• Open Access

Genetic alterations in systemic nodal and extranodal non-cutaneous lymphomas derived from mature T cells and natural killer cells

Authors


Abstract

Mature (peripheral) T-cell and natural killer (NK)-cell lymphomas comprise a series of rather different neoplasms. Based on morphologic, immunophenotypic, genetic, and clinical data, the World Health Organization classification recognizes more than 20 entities or provisional entities. The variable clinical presentations, the objective recognition and pathological stratification, the difficulties regarding treatment, and the hardly predictable response to therapy indicate that the management of these entities requires novel tools. In contrast to B-cell lymphomas or precursor T-cell neoplasms, few recurrent translocations have been identified so far in T-cell non-Hodgkin's and NK-cell lymphomas. Additionally, some of the entities recognized by the World Health Organization classification are very rare and very scarce molecular data are available for T-cell lymphomas. Here, we have reviewed published reports focusing on the genetic lesions and gene expression profiling underlying systemic nodal and extranodal non-cutaneous mature T-cell and NK-cell lymphomas. We also provide a summary of new agents in clinical development and outline some future directions. (Cancer Sci, doi: 10.1111/j.1349-7006.2012.02321.x, 2012)

Mature (peripheral) T-cell and NK-cell lymphomas comprise a series of rather different neoplasms. Based on morphologic, immunophenotypic, genetic, and clinical data, the WHO classification recognizes more than 20 entities or provisional entities (Table 1).[1] The variable clinical presentations, the objective recognition and pathological stratification, the difficulties regarding treatment, and the unpredictable responses to therapy[2-4] indicate that the management of these entities requires novel tools. For the most part, the outcome is unsatisfactory after CHOP chemotherapy; therapies tailored specifically for T-cell and NK-cell lymphomas are in their infancy and thus urgently needed.

Table 1. World Health Organization histological classification of mature T-cell and natural killer (NK)-cell neoplasms*
  1. *According to World Health Organization (WHO). WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon: IARC Press, [1]. †Provisional entity. ALK, anaplastic lymphoma kinase; EBV, Epstein–Barr virus.

Nodal
Peripheral T-cell lymphoma, unspecified
Angioimmunoblastic T-cell lymphoma
Anaplastic large-cell lymphoma, ALK-positive
Anaplastic large-cell lymphoma, ALK-negative
Extranodal
Extranodal NK/T-cell lymphoma, nasal type
Enteropathy-type T-cell lymphoma
Hepatosplenic T-cell lymphoma
Subcutaneous panniculitis-like T-cell lymphoma
Leukemic/disseminated
T-cell prolymphocytic leukemia
T-cell large granular lymphocytic leukemia
Aggressive NK-cell leukemia
Hydroa vacciniforme-like lymphoma
Adult T-cell leukemia /lymphoma (HTLV1-positive)
Systemic EBV+ T-cell lymphoproliferative disorders of childhood
Chronic lymphoproliferative disorders of NK cells
Extranodal cutaneous
Mycosis fungoides
Sézary syndrome

Primary cutaneous CD30-positive T-cell lymphoproliferative

disorders

Primary cutaneous anaplastic large-cell lymphoma
Lymphomatoid papulosis
Primary cutaneous γδ T-cell lymphoma

Primary cutaneous aggressive epidermotropic CD8-positive

cytotoxic T-cell lymphoma

Primary cutaneous small/medium CD4-positive T-cell lymphoma

In contrast to B-cell lymphomas or precursor T-cell neoplasms (i.e., T lymphoblastic leukemia/lymphoma), few recurrent translocations have been identified so far in T-NHL and NK-cell lymphomas. Additionally, some of the entities recognized by the WHO classification are very rare and very scarce molecular data are available. Here, we have focused on the genetic events underlying systemic nodal and extranodal non-cutaneous mature T-cell and NK-cell lymphomas. We also provide a summary of new agents in clinical development and outline some of the future directions.

Nodal Lymphomas

Peripheral T-cell lymphoma, not otherwise specified

The PTCL, NOS lymphomas constitute a heterogeneous category of mature T-NHL that do not correspond to any of the other specifically defined entities in the current classification.

