in peripheral T-cell lymphomas. Blood. 2012;120:2280–2289
© 2013 Lippincott Williams & Wilkins, Inc.