Adult T-cell leukemia/lymphoma (ATLL) is a mature T-cell neoplasm caused by the human T-lymphotropic virus 1 (HTLV-1, also called T-cell leukemia virus). This tumor is endemic in those parts of the world where HTLV-1 is prevalent in the population, including Japan, the Caribbean basin and parts of central Africa (Swerdlow et al. 2008). HTLV-1 is the first human retrovirus shown to cause malignant transformation, which is mediated by the viral protein Tax. One of its effects on cellular processes is to impair DNA repair mechanisms by repressing the expression of DNA polymerase-β, an enzyme involved in base excision repair, and by repressing nucleotide excision repair, involved in repairing UV irradiation and DNA replication-induced damage. Tax can also directly inactivate the TP53 protein (Kannian and Green 2010; Yasunaga and Matsuoka 2011). HTLV-1-induced destabilization of the genome may explain the complex karyotypes (3 or more cytogenetic abnormalities) that are typically observed in ATLL.

Chromosome abnormalities have been found by conventional cytogenetic analysis in almost all ATLL samples examined; the changes are typically complex and variable, with abnormalities affecting any chromosome pair. Nonetheless, several recurring abnormalities have been described. Translocations are identified in ~10 % of ATLL cases. The T-cell receptor α/Δ gene (TCRA and TCRD) locus at 14q11.2 is frequently rearranged, often as a t(14;14)(q11.2;q32), inv(14)(q11.2q32), or del(14)(q11.2q13) or as a rearrangement of 14q11.2 with one of several other chromosome arms, including Xq, 1p, 1q, 3p, 3q, 8q, 10p, 11p, 12q, and 18p (Kamada et al. 1992). The numerical abnormalities most frequently noted by conventional cytogenetic analysis comprised −X, −Y, −13, +X, +3, +5, and +7, whereas frequently observed segmental aneuploidies included deletions of 6q, 10p, 3q, 5q, 9q, 13q, 1p, or 7p (Schlegelberger et al. 1994a).

Genome profiles of ATLL have been extensively studied by conventional CGH. The most frequently observed imbalances were losses at 6q and 13q and gains at 7q and 3p. CGH analysis also showed abnormalities at numerous other chromosomal regions, including 1p, 1q, 3q, 5p, 5q, 9q, 10p, 10q, 11q, 12q, 18q, and Y. Comparison of genome profiles detected by CGH revealed differences between the acute and lymphoma types of ATLL, with the lymphoma type more frequently showing gains on 1q, 2p, 4q, 7p, and 7q and losses from 10p, 13q, 16q, and 18p and the acute type manifesting gains of 3p. Recurrent high-level amplifications were found at 1p36, 6p25, 7p22, 7q, and 14q32 in the lymphoma type, with CARD11 identified as a candidate oncogene in 7p22 (Tsukasaki et al. 2001; Oshiro et al. 2006).

Specific CGH profiles correlate with distinct clinical features. Aberrations of 1p, 1q, 1q10–21, 10p, 10p13, 12q, 14q, and 14q32 are associated with features such as hepatosplenomegaly, elevated LDH, and an unusual immunophenotype, which are all indicators of clinical severity in ATLL. Multiple changes; abnormalities of 1p, 1q22, 1q10–21, 2q, 3q, 3q10–12, 3q21, 14q, 14q32, and 17q; and partial losses from chromosome arms 2q, 9p, 14p, and 17q are correlated with shorter survival (Tsukasaki et al. 2001).

Nasal NK-cell lymphoma is characterized by complex karyotypes, which do not appear to be disease specific. Wong et al. (1997) reported a common deletion at 6q21–q25 in three of seven cases studied by conventional cytogenetic analysis, suggesting that this may be a nonrandom chromosomal aberration in this disease. Other recurring abnormalities detected by metaphase analysis include +X, i(1q), i(7q), +8, del(13q), del(17p), i(17q), and rearrangements of 11q23 (Wong et al. 1999). Conventional CGH studies identified del(6q), del(13q), del(17p), del(1p), del(12q), and partial gain of Xp, 2p, or 10q as recurrent abnormalities (Siu et al. 1999). Genome-wide array-based CGH studies identified a number of recurrent imbalances in nasal-type extranodal NK-/T-cell lymphoma, including gain of 2q and loss of 6q16–q27, 11q22–q23, 5p14, 5q34, 1p36, 2p16, 4q12, and 4q31 (Nakashima et al. 2005).