Although representing the majority of T-NHL cases, the genetics of PTCL, NOS are poorly characterized.[5-7] Studies evaluating genomic aberrations in PTCL, NOS by the use of classical cytogenetic analysis, CGH, or arrayCGH, have shown a complex karyotype, characterized by a high genomic instability (Table 2).[8-16] Most frequent gains occur on chromosomal regions 1q32-qter, 2p (2p15-p16), 7q22-ter, 8q (8q24), 9q33-qter, 11q (11q13), and 17q (17cen-q21). Recurrent losses mainly affect 6q (6q21), 9p21, 10 (10cen-p12, 10q23-q24), 13q (13q21), 14q (14q12-q21), 16q (16q11-q21), and 17p (17p13). A series of genes have been identified as deregulated. The 7q21 gains, seen in at least half of the cases, cause the overexpression of CDK6, which controls the cell cycle G1/S transition,[16, 17] although other transcripts could also be affected because the gains usually span large genomic regions. The 9p21 losses target the CDKN2A and CDKN2B genes, which negatively regulate CDK6 function. Additional cell cycle regulators appear affected by recurrent aberrations, such as RB1 (13q12) and TP53 (17p13).[16] The 2p15-p16 gains, observed in at least 10% of the cases, target REL, which encodes a member of the NF-κB family, often amplified[16] or involved in chromosomal translocations.[18] Other genes within the NF-κB pathway and targets of deletions are NFKBIA (14q13.2) and CYLD (16q12.1). These observations, alongside the data derived from GEP studies,[5, 7] support the clinical evaluation of agents targeting the NF-κB pathway, and thus the therapeutic efficacy of drugs such as bortezomib and/or new proteasome inhibitors. In contrast to B-cell non-Hodgkin's lymphoma, in T-NHL no data on somatic mutations affecting genes of the NF-κB pathway are currently available.

Table 2. Main characteristics of systemic nodal and extranodal non-cutaneous lymphomas derived from mature T cells and natural killer (NK) cells
Entity% of T-NHL% of all NHLMedian age*5-year OS*Recurrent translocationsMost common recurrent genomic aberrations
  1. *Vose J, Armitage J, Weisenburger D. International peripheral T-cell and natural killer/T-cell lymphoma study: Pathology findings and clinical outcomes. J Clin Oncol [2]; 26: 4124–30. †20% in Asia, 4–5% in Europe and North America. ‡2% in Asia, 9% in Europe, 6% in North America. ALCL, anaplastic large cell lymphoma; ALK, anaplastic lymphoma kinase; cALCL, primary cutaneous ALCL; cALK-ALCL, primary cutaneous ALK-negative ALCL; NHL, non-Hodgkin's lymphoma; n.d., not determined; OS, overall survival; T-NHL, T-cell non-Hodgkin's lymphoma.

Peripheral T-cell

lymphoma,

not otherwise

specified

30–40%3–5%60 years30%

t(5;9)(q33;q22) ITK-SYK,

raret(6;14)(p25;q11.2)

IRF4-TCRA, <5%

Other 6p25 rearrangements

involving IRF4, <5%

+1q32-qter, +2p15-p16 (REL), +7q22-ter (CDK6), +8q24, +9q33-qter, +11q13, +17q12-q21, −6q21-q22, −9p21 (CDKN2A), −13q21-q22, −14q12-q21 (NFKBIA), −16q11-q21 (CYLD), −17p13 (TP53)

Angioimmunoblastic

T-cell lymphoma

15–30%1–4%65 years30%n.d.+3, +5, +11q13, +19, +20q13,+22q, −8p22, −9p21 (CDKN2A), −13q22-q32,

Anaplastic large cell lymphoma,

ALK-positive

7

1–5% (10–20%

in childhood)

34 years90%

t(2;5)(p23;q35) ALK-NPM,

80–85%t(1;2)(q25;p23)

TPM3-ALK, 10–15%

Other 2p23 rearrangement

involving ALK, <15%

+7p, +17p11-pter, +17q, −4q13-q28, −6q13-q22, −11q14-q23, −13q

Anaplastic large cell

lymphoma,

ALK-negative

71–2%58 years74%

t(6;7)(p25;q32) DUSP2-FRA7H,

18% (29% in cALK−ALCL)

Other 6p25 rearrangements

involving IRF4, <5% (50% in cALK−ALCL)

+1q, +7q, +8q, +12q, +17q, −4q, −6q21, −11q, −13q, −17p13 (TP53)
Extranodal NK/T cell lymphoma5–20%1–2%44–52 years10%n.d.−6q21 (PRDM1/BLIMP1), −9p21 (CDKN2A),

Enteropathy-

associated T-cell

lymphoma

5%<1%61 years30%n.d.

