Cutaneous T cell lymphoma PDF

PRiMeR Cutaneous T cell lymphoma Reinhard Dummer1,2 ✉, Maarten H. Vermeer 3, Julia J. Scarisbrick 4, Youn H. Kim...

PRiMeR Cutaneous T cell lymphoma Reinhard Dummer1,2 ✉, Maarten H. Vermeer 3, Julia J. Scarisbrick 4, Youn H. Kim5, Connor Stonesifer6, Cornelis P. Tensen3, Larisa J. Geskin6, Pietro Quaglino7 and Egle Ramelyte1,2 Abstract | Primary cutaneous T cell lymphomas (CTCLs) are a heterogeneous group of lymphomas that present in the skin with no evidence of extracutaneous disease at the time of diagnosis. CTCL subtypes demonstrate a variety of clinical, histological, and molecular features, and can follow an indolent or a very aggressive course. The underlying pathogenetic mechanisms are not yet entirely understood. The pathophysiology of CTCL is complex and a single initiating factor has not yet been identified. Diagnosis is based on clinicopathological correlation and requires an interdisciplinary team. Treatment decision is made based on short-term and long-term goals. Therapy options comprise skin-directed therapies, such as topical steroids or phototherapy, and systemic therapies, such as monoclonal antibodies or chemotherapy. So far, the only curative treatment approach is allogeneic haematopoietic stem cell transplantation. Novel therapies, such as chimeric antigen receptor T cells, monoclonal antibodies or small molecules, are being investigated in clinical trials. Patients with CTCL have reduced quality of life and a lack of effective treatment options. Further research is needed to better identify the underlying mechanisms of CTCL development and course as well as to better tailor treatment strategies to individual patients. Primary cutaneous T cell lymphomas (CTCLs) are a be difficult as neoplastic cells constitute only a minority heterogeneous group of T cell lymphomas that initially of the infiltrate in the skin and clinical and histological present in the skin with no evidence of extracutaneous findings may overlap9. disease1. Subtypes of CTCL vary widely in terms of Sézary syndrome (SS) is a rare and aggressive type 1 Dermatology Department, biology, histopathology and clinical features (Table 1). of CTCL that is defined by the triad of erythroderma, University Hospital Zurich, Zurich, Switzerland. After multiple studies found that primary cutaneous generalized lymphadenopathy and the presence of neo- 2 Faculty of Medicine, lymphomas and primary nodal lymphomas of the same plastic clonal T cells (Sézary cells) in skin, lymph nodes University of Zurich, histological subtype differ in clinical behaviour and and peripheral blood. SS tends to progress more rap- Zurich, Switzerland. prognosis, a new classification for cutaneous lympho- idly than MF and has a worse prognosis2. Upon disease 3 Department of Dermatology, mas was established in 1997. The most recent iteration progression, patients with MF and SS typically experi- Leiden University Medical of this classification was published in 2018 in the fourth ence a worsening state of immune suppression with an Centre (LUMC), Leiden, edition of the WHO Classification of Skin Tumours increased risk for infections and a decreased antitumour The Netherlands. Blue Book2,3. immune response10. 4 Centre for Rare Diseases, University Hospital of Mycosis fungoides (MF), the most common type of CD30-positive lymphoproliferative disorders (CD30+ Birmingham, Birmingham, CTCL, usually presents with patches and plaques and LPDs) are a spectrum of disorders including self-limiting United Kingdom. generally has an indolent clinical course. MF has a ten- lymphomatoid papulosis (LyP) and primary cutaneous 5 Department of Dermatology, dency to progress over years or even decades to more anaplastic large-cell lymphoma (pcALCL) (Table 1). Stanford University School of infiltrated plaques and tumours4; progression occurs in CD30+ LPDs usually have an indolent clinical course; Medicine, Stanford, CA, USA. 25% of patients with early-stage disease5. Extracutaneous however, patients have a higher risk for second lymphoid 6 Department of Dermatology, disease develops in a minority of patients and varies malignancies compared with the general population and Columbia University, according to the stage of disease at diagnosis; risk of therefore require regular follow-up and assessment. New York, NY, USA. extracutaneous