Keywords

Introduction

The pathogenesis of B-cell lymphomas is assumed to be a multifactorial and multistep process. Known contributing factors include germline predisposition, processes of physiologic B-cell development, environmental factors (e.g. viruses), microenvironmental stimuli, and somatic alterations [1]. Remarkably, these factors do interact on various cellular levels, e.g. germline predisposition to pediatric B-cell lymphoma might cause altered response to viral infection or repair of DNA damage [2, 3]. Moreover, different pathogenetic means can substitute for each other, e.g. essential pathways contributing to lymphomagenesis can be altered by germline or somatic mutations on DNA level or by transcriptional changes on RNA level; the latter may be associated by epigenetic changes [4] which in turn may or may not be induced by external stimuli like viruses [3].

In the following we will outline key principles and general mechanisms underlying the pathogenesis of B-cell non-Hodgkin lymphomas in children and adolescents as far as not reviewed elsewhere in this book (regarding germline predisposition see Chap. 7; for details on the distinct lymphoma entities, see Chaps. 11, 12, 13, 14, 15, 16, 17, and 18). The basis for the understanding of B-cell lymphomagenesis is the normal B-cell development and its underlying cellular and genetic mechanisms, as these physiological processes become hijacked during lymphoma initiation and progression. Accordingly, in the following first the physiologic B-cell development and differentiation are summarized, before we subsequently give an overview on the various mechanisms contributing to B-cell lymphomagenesis.

Physiologic B-Cell Development

B-cells are key players in the physiological immune response including the humoral immune response as well as the immunological memory [5]. These functions are accomplished by the unique features of B-cells including antigen presentation, immune regulation, and provision of the cellular as well as humoral immune repertoire. The key function of B-cells is to produce immunoglobulins. Besides the signaling through these B-cell receptors, also mechanisms involved in determining immunoglobulin specificity, if mistaken, contribute to lymphomagenesis [6, 7]. Thus, in the following we first review the molecular processes underlying immunoglobulin determination, and then we will provide insights into the physiologic B-cell development.

Immunoglobulins and Immunoglobulin Genes

Immunoglobulin Structure

The immunoglobulins or antibodies are the effector molecules produced by the B-cells mediating the humoral immune response [8]. Furthermore, the membrane-bound form of the antibodies, the B-cell receptor, and its associated cofactors mediate intracellular signals important for B-cell development and differentiation. The gene loci encoding the B-cell receptor undergo complex rearrangements during B-cell development contributing to the antibody diversity [8]. Accordingly, every physiologic B-cell has a unique B-cell receptor harboring an individual specificity for a certain antigen. For a better understanding of the mechanisms leading to this diversity, the structure of the immunoglobulins is described in the following. Immunoglobulins are Y-shaped polypeptides consisting of two heavy and two light chains linked by disulfide bonds. The heavy chain consists of a variable (VH) and three constant domains (CH1-3), whereas the light chains are built of a variable (VL) and one constant domain (CL) [8]. The variable domains, composed of the variable domains of the light and heavy chains mediate the specific antigen binding. The constant region, which consists of the constant domains from heavy and light chains, interacts with the effector cells and molecules. The B-cell receptor carries, in contrast to the soluble antibody, a C-terminal polypeptide which anchors the receptor to the cell membrane [9, 10].

The immunoglobulins are encoded by a multi-gene family. Two light-chain types exist. On chromosome 2p11 maps the immunoglobulin κ (IGK) and on chromosome 22q11 the λ (IGL) gene locus. The heavy chain is encoded by a gene locus on chromosome 14q32. These three gene loci contain several coding and non-coding gene segments which are rearranged to form a functional immunoglobulin gene. The light-chain loci have multiple variable (V), joining (J), and constant (C) gene segments. The heavy chain locus harbors in addition diversity (D) gene segments as well as a series of C gene segments: Cμ, Cδ, Cγ, Cα, and Cε. These C gene segments encode the immunoglobulin isotypes: IgM, IgD, IgG, IgA, and IgE, respectively, which confer the effector functions of the respective antibodies [8].

Molecular Processes Remodeling Immunoglobulin Genes

Each B-cell harbors after complex, somatic rearrangements, which contribute to the antibody diversity, a unique immunoglobulin gene (Fig. 4.1). The first step in the generation of this diversity is the VDJ-gene rearrangement which takes place in precursor B-cells within the bone marrow. As outlined above, the immunoglobulin gene locus consists of multiple gene segments which need to be rearranged in order to give rise to a functional coding exon. First, in a random recombination process one of the D gene segments of the heavy chain locus is fused to one J gene segment, followed by a rearrangement of one V gene segment to the already fused DJ segment [11]. Lastly, the VDJ fusion is rearranged to the C gene segment. Due to the multiple gene segments of the heavy chain locus more than 104 different VDJ recombinations are possible, contributing to the combinatorial diversity of the antibody repertoire. After successful rearrangement, the heavy chain is expressed on the surface of the precursor B-cell with a surrogate light chain. In a next step, the light-chain locus is rearranged fusing one of the V gene segments to a J segment. The V(D)J recombination is mediated by a V(D)J recombinase complex, including the RAG1 and RAG2 proteins. Those proteins bind to the conserved recombination signal sequences adjacent to each gene segment. Upon binding, the double-stranded DNA is cut and a hairpin structure is built [12]. The hairpin can be removed in different ways leading either to the insertion of non-germline-encoded (N) nucleotides conferred by the terminal deoxynucleotidyl transferase (TdT) or the deletion of single nucleotides conferred by exonucleases at the recombination sites [13]. These sequence alterations confer the junctional diversity of the antibodies.

Fig. 4.1
figure 1

Schematic overview on the molecular processes remodeling the immuglobulin genes, using as example IGH locus. (a) Process on the VDJ recombination. First, the D gene segment is assembled to one of the J gene segments. After, one of the V gene segments is associated to the DJ segment. (b) Somatic hypermutation process is activated in the germinal center, introducing mutations in the V region of the heavy and the light chain (designated by the lollipops). (c) The latest step in the Ig remodeling is the class switch recombination process, which takes place only on the heavy chain locus. Modified from [1]

Full size image

Other mechanisms contributing to the diversity of the antibodies take place after the B-cell has encountered its antigen primarily during the germinal center (GC) reaction. By somatic hypermutation (SHM) point mutations are introduced within the gene segments encoding the V region of the antibody [14]. This is conducted by the activation-induced deaminase (AID) which upon binding to single-stranded DNA deaminates cytosine to uracil [15, 16]. The binding of AID is heavily dependent on the expression of the respective gene segment. By the introduction of an uracil within the DNA sequence, either the mismatch repair or the base excision repair mechanism pathways are activated which further introduce alterations within the DNA sequence [14]. Another mechanism taking place in the GC is the class switch recombination (CSR) [17]. By CSR the constant gene segment of the heavy chain locus is switched by an irreversible DNA recombination process including non-homologous DNA recombination [18]. This process is again conferred by the AID which is guided to specific switch regions located within the intron between the JH and upstream of all C gene segments [19]. Hence, B-cells initially expressing IgM or IgD switch their immunoglobulin isotype in the majority to an IgG but also to an IgA or IgE. This leads on the one hand to a change in the B-cell receptor signaling competence but alters as well the effector function of the antibody.

