Mutations in PLCG1 and CARD11 are two of the most frequently observed in MF/SS. They appear to be mutually exclusive and occur in almost 30% of SS cases (13). These gain of function mutations increase downstream T-cell signalling specifically through enhanced NFkB, NFAT and AP1 transcriptional activity (12, 14). These transcription factors regulate the expression of genes involved in cell proliferation, survival and differentiation. Crucially, there is evidence that many of these variants induce constitutive activation of downstream T-cell signalling without T-cell stimulation (12, 14). Furthermore, CTLA4-CD28 and ICOS-CD28 gene fusions enhance CD28 dependent T-cell signalling, and RLTPR variants activate the NFkB pathway thereby increasing downstream TCR signalling (7, 13). The high prevalence of these of NFkB pathway gene variants in CTCL supported by functional data indicates that there is a critical selection pressure for activation of the NFkB pathway in the transformation of mature T-cells.

PLCG1 mutations have also been detected in other mature T-cell malignancies, notably HTLV-1 associated ATLL (25), PTCL(NOS) (26), hepato-splenic T-cell lymphomas (27) and AITL (28) with the PLCG1 p.S345F and R48W variants being two of the most frequently reported. In addition, mutations of the JAK-STAT, CD28, VAV1, DNMT3A and TET2 genes are also reported in other mature T-cell malignancies (4–9, 25–28).

Enhanced T-cell activation via the T-cell receptor and NFkB pathway, leads to downstream activation of multiple pathways including the Janus tyrosine kinase (JAK) Signal transducers and activators of transcription (STAT) pathway. These proteins have a multitude of functions including TĤ cell proliferation and differentiation, as well as gene regulation and epigenetic modification. Unlike the STAT proteins, gain of function mutations in the JAK proteins are infrequently observed. However, copy number gains of both STAT3 and STAT5B are common and associated with constitutive expression of these key transcription factors (29, 30).

Epigenetic changes include DNA and histone modifications which affect gene transcription and regulate cell differentiation. Furthermore, epigenetic modification is critical for sustaining the transcriptional memory for T-cells allowing rapid transcription of inducible genes upon activation (31, 32). Global hypomethylation is a consistent feature of malignancy and contributes to genomic instability but DNA methylation of gene promoters can lead to gene silencing. Crucially these changes are observed in MF/SS with evidence for hypomethylation of 7.8% of CpG sites in SS, and hypermethylation of 3.2% of CpG sites, specifically in the proximal region of promoters (33). There is extensive evidence that promoter hypermethylation leads to the silencing of specific tumour suppressor genes in MF/SS including those involved in cell cycle regulation (CDKN2A/2B) (34), DNA repair (MLH1 and MGMT) ( 35), apoptosis (FAS) (36) and JAK-STAT signalling (SHP-1) (37). Methylation of cytosine residues to 5-methylcytosine is mediated by DNA methyltransferases (DNMTs) and gain of function mutations of DNMT3A have been frequently identified in haematologic malignancies including MF/SS (4, 5, 8, 9). A second type of DNA methylation involves 5-hydroxymethylation of cytosine which is mediated by ten-eleven translocation 1-3 enzymes (TET1-3) but, in contrast to 5-methylcytosine, this is associated with enhanced gene expression (38). Loss of function TET2 mutations are well documented in SS (5, 7–9, 30). In SS, there is also mutational selection pressure for genes involved in other epigenetic modifications including IDH encoding isocitrate dehydrogenases which inhibit TET proteins, ARID1A/1B which affect chromatin modelling and MLL genes which mediate histone methyltransferases (4, 8, 9, 30). Large epigenomic studies in MF/SS have shown that the methylation pattern of leukemic T-cells in SS can be similar to that of regulatory T-cells and that there is almost universal activation of NFkB (39). Recent data suggest that hypermethylation of the hTERT promoter in MF/SS may be associated with telomerase activation (40).