+9q33-q34, −16q12.1

Type 1: +1q22-q44, +5q

Type 2: +8q24 (MYC)

Hepatosplenic T-cell lymphoma1.5%<1%34 years0%n.d.i(7)(q10), +8, -Y

Subcutaneous panniculitis-like

T-cell lymphoma

0.5–1%<1%33 years64%n.d.+2q, +4q, +5q, +6q, +13q, −1p, −2p, −10q, −11q, −19, −20, −22q

Chromosomal translocations involving T-cell receptor genes are detected in only 1% or less of all PTCL, NOS.[19, 20] A t(5;9)(q33;q22) translocation causing the fusion of the tyrosine kinase domain of SYK to the N-terminal pleckstrin homology domain and proline-rich region of ITK[21-23] has been reported in a fraction of PTCL. It has been suggested to be associated to the rare PTCL-F. However, it is not clear whether PTCL-F represents a distinct subtype or a morphologic variant showing overlap with other PTCL, NOS or AITL. Importantly, the overexpression of SYK, important in proliferation and pro-survival signaling, can be seen in the majority of PTCL, despite the absence of SYK/ITK translocations in most cases.[24] This observation provides a rationale for the use of SYK inhibitors, such as fostamatinib disodium, currently under investigation in B-cell tumors.

Gene expression profiling data have provided some hints for better understanding this type of lymphoma.[5, 7, 25-28] Based on their gene expression signature, PTCL, NOS clearly differs from AITL, but it cannot be easily discriminated from ALCL.[27, 28] The PTCL, NOS appears heterogeneous in terms of activation of the NF-κB pathway, with some suggestions that these differences might affect patient outcome.[25] When compared to normal T-cells, gene expression profile observed in PTCL, NOS is characterized by the deregulation of genes involved in important cell functions such as matrix deposition, cytoskeleton organization, cell adhesion, apoptosis, proliferation, transcription, and signal transduction.[28] In particular, the overexpression of PDGFRα,[26] the deregulation of the NF-κB pathway, and the regulation of transcription provide the rationale for some of the clinical trials currently running for patients with PTCL, NOS (Table 3, Data S1). Indeed, PTCL, NOS is an aggressive lymphoma, with poor response to conventional therapy (CHOP-like regimens) and with frequent relapses, making mandatory the identification of more active regimens.

Table 3. Clinical trials in non-cutaneous peripheral T-cell lymphoma (PTCL) containing targeted agents alone or in combination with chemotherapy
RegimenTargetLine of treatmentPhasePatient populationNumber
  1. Included are trials open and recruiting patients in November 2011, according to the ClinicalTrials.gov registry of clinical trials (http://clinicaltrials.gov). Excluded are trials including patients with only cutaneous lymphomas, those containing only chemotherapy (pralatrexate also excluded), those containing radiation therapy, those containing both B- and T-cell lymphomas. AITL, angioimmunoblastic T-cell lymphoma; ALCL, anaplastic large cell lymphoma; CHOP, cyclophosphamide, doxorubicin, vincristine, and prednisolone; CTCL, cutaneous T-cell lymphoma; DNMT, DNA methyltransferase; EATL, enteropathy-associated T-cell lymphoma; HDAC, histone deacetylase; HSTL, hepatosplenic T-cell lymphoma; IL, interleukin; LGL, T-cell large granular lymphocytic leukemia; MF, mycosis fungoides; NK, natural killer; NOS, not otherwise specified; PTLD, post-transplant lymphoproliferative disorders; PLL, prolymphocytic leukemia; R-DA-EPOCH, rituximab, dose-adjusted doxorubicin, etoposide, vincristine, cyclophosphamide, and prednisone; RR, relapsed/refractory; SPTL, subcutaneous panniculitis-like T-cell lymphoma; SZ, Sezary syndrome.