spread was 0% in patients presenting This Primer discusses CTCL with a primary focus on 7 Department of Dermatology, University of Turin, Turin, Italy. with limited skin involvement at first diagnosis and was the most common subtype (MF) and the most common ✉e-mail: ~40% in those presenting with erythroderma at first aggressive subtype (SS). The Primer covers the epidemi- [email protected] diagnosis5–8. In the early stages of MF, that is, patch or ology, pathogenesis, diagnosis, management and qual- https://doi.org/10.1038/ plaque stage (or stage IA–IIA) disease, the differential ity of life (QOL) of patients with MF and SS and briefly s41572-021-00296-9 diagnosis with benign inflammatory skin conditions can discusses other CTCL subtypes such as LyP or pcALCL. NATure RevIewS | DISEASE PRIMERS | Article citation ID: (2021) 7:61 1 0123456789();: Primer Table 1 | Common clinical presentations of WHO-EORTC CTCL subtypes CTCL subtype Estimated Clinical presentation 5-year Refs frequency DSS2 (% of CTCL)2 MF 39 Erythematous patches, plaques, tumours or erythroderma often with fine scaling and 88 2,222 epidermal atrophy; may be hypopigmented and/or hyperpigmented lesions; pruritus is common; lesions occur in ‘Bathing suit’ distribution (non-sun-exposed areas) FMF 5 Grouped, erythematous, follicular papules, erythematous papules and plaques 75 2,223 with associated alopecia, and acneiform lesions; commonly affects face and neck (especially eyebrows) PR 6:1 in a lesion had a 90% not meeting criteria for leukaemia) (Table 1). MF lesions specificity for MF in retrospective studies130,145,146,148,149. most commonly present as erythematous patches with To aid assessment of histologically uncertain cases and to fine scaling (loss of the upper layer of the epidermis in inform prognosis, immunohistochemical assessment of small flakes) and epidermal atrophy16,133. Patches may be CD25, CD56, TIA1, granzyme B, CXCL13, ICOS and hypopigmented and/or hyperpigmented and often occur PD1 are also recommended130,145,146. These markers in non-sun-exposed areas, although presentation in any are important as some may help in prognostication skin area is possible. Lesions are often pruritic (itchy). (CD8+CD56+, TIA1 and granzyme B) and others may Of note, BSA may not always reflect the severity of MF be considered as targets for therapies (PD1 and PDL1, as the presence of few tumours may signify an inferior denileukin diftitox (CD25)); however, so far, apart from prognosis compared with patches and plaques taking CD30 in treatment with brentuximab vedotin, there up the same BSA. The Modified Severity Weighted are no universally accepted markers to guide treatment Assessment Tool (mSWAT) was created to overcome choices. TCR clonality should also be evaluated as part these shortcomings (Supplementary Table 4) and was of the diagnostic work-up for MF. found to better correlate with prognosis116,134,135. The A substantial reduction in the number of circulating mSWAT involves multiplying the measured BSA of CD8+ T cells (80% of BSA), general- 10% of cases, wherein the disease will typically spread ized lymphadenopathy and clonally related neoplastic to regional lymph nodes162. Although pcALCL gener- T cells in the skin, lymph nodes and peripheral blood. ally has an excellent prognosis, advanced-stage pcALCL Pruritic erythroderma presents with exfoliation, oedema (at least T3) has a 5-year survival of 77%163 and there- and lichenification. The clinical features of SS can be diffi- fore treatment rather than observatory monitoring is cult to distinguish from those of erythrodermic inflam- indicated. matory dermatoses; therefore, assessment for peripheral blood involvement is the gold standard for diagnosis. Biopsy and laboratory evaluation. By definition, CD30+ As patients with SS have an increased risk for relapse, LPDs present with CD30+ T cells detectable using immu- mostly cutaneous infections, this should also be assessed nohistochemistry. As subtypes of MF, LyP and pcALCL at the time of diagnosis and during each clinical visit. may all present with CD30+ T cell infiltration, further immunohistochemical labelling, such as cytotoxic mark- Laboratory evaluation. In addition to routine labora- ers TIA1 and granzyme B, both of which can be expressed tory assessments, flow cytometry must be sent to con- in pcALCL