Genomic landscape of mature B-cell non-Hodgkin lymphomas — an appraisal from lymphomagenesis to drug resistance

Background: Mature B-cell non-Hodgkin lymphomas are one of the most common hematological malignancies with a divergent clinical presentation, phenotype, and course of disease regulated by underlying genetic mechanism. Main body: Genetic and molecular alterations are not only critical for lymphomagenesis but also largely responsible for differing therapeutic response in these neoplasms. In recent years, advanced molecular tools have provided a deeper understanding regarding these oncogenic drives for predicting progression as well as refractory behavior in these diseases. The prognostic models based on gene expression profiling have also been proved effective in various clinical scenarios. However, considerable overlap does exist between the genotypes of individual lymphomas and at the same time where additional molecular lesions may be associated with each entity apart from the key genetic event. Therefore, genomics is one of the cornerstones in the multimodality approach essential for classification and risk stratification of B-cell non-Hodgkin lymphomas. Conclusion: We hereby in this review discuss the wide range of genetic aberrancies associated with tumorigenesis, immune escape, and chemoresistance in major B-cell non-Hodgkin lymphomas.


Background
Recent years have witnessed a significant advancement in the diagnostic modalities of mature B-cell neoplasms guided towards targeted therapeutic intervention leading to superior prognostication. This vast and divergent group of clonal malignancies requires a multifaceted approach including morphology, immunohistochemistry (IHC), flow cytometry, cytogenetics, and molecular studies for an accurate and pinpoint diagnosis. Simultaneously, correlation with clinical features and presentation is critical as these neoplasms manifest varied disease courses ranging from indolent to highly aggressive tumors [1]. Molecular biology in B-cell non-Hodgkin lymphomas (B-NHLs) too is quite heterogeneous and has a consequential impact on the phenotype as well as outcome. A broad spectrum of alterations in genetic profile including aneuploidies, structural rearrangement of chromosomes, copy number variations, and point mutations have been described in these lymphomas [1,2].
Nonetheless, the advent of advanced molecular techniques like targeted sequencing, RNA transcript quantification, cell-free DNA techniques, and multiplex ligation-dependent probe amplification (MLPA)-based assays continues to refine the genotype and risk stratification of individual disease [8,9]. With the advent of novel mutations such as BRAF V600E in hairy cell leukemia (HCL) and MYD88 L265P in lymphoplasmacytic lymphoma (LPL), the interpretation of these entities has become a lot easier [10,11]. Furthermore, the additional genetic mechanism such as missense and truncating mutations of chromatin modifiers like CREBBP, EZH2, and DDX3X as well as histone methylation such as KMT2D and SUZ12 in DLBCLs has given immense insight into the role of epigenetics in B-NHL pathogenesis [8,12,13].
The genetic alterations lead to dysregulation within the common intracellular pathways involved in ontogeny and maturation of B lymphocyte such as B-cell antigen receptor (BCR), nuclear factor kappa B (NF-κB), and PI3K/ AKT/mTOR signaling pathways [14]. It is noteworthy that automation as well as advanced technological tools in diagnostic genomics has revealed the complex biology of B-NHL. These cytogenetic and molecular techniques are now being routinely used on paraffin-embedded biopsies, frozen tumor sections, fine needle aspirate samples, and liquid hematological specimens like peripheral blood and bone marrow aspirate. Herein, we attempt to review the wide range of genetic aberrancies in individual B-NHLs and their key role in lymphomagenesis.

Overview of B-NHL diagnosis and classification
Mature B-cell neoplasms constitute approximately > 90% of lymphoid malignancies worldwide with a broad age distribution. The classification of B-NHLs is essentially based on the following: (1) ontogenic classification -based on B-cell ontogeny and differentiation; (2) morphological classification -based on B-cell size and morphology; (3) immunophenotypic classification -based on expression of various antigens; and (4) clinical classification -based on clinical behaviour, aggressiveness, and overall outcome.