Bortezomib + panobinostatProteasome, HDACRRIIPTCL NOS, AITL, ALCL, HSTL, EATL, extranodal NK/T cellNCT00901147
CarfilzomibProteasomeRRIPTCL NOS, AITL, ALCL, HSTL, EATL, extranodal NK/T cellNCT01336920
Bortezomib + azacitidineProteasome, DNMTRRIPTCL, CTCL, T/NK PTLD, LGL, T-PLLNCT01129180
BelinostatHDACRRIIPTCLNCT00865969
CHOP + vorinostatHDACRRI/IIPTCL NOS, AITL, ALCL ALK+/−, EATL, SPTLNCT00787527
Vorinostat + lenalinomide + dexatetasoneHDAC, multiple pathways (microenvironment, angiogenesis, immune)RRI/IIPTCLNCT00972842
RomidepsinHDACRRI/IIPTCL, MF, SZNCT01456039
CHOP + RAD001mTOR1st lineI/IIPTCL NOS, AITL, ALCL (ALK−), CTCLNCT01198665
CHOP14 +/− alemtuzumabCD521st lineIIIPTCL NOS, AITL, ALCL (ALK−), NK/T cellNCT00725231
R-DA-EPOCH + siplizumab (medi-507)CD21st lineIPTCL NOS, AITL, ALCL ALK−, EATL, SPTLNCT01445535
A-dmDT390-bisFv (UCHT1)CD3RRI/IIPTCL, TCL, CTCLNCT00611208
Brenduximab vedotin (SGN-35)CD30RRICD30+ mature T cell and NK cellNCT01309789
KW-0761CCR4RRIIPTCL, CTCLNCT01226472
Denileukin diftitox +/− CHOPIL-2R1st line/RRIPTCL NOS, ALCL, EATL, CTCL, MF/SZNCT01401530
Denileukin diftitox as in vivo purgingIL-2R1st or 2ndIIPTCL NOS, AITL, ALCL ALK−, extranodal NK/T, EATL, SPTL, HSTLNCT00632827
Endostatin + CHOPAngiogenesisRRIIAll T cellNCT00974324
Darinaparsin (SP-02L)MitochondriaRRIPTCL NOS, AITL, ALCL (ALK+/−)NCT01435863
LenalidomideMultiple pathways (microenvironment, angiogenesis, immune)RRIIAll T cellNCT00322985
LenalidomideMultiple pathways (microenvironment, angiogenesis, immune)RRIATL, PTCLNCT01169298

Angioimmunoblastic T-cell lymphoma

An aggressive systemic disease, AITL is characterized by a polymorphous infiltrate involving lymph nodes, with a prominent proliferation of high endothelial venules and follicular dendritic cells.[29-31] The consistent association with EBV has suggested a pathogenetic role for this virus, eliciting chronic antigen-driven T cell-mediated responses. TCR gene rearrangements are nearly always detectable, and the presence of an EBV+ B-cell expansion can result in the detection of rearrangement of clonal immunoglobulin genes in up to one-third of cases.[32]

Little is known about its genetics (Table 2).[12-14, 33] In addition to the high frequencies of trisomies of chromosomes 3 and 5, not specific for AITL,[8, 11] the most frequent gains have been reported at 11q13, 19, 20q13, and 22q, whereas losses have been reported at 13q22-q32, 8p22, and 9p21 (CDKN2A).

Despite that, in tumor specimens, neoplastic cells are largely overnumbered by reactive cells, making GEP studies difficult, AITL has been shown to clearly differ from other PTCL.[25, 34] Also, AITL appeared to derive from T follicular helper cells as normal counterparts,[27, 35] from which the lymphoma diverges, for example, for a lower expression of CD57 and higher CD10 expression.[35]

There is no standard therapy for AILT, and the results obtained with conventional regimens have been disappointing, underlying the importance of defining new treatment modalities (Table 3, Data S1). Gene expression profiling analysis highlighted that a set of genes involved in vascular biology was upregulated in AITL, notably the VEGF, suggesting a valuable target for therapeutic intervention given that several VEGF inhibitors are under investigation in lymphomas.[36-38] The NF-κB pathway is highly activated in AITL cases and this pathway is a strong candidate for therapeutic intervention. Inhibitors of the NF-κB pathway may provide significant therapeutic benefit by acting against both the tumor and microenvironmental components.

Anaplastic large cell lymphoma, ALK-positive

A T-cell lymphoma consisting of lymphoid cells that are usually large with abundant cytoplasm and pleomorphic, often horse-shoe-shaped nuclei, strongly positive for the CD30 antigen.[39, 40]

Anaplastic large cell lymphoma, ALK-positive is characterized by chromosomal translocations involving the ALK gene on chromosome 2, coding for a tyrosine kinase (Table 2).[41] The most common translocation is t(2;5)(p23;q35), generating the NPM-ALK fusion protein with transforming properties.[40, 42] Expression of NPM-ALK leads to the activation of several downstream signal transduction events that provide positive survival and proliferation signals. In approximately 20% of ALK+ALCL cases, alternative translocations have been described in which the ALK gene is fused to various partners, such as TPM3 (1q21.2), TFG (3q12.2), ATIC (2q35), TSPYL2 (Xp11.2), MSN (Xq11.1), RNF213 (17q25.3), and MYH9 (22q13.1) (Table 4). In terms of prognosis, no differences have been described in patients bearing different translocation variants.[43]

