Pathobiology of mantle cell lymphoma

INTRODUCTION — Mantle cell lymphoma (MCL) is a mature B cell non-Hodgkin lymphoma (NHL) that accounts for approximately 7 percent of adult NHLs in the United States and Europe. While it is often discussed together with the indolent forms of NHL, its behavior is more often that of an aggressive lymphoma. (See "Classification of hematopoietic neoplasms".)

Historically, MCL has been referred to as intermediate lymphocytic lymphoma, mantle zone lymphoma, centrocytic lymphoma, and lymphocytic lymphoma of intermediate differentiation [1,2].

The pathobiology of MCL is discussed in this topic.

Epidemiology, clinical presentation, diagnosis, and treatment of MCL are presented separately. (See "Mantle cell lymphoma: Epidemiology, pathobiology, clinical manifestations, diagnosis, and prognosis" and "Mantle cell lymphoma: Initial management".)

CELL OF ORIGIN — Most cases of MCL are postulated to derive from naïve pre-germinal center B cells, while a subset of MCL may originate from post-germinal center marginal zone or peripheral blood memory B cells. Those that appear to originate from post-germinal center B cells have unique clinicopathologic features, as they often involve marrow and peripheral blood and spare lymph nodes and are associated with an indolent clinical course.

The pre-germinal center ancestry of most MCLs is principally supported by the absence of somatic mutations in the immunoglobulin heavy chain variable region (IgVH) genes, which serve as a marker of germinal center transit [3]. However, several studies have found somatic hypermutation in approximately 20 to 30 percent of MCL cases [3-5]. IGHV3-21 and IGHV4-34 usage was significantly overrepresented in MCLs in these studies. The V3-21 expressing cases were highly associated with expression of the same light chain, IGLV3-19, suggesting that these tumor cells are stimulated by an autoantigen recognized by the Ig formed by pairing of these heavy and light chains [3].

PATHOGENESIS

Overview — The major oncogenic driver in the overwhelming majority of MCL is overexpression of the CCND1 gene (also called BCL1 or PRAD1) located at 11q13 that encodes the cyclin D1 protein [6,7]. Cyclin D1 is not expressed in normal B lymphocytes. Expression of cyclin D1 promotes cell cycle transition from G1 to S phase and proliferation. Cyclin D1 expression is not sufficient for the transformation of normal B lymphocytes, as transgenic mice that overexpress cyclin D1 alone do not develop lymphoma [8].

In addition to cyclin D1 expression, MCL tumors demonstrate a decreased response to DNA damage and enhanced cell survival (impaired apoptosis). Alterations in these three pathways (ie, cell cycle, DNA repair, and apoptosis) are described in the following sections (figure 1). Abnormalities in other pathways are also likely to contribute to MCL pathogenesis. Several groups have used unbiased genome-wide approaches to identify genes that are recurrently mutated or dysregulated in MCL besides CCND1. Three independent analyses showed that, in a subset of cases of MCL, the gene that encodes the NOTCH1 receptor is hypomethylated or has a gain-of-function mutation [9,10], suggesting that NOTCH signaling has an oncogenic role in some cases of MCL. One deep sequencing study identified 25 genes that are recurrently mutated in MCL, including genes involved in DNA damage responses (ATM, TP53), apoptosis (BIRC3, TLR2), and chromatin modification (WHSC1, MLL2, MEF2B), as well as the gene encoding the NOTCH2 receptor [11]. A second deep sequencing study of MCL exomes coupled with RNA sequencing from 134 patients, some with longitudinal samples, identified 29 recurrently mutated genes, including those identified in prior studies and novel genes such as LRP1B, PCLO, RYR2, PCDH10, OBSCN, TACC2, FAT3, LRP2, SVEP1, ZFHX4, MPDZ, DCDC1, IKBKB, and ARID1A [12]. Abnormal expression of RNA-binding proteins (eg, HNRNPH1) and associated perturbations of mRNA processing may also contribute to the pathobiology of MCL [13].