1.
Ontogenetic classification - Figure 1 shows the stages of normal B-cell ontogenetic differentiation ( Fig. 1). However, the B-NHL classification based on normal B-cell ontogenesis is inadequate considering the fact that the presence of lineage heterogeneity and, more rarely, lineage plasticity in few subgroups failed to support this approach. The latest World Health Organization Blue Book emphasizes on an integrated multiparametric approach for these lymphomas [1].   [1,15,16]. Interobserver variation also poses a difficulty in morphologically dubious cases. 3. Immunophenotypic classification -Immunophenotyping and IHC are so far the most important diagnostic tools for both classification and assessment of aggressiveness of individual disease (Fig. 2). For instance, CD5+ expression in a neoplastic B-NHL delineates a differential diagnosis between small lymphocytic lymphoma (SLL)/chronic lymphocytic leukemia (CLL) vs MCL based on CD23, CD200, cyclin D1 (CCND1) and antigenic intensity of surface heavy chain, CD79a, and CD20. However, entities like atypical CLL with bright surface heavy chain expression and loss of CD23 exist and likewise cases of CD10+ MCL and leukemic non-nodal MCL with CD23 and/ or CD200 expression [17][18][19]. Similarly, one poor prognostic subgroup of DLBCL has been described with CD5+ expression and need to be distinguished from the blastoid or pleomorphic variant of mantle cell lymphoma [1]. Distinction between LPL and marginal zone lymphoma (MZL) through immunophenotyping alone is hazy due to lack of specific antigenic markers. Double-/triple-hit lymphomas occasionally fail to demonstrate light chain restriction on immunophenotyping or IHC and may even show sparse Tdt expression creating confusion with lymphoblastic lymphomas [20]. 4. Clinical classification -A wide range of clinical behavior is seen within any entity of B-NHL, and histological transformation as well as clinical progression (e.g., FL into DLBCL or double-hit lymphoma and SLL/CLL with Richter's transformation) may be encountered during the course of the disease [21][22][23]. Therefore, stratification of these malignancies on the basis of histologic grade or clinical aggressiveness alone is unwarranted. Moreover, as a part of the clonal evolution process, lymphomas acquire additional genetic aberrancies and display both morphological and antigenic variation.
The genomic studies in this context have proven to be a crucial tool for restructuring B-NHL classification time and again. Polymerase chain reaction (PCR) of immunoglobulin (IG) gene rearrangement is a valuable technique for clonality assessment in neoplastic B cells. CCND1 rearrangement is seen not only in classical MCLs but also in leukemic non-nodal MCLs and is virtually absent in both SLL/CLLs as well as CD5+ DLBCLs [19,24]. Distinction between HCL and hairy cell leukemia variant (HCL-v) is based on BRAF mutation with the former being positive in most of the cases [10,25]. The identification of MYD88 L265P and CXCR4 S338X mutation is of immense importance for diagnosis of LPLs with none to rare cases of MZL showing these abnormalities [11,26,27]. Demonstration of MYC, BCL2, and/or BCL6 rearrangement is helpful for diagnosing HGBLs with unusual immunophenotypes or lacking light chain restriction [20,28]. Based on gene expression profile (GEP), newer prognostic molecular models have been introduced in DLBCLs [29,30]. Even the distinction between classical and leukemic non-nodal MCL is based on clinical features along with SOX11 mutation study with leukemic non-nodal MCL being typically SOX11 negative. The need for distinction between these two entities is essential as the patients with leukemic nonnodal MCL have been reported to have a longer treatment-free survival and OS compared to classical variants [31,32]. An overview of the critical genetic aberrancies associated with major B-NHLs is highlighted in Table 1.