During all the processes leading to the physiologic shaping of the IGs mistakes can occur which might result in fusion of part of the IG genes with other gene loci. If the latter contain oncogenes, these can be driven by the strong enhancers at the IG loci which physiologically ensure sufficient BCR/IG production in B-cells. Thus, as a consequence of such aberrant IG rearrangements, the oncogenes on the translocation partners are deregulated. Well known examples in pediatric B-cell lymphomas for such IG enhancer hijacking are the oncogenes MYC or IRF4. The three molecular mechanisms described, V(D)J recombination, CSR, and SHM, involved in the Ig remodeling have all been shown to be implicated in the generation of such aberrant IG rearrangements. Remarkably, the identification of the mechanism leading to an IG translocation provides evidence at which B-cell development stage it took most likely place, as V(D)J recombination usually is restricted to B-cell precursors in the bone marrow whereas CSR and SHM involve different compartments of the germinal center [6]. The IG translocations occurring as a consequence of V(D)J recombination typically have breakpoints that involve RAG recognition sites (RSS) and are directly adjacent to Ig heavy chain J-regions (JH) gene segments or that are adjacent to regions where the Ig heavy chain D-region (DH) joins the J-region (DHJH) [20,21,22]. The presence of N-nucleotides or nucleotides removed in the junctional sequence are typical features of this molecular mechanism. In contrast, the typical features of IG translocations due to aberrant SHM are that the breakpoints are located within or adjacent to rearranged V(D)J genes and that mutations in the V regions are present. Breakpoints in the IGH constant genes particularly affect switch regions and indicate that they derived from mistaken class switching.

B-Cell Differentiation

Early B-Cell Development

The development of B-cells, initiated in fetal liver, relocates to the bone marrow during the maturation of the embryo. B-cells derive from hematopoietic stem cells which give rise to cells of the myeloid as well as lymphoid lineage, the latter including the B- and T-cell progeny (Fig. 4.2). The commitment to the B-cell lineage is mediated through different transcription factors including EBF1, E2A, and PAX5 [23]. The early precursor B-cells differentiate within the bone marrow in which the B-cell receptor gene is formed by the V(D)J recombination process which is a central process for the generation of mature B-cells [24]. After rearrangement of the heavy chain locus, the pro B-cells express a precursor B-cell receptor on its surface together with a surrogate light chain [25, 26]. In case the rearrangement was non-functional, the second allele can be rearranged or the cells undergo apoptosis [27]. If the rearrangement was functional, the second heavy chain allele is suppressed (allelic exclusion) and the light chain is rearranged starting at the IGK locus [28, 29]. In case this rearrangement is non-functional, the IGL locus can be rearranged [30, 31]. Accordingly, due to isotype exclusion B-cells express usually either the Igκ or Igλ light chains. A functional rearrangement of both, the heavy and light chains leads to the expression of a B-cell receptor of the IgM isotype on the cell surface in now immature B-cells. These B-cells are counter-selected for autoreactivity [24]. In case the B-cells recognize self-antigens, cells either undergo apoptosis, the receptor can get edited using the non-rearranged allele or the cells enter a stage of immunological tolerance. The self-tolerant B-cells leave the bone marrow and pass through the spleen for further negative selection. The now mature naïve B-cells either reside in the spleen within the marginal zone or in the majority of cases circulate through the peripheral blood, lymph, and secondary lymphoid organs until they die or encounter their cognate antigen.

Fig. 4.2
figure 2

Overview of B-cell development. Precursor B-cells develop and mature in the bone marrow, developing to naïve B-cells. After exposure of the naive B-cells with their antigen and blast transformation, they evolve into short-lived plasma cells or enter the germinal center (GC). Centroblasts can undergo apoptosis or develop into centrocytes. Post-GC cells comprise plasma cells and memory/marginal zone B-cells. Modified from [1]

Full size image

The Germinal Center Reaction

The term germinal center describes a histological structure located within secondary lymphoid tissue in which B-cells selected for production of high-affinity antibodies reside. When the naïve B-cell encounters its cognate antigen, these cells migrate into the T-cell zone. The interaction with the T-cells stimulates the proliferation leading to the formation of primary foci as an early GC reaction. The GC initiation relies on the induction of several transcriptional modulators. Among those modulators is BCL6, which is expressed as soon as the activated, naïve B-cells interact with a T-cell in the T-cell zone. BCL6 functions as a transcriptional repressor regulating the GC formation and maintenance including silencing of the anti-apoptotic BCL2 protein [32]. Downregulation of BCL2 ensures that the B-cells maintain in a proliferative state during SHM-based introduction of mutations [33]. In addition, it is important for maintenance of the GC B-cell state, as it downregulates factors including BLIMP1 which is a master regulator for terminal differentiation [34]. Other important modulators for GC formation include MYC and MEF2B. After a couple of days the full germinal center structure has formed consisting of a dark zone, which is a cell dense zone consisting mainly of highly proliferative centroblasts harboring a dense nucleus. Furthermore, a light zone can be observed, in which centrocytes interact with follicular dendritic cells (FDC) as well as T-helper (TH) cells. These zones are surrounded by a mantle zone which consists of locally, resting B-cells which were not activated by their antigen [24, 35]. The polarization of the germinal center B-cells into the dark and light zone relies on chemokine gradients of CXCL13 and CXCL12, whose receptors are upregulated in light zone (CXCR5) and dark zone (CXCR4) B-cells [36, 37]. Initially, the centroblasts within the dark zone proliferate and accumulate mutations which are inserted within the first 1–2 kb downstream of the V gene segment transcriptional start [24, 38, 39]. Mutations within this site, encoding for the variable domain of the antibody which interacts with the antigen, might increase the affinity to the cognate antigen. After a couple of proliferative cycles, the cells enter the light zone in which the affinity of the now modified B-cell receptor is tested by FDCs and TH-cells [24, 39]. Accordingly, the centrocytes compete with others for the most specific antigen recognition. Only those with the highest affinity receive pro-survival signals by which they either undergo further rounds of proliferation within the dark zone or undergo the terminal differentiation steps [40]. One of the factors regulating the reentry into the dark zone is MYC. MYC, being repressed by BCL6 after the GC initiation, becomes reactivated in a subset of light zone B-cells that reenter the dark zone for further cycles of proliferation and SHM [41, 42]. The iterative process of the cyclic reentry back and forth from dark to light zone is a stepwise process further improving the affinity of the B-cell receptor to its antigen. Cells which are not positively selected for undergo apoptosis. During the time the centrocytes reside within the light zone, CSR takes place further diversifying the immunoglobulin repertoire.