Cell cycle progression — All cases of MCL demonstrate increased cell division and replication. In the vast majority of cases, cyclin D1 overexpression drives increased proliferation. A small number of cases that resemble MCL but do not express cyclin D1 appear to result from deregulation of cyclin D2 or cyclin D3. NOTCH signaling in MCL appears to directly upregulate transcription of MYC through B-cell-specific enhancers, which also stimulates cell growth and proliferation [14]. An additional potential mechanism for increased MYC activity is stabilization of MYC protein through signals that require MALT1, a component of the CARD11-BCL10-MALT1 complex that links B cell receptor (BCR) signaling to NF-kB [15].

Cyclin D1 expression — Cyclin D1 is not expressed in normal B lymphocytes and its expression appears to contribute to the development of a large majority of MCL cases. Cyclin D1 is an activating regulatory subunit for cyclin dependent kinase (CDK) 4 and 6 [16]. Activated cyclin D-CDK complexes phosphorylate and inactivate retinoblastoma protein (Rb), a potent inhibitor of G1 to S phase progression. This complex also binds cyclin dependent kinase inhibitor 1B (CDKI1B, also known as Kip1 and p27) [17]. The binding of cyclin D1/CDK4 complexes to CDKI1B stabilizes and enhances the activity of cyclin D1/CDK4 while also inhibiting CDKI1B, which when free acts as an inhibitor of other CDK complexes that are required at slightly later phases of the cell cycle.

Aberrant expression of cyclin D1 may occur through at least one of the following mechanisms:

●The (11;14)(q13;q32) translocation juxtaposes the DNA coding sequence for cyclin D1 at 11q13 with the immunoglobulin heavy chain joining region at 14q32 [18,19]. This translocation appears to be mediated by recombinase-activating genes (RAG) and activation-induced cytidine deaminase (AID) [20]. While found in the majority of MCL cases, t(11;14) is not specific for MCL. While the nontranslocated CCND1 allele is usually located far from the nucleolus, the translocated allele is preferentially localized to the perinucleolar area, an area with an increased density of transcriptional activators [21]. (See "Mantle cell lymphoma: Epidemiology, pathobiology, clinical manifestations, diagnosis, and prognosis", section on 'Genetic features'.)

●Point mutations and/or genomic deletions that affect the 3' untranslated region of CCND1 result in the expression of truncated mRNA [22]. These truncations stabilize the CCND1 RNA transcript, resulting in increased CCND1 mRNA and increased protein levels [23,24]. An increase in truncated relative to full-length cyclin D1 mRNAs is associated with poor clinical outcomes [22-24]. (See "Mantle cell lymphoma: Epidemiology, pathobiology, clinical manifestations, diagnosis, and prognosis", section on 'Genetic features'.)

●Up to 50 percent of patients with MCL demonstrate activation of the Wnt signaling pathway, which stabilizes beta-catenin, allowing it to translocate into the nucleus and form a transcription complex that upregulates the expression of multiple target genes, including CCND1 [25].

The pathogenic role of cyclin D1 activation in MCL is suggested by the ability of cyclin D1 overexpression to transform cells in vitro and contribute to B cell lymphomagenesis in transgenic mice [1,2]. In addition, gene expression profile analyses have confirmed the central role of cyclin D1 in MCL pathogenesis. (See 'Heterogeneity of MCL' below.)