Small lymphocytic lymphoma/chronic lymphocytic leukemia
The B-cell receptor consisting of IG molecule and CD79A/B subunits is central to CLL pathogenesis [33]. Two major molecular subgroups have been identified based on mutational status of immunoglobulin heavy chain variable region (IGHV) genes, i.e., unmutated IGHV (≤ 98% germline identity) in 50-70% patients and mutated IGHV (≥ 98% identity) in 30-50% patients. B-cell receptor stereotype with very similar, but not identical IG sequences is also noted in approximately 30% of the CLL cases [34,35]. Also, 80-90% of SLL/CLL cases have cytogenetic abnormalities detected by interphase FISH and/or copy number arrays, but disease-specific aberrancies are still unknown. The most common genetic anomalies include deletions in 13q14.3 (miR16-1 and miR15a), followed by trisomy 12 or partial trisomy of 12q13 and less frequently deletion in 11q22-23 (ATM and BIRC3), 17p13 (TP53), or 6q21 [1]. Gene mutations are also seen in 3-15% of CLL cases most commonly involving NOTCH1, SF3B1, TP53, ATM, BIRC3, POT1, and MYD88 genes [36,37]. Mutated NOTCH1, ATM, SF3B1, and TP53 are typically associated with a shorter OS and progression-free survival in treatment-naïve CLL patients [38]. Altered expression patterns of other regulatory microRNAs apart from miR16-1 and miR15a also play a dynamic role in CLL development and progression. Differential expression of miR-34a, miR-223, miR-150, miR-181, and miR-33b has been observed in 17p13 deleted subgroups. Moreover, miR-4524a and miR-744 have shown to be associated with shorter time to first treatment [39,40]. DNA methylation plays variable functional roles linked to CLL pathogenesis and disease outcome [41]. One study proposed a classification of CLL cases into three groups, based on their DNA methylation profiles, i.e., naive B-cell-like, memory B-cell-like, and intermediate CLL. Prognostically, memory-like cases have the best outcome followed by intermediate, whereas naivelike cases are associated with an unfavorable prognosis [42]. Few signature epigenomic events include DNA hypomethylation-induced upregulation of TP63, ZAP70, and NFATc1 and downregulation of DUSP22 and KLF4 responsible for aggressive course of disease [43][44][45][46]. Also, upregulated expression of PAX9 and CRY1 has been shown to have a shorter treatment initiation interval, and PAX9 particularly is associated with a shorter OS [47]. Recent epigenetic models point towards an aging-related increase in DNA methylation at specific genomic regions, particularly single CpG sites that give rise to different clones with divergent outcomes. One such example is a single non-promoter CpG+223 methylation critical for ZAP70 expression [48].

Mantle cell lymphoma
The crucial molecular mechanism in MCL lies in the overexpression of CCND1 molecule in the naive pre-germinal-center B cells. Aberrant truncated mRNA transcripts of CCND1 are particularly associated with aggressive course and poor prognosis. Majority of MCL cases are associated with t(11;14) (q13;q32) translocation between the IGH gene and CCND1 [24]. Cyclin D1/ CDK4/6 complex-induced phosphorylation and subsequent inactivation of RB gene lead to progression of cells from G1 to the S phase resulting in rapid cell proliferation. One key event in this process is the perinucleolar repositioning of the rearranged IgH-CCND1 segment in nucleolin transcription factor-rich areas [49][50][51]. In addition, abnormal overexpression of SOX11 is specific for MCLs independent of t (11;14). In CCND1negative patients, SOX11 positivity as well as cyclin D2 (CCND2)/cyclin D3 (CCND3) translocations is of diagnostic utility, and such patients behave identically to that of CCND1-positive ones [52,53]. Additional alterations in genes targeting cell cycle regulatory elements, the DNA damage response pathway, and cell survival most notably ATM, KMT2D, NOTCH1/2, NSD2, BIRC3, TRAF2, MAP2K14, CARD11, SMARCA4, UBR5, and BTK have been demonstrated in various morphological subgroups. Moreover, TP53 mutations in particular are associated with progression to blastoid or pleomorphic MCL [54][55][56]. Leukemic non-nodal variant of MCL is also associated with CCND1 translocation; however, SOX11 mutation has never been reported [31,32].