Post-germinal Center B-Cell Differentiation

After the final steps of affinity maturation of B-cells, the GC B-cells leave the GC as either differentiated memory B-cells or plasma cells. Both cell types play important roles in the adaptive immunity, as they produce high-affinity antibodies and confer the humoral immunological memory which is the first-line immune reaction upon reencounter of the antigen. The factors leading to the terminal differentiation of GC B-cells is yet not fully understood. Current theories convey that the strength of the B-cell receptor interaction with the antigen is a determinant, another, that a developmental switch in the GC reactions exists or certain cytokines stimuli promote the cell fate.

A prerequisite for the differentiation from a GC B-cell to a plasma cell is the termination of the GC transcription program. This includes the inactivation of PAX5 which is an essential maintenance factor of mature B-cell identity. After suppression of PAX5, cells differentiate into pre-plasmablast cells, which secrete low amounts of antibodies. Furthermore, the downregulation of PAX5 orchestrates the regulation of other factors important for terminal differentiation [43]. Additionally, B-cell receptor and CD40 signaling triggers the activation of BLIMP1/PRDM1, which promotes the plasma cell fate [34]. Parallel to the BLIMP1 activation or even upstream, IRF4 is activated which suppresses the expression of BCL6 which is the key GC identity factor. Hence, taken together, low PAX5 and BCL6 but high IRF4 and BLIMP1 expression switch off genes required for proliferation as well as affinity maturation and promote the reprogramming to the plasma cell transcription program [34, 44, 45]. Some plasma cells migrate to the bone marrow, where these long-lived cells produce high-affinity antibodies [46]. Other plasma cells enter the medullary cords of the lymph nodes or spleen, where they express high titers of antibodies, but are depleted within 2 weeks after infection.

Memory B-cells are long-lived B-cells which can divide if at all very slowly and circle through the blood or reside in the bone marrow or spleen. They represent the initial phase of secondary immune response [47]. Upon antigen encounter, the memory B-cells become a proliferative burst and rapidly differentiate into plasma cells producing high-affinity antibodies as the first line against pathogens. In contrast to plasma cells, the factors leading to differentiation into memory B-cells are unclear. A role for phosphorylated STAT5 and BCL6 has been proposed. In addition, a CD40 stimulation in the centrocytes directing the memory B-cell differentiation has been described [48].

Extra-germinal Center Plasma and Memory B-Cell Differentiation

Within the recent years, several publications have shown that plasma cells as well as memory B-cells do not all derive from mature B-cells that have went through the GC reaction. Instead, plasma cells can as well derive from naïve marginal zone B-cells and mature, naïve B-cells circulating through blood and lymph system. Which of the cells differentiate into plasma cells depends on the nature, dose, and form of the antigen as well as the location at which the antigen was encountered. The differentiation into a plasma cell is also dependent on the interaction with a T-cell, hence if the activation of the B-cells is T-cell dependent (TD) or independent (TI) [49]. The TD immune response leads usually to the differentiation to plasma cells via the GC reaction whereas the TI independent immune response is independent from this. The TI immune response can be induced by antigens which activate conserved pattern recognition receptors as, for example, the toll-like receptors leading to a polyclonal B-cell response (TI-1) or by antigens which have a repetitive structure as bacterial capsules which active the B-cells by B-cell receptor cross-linking (TI-2) [50]. Generally, plasma cells generated by a TI immune response are in comparison to plasma cells induced by a TD response short lived [51]. Moreover, the affinity of their B-cell receptor to the antigen is lower than from TD plasma cells, as the cells are not selected for higher affinity. Interestingly, it has been reported that class switch recombination can, although the cells did not undergo the germinal center reaction, take place, mainly switching the B-cell receptor isotype to IgG2.

The earliest antibody response to a couple of pathogens already takes place in the fetal liver. At this time point B1 cells, which are located within the peritoneal and pleural cavities, as well as, the lamina propria of the gut, can produce natural antibodies as response to some pathogens. The B-cell receptor repertoire is skewed toward antigens which induce a TI-2 immune response. When the B1 encounter such a pathogen, they migrate into the spleen or gut and produce, as the earliest plasma cells, natural IgM antibody [52].

Another source for an early plasma cell response are marginal zone B-cells. These B-cells, localized within the marginal zone of the spleen are the first B-cells to respond to pathogens and differentiate into plasma cells [53]. Mostly, the marginal zone B-cells recognize TI-2 antigens, which circulate through the blood. Hence, due to the permanent localization within the marginal zone of the spleen, the B-cells respond more rapidly than naïve B-cells to antigens localized within the blood. Activated B-cells move to the red pulp of the spleen where they undergo massive proliferation while in parallel differentiating into plasmablasts secreting immunoglobulins. In contrast to mature, naïve B-cells, they react faster to the presence of antigens and have an increased responsiveness [54].

Lastly, early plasma cell response can be conducted by mature naïve B-cells which have encountered their antigen in a TD response. But instead of inducing a GC reaction, these cells can proliferate and form extra-follicular foci of plasmablasts and plasma cells at the periphery of peri-arteriolar lymphoid sheaths. Within these foci the activated cells differentiate to plasmablasts and plasma cells producing antibodies. These cells are in contrast to post-GC short lived as the foci disappear 8 days after antigen encounter [55].

Taken together, plasma cells generated from marginal zone or mature naïve B-cells are short lived, respond rapidly as initial immune response, and have B-cell receptors which are somatically not mutated, leading to production of low-affinity antibodies to their cognate antigen.

In line with the extra germinal center plasma cell differentiation, several studies reported on the existence of memory B-cells which express IgM or have somatically unmutated B-cell receptor which indicates that they do not derive from germinal center B-cells. Hence, memory B-cells can derive early after immunization from naïve B-cells which become activated by their cognate antigen in a T-cell-dependent manner and differentiate outside the germinal center [56]. In principle the GC-independent memory B-cells harbor the same features as the GC-dependent memory B-cells including a long life, rapid proliferation capacity, high sensitivity to low-dose antibody, as well as the capability to differentiate fast into plasma cells during second immune response. In contrast, the memory quality is not as good as in GC-dependent memory B-cells, as there are a fewer isotype-switched memory B-cells, less somatic mutations, and no affinity maturation. But the latter is not a disadvantage, since GC-independent memory B-cells are already preselected, giving rise in a secondary immune response to progeny which acquire mutations and, hence, can act quicker in a second pathogen encounter.