Enteropathy-associated lymphoma (ETL) is an intestinal tumor of intraepithelial T lymphocytes usually presenting as a tumor composed of large lymphoid cells that show varying degrees of transformation (Swerdlow et al. 2008). Genetic studies of ETL are limited, but most ETL cases (58–70 %) are characterized by frequent complex gains of the 9q31.3-qter chromosome region or by deletions of 16q12.1 (Deleeuw et al. 2007; Zettl et al. 2002). The affected 9q region harbors several candidate genes, including ABL1 and NOTCH1, which are preferentially amplified in ETL (Cejkova et al. 2005). The 9q33–34 imbalances, however, are not specific for ETL, being found in almost 20 % of PTCL-NOS (Zettl et al. 2004). Other recurrent chromosomal aberrations include a partial trisomy of 1q22–q44 and gains of 5q reported in the classical form of ETL (Verkarre et al. 2003; Zettl et al. 2004) and amplifications of 8q24.2/MYC which are recurrently observed in the monomorphic variant (Zettl et al. 2002; Deleeuw et al. 2007). Furthermore, loss of heterozygosity at 9p21 is recurrent in classical ETL, and one or more genes in this region have been postulated to be involved in the pathogenesis of ETL (Obermann et al. 2004).

Hepatosplenic T-cell lymphoma (HSTCL) is a clinically aggressive subtype of PTCL, and the affected patients have a dismal outcome. This lymphoma accounts for 1.4 % of all mature T-cell neoplasms and represents <1 % of all non-Hodgkin lymphomas. The vast majority of HSTCLs express the γδ T-cell receptor (TCR) (HSγδTCL), but rare cases with the αβ TCR phenotype (HSαβTCL) have been also described (Lai et al. 2000; Suarez et al. 2000; Macon et al. 2001; Rashidi et al. 2012).

HSγδTCL is characterized by an isochromosome 7q [i(7)(q10)] (Fig. 2.3a), which has been identified by conventional cytogenetic and/or FISH analysis in almost all cases analyzed (Wang et al. 1995; Cooke et al. 1996; Jonveaux et al. 1996; Alonsozana et al. 1997). The i(7)(q10) can appear as the sole karyotypic abnormality, what suggests its primary and critical role in the development of HSγδTCL. The molecular consequences of i(7)(q10) are largely unknown, however, gene dosage effect resulting from the associated loss of 7p and gain of 7q (Fig. 2.3b) has been postulated. The tendency of HSγδTCL to select clones with increased copies of i(7)(q10) (Wlodarska et al. 2002) (Fig. 2.3c) or with the amplified 7q sequences (Shetty et al. 2006; Tamaska et al. 2006) suggests that this chromosome harbors genes potentially important for its pathogenesis. One of the candidate genes is cyclin-dependent kinase 6 (CDK6), known as a target of the t(2;7)(p12;q21) in splenic marginal zone lymphoma (Corcoran et al. 1999). Despite a strong association of the i(7)(q10) with HSγδTCL, this aberration is not specific for the entity, being observed at a lower frequency in a broad spectrum of hematological malignancies (Mertens et al. 1994). Secondary aberrations frequently associated with i(7)(q10) in HSγδTCL include trisomy 8 and loss of the Y chromosome (Alonsozana et al. 1997; Jonveaux et al. 1996; Wlodarska et al. 2002). The role of these abnormalities in the pathogenesis of HSγδTCL is unknown. Early cytogenetic studies reported rare HSγδTCL cases with abnormal karyotypes, but lacking i(7)(q10) (Ross et al. 1994; Salhany et al. 1997a, b; Weidmann et al. 2000); these findings, however, have not been validated by FISH.

Despite a different immunophenotype, HSαβTCL is also characterized by a recurrent i(7)(q10) (Suarez et al. 2000; Macon et al. 2001), suggesting a common pathogenesis of both lymphomas. Hence, it has been postulated that HSγδTCL and HSαβTCL represent phenotypic variants of the same disease entity.