Cyclin D1 negative cases — A small number of lymphomas that morphologically resemble MCL and express the characteristic gene expression signature are negative for cyclin D1 [26]. Of six cases identified in one study, two overexpressed CCND2 and four overexpressed CCND3, typically without chromosomal rearrangements [27]. In another study, 22 of 40 cases of cyclin D1 negative MCL displayed a CCND2 gene rearrangement [28]. These tumors expressed the transcription factor SOX11 and had an otherwise typical morphology and immunophenotype (CD5+, CD10–, CD23–) for MCL. Cyclins D1, D2, and D3 are highly homologous and functionally interchangeable, and MCLs with cyclin D1 or cyclin D2 dysregulation appear to be very similar in terms of their presentation and clinical course [28]. Recognition that cyclin D1 negative MCL usually expresses SOX11, which is only rarely expressed in other lymphoid neoplasms, has provided a generally reliable means to recognize these variants [29]. (See "Mantle cell lymphoma: Epidemiology, pathobiology, clinical manifestations, diagnosis, and prognosis", section on 'Clinical features'.)

Other mechanisms — In addition to the overexpression of cyclin D1, MCLs often have other acquired alterations that contribute to increased proliferation, including:

●Decreased CDKI1B – The cyclin E/CDK2 complex is similar to the cyclin D1-CDK complexes in that it phosphorylates and inactivates the retinoblastoma protein that normally impairs cell cycle progression. The cyclin E/CDK2 complex is normally inhibited by cyclin-dependent kinase inhibitor 1B (CDKI1B, also known as Kip1 and p27), levels of which are decreased in MCL [30]. CDKI1B is typically expressed at lower levels in highly proliferative subtypes of non-Hodgkin lymphoma. In MCL, there is a wide range of CDKI1B expression with greater expression in more highly proliferative tumors [17]. This is likely because the CDKI1B is sequestered in complexes containing cyclin D1, which blocks its cell cycle inhibitory function.

●Deletions of INK4a/ARF – The INK4a/ARF locus on 9p21 encodes the tumor suppressors p14 and p16 that, like p27, inhibit cell cycle progression by binding to cyclin/CDK complexes [26,31]. Deletion of this locus occurs in 14 to 20 percent of cases and has been associated with more aggressive (blastoid) histology.

●Abnormalities in TP53 – TP53 is involved in the regulation of the cell cycle primarily at the G1-S transition and has an important role in DNA repair. A minority of MCLs has mutations or deletions involving the TP53 gene, an event that has been consistently associated with a poor prognosis [12,32,33]. (See 'TP53' below.)

●Expression of SOX11 – SOX11 is a transcription factor involved in central nervous system development. While studies of SOX11 expression in MCL have produced conflicting results, overexpression of SOX11 appears to confer a worse prognosis [28,34,35]. In addition, cases of cyclin D1 negative, cyclin D2 negative, SOX11-positive MCL with an aggressive clinical course have been described [28]. In SOX11-positive MCL, SOX11 appears to upregulate the expression of PAX5, which in turn impairs the expression of PRDM1/BLIMP1, resulting in a block in plasmacytic differentiation [36]. SOX11 also appears to regulate the expression of platelet-derived growth factor alpha (PDGFA), upregulation of which may contribute to tumor-induced angiogenesis [37].

Impaired DNA damage response — At least a subset of MCL cases appear to have an impaired response to DNA damage that produces increased chromosomal instability. This impaired response is largely due to defects in the TP53 and ataxia-telangiectasia mutated (ATM) genes. In addition, the cell cycle checkpoint kinases (CHK) 1 and 2 are downregulated in some MCL cases [30].

TP53 — Mutations of TP53 are seen in 10 to 25 percent of MCL cases and are associated with more aggressive histology and shorter survival [12,31,38-40]. The TP53 gene, located at 17p13.1, encodes a tumor suppressor protein that binds DNA and activates transcription of growth inhibitory genes. Normally, activated TP53 induces a variety of growth-limiting responses, including cell cycle arrest (thereby facilitating DNA repair), apoptosis, senescence, and differentiation.

Downregulation of TP53 expression or expression of mutant forms of p53 protein result in a loss of the normal growth-limiting activities of this gene. Interestingly, some studies suggest that mutation of TP53 is more strongly associated with poor outcome than its deletion, raising the possibility that gain of function mutations are involved [40]. A second study suggests that the impact of TP53 mutations on prognosis may be modified by the presence of other driver mutations, an observation that requires additional study [12].