Follicular lymphoma and variants
Follicular lymphoma is a heterogeneous neoplasm, and the latest WHO classification describes different clinicobiological variants, namely nodal FL, in situ follicular neoplasia (ISFN), duodenal-type FL, testicular FL, diffuse FL, and pediatric type. Through multiplex PCR, somatic hypermutation of IGH VDJ gene through acquisition of asparagine (N)-linked glycosylation sites is detected in almost all the cases of FL [57,58]. The aberrant genetic hallmark characterized by the t(14;18) (q32;q21) is identified in as many as 90% of grades 1-2 nodal FLs and involves translocation between the IGH and BCL2 genes resulting in the antiapoptotic protein BCL2 overexpression [57]. Nevertheless, BCL2 translocation is generally absent in testicular, diffuse, and in pediatric variants [1,59,60]. Additional genetic alterations in nodal FL include loss of 1p, 6q, 10q, and 17p and gains of chromosomes 1, 6p, 7, 8, 12q, X, and 18q [1]. Specifically, in ISFN, comparative genomic hybridization array studies have identified low amplitude copy number amplifications of chromosomes 1 and 18 which could be an early step in lymphomagenesis. Apart from the driver oncogenes, few recently discovered genes particularly related to tumor biology in ISFN include BACH2, TOX, AFF3, and EBF1 [61]. One noteworthy mutation  [62,63]. Gene expression profiling also recognizes specific mutations within individual variants of FL. Particularly in pediatric-type FLs, MAP2K1 gene encoding the MEK1 protein is the most commonly mutated gene followed by TNFRSF14 [64]. Similarly, duodenal-type FLs typically overexpress CCL20 and MADCAM1 genes resembling the mutational profile of MALT lymphoma [65]. Epigenetic aberrancy in FLs tumorigenesis is a wellestablished phenomenon. Amplifications of histone modifiers EZH2, ARID2, and HDAC7 as well as chromatin regulators CREBBP and KMT2D are commonly noted in FLs and appear to be an early driving event [61,66]. However, one exception to this finding is pediatric-type FL which usually lacks recurrent mutations of epigenetic modifiers [58,61,64].

Lymphoplasmacytic lymphoma
A diagnostic challenge remains to differentiate between LPL and marginal zone lymphoma on the basis of morphology and immunophenotype. However, with the detection of MYD88 L265P mutation in > 90% cases of LPLs, this distinction is now easier. MYD88 L265P mutation activates the NF-κB pathway through phosphorylation of Bruton's tyrosine kinase (BTK) in the B-cell receptor [26,27,67]. In addition, approximately 30% LPL patients have truncating CXCR4 mutations most frequently S338X (nonsense), followed by S341fs (frameshift) and R334X (nonsense) mutation [27]. Other somatic mutations, such as mutations of ARID1A, TP53, CD79B, KMT2D, and MYBBP1A, are encountered with lower frequency. Mutations and deletion of TP53 are typically associated with an unfavorable prognosis. 6q deletion and trisomy 4 are the most common cytogenetic abnormalities demonstrated in 7-54% and 20% of patients with Waldenstrom macroglobulinemia (WM) respectively which can be of additional diagnostic utility [1,26,27].

Hairy cell leukemia and hairy cell leukemia variant
The high frequency of BRAF V600E mutation, described by multiple studies, plays a key role in the pathogenesis of HCL. The mutation is located at exon 15 of BRAF gene on chromosome 7q34 leading to valine substitution by glutamate at codon 600 (V600E) of the BRAF protein. This promotes autophosphorylation of BRAF protein and constitutive activation downstream MEK-ERK signaling pathway [10,25]. BRAF mutation is reported to be absent in 10 to 20% of patients with HCL. However, such patients may demonstrate mutations at alternate exon sites such as exon 11. In BRAF wild-type patients, mutations in MAP2K1 gene encoding MEK are commonly encountered [68].
Apart from these signature aberrancies, few cases show recurrent loss-of-function mutation in CDKN1B/ p27 and KLF2; however, TP53 mutations or del17p are rare. Among the epigenetic regulators, mutations in the histone methyltransferase KMT2C are found in approximately 15% of patients [69,70].
In contrast to HCL, mutations of BRAF are not seen in HCL-v, but activating mutations in MAP2K1 are detected in half of the cases. Other less commonly identified mutations are noted in cell cycle regulator CCND3 and spliceosome encoding gene U2AF1. Compared to HCL, HCL-V shows more frequent recurrent mutations in epigenetic modifiers KMT2C, CREBBP, KDM6A, and ARID1A [69][70][71].