and LyP, and clinicopathological correlation firm clonality in the peripheral blood and assess the are key to a correct diagnosis. LyP and pcALCL do not phenotype of potential neoplastic cells. Malignant SS show blood involvement; however, patients with these cells are clonal and often have a CD3+CD4+ and CD8– disorders should undergo regular monitoring (compris- phenotype. They also commonly lose expression of CD7 ing complete blood count, a comprehensive metabolic and/or CD26 (refs155–157). In fact, in peripheral blood, panel and LDH) owing to their increased risk of second a loss of CD26 in >80% of CD4+ T cells has a sensi- lymphoproliferative malignancies. tivity of 83% for SS, whereas loss of CD7 in >40% of For a diagnosis of pcALCL, histopathologically, at CD4+ T cells has a sensitivity of 98%, with both metrics least 75% of the neoplastic cells should express CD30 achieving 100% specificity158. in tissue from lesion biopsy164. Most neoplastic cells in KIRDL2 (CD158k) is a natural killer cell marker pcALCL are CD4+165 but, in rare cases, the neoplastic recently found to have a sensitivity of 88.6% and spec- cells are CD8+CD30+166. In contrast to systemic ana- ificity of 96.3% in detecting SS cells and may be use- plastic large-cell lymphoma, pcALCL T cells are neg- ful in supporting a diagnosis159. In addition, numerous ative for CD15 and epithelial membrane antigen162,167. driver genes have been identified that may provide Importantly, pcALCL usually does not express anaplastic future opportunities for further diagnostic specificity lymphoma kinase 1 (ALK1) or the t(2;5) chromosomal (see ‘Genetic and molecular alterations’ above). For the translocation found in nodal ALCL168. diagnosis and staging of SS, WHO-EORTC and ISCL The three main histological subtypes of LyP are A, B guidelines require both confirmation of a TCR clone in and C. The infiltrate is usually wedge-shaped with ulcer the blood or skin as well as the following findings: an formation. The large, atypical cells of type A LyP resem- absolute Sézary cell count of >1,000/µl or an expanded ble Reed–Sternberg cells. These cells are surrounded by CD4+ T cell population with a CD4 to CD8 ratio of ≥10:1 neutrophils and eosinophils. Type B cells resemble MF, and either ≥40% CD4+CD7− cells or ≥30% CD4+CD26− with lichenoid lymphocytic infiltration of cells with cells in peripheral blood (Supplementary Table 1). As cerebriform nuclei and some epidermotropism169. The the clinical and histological presentation of SS is not neoplastic cells in type C LyP resemble ALCL, with specific for SS, flow cytometry and identification of sheets of large CD30+ cells in the infiltrate169. The histo­ clonally related neoplastic T cell populations are crucial logical distinction between LyP and other histologically for diagnosis. similar lymphoproliferative conditions may be difficult and clinical correlation is required169. About 10% of CD30+ LPDs patients with LyP develop another lymphoma, often Clinical features. LyP typically presents as recurrent, ery- MF. In addition, a higher incidence of lymphoid and thematous dome-shaped papules occurring in cropped non-lymphoid malignancies is observed in patients or generalized eruptions on the trunk or proximal with LyP170–172. In some cases, the distinction between extremities. The lesions often spontaneously regress after LyP and pcALCL cannot be made because of discrep- a few weeks, during which time they may involute, ulcer- ancies between clinical features and histological appear- ate and possibly scar. LyP typically has a chronic course. ance. These cases are referred to as borderline lesions 10 | Article citation ID: (2021) 7:61 www.nature.com/nrdp 0123456789();:  Primer and their classification should consider their clinical undergo irradiation in the presence of a photosensitiz- behaviour and appearance. ing agent, followed by reinfusion into the patient), reti- noids, low-dose methotrexate, total-skin radiotherapy, Rare aggressive CTCLs monochemotherapy (gemcitabine, pegylated liposo- Other CD30– non-MF/SS subtypes of CTCL are rare mal doxorubicin), multi-agent chemotherapy (CHOP/ (Table 1). These disorders comprise less than 1% of total CHOEP regimens) and allogeneic haematopoietic stem CTCL diagnoses but may be highly aggressive and there- cell transplantation (HSCT). Treatment choice