Assumed Cell of Origin of B-Cell Lymphomas in Children and Adolescents

Different stages of B-cell differentiation are characterized by the specific structure of the BCR, the transcriptional and epigenetic profiles, as well as, the expression pattern of differentiation markers. When B-cells go through malignant transformation, they usually keep the features of the respective differentiation stages. Nevertheless, the initial transforming event might occur on a different B-cell maturation stage than the one the tumor cells are ultimately frozen in. The best known example is follicular lymphoma in adults, where the pathognomonic IGH-BCL2 translocation derives from a mislead V(D)J rearrangement in bone marrow B-cell precursor cells, whereas the actual tumor cells clearly show morphologic, transcriptional, and immunologic features of germinal center B-cells with ongoing somatic hypermutation of IG genes. Despite such discrepancies, the supposed cell of origin or the frozen maturation state are used to determine the origin of the human B-cell lymphomas [1, 57].

Taking these pitfalls and limitations into account, B-cell lymphomas can be divided into the following groups (Fig. 4.3):

  • B-cell precursor neoplasms : B-cell lymphoblastic lymphoma in many aspects resembles precursor B-cell lymphoblastic leukemia. The cell of origin based on the marker profile (particularly TdT) and immune gene analysis is supposed to be a B-cell precursor in the bone marrow. There might be some heterogeneity regarding the different pro- and pre-B-cell stages.

  • Germinal center-derived B-cell lymphomas : The vast majority of pediatric and adolescent lymphomas derive from cells which had entered the germinal center. Both endemic and sporadic Burkitt lymphomas (BL) show several features of dark zone cells (centroblasts). At least sporadic BL might be derived from B-cells poised to become IgA-expressing cells and a link to a primary immune response is discussed. Diffuse large B-cell lymphoma based on gene expression data are divided into those with similarity to GC B-cells (GCB-DLBCL) or to in vitro-activated B-cells (ABC-DLBCL). ABC-DLBCL are rare in young patients, their incidence increases with age. Thus, most pediatric and adolescent DLBCL are of GCB type. Primary mediastinal and central nervous system B-cell lymphomas are also derived from cells with contact to the germinal center but might be more advanced in the GC reaction than typical GCB.

  • Marginal zone B-cell lymphomas : Nodal marginal zone B-cell lymphomas (NMZL) derive from marginal zone B-cells, a subset of those might also derive from naïve B-cells [58].

Fig. 4.3
figure 3

Diagrammatic representation of cellular origin of the most common B-cell lymphomas in children and adolescents. B-cell tumors develop from stages of B-cell maturation. Most lymphomas are derived from germinal center (GC) B-cells or from B-cells that went through the GC. Nevertheless, precursor B-lymphoblastic leukemia and lymphoma are derived from precursor B-cell from the bone marrow compartment

Full size image

Mechanisms Contributing to the Transformation of B-Cells

Somatic Genetic Alterations

The theory of cancer development postulates that a normal cell requires several “hits” that change its normal functions. Cellular functions typically altered by such hits are cell cycle control, proliferation and apoptosis, B-cell inherent signaling pathways, and epigenetic modifiers. Various mutational mechanisms can contribute to the generation of such mutations. Besides the B-cell inherent mechanisms outlined above, mutational signature analyses have pointed to the role of aging or antiviral responses triggered by the APOBEC family [59, 60]. These hits can occur either in cancer drivers, which are typically divided into oncogenes and tumor suppressor genes, but also in passengers, which only have a minor importance in the transformation or clonal evolution. Cells with genetic abnormalities are selected based on the fitness and survival and have the opportunity to acquire further aberrations. Different genomic aberrations, like chromosomal translocations, single nucleotide variants, or copy number changes lead to oncogene activation and tumor suppressor gene inactivation.

Oncogene Activation via Chromosomal Translocations

Chromosomal translocations usually activate oncogenes either via enhancer hijacking or via the production of fusion transcripts.

In case of enhancer hijacking, the complete (coding part) of a gene is usually brought in the vicinity of an enhancer of another gene expressed in the (cell of origin of the) tumor cells. Typical events associated with enhancer hijacking in pediatric and adolescent B-cell lymphomas are chromosomal translocations affecting the IGH locus on 14q32.33 or the IGK and IGL light-chain loci on 2p12 and 22q11, respectively. The intact oncogenes, encoding, e.g. MYC or IRF4, are juxtaposed to the various enhancer elements of the IG loci resulting in deregulated expression of the oncoproteins [61] (Table 4.1). Besides, the SHM machinery might further activate the oncogenes through mutations, e.g. in the MYC boxes after the translocation, as the machinery “jumps” over from the IG locus to the partner chromosome after the juxtaposition. The probably most promiscuous oncogene in B-cell lymphomas, which can be activated by hijacking of enhancers from a wide set of genes, is BCL6, which encodes a key regulator of germinal center B-cells. Remarkably, the set of genes deregulated by enhancer hijacking in mature B-cell lymphomas shows some dependence of age: whereas BCL2 deregulation through translocation to the IGH locus is present in around 85% of follicular lymphomas and 30% of diffuse large B-cell lymphoma of adulthood, this change is rare in the same disease in the adolescent population and almost completely absent below the age of 18 years (with the notable exception of IG-MYC translocated Burkitt like lymphoma with precursor phenotype). Similarly, the frequency of BCL6 translocations is lower in children than in (elderly) adults, which might be associated with the age-associated changes in cell of origin of these neoplasms. In contrast, IG-IRF4 translocations are more frequent in younger patients; the same seems to hold true for Burkitt lymphoma with IG-MYC translocation in general, which might point to an age-dependent susceptibility to these diseases. It is intriguing to speculate that the latter is associated with the frequency of primary immune response particularly those driven by enteric microbiota.

Table 4.1 Chromosomal translocations involving the IG loci in mature B-cell neoplasm and children and adolescents
Full size table

Whereas enhancer hijacking is quite common in mature B-cell lymphoma and associated with many of the hallmark chromosomal alterations, the development of fusion transcripts from two separate genes seems to be rather rare in mature B-cell neoplasms [61]. In precursor B-cell neoplasms, i.e., lymphoblastic leukemia/lymphoma, a set of fusion genes has been described, including e.g. BCR/ABL, ETV6/RUNX1, or MLL-fusions, to name the most common. In mature B-cell lymphomas, the probably best known are ALK-fusion genes. ALK-positive large B-cell lymphoma is a very rare lymphoma, characterized by the chromosomal translocation t(2;17)(p23;q23), involving Clathrin (CLTCL) gene in 17q23 and the ALK gene in 2p23, generating a CLTCL-ALK fusion protein [62, 63]. Some cases are associated with the t(2;5)(p23;q35), generating the NPM1-ALK fusion protein. Fusion transcripts encode fusion proteins containing parts of both involved proteins. The part of the partner protein changes the protein function of the oncoprotein, e.g. in case of ALK-fusion leads to constitutive activation of kinase activity due to (aberrant) homodimerization.

