New insights into the biology of T-cell lymphomas

Abstract

Peripheral T-cell lymphomas (PTCLs) encompass a heterogeneous group of postthymic T-cell lymphomas with >30 distinct subtypes associated with varied clinicopathological features. Unfortunately, the overall survival of the major PTCL subtypes is dismal and has not improved for decades; thus, there is an urgent unmet clinical need to improve diagnosis, therapies, and clinical outcomes. The diagnosis is often challenging, requiring a combinatorial evaluation of clinical, morphologic, and immunophenotypic features. PTCL pathobiology is difficult to investigate due to enormous intertumor and intratumor heterogeneity, limited tissue availability, and the paucity of authentic T-cell lymphoma cell lines or genetically faithful animal models. The application of transcriptomic profiling and genomic sequencing has markedly accelerated the discovery of new biomarkers, molecular signatures, and genetic lesions, and some of the discoveries have been included in the revised World Health Organization or International Consensus Classification. Genome-wide investigations have revealed the mutational landscape and transcriptomic profiles of PTCL entities, defined the cell of origin as a major determinant of T-cell lymphoma biology, and allowed for the refinement of biologically and clinically meaningful entities for precision therapy. In this review, we prioritize the discussion on common nodal PTCL subtypes together with 2 virus-associated T-cell and natural killer cell lymphomas. We succinctly review normal T-cell development, differentiation, and T-cell receptor signaling as they relate to PTCL pathogenesis and biology. This review will facilitate a better biological understanding of the different PTCL entities and their stratification for additional studies and target-directed clinical trials.

Graphical abstract

Figure 1.

Classification, incidence, and overall survival of PTCL subtypes. (A) Classification of peripheral T/NK cell lymphoma. PTCL classification is based on several parameters including clinical presentation, pathological, genetic features, and their association with normal cellular counterparts. Of the >30 PTCL entities recognized as distinct entities either by WHO or International Consensus Classification, the major subtypes present predominantly as nodal, extranodal, disseminated (leukemic), or cutaneous diseases, indicating that disease localization represents relevant diagnostic criteria for major PTCL entities. (B) Frequency of the PTCL entities. Epidemiological findings from published sources are reported (from North America [Bellei et al and Hsi et al¹⁷²], the United States [Adams et al and Ruan et al¹⁷⁴], South America [Fischer et al¹⁷⁵], Africa [Fitzpatrick et al and Belarbi et al¹⁷⁷], Western Europe [Laurent et al⁷], Central Europe [Janikova et al¹⁷⁸], Nordic [Ellin et al⁸], India [Park et al and Nemani et al¹⁸⁰], China [Park et al and Liu et al¹⁸¹], and Asia [Park et al and Yoon et al¹⁸²]) (see citations in supplementary Table 2). The average frequencies from these studies were estimated and presented as pie charts for each specified region as indicated. ALK⁺ALCL vs ALK^–ALCL cases were compared when data were reported. It must be noted that the frequencies presented may not represent the full spectrum of PTCL subtypes due to variations in data reporting across countries and also lack of requisite immunostains for current classification due to limited resources. Because of the exclusion of “other T-NHLs in individual regions or countries,” pie graphs may overrepresent the frequency of some PTCL subtypes. Different countries may have different reporting standards, subtype classifications, methodologies, or levels of data accuracy, which can affect the comprehensiveness and comparability of the data. However, for the worldwide frequency pie graph, the “other T-NHLs” were included to reflect the actual frequency of PTCL subtypes (see supplemental Table 2). (C) Overall survival of histological PTCL subtypes. These patients were generally treated with an anthracycline-containing regimen (adapted from Vose et al¹). Regional 5-year survival is displayed in North America (Hsi et al¹⁷²), Czech (Janikova et al¹⁷⁸), and Asia (Yoon et al¹⁸²). AITL, angioimmunoblastic T-cell lymphoma; CTCL, cutaneous T-cell lymphoma; EATL, enteropathy-associated T-cell lymphoma; HSTL, hepatosplenic T-cell lymphoma; MEITL, monomorphic epitheliotropic intestinal T-cell lymphoma; T-NHL, T-cell non-Hodgkin lymphoma.

Figure 2.

Stages of T-cell development and differentiation. (A) T-cell lymphopoiesis initiates with precursor HSC in the bone marrow (BM), the primary lymphoid organ; and CLPs in BM are directed to T-cell lineage fate via IL7/NOTCH signaling. CLP enters the cortex region of the thymus and undergoes a series of differentiation stages including β-selection and TCR repertoire phase, which can be identified through cell surface receptors (CD44 and CD25). T cells undergo a dual selection process, positive selection for compatibility of their TCR with self-MHC molecules and negative selection against autoantigenic peptides before leaving the thymus to form the peripheral T-cell repertoire. Double-positive cells differentiate toward CD4⁺ T cells through ThPOK and GATA3 expression, as well as a strong TCR signal, as well as toward CD8⁺ T cells through RUNX3 and NOTCH expression and a weak TCR signal. Although the vast majority of cells express an αβ TCR, a small subset expresses an alternative TCR composed of γ and δ chains (γδ T cells) and diverge early place at the DN3A stage in T-cell development and mostly become residents of mucosal sites. (B) In the peripheral secondary organs, the CD4⁺ T cells can differentiate into distinct TH subsets (T_H1, T_H2, T_H9, T_H17, Treg, and T_FH), dependent upon stimulation from APC, their coreceptors, and cytokine milieu. Several PTCL subtypes show a similar cell state or transcriptomics associated with these TH subsets precursors. T_H1 cells are characterized by the expression of key master regulators or TFs (ie, STAT4 and TBX21). These cells require interleukin-12 (IL-12) signaling for their differentiation and secrete IFN-γ, IL-2, and lymphotoxin. T_H2 cells express TFs, STAT6 and GATA3, require IL-4, and produce IL-4, IL-5, and IL-13, important for T_H2 differentiation. T_H9 cells secrete IL-9, require TGF-β and IL-4, and are characterized by the expression of PU.1 and IRF4. T_H17 expresses STAT3 and RORγt as key TFs, produces IL-17 and TNF-α, and requires IL-1β, TGF-β, IL-6, and IL-23. Tregs differentiate with TGF-β and IL-2, express FOXP3 and STAT5, and release TGF-β and IL-10 to suppress immune response. T_FH cells produce IL21, with key TFs being BCL6 and STAT3, and require IL-6 signaling and ICOS-L. CD8⁺ T cells serve a cytotoxic function, regulated by PRMD1 and TBX21, and release IL-2, IFN-γ, and TNF-α. These cells can be further differentiated into CD8⁺ T_EM and T_CM via EOMES and BCL6, respectively. APC, antigen-presenting cell; CLP, common lymphoid progenitor cell; IFN-γ, interferon gamma; T_CM, central memory T cell; T_EM, effector memory T cell; TGF-β, transforming growth factor β; TNF-α; tumor necrosis factor α.