MicroRNA (miRs), one of a group of non-coding RNA transcripts, are key post-transcriptional regulators of mRNA and are known to affect both the stability and translation of mRNA. In MF/SS, miR dysregulation has been observed with aberrant expression linked to abnormal DNA methylation of miR promoters as well as copy number changes (41).In addition, constitutive activation of STAT3/5 has been shown to enhance miR-155 and miR-21 expression leading to increased apoptosis resistance and Th2 proliferation (41).

The DNA damage response (DDR) consists of numerous complex and inter-dependent signalling pathways which either maintain cell viability by repair of DNA or direct the damaged cell to undergo senescence or programmed cell death. Inevitably this complex process is closely linked with pathways regulating the cell cycle, chromatin remodelling and apoptosis (42). Previous cytogenetic and array CGH studies in MF/SS identified complex structural and numerical chromosomal abnormalities (24). More recent WGS and WES studies have confirmed a high degree of genomic instability with over 60 gene aberrations reported across all 5 DDR pathways (5, 8–10). One study of 101 SS samples identified SNVs and/or CNVs affecting genes involved in DNA repair and telomere maintenance in over 50% of cases. Notably, SNVs and CNVs affecting genes involved in homologous recombination such as RAD51C, BRAC2 and POLD1, are also detected in MF/SS (9). TP53 is the most commonly mutated gene in CTCL with loss of function SNVs and deletions which lead to a significant detrimental effect on the DNA damage response and telomere stability.

Telomere dysregulation is a recognised feature of MF/SS where shortened telomeres have been observed (43, 44). Furthermore, mutations in POT1, encoding a telomere binding protein are also frequently detected in MF/SS and ATLL (4, 5, 8–11, 25, 30, 45) and studies suggest that these loss of function variants likely contribute to telomere dysfunction by abolishing telomere binding and inducing DNA damage at telomeres in the form of telomere induced foci (TIFs) (3, 46–51). Cell cycle dysregulation is a major contributor to tumorigenesis in MF/SS. Mutation or deletions have been reported in several cell cycle checkpoint and tumour suppressor genes including CDKN1, CDKN2A, CDKN2B, ATM, ATR, TP53, RB1 and PTEN (4, 5, 8–10). Loss of function ATM mutations have been reported in several T-cell lymphomas (25, 26, 52–60) including MF/SS (9–11) and ATR mutations are also observed in MF/SS and NK T-cell lymphoma (5, 8, 10, 11, 61). In view of the key role of these kinases in the DDR and in cell cycle regulation, pathway disruption either by gene mutation or as a result of telomere dysfunction is likely to contribute to the genomic instability seen in MF/SS.

Primary cutaneous CD30+ anaplastic large cell lymphoma (pcALCL) has similar genomic alterations to its systemic counterpart albeit at much lower frequency. Most fail to express anaplastic lymphoma kinase (ALK), but have an excellent prognosis (67). In a small proportion of patients mutations of JAK1 and/or STAT3 and NPM1-TYK2 gene fusions been reported in pcCD30+ALCL (68).

Chromosomal rearrangements involving the DUSP22-IRF4 (MUM1) locus on 6p25.3 have also been identified in both pcCD30+ ALCL (25%) and less commonly (5%) in lymphomatoid papulosis (LYP) (69), but MUM1 expression is not specific for this rearrangement.

In view of the immune dysregulation profile seen in advanced MF/SS with a diminished Th1 immune response (IFN-gamma, IL-12) and skewing towards a Th2 immune profile (increased IL-4, IL-5 and/or IL-13) (80), interferons (IFN-alpha and IFN-gamma) were amongst the first immunotherapies to be used. IFN-alpha has been shown to stimulate antitumour cytotoxicity by activating CD8+ T-cells and NK cells and restores the Th1/Th2 balance by reducing IL-4 and IL-5 production by malignant T-cells (81, 82). IFN-gamma acts similarly to IFN-alpha by activating CD8+ and NK cells and increasing Th1 cytokine profile (80). IFN-alpha is EMA approved and effective in early stage patients who are refractory to skin-directed therapies (83), in which case it can be combined with phototherapy (84). Several clinical trials have highlighted the potential use of recombinant IL-12 as a novel immunotherapy with encouraging results (85–87).