ATM gene — Deletion or point mutations in the ataxia telangiectasia mutated (ATM) tumor suppressor gene (11q22-q23) are seen in approximately one-third to one-half of MCL cases [26,38,41]. ATM is involved in the detection of DNA damage and plays an important role in the regulation of cell cycle progression. In one study, ATM deletions were not correlated with tumor proliferation, cyclin D1 levels, or INK4a mutations [26]. ATM mutations and TP53 mutations are seen together in only 10 percent of cases [38], as might be expected, given that ATM protein functions upstream of p53. (See "Ataxia-telangiectasia", section on 'Genetics and pathogenesis'.)

Enhanced cell survival — Microarray studies have suggested that MCL cases display disturbances of pathways associated with apoptosis [42]. Specifically, MCL cells appear to avoid programmed cell death (apoptosis) by the expression of B cell lymphoma 2 (BCL2), upregulation of the PI-3 kinase/AKT prosurvival signaling pathway, activation of nuclear factor-kB (NF-kB), and mutations in TP53.

●B cell leukemia/lymphoma 2 (BCL2) is an oncogene that blocks programmed cell death, leading to prolonged cell survival. BCL2 overexpression is common in follicular lymphoma and chronic lymphocytic leukemia (CLL), but is also seen in other non-Hodgkin lymphomas, including more than half of MCL [43]. (See "Pathobiology of follicular lymphoma", section on 'Translocations involving BCL-2'.)

●The PI-3 kinase (PI3K)/AKT signaling pathway promotes cell growth, cell survival and tumorigenesis by inactivating several proteins including CDKI1B (p27) BAD (a member of the BCL2 family) [44], and FOXO1, a transcription factor that regulates the expression of pro-apoptotic genes [45]. The PI3K/AKT pathway also increases the levels of nuclear factor-kappa B and mammalian target of rapamycin (mTOR), resulting in a decrease in functional TP53 and subsequent cell survival. MCLs, especially those with blastoid morphology, have active PI3K/AKT signaling [46]. Of note, as with other relatively indolent B cell tumors such as CLL, Bruton tyrosine kinase (BTK) inhibitors such as ibrutinib have shown excellent, durable single agent activity in MCL [47]. BTK functions downstream of the membrane-bound B cell receptor (BCR) and links constitutive BCR signaling to PI3K/AKT activation and nuclear factor-kappa B activation, consistent with the view that these pathways are critical for MCL growth and survival. To date, PI3K inhibitors have had lower response rates and poor durability [48]. (See 'Other mechanisms' above.)

●Nuclear factor-kappa B (NF-kB) polypeptides comprise a family of transcription factors that have important roles in regulation of normal immune function and are often dysregulated in lymphoid malignancies. Early studies suggested that NF-kB is constitutively activated in at least a subset of MCL and that ex vivo treatment with an NF-kB inhibitor results in cell cycle arrest and apoptosis [49]. More recent work using pharmacological profiling identified a subset of MCL cell lines that are insensitive to BCR inhibitors, such as ibrutinib [50]. While ibrutinib-sensitive MCL models exhibited activation of NF-kB through the so-called classical pathway, in insensitive cell lines NF-kB was activated through an alternative pathway involving the protein kinase NIK (also called mitogen-activated protein 3 kinase 14 or MAP3K14) both in vitro and in vivo. RNA sequencing revealed genetic lesions in alternative NF-kB pathway signaling components in ibrutinib-insensitive cell lines, and sequencing of 165 primary MCLs identified recurrent mutations in the alternative NF-kB pathway components TRAF2 or BIRC3 in 15 percent of tumors. This study suggests that NIK may be a therapeutic target in MCLs that are refractory to BCR pathway inhibitors.