Burkitt lymphoma and Burkitt-like lymphoma with 11q aberration
Burkitt lymphoma is a highly aggressive lymphoma, and the key oncogenic event is the translocation between c-MYC gene at band 8q24 and IGH region on chromosome 14q32 [t (8;14) [72,73]. In approximately 70% of sporadic BL cases, factor TCF3 or its negative regulator ID3 mutation is demonstrated by whole-genome or transcriptome sequencing. These mutations activate antigen-independent B-cell receptor signaling through the Pl3K pathway which promotes BL cell survival [73,74]. Other less commonly encountered genes involved in BL tumorigenesis include CCND3, TP53, RHOA, SMARCA4, ARID1A, IGLL5, BACH2, SIN3A, and DNMT1. Immunodeficiency-associated BL share an identical genetic profile found in sporadic type. However, endemic BL often harbors mutations in BCL7A as well as BCL6 and less commonly alterations in DNMT1, CCND3, ARID1A, and RHOA genes. The incidence of TCF3 or ID3 driver mutations is also lower than the sporadic type [74]. Moreover, individuals with SH2D1A gene-mutated X-linked lymphoproliferative syndrome have shown to be highly susceptible for developing BL [75].
The latest WHO classification has subdivided the previous "Burkitt-like" lymphoma category into more precisely defined entities like Burkitt-like lymphoma with 11q aberration, high-grade B-cell lymphoma with MYC and BCL2 and/or BCL6 rearrangement, and high-grade B-cell lymphoma, not otherwise specified. Burkitt-like lymphoma with 11q aberration is a new entity with a complex karyotype and lacking MYC rearrangement or 1q gain typical of BL. These lymphomas often display mild to moderate cytologic pleomorphism and a nodal presentation, predominantly with a single bulky tumor mass. The alteration of chromosome 11q is characterized by interstitial amplification in 11q23. 2-23.3 region and telomeric losses of 11q24.1-qter [76,77]. Studies have demonstrated that these lymphomas have high-grade gene expression profiling, and their mutational profile resembles that of germinal center-derived lymphomas (also supported by LMO2 expression). Most commonly occurring recurrent gene mutations include BTG2, DDX3X, and ETS1; however, mutations in ID3, TCF3, or CCND3 genes typical of BL are primarily absent [77,78]. A significant association has been demonstrated with mutations in epigenetic modifier genes such as CREBBP, KMT2C, ARID1A, EP300, CREBBP, KMT2C, EZH2, and KMT2D; however, the data is limited owing to the rarity of the disease [78].

Diffuse large B-cell lymphoma
Diffuse large B-cell lymphoma is the most commonly occurring NHL with a high refractoriness and relapse rate. Molecular heterogeneity is a hallmark of DLBCL, and several studies have demonstrated a diverse gene expression profile. In the cell-of-origin classification, transcriptional subtypes of DLBCL encompassing activated B-cell like (ABC) and germinal center B-cell like (GCB), and molecular high-grade variants, have been described [79]. Recurrent somatic gain-of-function mutations in histone-lysine N-methyltransferase EZH2 as well as loss-of-function mutations GNA13 genes are fundamental molecular aberrancies demonstrated in GCB DLBCL [1,80]. Similarly, 40% cases of GCB DLBCL show translocation of the BCL2 gene, a characteristic of FL. In contrast, ABC DLBCLs frequently demonstrate mutations of genes linked to BCR signaling and NF-κB pathways, i.e., CD79A, CD79B, CARD11, MYD88, TNFAIP3, TRAF2, and TRAF5. Translocation of the 3q27 region involving BCL6 gene tends to occur more commonly in ABC type as well [81].
Apart from these genetic aberrancies, GCB DLBCL express copy number alterations in terms of amplifications of REL, MDM2, and MIRHG1 genes as well as deletions in TNFRSF14, PTEN, and ING1. Loss of PTEN particularly provides an oncogenic signal through constitutive activation of the PI3K/AKT pathway. At the same time, ABC DLBCL cases show gain of FOXP1, NFKBIZ, BCL2, and NFATc1 and deletion of PRDM1 as well as INK4/ARF locus encoding CDKN2A tumor suppressor gene [82,83].
A recent study has classified DLBCLs through targeted sequencing into five molecular subgroups based on their genomic profile and different outcomes, i.e., MYD88, BCL2, SOCS1/SGK1, TET2/SGK1, and NOTCH2 types [29]. The MYD88 subgroup comprises mutations in MYD88 L265P, PIM1, CD79B, and ETV6, and loss of CDKN2A belongs to the ABC type and is associated with a poor prognosis. BCL2 type demonstrates recurrent mutations of EZH2, BCL2, CREBBP, TNFRSF14, KMT2D, and MEF2B, belongs to GCB type, and generally has a favorable prognosis. The most favorable outcome is associated with SOCS1/SGK1 subgroup harboring mutations of SOCS1, CD83, SGK1, NFKBIA, and HIST1H1E and predominantly of GCB type. TET2/SGK1 is a less commonly delineated subtype having a similar genomic profile as that of SOCS1/SGK1 but lacks SOCS1 as well as CD83 mutations and instead carries additional mutations of TET2 and BRAF. This too is predominantly of GCB origin and shows a favorable prognosis. The 5th subgroup, i.e., NOTCH2 type dominated by mutations of NOTCH2, BCL10, TNFAIP3, CCND3, and SPEN, is not associated with any cell of origin and in fact resembles the tumor biology of MZLs. Another study too suggested a novel molecular classification for DLBCLs and categorized them into five distinctive genetic clusters (C1-C5) [30]. A detailed description of individual subtypes is beyond the scope of this article. Roughly, the MYD88 cluster strongly recapitulates with the C5 subgroup, BCL2 with C3, TET2/SGK1 with C4, and NOTCH2 with that of C1. The distinct C2 subtype shows biallelic TP53 inactivation, loss of CDKN2A, and widespread copy number changes and carries a dismal prognosis.
Moreover, DLBCLs predominantly arising in testis and central nervous system demonstrate immune escape phenomenon through downregulation of the MHC class II transactivator CIITA which encodes for HLA I and HLA II and thereby leading to MHC II silencing [84]. Loss of function of beta2-microglobulin gene, a component of HLA heavy chain, is also noted in one-third of DLBCL cases facilitating immune escape [85].