Table 4. Recurrent chromosomal translocations involving anaplastic lymphoma kinase (ALK)
Chromosomal translocationALK partnerFrequency (%)Cellular localization
  1. ALO17, ALK lymphoma oligomerization partner on chromosome 17; ATIC, amino-terminus of 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase gene; CLTC, clathrin heavy polypeptide gene; MSN, moesin gene; MYH9, myosin heavy chain 9 gene; NPM, nucleophosmin gene; TFG, TRK-fused gene, three different fusion proteins of 85, 97, and 113 kD are associated with the t(2;3)(p23;q35) which involves TFG; TPM3 and TPM4, non-muscolar tropomyosin gene.

t(2;5)(p23;q35)NPM80–84Nuclear, diffuse cytoplasmic
t(1;2)(q25;p23)TPM313Diffuse cytoplasmic with peripheral intensification
t(2;3)(p23;q21)TFG 2Diffuse cytoplasmic
inv(2)(p23;q35)ATIC 2Diffuse cytoplasmic
t(2;17)(p23;q23)CLTC1 2Granular cytoplasmic
t(2;X)(p23;q11-12)MSN<1Cell membrane-associated
t(2;19)(p23;p13)TPM4<1Diffuse cytoplasmic
t(2;17)(p23;q25)ALO17<1Diffuse cytoplasmic
t(2;22)(p23;q11.2)MYH9<1Diffuse cytoplasmic
Others?<1Nuclear or cytoplasmic

Besides the well-defined primary chromosomal translocations, ALK+ALCL carries frequent secondary chromosomal imbalances including losses of chromosome 4 (4q13-q28), 6q (6q13-q22), 11q (11q14-q23), and 13q (13q21-q31; 13q32-q33), and gains at 7p11-pter and 17 (Table 2).[11, 44-46] Interestingly, ALK+ALCL is not only virtually devoid of losses affecting TP53, but the 17p13 locus has also been reported as a target of genomic gains. This represents a unique example among lymphomas and tumors in general. Indeed, gains of 17p, together with losses at chromosome 4, are uncommon among other types of T-cell lymphomas, including ALK−ALCL, making them a characteristic of ALK+ALCL.[11, 47] So far, no genes have been clearly identified as altered by these lesions.

Independently from the expression of ALK, ALCLs share a cluster of transcripts that allow their stratification and distinction from other T-cell lymphomas, suggesting a common signature for both ALK+ and ALK− cases and, possibly, a unique origin for both subtypes.[35, 48] Importantly, cases of primary systemic ALK+ALCL express a STAT3-dependent gene expression signature, which can discriminate them from other T-cell NHL, but, again, not from ALK−ALCL.[48] However, although ALK+ALCL and ALK−ALCL are largely overlapping in terms of their gene expression profiles, different groups have reported a series of genes with a differential expression between the two lymphoma subtypes.[35, 48-50] Excluding ALK, Lamant et al.[50] reported BCL6, PTPN12, CEBPB, and SERPINA1 as the most discriminating genes between the two subsets. In another work, the transcripts overexpressed in ALK+ tumors comprised CCND3 and genes involved in immune response, the NF-κB pathway, leucocyte transendothelial migration, focal adhesion, and signal transduction (SYK, LYN, CDC37), whereas the transcription factor genes HOXC6 and HOXA3 appeared more expressed in ALK−ALCL.[49] Piva et al.[48] identified GAS1 as selectively expressed in ALK+ALCL and as the most highly ALK signaling-dependent gene. Finally, in the study carried out by Iqbal et al.[35] among the top-ranked genes were ALK, TNFRSF8 (CD30), MUC1, and TH17, cell-associated molecules (IL-17A, IL-17F, GATA3, RORC), and a small group of immunoregulatory cytokines/receptors regulating STAT3 (IL-26, IL-31RA) or JAK3 (IL-9) activation. ALK+ALCL also showed a lower expression of transcripts related to TCR components and TCR signaling or activation, alongside a higher expression of the cytotoxic molecules GZMB and PRF1.[35] The coexistence of transcripts characteristic of TH17 cells with cytotoxic molecules and enrichment of the proto-oncogene ETS1 target gene signature indicate a profoundly deranged differentiation program.[35]