There are currently several trials underway targeting checkpoint molecules (88, 89). Mogamulizumab, an anti-CCR4 monoclonal antibody has recently been approved by FDA and EMA following a phase III clinical trial which showed increased progression free survival (PFS) compared with the HDAC inhibitor, Vorinostat (21). Recently, it has emerged that Mogamulizumab also contributes to efficient immune restoration involving CD8+ as well as stem and memory CD4+ cells (90).

Pembrolizumab and Nivolumab inhibit the PD-1 receptor which enhances cytotoxic T-cell killing and have shown clinical responses in phase I and II trials (23, 91, 92). Whilst the high mutational burden of MF/SS would suggest that immunotherapies should be successful, the modest responses from these phase II trials (93) highlight the challenges of using a PD-1 inhibitor on a T-cell lymphoma which can express PD-1. Specifically there are two scenarios: PD-1 expression might reflect underlying gain of function mutations which would benefit from inhibition and there is emerging data suggesting that PD-1 mutations can be detected in some MF/SS patients (94), or PD-1 inhibition might reverse the tumour cell exhaustion leading to activation and proliferation of malignant T-cells. Whilst this second scenario has not yet been seen in MF/SS patients receiving PD-1 inhibitors, a phase II trial of Nivomumab in ATLL was discontinued because of rapid disease progression (95).

Targeting of CD47 with intralesional or intravenous TTI-621 has been used in patients with relapsed or refractory MF/SS in recent phase I trials with encouraging results (96, 97). Recent reports have shown that the CD39-CD73-adenosine pathway generates an immunosuppressive tumour microenvironment in SS and this provides a potential option to use emerging novel therapeutic approaches to target this pathway possibly in combination with checkpoint inhibitors or Mogamulizumab (98, 99).

Perhaps the most interesting option would be tumour agnostic therapies targeting the DNA damage response (DDR) pathways which show considerable promise in various solid malignancies often as maintenance therapies after platinum-based chemo regimens (42). The DDR pathways are a vast network of over 450 proteins. These therapeutic approaches build on the success of PARP inhibitors inducing synthetic lethality in homologous recombination (HR) deficient malignancies (due to BRCA1 or BRCA2 loss). In view of the somatic mutations or deletions of HR genes (including ATM, BRCA1, BRCA2, Chk2, RAD50, RAD51C), these could be repositioned for use in MF/SS patients ( Figure 1 ). Several ATR inhibitors are in early clinical development for use in both solid and haematological malignancies (100, 101). There is increasing evidence that ATR inhibition may be a potential therapeutic target in MF/SS. Small molecule inhibitors of ATR and Chk1 (VE-821/2 and Chir-124) have been shown to sensitise MF/SS cell lines to phototherapy by inducing apoptosis (17). In addition, cells overexpressing POT1 mutants (p.F62V and p.K90E) treated with an ATR inhibitor (ETP-46464) resulted in significant abrogation of TIFs (3). ATM inhibition has also been shown to overcome HDAC inhibitor resistance in both B and T-cell derived lymphomas including MF/SS, providing a rationale for combination therapy (102). Acting immediately downstream of ATR, targeting of Chk1 is effective in MYC-driven tumours including B-cell lymphomas owing to MYC-induced replication stress (103). Amplification of the MYC oncogene is one of the most commonly observed aberrations in MF/SS (4) which could increase replication stress in these cells and hence increase their sensitivity to ATR and/or Chk1 inhibitors. Furthermore, Chk1 inhibitors synergise with a number of therapeutic agents to induce cell death in MF/SS including the proteasome inhibitor, Ixazomib (104) and phototherapy (17).