Primary mediastinal (thymic) large B-cell lymphoma
Primary mediastinal (thymic) large B-cell lymphoma (PMBL) is an aggressive mature B-cell NHL with unique clinicopathologic features. The key biological aberrancy is mutations or rearrangements of transactivator CIITA gene at chromosome 16p13 leading to MHC class II molecule downregulation [84,86]. Most common partner genes for rearrangement of CIITA are programmed cell death ligands 1 and 2 (PDL1 and PDL2). REL and BCL11A gene amplifications are identified in around 50% cases of PMBL. Constitutive activation of NF-κB and JAK/STAT pathways is essential for PMBL pathogenesis [87]. Recent studies suggest that deletion of TNFAIP3 tumor suppressor gene in turn promotes NF-κB signaling in approximately 60% patients [88]. On the other hand, amplification of the subtelomeric region of chromosome 9p24.1 together with loss-of-function mutation of SOCS1 and PTPN1 activates the JAK/STAT pathway [87,89]. However, rearrangements of MYC, BCL2, and BCL6 are generally not reported in PMBL.

Large B-cell lymphoma with IRF4 rearrangement
Large B-cell lymphoma (LBCL) with IRF4 rearrangement is a rare subtype of LBCL with a favorable outcome predominantly occurring in pediatric and young adolescents. It often involves the Waldeyer's ring or head and neck lymph node and shows a characteristic cryptic rearrangement of the IRF gene at chromosome 6p25 locus [90,91]. The partner gene is most often IGH, and light chains are involved rarely. In situ hybridization using a break-apart probe for IRF4 is a simple and easy technique for diagnosing such cases [1,91]. Recent studies have identified frequent mutations in IRF4, CARD11, CD79B, and MYD88 genes and possible activation of NF-κB pathway responsible for tumorigenesis. Almost all cases lack MYC or BCL2 rearrangement, whereas BCL6 breakpoints may be detected occasionally [92,93].