T-cell lymphomas showing primary localization in the skin include mycosis fungoides and Sézary syndrome, primary cutaneous CD30+ lymphoproliferative disorders including primary cutaneous anaplastic large cell lymphoma and lymphomatoid papulosis, subcutaneous panniculitis-like T-cell lymphoma, and some rare lymphomas such as primary cutaneous aggressive epidermotropic CD8+ T-cell lymphoma (provisional), cutaneous γ/δ T-cell lymphoma (provisional), and primary cutaneous CD4+ small-/medium-sized pleomorphic T-cell lymphoma (provisional) (Swerdlow et al. 2008).

Complex karyotypes have been observed in mycosis fungoides (MF) and Sézary syndrome (SS), but no disease-specific abnormalities have been described (Prunieras 1974; Thangavelu et al. 1997). Karenko et al. reported recurrent rearrangements with a breakpoint at 12q21, targeting the NAV3 (POMF1L1) gene (Karenko et al. 2007). Translocations involving TCR loci are notably absent in MF/SS (Salgado et al. 2011). CGH studies have revealed common deletions at 1p, 6q, 10q, 13q, and 17p and on chromosome 19 and gains of chromosomes 7 and 18 and at 8q and 17q (Mao et al. 2002).

Complex karyotypes are also characteristic for primary cutaneous anaplastic large cell lymphoma (C-ALCL). The t(2;5)(p23;q35), characteristic for ALK-positive ALCL, is detected in only rare cases. NPM1-ALK transcripts were detected by PCR in the absence of ALK protein expression in these tumors; thus, their pathogenetic significance is uncertain (Wood 1998). Deletions at 9p21, affecting the CDKN2A/p16 locus, are present in some C-ALCLs. CGH studies also demonstrated oncogene amplifications involving CTSB at 8p22, RAF1 at 3p25, REL at 2p16, and JUNB at 19p13.2 (Mao et al. 2003; van Kester et al. 2010).

Cytogenetic studies of regressing lymphomatoid papulosis lesions demonstrated either a normal karyotype or abnormalities of chromosomes 7, 10, and 12. The t(2;5) has not been found (Mao et al. 2002).

Peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS) is the largest and most common category of PTCL, accounting for approximately 30 % of all PTCLs in the Western world. This entity is very heterogeneous comprising cases that do not correspond to any of the PTCL subtypes recognized in the current WHO classification (Swerdlow et al. 2008). Cytogenetic analysis showed that karyotypes of PTCL-NOS are typically very complex. Both conventional and array CGH studies detected recurrent gains of 7q/CDK6, 8q/MYC, 17q, and 22q and recurrent losses of 4q, 5q, 6q, 9q, 10q, 12q, and 13q (Thorns et al. 2007; Nagel et al. 2008). Deletions of 5q, 10q, and 12q correlate with a better prognosis (Zettl et al. 2004). Notably, genomic profiles of PTCL-NOS differ from those observed in AITL and ALCL. The regions 6q16–q22, 9p21, and 11p11.2 are predominantly lost in PTCL-NOS when compared to AITL and gains of 7q22 and 8q24.1–q24.3 are more frequent in PTCL-NOS than in AITL- and ALK-negative ALCL. Common genetic events identified in these entities include a recurrent gain of 11q13 in both AITL and PTCL-NOS and loss of 6q21 in both ALK-negative ALCL and PTCL-NOS (Zettl et al. 2004; Thorns et al. 2007; Nelson et al. 2008).

The WHO classification recognized three variants of PTCL-NOS: lymphoepithelioid (Lennert’s lymphoma), follicular, and T-zone variants (Swerdlow et al. 2008). Genetic data are mainly available for the follicular variant (PTCL-F), in which the growth pattern mimics follicular B-cell lymphoma or T-cell lymphomas with a perifollicular growth pattern (Rudiger et al. 2000; Ikonomou et al. 2006; Huang et al. 2009). This lymphoma is characterized by a recurrent t(5;9)(q33;q22), which involves two protein tyrosine kinase genes: ITK, the IL-2-inducible T-cell kinase gene located at 5q33, and SYK, the spleen tyrosine kinase gene mapped at 9q22 (Streubel et al. 2006b). The resulting ITK-SYK fusion protein revealed a constitutively active SYK tyrosine kinase, which has shown to be transforming both in vitro and in vivo (Rigby et al. 2009; Dierks et al. 2010; Pechloff et al. 2010). SYK plays an important key role in TCR signaling; overexpression and activation of SYK is a common feature of PTCL, and therefore, the gene represents a potential therapeutic target (Feldman et al. 2008; Wilcox et al. 2010). Although the t(5;9)/ITK-SYK has been detected in 17 % of PTCL-NOS, it seems to be restricted to PTCL-F (Streubel et al. 2006b; Feldman et al. 2008). Given the rarity of PTCL-F, the prognostic significance of the t(5;9) is unknown.