depends fore important to consider in any CTCL work-up. The on disease subtype, patient age, the presence of comor- workflow for diagnosis of these subtypes is generally bidities, disease extension, and staging and treatment consistent with that described above. Staging occurs availability21,174 (Box 1; Fig. 4). A multidisciplinary group via ISCL/EORTC TNM guidelines for non-MF/SS of different specialties (dermatologists, pathologists, subtypes173. onco-haematologists and radiotherapists) is required particularly in advanced stages and aggressive subtypes. Management Basic principles Mycosis fungoides Treatment strategies for CTCL vary widely from International guidelines for the treatment of MF ‘wait-and-see’ approaches, skin-directed therapies (EORTC 2017, ESMO 2018, BAD2018 and NCCN) (SDTs; namely, topical steroids, chlormethine gel, photo- divide treatments into first-line options and treatments therapy and local radiotherapy) and systemic therapies. to be considered after but do not recommend any par- Systemic therapies include immune-modifiers (inter- ticular order of individual therapies owing to lack of feron, extracorporeal photopheresis (ECP) wherein evidence from clinical trials16,59,175,176 (Fig. 5; Table 2). patient blood is extracted and separated leucocytes The distinct features of early versus advanced MF drives the treatment strategy. In early MF, the decision to treat relies on patient age, lesion site, symptoms and Box 1 | The cornerstones of CTCL management disease kinetics. The aim of therapy is to substantially To treat or wait-and-see improve clinical and health-related QOL (HRQOL), In indolent cases, in the presence of limited skin involvement, for example, early- rather than the complete eradication of skin lesions. stage T1a mycosis fungoides (MF) or lymphomatoid papulosis, the decision can be Real-world data from patients with early-stage MF13 ‘wait-and-see’ also based on the site of the lesions, disease evolution, age of the patient showed that the SDTs with topical steroids and photo- and symptoms associated. A wait-and-see approach has not been demonstrated to be therapy were the first-line therapy in 81.6% of patients, associated with a worsening of disease course and survival. whereas the wait-and-see approach was used in 7.3% of Treat what you see patients. Other patients (11.1%) received first-line reti- There is no evidence that aggressive treatments could modify the disease course and noids, IFN or combination with phototherapy, most of survival when given in the early stage of disease; therefore, the indication is to tailor whom had plaque stage or folliculotropic MF13. Of note, the treatment focusing on disease extension in the skin and extracutaneous sites. the response rate to first-line SDTs (73%) is superior to Thus, accurate staging procedures should be developed according to the disease entity to allow the correct management of the patient. that of systemic treatments (57%), meaning that it is conceivable to start with an SDT even in patients with SDTs versus systemic therapies worse prognostic factors18,25. Skin-directed therapies (SDTs) represent very active and well-tolerated treatment Advanced MF has a lack of effective treatments, as options that need to be performed in all cases whenever indicated, alone and before systemic therapies. In advanced stages or aggressive subtypes, SDTs can be given available treatment options have a low response rate, in association with systemic therapies to improve the response or to better manage and a short duration of response. Moreover, there is a localized cutaneous lesions. lack of effective maintenance treatments, and the avail- able treatments should be assessed for the relevance of Up-grade and down-grade of treatments adverse effects in elderly patients and severe HRQOL The majority of cutaneous T cell lymphoma (CTCL) (in particular MF) show an indolent disease course spanning over decades; therefore, it is important to carefully evaluate impairment177. A wide range of treatments for advanced treatment options to preserve potentially active treatments when they are really MF was reported in one retrospective international needed. In progressive cases, an up-grade of treatments needs to be conducted study, with more than 25 different first-line treatments (that is, in MF, from phototherapy to