Though it is usually assumed that chromosomal translocations lead to oncogene activation, it needs to be emphasized that tumor suppressor gene inactivation due to gene disruption is in part associated with intron retention in fusion transcripts and seems to be a rather much more common consequence of these structural chromosomal aberrations.

Tumor Suppressor Inactivation

Tumor suppressor gene inactivation usually occurs via biallelic deletion and/or mutation. An alternative mechanism to deletion is copy neutral loss of heterozygosity (CNN-LOH), sometimes described as (partial) uniparental (iso) disomy. In addition, tumor suppressor gene function might also be altered by mono-allelic inactivation and haploinsufficiency or by expression of a mutant dominant-negative protein form.

Tumor suppressor genes commonly altered in many tumors including various subtypes of pediatric and adolescent B-cell lymphomas are TP53 and CDNK2A. Mostly, changes in these genes are secondary or even late events in clonal evolution and particularly TP53 inactivation has been frequently linked to unfavorable prognosis. In case of Li Fraumeni syndrome, monoallelic inactivation can predispose to lymphomas, similarly to ATM inactivation in Ataxia telangiectasia. Whereas both latter genes function particularly in DNA repair, another class of tumor suppressors commonly hit in mature B-cell lymphomas both in younger and older patients are genes involved in epigenetic modifications, like KMT2D (formerly known as MLL2), encoding a histone methyltransferase, CREBBP, encoding a histone acetyltransferase, and SMARCA4, being a member of the SWI/SNF chromatin remodeling complex.

Other tumor suppressor genes recurrently targeted in particular subtypes of pediatric and adolescent B-cell lymphomas are TNFRSF14A, frequently targeted in pediatric follicular lymphoma, and ID3, mutated in around 60–70% of sporadic Burkitt lymphomas [64, 65]. The ID3 mutations free its binding partner, namely, TCF3, from the heterodimeric complex and allow TCF3 to bind the DNA and activate its targets. Remarkably, activating mutations of the TCF3 oncogene do have the same effect showing the strong interplay of tumor suppressors and oncogenes.

Oncogene Activation

As shown above oncogene activation cannot only occur though chromosomal translocations as detailed before, but also like in case of TCF3 through activating mutations. Similarly, a high proportion of FL particularly in older patients carry activating mutations of the methyltransferase gene EZH2. Additional means of oncogene activation are copy number gains up to amplifications. The oncogenes outlined above, like MYC or IRF4, are also recurrently hit by such copy number gains. There are several other examples of oncogene activation described in B-cell lymphoma, e.g. gains in REL, a component of the NF-κB complex [66, 67]. Moreover, gains in the 9p region, typically detected in primary mediastinal B-cell lymphoma (PMBL), lead to activation of JAK2, PDL1, and PDL2. The prior is involved in activation of the JAK/STAT pathway, the latter two, also involved in chromosomal translocations with various partners, play a role in the immune detection of the tumor cells. Besides coding genes, also non-coding genes can be targeted by activating oncogenic mechanisms. Probably the best known is gain of oncogenic miR-17-92 cluster, which is a transcriptional target of MYC [68] frequently altered in BL.

Oncogenic activation through mutations can also be a side effect of SHM and CSR at non-Ig genes [69]. Such off-target activity of SHM produces point mutations in proto-oncogenes like BCL6 and CD95 which are also mutated in a considerable fraction of normal GC and memory B-cells.

Signaling Through the B-Cell Receptor

The selection of cells expressing a BCR is also a common feature in malignant B-cells. The majority of B-cell lymphomas express a BCR, however sometimes at low levels [1, 69,70,72]. In case of an IG translocation, this usually affects a non-productive IG locus and leaves the productive IG locus intact [73, 74] indicating that at least when the IG translocation occurred, the BCR signaling has been necessary for B-cell survival and development of the B-cell neoplasm. Moreover, several types of B-cell lymphomas show ongoing mutations in the IGH V-region during tumor clone progression [77,78,80]. Though such mutations if deleterious could prevent functional heavy and light-chain pairing, those tumors still express the BCR indicating selection against damaging mutations [64, 66]. In fact, the low frequency of BCR loss in subtypes of B-cell lymphomas with ongoing somatic hypermutation, like FL and MALT lymphoma, shows the importance of BCR expression in these B-cell lymphomas. Consequently, the survival signals mediated by the BCR expression in normal B-cells likely also play a role in the survival of at least a subset of B-cell lymphoma cells.

BCR-Dependent Lymphomas

Some B-cell lymphomas use the IgM constant regions to form their BCR; however, the majority of B-cell lymphomas derives from germinal center cells that usually switch their BCRs from IgM to IgG. Notably, IgM and IgG-associated BCRs are linked to different downstream signaling: IgM-BCR signaling promotes the survival and proliferation of B-cells by activating pathways like the NF-KB pathway; in contrast, IgG-BCR signaling promotes plasmacytic differentiation by the activation of ERK and MAPK pathways [74,75,77].

A prototype of lymphomas retaining IgM-BCR expression is FL [78]. This lymphoma is genetically characterized by a t(14;18) translocation, involving the IGH locus and the BCL2 gene. In FL, the productive IGH allele is never translocated to BCL2 and assures expression of IgM. In contrast, the non-productive allele is translocated and undergoes CSR to IgG, showing the selective pressure on the cell to retain IgM expression [79]. Another example is DLBCL, where the GCB subtype usually expresses IgG-BCR but does not require BCR signaling for survival [80], whereas the ABC subtype retains the IgM-BCR [81] in part due to deletions within the IGH “switch” regions (Sμ and Sɣ), needed for class switch recombination [82]. These deletions take place on the productive IGH allele, and blocking class switch recombination [83].

There are two forms of pathological BCR signaling in B-cell malignancies: Chronic active BCR signaling involves diverse downstream pathways, including MAPK, PI3K, NFAT, and NF-kB pathways. This form is typical for ABC-DLBCL, where the presence of mutations in CD79A and CD79B is reported in over 20% of the cases. Functional analyses showed that knockdown of proximal BCR subunits, including IgM, IgK, CD79A, and CD79B, is lethal for ABC-DLBCL [81]. On the other hand, tonic BCR signaling activates only the PI3K pathway, like in BL. As outlined above, the majority of BL cases show activating mutations in TCF3 or inactivating mutations in its negative regulator, ID3. TCF3 has been reported as a factor required for expression of all immunoglobulin genes. In addition, TCF3 represses the expression of SHP1, a negative regulator of BCR signaling. Hence, TCF3 enhances BCR signaling by two different ways. Correspondingly, knockdown of TCF3 decreased PI3K activity in BL cell lines and was lethal. Moreover, HSP90 induces apoptosis in BL cells due to the disruption of tonic BCR signaling. HSP90 impairs SYK kinase which is required for the efficient activation of BCR complex [84].