Angioimmunoblastic T-cell lymphoma (AITL) accounts for 1–2 % of all non-Hodgkin lymphomas and for 25–30 % of PTCL cases in Europe and North America (Rudiger et al. 2002). Clonal chromosomal aberrations have been detected in up to 90 % of the cases analyzed (Weiss et al. 1986; Tan et al. 2006; Attygalle et al. 2007). The recurrent aberrations include trisomies of chromosomes 3, 5, and 21; gain of X; and loss of 6q and 13q (Schlegelberger et al. 1994b; Thorns et al. 2007; Nelson et al. 2008; reviewed by Dogan et al. 2003). Chromosomal breakpoints affecting the TCR gene loci seem to be very scarce (Leich et al. 2007). Although the cellular derivation is uncertain, recent findings suggest that MAF-expressing follicular helper T cells (T_FH) represent the normal counterpart of AITL (Thielen et al. 2011).

B-cell prolymphocytic leukemia (B-PLL) is a rare lymphoid neoplasm characterized by peripheral blood, bone marrow, and splenic involvement by a clonal proliferation of prolymphocytes. The cytogenetic pattern of B-PLL has been poorly characterized due to the rarity of the disease, the difficulty in obtaining mitoses of the rather mature leukemic cells, and the overlap with other entities that previously may have been considered to be B-PLL. The t(11;14)(q13;q32)/CCND1/IGH rearrangement has previously been reported in up to 20 % of B-cell PLL cases (Brito-Babapulle et al. 1987); however, these cases are now considered leukemic manifestations of MCL (Ruchlemer et al. 2004). More recent studies that focused on confirmed cases of B-PLL showed the presence of complex karyotypes, with frequent abnormalities of chromosome 7 (Schlette et al. 2001). FISH analysis has been used in B-PLL to test for abnormalities at specific genetic loci; this resulted in detection of 13q14 deletions involving the RB1 gene in approximately 50 % of the cases. Additionally, deletions were frequently noted at band 11q23 and at 17p (TP53 locus) (Lens et al. 2000). None of the aberrations reported thus far in B-PLL have been disease specific.

Marginal zone lymphoma (MZL) encompasses three distinct entities: extranodal MZL of the mucosa-associated lymphoid tissue (MALT lymphoma), nodal MZL (NMZL), and splenic MZL (SMZL) (Swerdlow et al. 2008). It is believed that these indolent malignancies, which develop in different anatomical sites, originate from post-follicular memory B cells. MZL share several morphologic, immunophenotypic, and genetic features, including trisomies 3 and 18; however, the pathogenesis of these lymphomas is poorly understood.

Mantle cell lymphoma (MCL) is a well-defined neoplasm originating from naïve pre-germinal-center B cells, associated with an aggressive clinical presentation, poor response to therapy, and a short survival (Swerdlow et al. 2008). Cytogenetically, MCL is defined by the t(11;14)(q13;q32), a translocation that brings the CCND1 gene (the breakpoint region was previously known as BCL1) under transcriptional control of the regulatory sequences of IGH (Fig. 2.6a–b) (Tsujimoto et al. 1984; Williams et al. 1991). CCND1 encodes cyclin D1 which plays an important role in the regulation of the G1-S transition following mitotic growth factor signaling (Hunter and Pines 1994); its aberrant expression in MCL is routinely detected by IHC (Fig. 2.6c). Of note, the t(11;14)(q13;q32) is also noted in a subset of multiple myeloma and related disorders (Fonseca et al. 2003). Conventional cytogenetic analysis identifies the t(11;14) in 60–80 % of MCL (Li et al. 1999; Wlodarska et al. 1999; Au et al. 2002); however, FISH detects the IGH-CCND1 rearrangement in almost all MCL cases examined (Li et al. 1999; Bentz et al. 2000, 2004; Frater et al. 2001). FISH analysis is particularly recommended in MCL cases with atypical karyotypes, because the t(11;14) may be cryptic or masked by complex secondary alterations (Gruszka-Westwood et al. 2002; Aventin et al. 2003; Gazzo et al. 2005) (Fig. 2.6d–e). Sporadic MCL cases display variant 11q13/CCND1 translocations involving either IGK/2p12 or IGL/22q11.2 (Fig. 2.7) (Komatsu et al. 1994; Wlodarska et al. 2004a; Espinet et al. 2010; Rocha et al. 2011). Rare cases of MCL that are negative for the translocation and cyclin D1 have been identified. Interestingly, these lymphomas are similar to classical MCL with a typical morphology, transcriptome profile, and pattern of secondary genetic alterations, but they express high levels of either cyclin D2 or cyclin D3 (Fu et al. 2005; Salaverria et al. 2007; Hartmann et al. 2010) and do not express SOX11 (Mozos et al. 2009). As documented in several cases, an aberrant expression of cyclin D2 and D3 in t(11;14)-negative MCL is a consequence of IG translocations targeting CCND2 and CCND3, respectively (Gesk et al. 2006; Herens et al. 2008; Wlodarska et al. 2008; Quintanilla-Martinez et al. 2009; Shiller et al. 2011) (Fig. 2.8).