Figure 3.

TCR signaling. T-cell activation is initiated by the interaction of TCR with peptide-MHC and coreceptors (ie, CD4 or CD8) and other costimulatory molecules including CD28, resulting in the multimolecular signalosomes, followed by the activation of several distal signaling pathways such as Ca²⁺-calcineurin-NFATs, PKCθ–IKK–NF-κβ, RASGRP1-RAS-ERK1/2, and TSC1/2-mTOR. The signaling complex containing scaffold adapter molecules, SLP-76 and GADs, which recruits PLCγ1 and tyrosine kinase ITKs. The tyrosine kinase ITK phosphorylates and activates PLCγ1, which then cleaves phosphatidylinositol (4,5) bisphosphate to generate 2 important second messengers, IP3 and DAG. IP3 binds to receptors on the ER, leading to an initial phase of calcium release, which subsequently activates the NFAT family of TFs, whose targets include many important cytokines, including IL-2 that are activated. Upon TCR activation, IL-2 receptor (IL-2R) activates JAK3, which in turn phosphorylates the beta-chain of IL-2R, recruiting JAK1. Both JAK1 and JAK3 in their active form phosphorylates (STAT), resulting in the nucleus transport. DAG activates several important proteins including several isoforms of PKC and RASGRP1 and RASGRP2, that are responsible for an initial phase of RAS activation, which is then sustained and amplified by SOS1. PKCθ binds to DAG and is recruited to the lipid rafts and triggers the formation of a trimolecular complex of adapter proteins in the cytoplasm called the CBM complex (CARMA1, BCL10, and MALT1), resulting in PKCθ–IKKβ–NF-κβ pathway activation. DAG induces the activation of another key molecule, RASGRP1, responsible for RAS activation in T cells, and initiates the RAS-MAPK cascade by activating the serine/threonine kinase Raf1 and activates MAPK ERK1 and 2. The costimulatory molecules CD28 mediates PI3K and VAV1 activation and further increases NF-κB and NFAT nuclear translocation, augmenting T-cell survival and production of the proliferative cytokine IL-2. Other TFs or pathways (p38MAPK, ERK1/2, and STAT3) are also crucial for T-cell activation and distinct functions associated with T cells. Several key genes involved in TCR are either genetically aberrant (highlighted in yellow text) or indirectly affected by other alterations in several PTCL subtypes as indicated below. Pathways that are relevant to PTCL entities are highlighted by brackets, followed by the PTCL entity in which the pathway is implicated in. CARMA1, caspase recruitment domain–containing membrane–associated guanylate kinase protein 1; DAG, diacylglycerol; ER, endoplasmic reticulum; ERK, extracellular signal–regulated kinase; IP3, inosine trisphosphate; MALT1, mucosa-associated lymphoid tissue translocation protein 1; mTOR, mammalian target of rapamycin; NFAT, nuclear factor of activated T cell; PKCθ, protein kinase C θ; PLCγ1, phospholipase Cγ1; RASGRP1, RAS guanyl nucleotide–releasing protein 1.

Figure 4.

Major PTCL subtypes and association with T_H cell states/cell-of-origin and genetic alteration. Although there is evidence from molecular or GEP studies indicating the association of transcriptional programs of major T_H subset with PTCL subtypes, the genetic data suggest that specific driver genetic lesions may skew a naïve CD4⁺ T cell toward a T_H differentiation to maintain that cell state and may take part with other genetic aberrations crucial for lymphomagenesis. It is also plausible that the genetic defects acquired along with transformation can modulate differentiation profiles and forced plasticity. Each colored box depicts mutation landscape and copy number data summarized from several recent genomic studies. The frequency of recurrent mutations/CN is arranged from left to right based on the frequency reported in the literature. Currently in-use and potential therapeutic targets are listed per cell state (Mereu et al, Feldman et al, Iqbal et al, Iqbal et al, Herek et al, Yu and Zhang, Sun et al, Liu et al, Bongiovanni et al, and Thakral et al¹⁸⁹). ALKi, ALK inhibitor; BV, brentuximab vedotin; CN, copy number.

Figure 5.

Gene expression–based molecular predictor for major PTCL entities. Amador et al used gene expression profiling for major PTCL subsets to differentiate each PTCL subtype including the novel molecular subtypes (PTCL-GATA3 and PTCL-TBX21). Shown are the expressions of known distinct transcripts for each patient in a molecular PTCL subtype (the detailed characteristics of these transcripts have been described in Amador et al¹⁰⁷). Key histopathological features and immunohistochemistry biomarkers for each PTCL entity are shown under each molecular subtype.