Patients with ALCL are generally managed in a manner similar to patients affected by PTCL, NOS: CHOP chemotherapy represents the standard first line treatment, but relapses are frequent. Recently, some new agents have shown significant clinical activity, such as romidepsin, a histone deacetylase inhibitor, or brentuximab vedotin, an anti-CD30 antibody coupled to the anti-tubulin agent monomethyl auristatin E (SGN-35). The latter has very recently obtained the FDA accelerated approval for patients with systemic ALCL who have failed at least one prior multi-agent chemotherapy regimen (Data S1, Table 3). In addition to CD30, ALK also represents a target for the development of new agents.[51, 52] The success of other tyrosine kinase inhibitors has encouraged the search for ALK-selective small molecule inhibitors and some very effective compounds have emerged. Also, ALK is under investigation as a target for antitumor vaccination.

Anaplastic large cell lymphoma, ALK-negative

Considered a provisional entity, ALK−ALCL is defined as a CD30+ T-cell neoplasm that is not reproducibly distinguishable on morphological grounds from ALK+ALCL, but lacks the expression of ALK and/or chromosomal translocations affecting its gene locus. These lymphomas have a different clinical presentation involving predominantly adults with advanced age, and a more aggressive clinical course.[39, 53]

Anaplastic large cell lymphoma, ALK-negative tends to differ in terms of losses and gains from both PTCL, NOS and from ALK+ALCL, although overlapping features can be found (Table 2).[11, 45] The most frequent genetic alterations are gains of 1q41-qter, 5q, 6p, 7p 8q, 12q and 17q (17q12-q21), and losses at 6q (6q21-q22) and 13q (13q21-q22). Very recent data[54] indicate that the 6q losses would target the PRDM1 gene coding the BLIMP1 transcription factor, also shown to be deregulated in other B- and T-NHLs.[55, 56] Importantly, these chromosomal imbalances differ from those identified in ALK+ALCL, supporting the concept that they are different biological entities. The abnormalities of ALK−ALCL do not fully overlap with whose reported for other T-cell neoplasms, but can be shared with primary cutaneous ALCL.[11, 45, 57, 58] The last observation is intriguing, indicating a high similarity between these two types, despite their diverse clinical features and outcomes, with a much better prognosis observed for primary cutaneous ALCL.[59]

Two translocations have been recently reported in ALK−ALCL. The t(6;7)(p25.3;q32.3), involving the DUSP22 gene and the FRA7H fragile site, has been reported in both systemic and cutaneous ALK−ALCL cases.[60] The lesion seems to cause a downregulation of the DUSP22 expression level and an upregulation of MIR29A, mapped within FRA7H.[60] Their biologic and clinical significance is to be determined, as MIR29A is believed to act as tumor suppressor gene in ALK+ALCL, in which it appears to be epigenetically silenced.[61] The second chromosomal translocation affects the gene coding for IRF4, which is located less than 50 kb telomeric to the DUSP22 gene.[62, 63]

Anaplastic large cell lymphoma, ALK-negative could not be accurately classified molecularly. However, from gene expression data, ALK−ALCL appeared distinct from PTCL, NOS and ALK+ALCL.[35] Compared with PTCL, NOS, genes associated with TCR signaling were expressed at a lower level, whereas two cytokines, IL-20, which promotes angiogenesis, and IL-9, which activates JAK3, were highly expressed. Compared with ALK+ALCL, lower expression of ALK, cytotoxic molecules (PRF1, GZMB), cathepsins (CTS-W, -D, -L1, -B), TH17-cell associated molecules (IL-17-F, IL-17-A, RORC), and B-cell associated transcripts (Ig-H, -K, -L) was noted. However, ALK−ALCL showed greater expression of a set of cytokine/receptors (CCL1, CCL22, CCR8, CCR4, IL-13RA2, CXCL14, TGFBR1) and several anti-apoptotic factors (BCL2, BIRC6, BIC), but low expression of certain pro-apoptotic genes (BAX, BCL2L1, BNIP3) as shown in a previous study.[50]

Although ALK+ALCL lymphomas have a favorable prognosis, ALK−ALCL tumors are more heterogeneous and more frequently have an aggressive course, with poorer response rates and a high relapse rate. The standard regime is CHOP, but, even with intensification, prognosis remains poor for most patients. Several new drugs are under clinical evaluation with the above-mentioned anti-CD30 conjugated antibody having shown the most promising results (Table 3, Data S1).