Approximately one-third of MCLs have the t(11;14)(q13;q32) as a sole cytogenetic aberration or together with 1–2 additional chromosomal changes. Remarkably, these cases are usually associated with a leukemic (non-nodal) disease, an indolent clinical course, and a long survival (Fernandez et al. 2010; Royo et al. 2012). In contrast, classical aggressive MCLs display complex karyotypes characterized by a high number of nonrandom secondary chromosomal changes, initially shown by cytogenetic studies (Li et al. 1999; Wlodarska et al. 1999; Au et al. 2002) and further demonstrated by genome-wide screening approaches (reviewed by Royo et al. 2011). These aberrations are commonly associated with genomic gains and losses. Balanced translocations are infrequent in MCL. An exception is the t(8;14)(q24.2;q32)/IGH-MYC and variant MYC aberrations, which have been recurrently seen in MCLs with blastoid morphology (Tirier et al. 1996; Au et al. 2000; Vaishampayan et al. 2001; Hao et al. 2002; Michaux et al. 2004). As in the well-known MYC/BCL2 “double-hit” lymphomas, survival of patients with MCLs with MYC/8q24.2 alterations is extremely short.

The genomic profile of MCL was initially investigated using conventional CGH (Monni et al. 1998; Bea et al. 1999; Bentz et al. 2000; Martinez-Climent et al. 2001; Allen et al. 2002; Jarosova et al. 2004; Salaverria et al. 2007). These studies identified numerous recurrent genomic imbalances in MCL, of which the most frequent were losses of 1p, 6q, 8p, 9p, 10p, 11q, 13q, and 17p and gains of 3q, 7p, 8q, 12q, 15q, and 18q (Table 2.4, adapted from Royo et al. 2011). These findings were further confirmed using high-resolution CGH and SNP arrays that mapped the minimal regions of loss and gain and identified new submicroscopic deletions and duplications (Schaffner et al. 2000; Kohlhammer et al. 2004; Rubio-Moscardo et al. 2005; Schraders et al. 2005; Tagawa et al. 2005; Rinaldi et al. 2006; Flordal et al. 2007; Bea et al. 2009; Kawamata et al. 2009; Vater et al. 2009; Halldorsdottir et al. 2011). The results and candidate target genes are shown in Table 2.5. The genes postulated to be involved in the pathogenesis of MCL include TNFAIP3 (6q23), CDKN2A (9p21), ATM (11q22.3), RB1 (13q14), and TP53 (17p13) located in the recurrently deleted regions and BMI1(10p12), CDK4/MDM2 (12q14), and BCL2 (18q21.3), which are found to be gained and/or amplified in MCL. In addition, SNP array analysis identified CN-LOH events in up to 60 % of MCL (Bea et al. 2009; Kawamata et al. 2009; Vater et al. 2009; Fernandez et al. 2010; Hartmann et al. 2010; Halldorsdottir et al. 2011). CN-LOH frequently affects 6p, 9p, 11q, 17p, and 20q, which are recurrently deleted regions in MCL. Whether CN-LOH represents an alternative mechanism of biallelic inactivation of tumor suppressor genes remains to be determined.