An atypical form of BCR signaling is observed in follicular lymphoma caused by the presence of N-linked glycosylation acceptor sites in the V region induced by SHM [85]. Due to these, BCRs are modified by high-mannose oligosaccharides, which can interact with mannose-binding lectins present on the stromal cells in the tumor microenvironment, leading to cross-linking of the BCR and initiation of the BCR signaling [86, 87].

BCR-Independent Lymphomas

Despite the essential role of BCR expression and signaling in many B-cell lymphomas, there are a considerable number of exceptions. For example, inactivating IGH V-region gene mutations have been described in at least 10–20% of the cases of post-transplant lymphomas [87,88,90]. As these are EBV-positive lymphomas, it is assumed that expression of the EBV-encoded latent membrane protein 2A (LMP2A) can replace for the BCR expression and signaling. More controversial is the role of BCR function in primary mediastinal B-cell lymphomas. These lymphomas usually lack expression of a BCR, and the components of the BCR signaling cascade are downregulated. Nevertheless, these lymphomas are usually not associated with inactivating IGH V-region gene mutations [90,91,93].

The Role of Pathogens in Pediatric and Adolescent B-Cell Lymphomas

Some viruses clearly seem to contribute to the pathogenesis of pediatric and adolescent B-cell lymphomas. The best documented examples are two members of the ɣ herpes virus family, human herpes virus (HHV) 4, better known as EBV, and human herpes virus 8 (HHV8) [3, 94].

EBV has been shown to contribute to the pathogenesis of Burkitt lymphoma (BL) and post-transplant lymphoproliferative disorders (PTLD). Of BL, three epidemiologic variants are described: endemic, sporadic, and immunodeficiency-associated, with the frequency of EBV infection differing significantly between the variants. The endemic BL variant shows EBV infection in the tumor cells in nearly all of the cases [95]. In contrast, EBV infection of the tumor cells is only described in 25–40% of patients with immunodeficiency-associated BL [96, 97] and even less frequently in sporadic BL, where it can be detected in less than 20% of the cases. Importantly, EBV seronegativity at the time of transplantation is the most important risk factor for EBV-driven PTLD. About 60% of the PTLDs are EBV-positive. Nevertheless, EBV-negative PTLDs are also reported in around 20–40% of the cases and more common in adults.

With regard to EBV, three latency types specific to individual EBV-associated tumors have been described. Latency type I is associated with BL and shows restricted expression of EBV-encoded nuclear antigen 1 (EBNA1), the EBV-encoded small RNAs (EBERs), and BAMHI A rightward transcripts (BARTs) [97,98,100]. Latency type II infected cells typically express EBNA1, EBERs, BARTS, and the latent membrane proteins (LMP1, LMP2a, and LMP2B). This latency type is associated with Hodgkin lymphoma. LMP1 and LMP2a mimic an active CD40 receptor and BCR, respectively [101]. Both latent membrane proteins provide survival signals for B-cells in the GC. Latency type III, which is associated with PTLD and EBV-transformed lymphoblastoid cell lines (LCLs), expresses both transcripts and all the EBV latent proteins, containing the six nuclear antigens (EBNA1, EBNA2, EBNA3A, EBNA3B, EBNA3C, and EBNA-LP), and the three membrane proteins (LMP1, LMP2A, and LMP2B). EBNA2 is important to drive the proliferation of the transformed B-cells [3]. On the other hand, EBERs are expressed by all EBV-infected cells, which is used for the EBV detection by means of in situ hybridization for these transcripts.

HHV8 is detected in few DLBCL, not otherwise specified, and in primary effusion lymphoma (PEL). PEL is a very rare lymphoma mainly found in acquired immune deficiency syndrome (AIDS) patients. HHV8 establishes a latent infection in B-cells and encodes several homologues to human proteins including cytokines (interleukin-6, macrophage inflammatory proteins, interferon-regulatory factors) and regulatory genes (cyclin D, G-protein coupled receptor, etc.) [94, 102].

In addition to HHV4 and HHV8, a pathogenetic role of other viruses like hepatitis C virus (HCV) [103] and HHV6 is being discussed for lymphomagenesis [104]. Future studies using next-generation sequencing (NGS) will probably provide further insights into the role of pathogens in B-cell lymphomagenesis.

Besides viruses, also other pathogens have been linked to the development of B-cell lymphomas. In particular, subtypes of marginal zone lymphomas of MALT type have been associated with infections with H. pylori, C. psittaci, or B. burgdorferi. The best evidence of a pathogenic role for foreign antigens in these lymphomas derives from MALT lymphomas of the stomach [105]. The vast majority of MALT lymphoma patients are infected by H. pylori, where the antibiotic treatment targeted to the pathogen can cure these patients [106, 107]. Besides, some patients with DLBCL of the stomach infected by H. pylori can be also cured using antibiotic treatment [108]. Whether such pathogens also play a role in pediatric marginal zone lymphomas needs to be investigated.

On a more general level, various subtypes of lymphomas are assumed to be polymicrobial diseases. A clear link to malaria infections caused by P. falciparum has been established for endemic BL where an immune stimulatory and probably AID-mediated DNA damaging role of the parasite infection is discussed. In sporadic BL the localization of the tumors in the ileoceacal region, a cell of origin being a B-cell poised to IgA expression and an incidence curve with intriguing parallels to the IgA-expression and Peyer patch development might indicate a role of the primary immune response to the microbiome colonialization of the gut. In analogy, also B-cell lymphomas predominately presenting in the Waldeyer’s ring like IRF4-rearrangement positive LBCL might be triggered by pathogenic infection. Antigenic stimulation seems also to be involved in splenic marginal zone lymphoma (SMZL). Some patients with a SMZL with villous lymphocytes are infected with hepatitis C virus (HCV), and treatment against this virus using interferon-α (IFN-α) abolishes the lymphoma in around 75% of these patients [109]. In contrast, IFN-α has no effect in other patients with disease but without HCV infection. Thus, the overall role of pathogens besides direct oncogenic function but rather as promoting factor in B-cell lymphomas needs further investigation, particularly in the light of BCR signaling and primary immune responses like in children, young adults, and after transplantation.