Extranodal Lymphomas

Extranodal NK/T-cell lymphoma, nasal type

Extranodal NK/T-cell lymphoma is a clinically aggressive entity, histologically characterized by EBV+ atypical lymphoid cytotoxic cells, associated with extensive vascular destruction, and prominent tissue necrosis.[1, 30, 31, 64] The upper aero-digestive tract is most commonly involved, with the nasal cavity being the prototypic site of involvement. If occurring outside this tract, ENKTL frequently involves the skin, soft tissue, gastrointestinal tract, and also testis.

A variety of cytogenetic aberrations have been reported, but no specific chromosomal translocations have been identified. The commonest cytogenetic abnormality is del(6q21q25) or i(6)(p10), but it is currently unclear whether this is a primary or progression-associated event (Table 2).[65, 66] Only few reports using array-CGH have been published and many of them did not distinguish between aggressive NK-cell leukemias and ENKTL. They have documented chromosome losses of 1p, 4q, 5, 6q, 7q, 11q, 12q, 15q, and 17q, and gain of 2q, 10q, and 13q. Several investigators have focused on the loss of 6q, a region that includes many tumor suppressor genes.[67-69] Very recent data indicate PRDM1, coding for BLIMP1, possibly together with FOXO3, as the genes inactivated after the 6q21 deletions,[55, 56] similar to that observed in ALCL.[54] In order to better characterize the genomic alterations of ENKTL, the array-CGH technique has been used.[70] Gain of 1q was found in both primary cutaneous ENKTL and non-cutaneous ENKTL.[67, 69, 70] Deletions at 9p, a region including CDKN2A and CDKN2B, 12p, and 12q are present only in ENKTL. A proportion of cases show the partial deletion of FAS or mutations of TP53, b-catenin, KRAS, or KIT.[71] Although still to be confirmed in independent studies, the latter lesion might provide an important therapeutic target.

There is no standard therapy for ENKTL, and investigation of new drugs is fundamental (Table 3, Data S1). With this regard, different gene expression studies[72-74] evaluating both RNA and miRNAs and comparing ENKTL and normal NK cells, have identified a series of deregulated genes and pathways, many of which could represent the rationale for targeted therapies, including the PDGFRA, Aurora-A, Survivin, AKT, JAK-STAT, MYC, TP53, and NF-κB pathways. In particular, in vitro data confirmed the potential usefulness of imatinib, targeting PDGFRA,[72-74] and of a Survivin inhibitor (terameprocol, EM-1421).[74]

Enteropathy-associated T-cell lymphoma

Enteropathy-associated T-cell lymphoma is defined as an intestinal lymphoma of intraepithelial T lymphocytes. It is uncommon in most parts of the world, but has been observed with increasing frequency in areas in which there is a high prevalence of celiac sprue, particularly in northern Europe.[30, 75, 76] Enteropathy-associated T-cell lymphoma has been recently subdivided in two types: EATL Type 1, strongly associated with celiac sprue and has the HLA-DQ2 haplotype, and the neoplastic cells are often large or pleomorphic, usually CD56−; and EATL Type 2, less strongly linked to celiac sprue and HLA-DQ2, and the lymphoma cells are medium-sized and monomorphic and often CD56+. The geographical distribution of the two EATL types is characterized by a higher prevalence of Type 1 in Europe, and Type 2 in Asia.

The two types of EATL are associated with similar, but not identical, genetic alterations, which might also present geographical differences (Table 2).[75, 77-79] Gains of 9q33-q34 and 16q12.1 deletions are frequent in both types and they seem to form a common genetic link between the two subtypes. EATL Type 1 frequently displays 1q22-q44 and 5q gains, whereas EATL Type 2 is more often characterized by 8q24 (MYC) gains. NOTCH1 or NEK6 have been proposed as genes possibly affected by the 9q gains.[78, 80]

The prognosis of both subtypes of EATL is very poor with conventional chemotherapy, also due to local complications and the underlying poor nutritional and immunological conditions (Table 3, Data S1).

Hepatosplenic T-cell lymphoma

Hepatosplenic T-cell lymphoma is a rare entity with peculiar clinical, morphological, and phenotypic features.[30, 81-83] It usually consists of monomorphic medium-sized T cells, mostly expressing the γδ-T-cell receptor. The cells express the cytotoxic granule-associated proteins TIA1 and granzyme M, but are usually negative for granzyme B and perforin, suggesting that the cell of origin is a functionally immature cytotoxic T cell. Most tumors are characterized by the presence of an isochromosome 7q, which can be the sole karyotypic abnormality, suggesting its primary role in pathogenesis (Table 2).[84, 85] Trisomy 8 and loss of chromosome Y seem to be associated with progressive disease.