High -grade B-cell lymphoma with MYC and BCL2 and/ or BCL6 rearrangements
High-grade B-cell lymphoma with MYC and BCL2 and/ or BCL6 rearrangements is an aggressive form of B-cell lymphoma carrying a MYC rearrangement at chromosome 8q24 accompanied by additional translocations of BCL2 at chromosome 18q21 and/or BCL6 at chromosome 3q27, also known as double-hit or triple-hit lymphoma [28,94]. In approximately 65% of cases, the partner gene for MYC is one of the IG genes frequently IGH followed by IGK or IGL [94]. These lymphomas are often associated with a complex karyotype and may demonstrate hemizygous mutations of ID3 [1,95]. Recent studies suggested that double-hit lymphomas have an identical mutation profile (recurrent mutations in BCL2, KMT2D, CREBBP, EZH2, and TNFRSF14) seen in FLs, and therefore, the possible origin might be FLs or its precursor lesion. It also emphasizes the fact that additional mutations in MYC, TP53, GNA13, P2RY8, PIM1, and B2M genes displayed by FLs during high-grade transformation are detected in double-hit HGBLs as well [96,97].

Overlapping molecular aberrancies in B-NHLs
More than 20 lymphoproliferative disorders are classified under B-cell NHLs. Nonetheless, these share considerable resemblance not only in terms of morphology and immunophenotype alone but genomic profile as well. The hallmark MYD88 L265P driver mutation for LPLs is also detected in CLLs, MZLs, MCLs, ABC DLBCLs, and primary CNS lymphomas with a much lower frequency [98]. It should be emphasized here that the cell of origin for all these lymphomas is not the same. Similarly, c-MYC rearrangements tend to occur in indolent low-grade B-cell lymphoma as well as highly aggressive BL, plasmablastic, and double-hit lymphomas [99]. At the same time, some GCB DLBCLs also demonstrate t (14;18) observed in FLs as already mentioned above, and in this case, both of these lymphomas originate from germinal center B cells. Moreover, CD30+ DLBCLs show an overlapping gene expression profile with that of PMBL. Similar patterns of mutation in KMT2C, TP53, and LAMA3 have been described in BLs and DLBCLs. Loss-of-function mutation of TNFAIP3 is found in ABC DLBCLs, PMBLs, classical Hodgkin lymphomas, and MCLs [88]. Also, the acquisition of additional hallmark genetic abnormalities during high-grade transformation of indolent lymphomas is a well-documented phenomenon. For example, CLL transformation to DLBCL shows a number of acquired anomalies including NOTCH1 and TP53 mutations, MYC translocations, and loss of function of CDKN2A [100,101]. Likewise, FLs may progress to double-hit lymphomas with additional rearrangements of BCL2 and MYC. Another aspect is that neither all cases of a specific lymphoma express its hallmark genetic anomaly nor the gene expression profile is the same for each case. These molecular features are very similar to the immunophenotyping findings observed in these lymphomas.

Molecular alterations in tumor microenvironment
The tumor microenvironment (TME) in B-NHLs is heterogeneous and composed of the cellular compartment of immune and inflammatory cells, fibroblasts, endothelial cells, and lymphovascular networks and the extracellular matrix [102]. Lymphoma cells are dependent on TME for the regulation of tumor cell survival. Figure 3 demonstrates the detailed interaction of various TME cells with tumor cells in B-NHLs. Nonetheless, the interaction between TME and lymphoma cells plays a critical role in the lymphomagenesis in one of the following ways: (1) escape of lymphoma cells from immune surveillance and (2) emergence of drug resistance. Immune escape mechanism is a well-recognized phenomenon in both DLBCLs and PMBLs. The MHC class II transactivator CIITA is commonly mutated in DLBCLs and PMBLs resulting in T-cell exhaustion [87]. In addition, copy number gains of PDL1 and PDL2 appear to regulate immune escape in PMBLs. PDL1 overexpression on tumor-associated macrophages has been linked to adverse prognostic subgroups in DLBCLs. The same PDL1 gene mutation in MCL results in dysregulated T-cell proliferation and impaired antitumor response [103,104]. B2M and CD58 genes are critical for recognition of tumor antigen by circulating T lymphocytes and NK cells, respectively. Lossof-function mutation or deletion of B2M and CD58 genes is responsible for evading this immune cell recognition in cases of DLBCL [85]. Also, there is an increased VEGF expression and higher microvascular density found in CD5+ABC DLBCLs than GCB DLBCLs [105,106]. This finding is in accordance with increased pro-angiogenic microRNAs expression, i.e., miR-126 and miR-130a, in both DLBCL tumor cells and TME cells [107]. In FLs too, GEP of tumor-infiltrating lymphocytes (TILs) has shown that patients with ≤ 5% PD1+ TILs are more likely to progress to high-grade lymphomas [108]. For DLBCL, an international prognostic system (IPS) incorporating simple clinical parameters (age, lactate dehydrogenase, number/sites of involvement, stage, performance status) is widely used. However, IPI is inadequate to identify patients with poor survival; thus, integration of molecular and genetic features of the tumor and its microenvironment into existing scoring systems is strongly recommended for better risk stratification. In this context, a risk model combining clinical parameters of the IPI with genetic aberrancies is (KMT2D, PIM1, and MEF2B) is proposed which identifies patients with a poorer prognosis than the highestrisk IPI category, but it has not been externally validated in a large group of patients [109].