The Role of the Microenvironment in the Pathogenesis of B-Cell Lymphomas

The interaction of tumor cells with cells in the tumoral microenvironment can also affect survival and proliferation of the malignant B-cells in various lymphomas. For example, follicular lymphoma cells proliferate in follicular structures associated with T-helper cells and follicular dendritic cells, resembling normal GC B-cells. In vitro studies have shown that follicular lymphoma cells, among others, receive stimulation via the CD40 receptor from the microenvironment. Expression and signaling via CD40 is a main survival signal also for normal GC B-cells [110, 111].

In addition, the macrophage (MP) infiltration has been described as pathogenetic factor in various lymphomas even linked to prognosis in several studies [112, 113]. Macrophages are heterogeneous, multifunctional, myeloid-derived leukocytes that are part of the innate immune system. Tumor-associated MPs (TAMs) are MPs with specific M2 phenotype that play a central role in the pathophysiology of tumors [114].

Related to B- and T-cell neoplasms, TAMs are involved in tumor progression, often associated with poor prognosis owing to the secretion of chemokines and cytokines. Different active proteases stimulate tumor growth, angiogenesis, metastasis, and immunosuppression [115]. In DLBCL, the implication of MPs has been related to the ability of DLBCL cells to escape the immune surveillance of tumor-specific cytotoxic T-cells recruiting M2 TAMs that highly express immune checkpoint molecules, such as PD-L1 and PD-L2, on their surfaces. These interact with PD-1 receptors expressed on intratumoral T-cells and provide inhibitory signals. This could be an explanation for the effective therapy with anti-PD-/PD-L1 in some cases of DLBCL [116].

Epigenetic Alterations Leading to Tumorigenesis

In addition to the genomic alterations contributing to the initiation and progression of B-cell lymphoma and leukemia, the importance of epigenetic alterations in the pathogenesis of these neoplasias has been acknowledged over the last years. The best studied epigenetic marks in B-cell neoplasias are DNA methylation and histone modifications, which are involved in basic biological mechanisms including regulation of gene expression, replication, and DNA repair (as reviewed in [117, 118]).

Among the histone modifiers most frequently altered in B-cell lymphomas are EZH2, CREBBP, and EP300 as well as KMT2D. The methyltransferase EZH2 is part of the Polycomb Repression Complex 2 (PRC2), which methylates histone 3 lysine 27 (H3K27) a marker for repressed chromatin (heterochromatin). In B-cells, EZH2 is expressed during early B-cell development where it plays a role in VDJ recombination [119]. In naïve B-cells, the expression of EZH2 is downregulated. During the GC reaction, EZH2 is upregulated and establishes repressive H3K27 marks at promoters of genes which are involved in differentiation and cell cycle regulation [120, 121]. EZH2 mutations are frequently detected in DLBCL as well as FL, with the most recurrent alteration affecting a tyrosine in the functional SET domain of the protein (Tyr641) being mutated in 21.7% of DLBCL and 7.2% of FL [122]. This mutation is predicted to affect the enzymatic activity by shifting the efficiency to methylate H3K27 to trimethylation instead of mono- or demethylation [123, 124].

The Mixed Lineage Leukemia (MLL) gene family belongs also to the histone methyltransferases. In contrast to the PRC2 complex, members of this family methylate histone 3 lysine 4 (H3K4) which is a marker for transcriptional activation [125]. Diverse alterations affecting these modifiers have been reported. Accordingly, KMT2A (formerly known as MLL1 ) is translocated in about 70% of infant leukemias [126]. The catalytic SET domain conducting the methyltransferase activity is frequently lost due to the translocation. Nevertheless, it is believed that the MLL1 alterations are associated with aberrant histone methylation and hence, with the overexpression of target genes. Truncating and frameshift mutations of KMT2D  have been reported to occur in up to ~90% of FL [66, 127] and ~30% of DLBCL [128, 129]. These mutations affecting the SET domain lead to a loss of function and, hence, to a deficiency of H3K4 methylation, suggesting a tumor-suppressive function for KMT2D [130].

In addition to histone methylation, frequent alterations in B-cell neoplasia affect histone acetylation via changes in histone acetyltransferases (HAT). Histone acetylation mediates an open chromatin structure allowing bromodomain proteins to be recruited which induce the transcriptional activation. The most recurrently altered HAT in malignant B-cell lymphomas are CBP (encoded by CREBBP) and p300 (encoded by EP300) which are described to have tumor-suppressive functions in B-cell lymphoma [67, 131]. About ~30% of DLBCL harbor alterations leading to a loss of CREBBP HAT domain function, with GCB-DLBCL being more frequently affected than ABC-DLBCL [67]. The HAT domain is also inactivated in ~30% of FL [67] and 18% of relapsed pediatric B-ALL [131]. The inactivation of CBP leads to an expansion of the GC B-cell compartment, downregulates MHC class II expression, and promotes tumor cell growth [132]. Mutations disrupting the HAT domain of p300 were reported in ~10% of DLBCL [67, 133] and FL [67]. P300 inactivation confers resistance against BCL6 inhibitors [133].

Apart from the changes affecting the histone modification, changes in DNA methylation take place during B-cell development and contribute to the physiological processes linked to the differentiation. Hence, each developmental stage of a B-cell is composed of an unique epigenetic pattern [4, 134]. B-cell neoplasias maintain a certain degree of similarity to their assumed normal B-cell counterpart; thus, the DNA methylation pattern can be used for the determination of the cell of origin and for classification purposes [134, 135]. On the other hand, the neoplastic B-cells are characterized by a number of DNA methylation changes which in part interact with genomic and transcriptional changes in the deregulation of key transforming processes [134,135,136,137,138,140].

The establishment of DNA methylation patterns involves the DNA methyltransferases (DNMT) DNMT1, DNMT3A, and DNMT3B. DNMT1 has been shown to be significantly upregulated in GC B-cells suggesting a role in GC reaction and differentiation [141]. Indeed, experiments with Dnmt1 hypomorphic mice have shown that the GC formation upon immunization is impaired [141]. In line, DNMT1 as well as DNMT3B overexpression has been described in 69% and 86% of BL, respectively [142].

Pathogenetic Hallmarks of Common Subtypes of B-Cell Lymphomas in Children and Adolescents

Burkitt Lymphoma and Burkitt-Like Lymphoma with 11q Aberration

The hallmark genetic aberration as well as the assumed primary event in all three epidemiologic subtypes of BL is t(8;14)(q24;q32) or its variants t(2;8)(p12;q24) and t(8;22)(q24;q11). All these changes lead to deregulation of the MYC oncogene by its juxtaposition next to one of the enhancer elements in the IGH (14q32), IGK (2p12), or IGL (22q11) locus (Table 4.2).