treat patches and plaques to interferons, retinoids and 38.9% of patients receiving 4 or more therapies28. and total skin electron therapy, up to monoclonal antibodies and chemotherapy Stage IIB disease was most frequently treated by SDTs, for treatment in the tumour stage); however, in case of response, a down-grade of bexarotene and gemcitabine, patients with erythroder- treatments is warranted (that is, phototherapy to treat residual patches after response). mic MF or SS were most frequently treated with ECP, and Re-challenge/re-treatment patients with stage IVA2 were most frequently treated by In contrast to what occurs in nodal lymphomas, in CTCLs, there is the possibility of polychemotherapy28. Multivariate analysis identified age re-challenge (that is, the use of a treatments already performed in the clinical history and stage as prognostic factors and monochemotherapy of the patient after other approaches) or simply re-treatment (the use of the same last or polychemotherapy as first-line treatment were asso- treatment performed) (Fig. 3). ciated with an increased risk of death and/or a change Quality of life of therapy, suggesting that chemotherapy should be used As these diseases primarily involve the skin and have a long duration and, in the only after failure or relapse of previous approaches in majority of patients, skin represents the only site of involvement, the impact of those with aggressive disease and where other options cutaneous lesions on visible parts of the body is fundamental for the patient and are not indicated28. However, the retrospective nature of implies that the maintenance and preservation of quality of life is a cornerstone this analysis poses a risk of selection bias and should be in the treatment strategy. interpreted with caution. NATure RevIewS | DISEASE PRIMERS | Article citation ID: (2021) 7:61 11 0123456789();: Primer a b Topical c PUVA steroids Debulking PUVA chemotherapy f TSET d e Fig. 4 | The re-challenge paradigm of mycosis fungoides therapy. a | Mycosis fungoides (MF) presenting as patch and solitary papules. b | Re-challenge paradigm of MF therapy. c | MF in patch stage. d | MF in tumour stage. e | Ulcerated MF patches. f | Widespread MF patches and plaques. Re-challenge (that is, the use of treatments already performed in the clinical history of the patient after other approaches) or simply re-treatment (the use of the same last treatment performed) can be used for cutaneous T cell lymphoma when deemed suitable. PUVA, psoralen plus UVA; TSET, total skin electron therapy. Two recent multicenter randomized trials led to the regimens), itching and severe QOL impairment2,59,175,176,182. approval of two new drugs for CTCL: brentuximab and The recommendations for the treatment of SS include mogamulizumab. The ALCANZA trial enrolled 128 pre- ECP as first-line therapy, where possible, in patients with treated patients with CD30+ MF or pcALCL to receive a low tumour burden in the blood and as maintenance brentuximab, methotrexate or bexarotene178. This study therapy after remission with more aggressive therapies183, found a higher ORR4 (that is, overall response lasting for alone or in combination with other agents such as bex- at least 4 months) in the brentuximab group (56.3% vs arotene or other retinoids, IFN, and TSET. The recently 12.5%), with a higher response in a subgroup analysis of approved drug mogamulizumab was demonstrated to those with tumour-stage MF. The other multicentre reg- induce significant responses particularly in patients with istrative study was the phase III MAVORIC, which com- SS and in the blood compartment179 and can thus be used pared the anti-CCR4 treatment mogamulizumab with as a debulking agent and as a bridge towards transplan- vorinostat179. In this study, progression-free survival was tation. Other therapies for SS include mogamulizumab significantly higher in the mogamulizumab group com- (which can induce a high response rate particularly in pared with the vorinostat group, with better responses in the blood) and chemotherapy (namely, gemcitabine patients with SS (37%) and patients developing a response and liposomal doxorubicin). Multi-agent chemother- in the blood (68%) associated with a significant duration apy treatments do not lead to a significant increase in (median 25.5 months in the blood and 20.6 