Role of genetic aberrancies in drug resistance in B-NHLs
Molecular aberrancies and genomic instability in tumor cells leading to emergence of heterogeneous subclones underlie the primary mechanism of chemoresistance in B-NHLs (Fig. 4). BTK, a key intermediate of BCR signaling and NF-κB anti-apoptotic pathways, is widely targeted in lymphomas by ibrutinib, a novel BTK inhibitor. The C481S mutation in BTK gene as well as mutation in Pim-1 proto-oncogene (PIM1), a serine/threonine kinase, led to ibrutinib resistance in ABC DLBCLs [110,111]. Similarly, concurrent mutation in CXCR4 S338X driving the PI3K/AKT and ERK pathways is responsible for acquired ibrutinib resistance in one-third of MYD88 L265P-mutated WM patients. In contrast, the presence of wild-type MYD88 in WM patients displays similar drug resistance and shorter progression-free survival as well as OS [112,113]. In DLBCL patients, mutations of genes regulating BCR signaling pathway especially CD79A/B and CARD11 show refractory disease course [114]. In ibrutinib-treated MCL patients, mutations of TP53 and NSD2 as well as loss of function of PTEN and FOXO3a have shown to be associated with blastoid transformation [115]. One noteworthy example is gain-of-function rearrangement of BMI-1 gene at chromosome 10p12 that leads to development of drug-resistant phenotype through transcriptional repression of pro-apoptotic genes in indolent lymphomas like CLLs, FLs, and MCLs. Loss-of-function mutation of CDKN2A was shown to be responsible for chemoresistance in MCLs and DLBCLs. Also, in CLL/SLL patients, mutation or deletion in BIRC3 and SF3B1 genes has been associated with fludarabine resistance [116]. IL6-induced JAK/STAT signaling has been linked to acquired resistance to PI3K pathway inhibitors copanlisib and duvelisib [117]. Similarly, one hypothesis suggests a hyperactivated insulin-like growth factor 1 receptor-related mechanism for idelalisib refractoriness in CLL patients with trisomy 12 [118,119].
It is widely accepted that BCL2 anti-apoptotic protein overexpression either through t (14;18) or silencing of microRNAs miR15a and miR16-1 is associated with inherent as well as acquired chemoresistant subclones within B-NHLs [119]. In addition, there is increasing evidence to class I BCL2 inhibitor venetoclax recently in MCL and CLL patients which has been related to mutation in BH3 drug binding domain of BCL2 gene, loss of BIM, or overexpression of MCL-1 and BCL-XL genes [120,121]. In the era of targeted therapies, a deeper understanding of all these molecular mechanisms regarding drug resistance will indeed help in optimizing treatment protocols.

Conclusion
In conclusion, the spectrum of molecular and genomic landscapes of B-NHLs is ever evolving. Considerable genetic variations and overlapping molecular traits observed across the various subgroups are highlighting the role of emerging techniques, viz., molecular analysis at single-cell level and proteomics in precise diagnosis, classification, and risk stratification of B-NHLs. Single-cell sequencing may open new facets of molecular mechanisms which may prove to be a key element for translational research. Furthermore, with the increasing evidence of the potential role of tumor microenvironment on lymphomagenesis as well as in drug resistance, integration of the genomic landscape with microenvironment composition is essential to get better insight of the therapeutic modalities of B-NHL improving the overall treatment outcome.