●TP53 can induce apoptotic cell death. As mentioned earlier, a minority of patients with MCL will demonstrate mutations or deletions involving the TP53 gene. (See 'TP53' above.)

HETEROGENEITY OF MCL — The diagnostic category of "mantle cell lymphoma" is heterogeneous in terms of morphology, genetics, and biologic behavior. As an example, indolent MCL often presents as non-nodal leukemic disease with hypermutated IgVH and a noncomplex karyotype [51]. The role that gene expression profiling and microRNA expression profiling play in the diagnosis and treatment of MCL is a matter of ongoing research [9,26,51-53].

Gene expression has been studied in MCL using microarray technology. As an example, a gene expression profile primarily composed of genes required for cell proliferation was predictive of aggressive disease and worse survival [26]. Higher expression of CCND1 is also associated with worse survival and correlated with a proliferation signature. Two large subsequent studies confirmed the clinical prognostic importance of these cell proliferation rates [54,55]. Another study compared the gene expression profiles of indolent MCL with conventional MCL and identified a 13-gene signature that was highly expressed in conventional MCL, but underexpressed in indolent MCL [51]. In both the test and validation cohorts, lack of expression of SOX11 was associated with an indolent course. MCLs with activating NOTCH1 mutations appear to have a more aggressive course [10,12]. (See "Mantle cell lymphoma: Epidemiology, pathobiology, clinical manifestations, diagnosis, and prognosis", section on 'Prognostic features'.)

A study of whole exome and RNA sequencing of 134 cases of MCL identified four distinct MCL groups, termed C1-C4 [12]. The C1 group was enriched for "leukemic," non-nodal cases of MCL and was associated with low SOX11 expression, mutant CCND1, mutant TP53, and amp(11q13). Unexpectedly, within this group, TP53 mutation was not associated with worse outcome. The C2 group was strongly associated with del(11q) (the region that contains ATM) and co-occurring ATM mutation and upregulation of genes involved in DNA replication, DNA repair, hyperproliferation, TNF-alpha signaling, and IFN-alpha and IFN-gamma response. The C3 group was enriched for NOTCH1, WHSC1, KMT2D, and SP140 mutations, amp(13q), and del(6q). The C4 group was enriched for del(17p), del(13q), and del(9p); mutations of TP53 and TRAF2; and gene signatures of associated with MYC pathway activation and hyperproliferation. The C4 group was associated with the highest incidence of blastoid or pleomorphic MCL morphology and the worst clinical outcomes.

SUMMARY

●Mantle cell lymphoma (MCL) is a mature B cell non-Hodgkin lymphoma (NHL), which accounts for approximately 7 percent of adult NHLs in the United States and Europe. While it is often discussed together with the indolent forms of NHL, its behavior is more often that of an aggressive lymphoma.

●Most MCL cases are postulated to derive from naïve pre-germinal center B cells, while a subset of MCL may originate from marginal zone or peripheral blood memory B cells. (See 'Cell of origin' above.)

●All cases of MCL demonstrate increased cell division and replication. In the vast majority of cases, expression of the CCND1 gene, located at 11q13, drives cell cycle progression from G1 to S phase. A minority of cases does not demonstrate cyclin D1, but often overexpresses other cell cycle mediators, particularly cyclin D2 and cyclin D3 (figure 1). (See 'Cell cycle progression' above.)

●MCL, including cyclin D1 negative cases, is frequently associated with altered expression of SOX11, a transcription factor that may contribute to a block in differentiation in MCL cells.

●In addition to cyclin D1 expression, MCL tumors demonstrate a decreased response to DNA damage and enhanced cell survival (impaired apoptosis). (See 'Impaired DNA damage response' above and 'Enhanced cell survival' above.)

●Mutation and/or deletion TP53 is generally associated with poor response to therapy and worse outcome, but this may not hold for "leukemic," non-nodal MCL originating from post-germinal center B cells.