The disease is aggressive with a median survival of <2 years (Table 3, Doc. S1). A very recent work has reported GEP data obtained from nine hepatosplenic lymphomas highlighting the activation of many, potentially targetable pathways, such as VEGF, MAPK, JAK-STAT, mTOR, Notch signaling, cytokine–cytokine receptor interaction, cell adhesion, and also Syk.[85]

Subcutaneous panniculitis-like T-cell lymphoma

Subcutaneous panniculitis-like T-cell lymphoma represents a rare, difficult-to-diagnose, and poorly characterized subtype of extranodal cytotoxic T-cell lymphoma.[86]

There are only a few publications on SPTL. Comparative genomic hybridization analysis revealed large numbers of DNA copy number changes in this disease.[87] The most common alterations were losses of chromosomes 1pter, 2pter, 10qter, 11qter, 12qter, 16, 19, 20, and 22, and gains of chromosomes 2q and 4q (Table 2). DNA copy number aberrations in SPTL, such as loss of 10q, 17q, and chromosome 19, partially overlap with those seen in other cutaneous T-cell lymphomas, such as mycosis fungoides or Sézary syndrome, whereas 5q and 13q gains characterize SPTL. Allelic NAV3 aberrations, previously found in mycosis fungoides and Sézary syndrome, were identified in 44% of SPTL patients.[87] The clinical course is variable, ranging from an indolent disease to rapidly fatal hemophagocytosis.

Conclusions

In summary, systemic and extranodal T-NHL make up a very heterogeneous group of disorders with few characteristic genomic lesions. Most patients are not cured and major efforts are still needed in order to improve treatment results. A series of possible therapeutic targets is currently being evaluated, also supported by data derived from gene expression studies. The rarity of the individual subtypes does not affect only the possibility of running dedicated prospective clinical trials, but it also results in the lack of published large series of homogeneously studied cases. Also, no data derived from deep sequencing of the genome of systemic and extranodal T-NHL reporting recurrently mutated genes have been published. International cooperation at all levels is strongly needed to improve our knowledge of the underlying pathogenesis of the individual T-cell lymphoma subtypes, providing rational bases for identifying new therapeutic targets and improving the use of available antitumor compounds.

Acknowledgments

The authors would like to thanks our colleague Afua Adjeiwaa Mensah (Bellinzona, Switzerland) for manuscript editing. Supported by: Oncosuisse KLS-02403-02-2009; Fondazione per la Ricerca e la Cura sui Linfomi nel Ticino (Bellinzona, Switzerland); Nelia et Amadeo Barletta Foundation (Lausanne, Switzerland); AIRC 5x1000 (Genetics-driven targeted management of lymphoid malignancies Italy); F.E.S.R.2007-13 (Immonc); Converging Technologies (Modeling Oncogenic Pathways: From Bioinformatics to Diagnosis and Therapy (BIO_THER). M.B. is enrolled in the PhD program in Pharmaceutical Sciences, University of Geneva, Switzerland.

Disclosure Statement

The authors have no conflicts of interest.

Abbreviations
AITL

angioimmunoblastic T-cell lymphoma

ALCL

anaplastic large cell lymphoma

ALK

anaplastic lymphoma kinase

arrayCGH

microarray-based comparative genomic hybridization

CGH

comparative genomic hybridization

CHOP

cyclophosphamide, doxorubicin, vincristine, and prednisolone

EATL

enteropathy-associated T-cell lymphoma

EBV

Epstein–Barr virus

ENKTL

extranodal natural killer/T-cell lymphoma

GEP

gene expression profiling

HSTL

hepatosplenic T-cell lymphoma

IL

interleukin

NF-κB

nuclear factor-κB

NK

natural killer

NPM

nucleophosmin

PDGFRα

platelet-derived growth factor receptor α

PTCL, NOS

peripheral T-cell lymphoma, not otherwise specified

PTCL-F

peripheral T-cell lymphoma with follicular growth pattern

SPTL

subcutaneous panniculitis-like T-cell lymphoma

T-NHL

T-cell non-Hodgkin's lymphoma

VEGF

vascular endothelial growth factor

WHO

World Health Organization

Ancillary