Table 4.2 Overview of the frequent chromosomal aberrations in pathogenetic hallmark of B-cell lymphomas in children and adolescents
Full size table

BL is characterized by a low genomic complexity. Cytogenetically, an IG-MYC translocation is detected as sole abnormality in 40% of cases. The most frequent secondary alterations are structural aberrations involving chromosome 1 (>30% of the patients), especially of the long arm, and often resulting in a partial trisomy 1q. Aberrations affecting the chromosomal region 13q31, mainly involving the mir-17-92 miRNA cluster [68], have been reported in 15% of the cases [143]. Moreover, gains of chromosomes 7 and 12 and deletions in 6q and 17p are common. Besides the secondary chromosomal imbalances, recent genomic sequencing studies have identified recurrent somatic mutations in MYC, ID3, TCF3, CCND3, SMARCA4, TP53, FBXO11, ARID1A, DDX3X in both sporadic and endemic BL [64, 135, 144, 145] (Table 4.2).

The existence of BLs without an IG-MYC translocation has been subject to controversial discussion. A provisional entity of MYC-negative “Burkitt-like lymphoma with 11q aberration” has been recently included in the new WHO lymphoma classification. Those cases resemble BL based on gene expression profile and pathological characteristics but importantly lack a MYC translocation. Instead, 11q aberrations with proximal gains and telomeric losses are typical [146] (Table 4.2).

Diffuse Large B-Cell Lymphoma (Including ALK+ Large B-Cell Lymphoma)

DLBCL is a heterogeneous group of diseases with varying morphologic, immunophenotypic, and molecular features. Several of these features change with age [147]. DLBCL in children and adolescents, in contrast to those occurring at older age, are enriched for GCB-type cases and depleted for BCL2 and BCL6 translocations. The mutational landscape of DLBCL in children and young adults warrants further investigation.

A special subgroup of large B-cell lymphoma enriched in young patients is ALK+ large B-cell lymphoma (ALK+ LBCL). It is a rare and an aggressive neoplasm, accounting for <1% of DLBCLs. The key player of this lymphoma is the overexpression of ALK protein, as a result of fusion protein generated by a translocation of the ALK gene on chromosome 2, as described above. Typically these translocations are associated with complex karyotypes. The STAT3 pathway is constitutively activated in ALK+ LBCLs and the tumors respond to ALK inhibitors (Table 4.2).

Primary Mediastinal Large B-Cell Lymphoma

Primary mediastinal large B-cell lymphoma (PMBL) is a mature aggressive large B-cell lymphoma (LBCL). It affects mainly young adults with a predominance of females and a median age at diagnosis of 35 years. PMBL display a specific gene expression profile which is different from GCB or ABC DLBCLs but shows similarities to Hodgkin lymphoma. Chromosomal aberrations affecting the BCL6, MYC, and BCL2 loci are absent or rarely detected. Nevertheless, translocations involving CIITA locus in 16p13.3 have been described in more than 50% of cases. The partner genes of translocations are in the majority of cases PDL1 or PDL2 but these are not the only partners described. In addition, gains in the region of 9p, containing the genes JAK2, PDL1, PDL2, and gains in 2p16.1, including the genes REL and BCL11A, have been described as recurrent genetic events in PMBL [148, 149] (Table 4.2).

PMBL are characterized by a constitutively activated JAK/STAT signaling pathway. Furthermore, mutations in STAT6 and PTPN1, a negative regulator of JAK/STAT signaling, are detected in 72% and 25% of PMBL cases, respectively. Additional mutations affect ITPKB, MFHAS1, and XPO1 [148] (Table 4.2).

Large B-Cell Lymphoma with IRF4 Rearrangement

Large B-cell lymphoma (LBCL) with IRF4 rearrangement is an uncommon lymphoma, comprising for 0.05% of diffuse LBCLs. This neoplasm shows a decreasing incidence in older age groups.

Besides expression of germinal center markers, the immunophenotype is characterized by the strong expression of IRF4/MUM1. Moreover, BCL6 is co-expressed, while PRDM1 is frequently negative [148].

The key genomic event in this lymphoma is an often cryptic rearrangement of IRF4 with an IG locus, BCL6 rearrangement have also been detected in LBCL with IRF4 breaks, whereas MYC and BCL2 breaks were absent in the reported cases. In addition to IRF4 rearrangements, the genomic profile of this entity shows a complex pattern of genetic changes including TP53 deletions [148, 150] (Table 4.2). Nevertheless, patients have favorable outcome after treatment.

Pediatric-Type Follicular Lymphoma

Pediatric-type follicular lymphoma (PTFL) is an infrequent nodal follicular lymphoma (FL) that appears in children and young adults, but can also occur in the older population. This neoplasm is characterized by the lack of genomic rearrangements involving the BCL2, BCL6, or IRF4 locus. Moreover, the frequent mutations affecting KMT2D, CREBBP, and EZH2 described in adult FL, are absent in PTFL. Instead, deletions in the1p36 chromosomal region or mutations involving TNFRSF14 are the most common genetic aberrations in PTFL. Moreover, mutations in MAP2K1 have been reported in 40–50% of the cases [148] (Table 4.2).

Conclusions and Outlook

As detailed above and exemplified for B-cell lymphomas common in young patients, the pathogenesis of B-cell lymphomas is based on multifactorial grounds. The complexity, interdependency, and timely order of pathogenetic processes ultimately leading to clinically overt B-cell lymphomas are yet by far not completely understood. Even the distinction between driver and passenger events is still challenging, though some key lymphoma-initiating events like the Burkitt translocation t(8;14) have been known since more than 40 years. The reception of the above detailed mechanisms involved in the pathogenesis of B-cell lymphomas has to take into account that most current models of B-cell lymphomagenesis rely on simple, mostly mono-dimensional data and assumptions. Moreover, a linear and directional evolution of a B-cell lymphoma from a non-neoplastic precursor is usually assumed. Nevertheless, in fact to capture the overall complexity of B-cell lymphomagenesis, each neoplastic cell of a probably heterogenic tumor would have to be mapped at a single point in a space of n-dimensions, with several of the factors outlined above but also all different OMIC layers being separate dimensions. Moreover, the age of the patient and time since tumor initiation have to be considered as dimensions in this space. It needs to be clearly stated, that most of the concepts outlined above do not take such a “spacial” approach. Moreover, in many instances the experimental or observational procedures providing the data for the models likely modify features which appear different in the native host of the tumor, i.e., the patient. All this has to be taken into account if such observations are translated into clinics. Clearly, no diagnostic test yet captures the biologic complexity in its entirety (and if so probably best morphology which in essence provides a birds-eye view on the tumoral processes). Indeed, given the number of deregulated processes, altered genes, and changed signals, it seems rather surprising that the tumor takes advantage rather than disadvantage from those. Likely, this is due to the fact that many processes show a high grade of redundancy and alternatives. Thus, it will remain a challenge for the future to identify in the n-dimensional space of pathogenesis those events, which are indeed key to the pathogenesis and, thus, can be subject of novel treatment strategies.