months response rate compared with monochemotherapy, result- in the skin)179. In a post hoc analysis, patients with a ing instead in burden by greater systemic toxicity16,59,175,176. B1 and B2 blood-score responded better than those with Allogeneic HSCT, albeit in selected patients, is the only B0 scores in terms of cutaneous mSWAT180. A third com- available treatment with a truly curative intent184. pound, topical chemotherapy chlormethine gel, has also been approved by the EMA for the treatment of early CD30+ LPDs phases of disease or, in those with advanced disease, in Management of indolent CD30 + LPDs is highly combination with systemic therapies181. dependent on the spread of the lesions. In those with solitary and infrequent LyP, a wait-and-see approach, Sézary syndrome surgery or radiotherapy can be used185. In those with Treatment of SS is challenging owing to the aggressive widespread or recurrent LyP lesions, weekly low-dose disease course, risk of relapsing infections (which can methotrexate achieves responses in up to 100% with be increased by chemotherapy and immunosuppressive complete response in 34%; however, relapses occur 12 | Article citation ID: (2021) 7:61 www.nature.com/nrdp 0123456789();:  Primer in over half of the patients and methotrexate does not Allogeneic haematopoietic stem cell transplantation seem to reduce the recurrence rate185. Brentuximab, an Allogeneic HSCT is a potentially curative treatment for antibody–­drug conjugate, composed of CD30-antibody advanced-stage MF and SS. In one series of 113 patients, linked to an anti-tubulin agent monomethyl auristatin E, allogeneic HSCT induced a sustained clinical benefit showed an overall response rate of 100% and complete in a significant percentage of patients, with a 5-year over- response of 58% in refractory LyP185. all survival of 38% and 5-year progression-free survival of Similar to LyP, surgery or radiotherapy are the treat- 26%190. Adverse prognostic factors were advanced-phase ments of choice for those with solitary pcALCL lesions. disease, refractory, relapsing or progressive disease after Radiotherapy with a 30–40 Gy dose can achieve responses repeated therapy lines, and an unrelated donor190. Similar in up to 95% of patients2,16,59,175,176. Although responses results were reported in other studies191, with a 5-year can be durable, relapses are common185,186. In those with progression-free survival of 17% and overall survival of widespread or recurrent disease, systemic treatment is 32% in 129 patients with relapsed or refractory MF or SS recommended. As with LyP, low-dose weekly methotrex- and a 2-year progression-free survival of 31% and over- ate is a safe and effective strategy for pcALCL; however, all survival of 57% in 37 pretreated patients in another brentuximab demonstrated a 75% overall response rate study192. Patients with MF as well as SS were suggested to and a 31% complete response rate, compared with 33% present long-term outcome benefit191. overall response rate and 7% complete response rate for Owing to the high post-transplant relapse rate and either methotrexate or bexarotene178. the adverse effects of the procedure, allogeneic HSCT is a treatment option in young patients with aggressive Rare aggressive CTCLs advanced relapsed CTCL without comorbidities. Open Other rare CTCL subtypes, such as primary cutaneous challenges are the timing of allogeneic HSCT in the dis- aggressive epidermotropic cytotoxic CD8+ T cell lym- ease course, the early identification of best candidates and phoma or primary cutaneous γδ T cell lymphoma, are the relevance of disease control before transplantation184. aggressive and have a poor prognosis16. Interdisciplinary approaches with dermato-oncologists, haematologists New drugs and trials and radiation oncologists are crucial to achieve best As frequent targetable driver mutations in CTCL have outcomes in these patients. Multi-agent chemotherapy, not been identified, most of the new systemic ther- such as cyclophosphamide, doxorubicin, vincristine, and apeutic agents under investigation are monoclonal prednisone (CHOP) and CHOP-like regimens are the antibodies, histone deacetylase inhibitors, proteasome preferred first-line treatments and, in eligible patients, inhibitors, immune-checkpoint inhibitors or chimeric such as young patients with aggressive advanced relapsed antigen T cell receptor strategies, administered alone CTCL without comorbidities and a good performance or in combination (Table 3)193–195. Due to the rarity of status (ECOG), allogeneic HSCT should be considered. the disease, randomized clinical trials are still uncom- However, even with treatment, most cases are fatal2,187,188. mon in CTCL+; therefore, homogeneity among trials is In selected patients, high-throughput molecular analysis extremely important to allow consolidation or compar- with possible identification of molecular targets, such as ison of trial data on specific treatments. Accordingly, the CD30 or PD1, for off-label treatment attempts should be ISCL has proposed a consensus statement for clinical discussed. Palliation using radiotherapy of single lesions endpoints and response criteria in MF and SS196 and can be applied189. EORTC proposed new flow cytometry-based criteria Stage Wait and see Topical GCS CG Phototherapy Local RT TSET IFN Bexarotene BV MTX ECP Mogamulizumab Mono ChT Poly ChT Allo-HSCT IA IB IIA IIB III SS IVA–IVB Recommended therapies Alternative therapies Not recommended Fig. 5 | Recommendations for MF/SS treatment, based on EORTC approved for patients with relapsed or refractory disease who received guidelines. The dark green boxes identify treatments recommended for the at least one prior systemic therapy); CG, chlormethine gel; ChT, respective stage; lighter green boxes identify subsequent and alternative chemotherapy; ECP, extracorporeal photopheresis; GCS, glucocorticoster- choice; and grey boxes represent treatment options that are not generally oids; IFN, interferon; MTX, methotrexate; RT, radiotherapy; SS, Sézary recommended. The red bar separates early vs advanced-stage mycosis syndrome; TSET, total skin electron therapy. Mogamulizumab is approved fungoides (MF). Allo-HSCT, allogeneic haematopoietic stem cell for patients with relapsed or refractory disease who received at least one transplantation; BV, brentuximab vedotin (in cases with CD30 expression, prior systemic therapy. NATure RevIewS | DISEASE PRIMERS | Article citation ID: (2021) 7:61 13 0123456789();: Primer Table 2 | Summary of MF treatment results from the main studies Drug Phase No of Inclusion Response rate Disease outcome Drug patients approval Narrow-band Review236 251 IA–IIA 87.6%; CR 62.2 35% at average Not applicable UVB follow-up of 77 weeks Review237 >300 IA–IIA 54–90% (median 87%) 29–80% relapses Not applicable PUVA Review236 527 IA–IIA 90.9%; CR 73.8 20–100% relapse Not applicable Review237 400 IA–IIA 65–85% superior to UVB 40–50% relapses Not applicable in plaques Low-dose TSET Pooled three phase II 33 IB–IIIA 88%; CR 27% Median duration Not applicable trials238 of clinical benefit: 70.7 weeks PUVA plus IFNα2a Randomized clinical 98 IA–IIB (82 patients CR 70% in PUVA plus Shorter time to CR in FDA and EMA vs acitretin plus trial239 evaluable) IFNα2a vs 38% in IFNα2a PUVA plus IFNα2a IFNα2a plus acitretin Bexarotene Phases II–III240 58 IA–IIA 54–67%; CR 7–27% Duration of response up FDA and EMA to 64.7 weeks Bexarotene Phases II–III201 94 IIB–IV 45–55%; CR 2–13% Duration of response FDA and EMA 42–55 weeks Bexarotene/MTX Phase III randomized 64 CD30+ MF or 12.5% (ORR4) mPFS 3.5 months FDA and EMA control arm pcALCL (ALCANZA trial)29 Bexarotene plus Phase III randomized 93 IB–IIA 77% in combination mDOR 9.7 months for Combination PUVA vs PUVA alone241 vs 71% in PUVA alone; PUVA vs 5.8 months for not approved trend towards fewer the combination PUVA sessions to CR in the combination (median 22 sessions) Brentuximab Phase III randomized 64 CD30+ MF or 56.3% vs 12.5% (ORR4); mPFS: 16.7 months vs FDA and EMA vedotin vs bexarotene or MTX pcALCL MF IIB: 63%; CD30+ 3.5 months (ALCANZA trial)178 anaplastic: 75% Pegylated Prospective 49 IIB, IVA, IVB 40.8%; CR 6.1% mTTP: 7.43 months; FDA and EMA liposomal multicentre phase II242 mDOR: 6 months doxorubicin Gemcitabine Phase II multicentre243 32 IB–IV MF/SS, 75%; CR 2% Median duration of CR Not approved PTCLU was 10 months (range, 4–22 months) ECP Review183 407 SS/erythrodermic Median response rate mDOR 22 months Not applicable MF 63% (range 31–86%); (up to 11 years) median CR 20% (0–62%) Mogamulizumab Phase III randomized 372 MF/SS stage IB–IV 28% vs 5%; response mPFS 7.7 months vs 3.1 FDA and EMA vs vorinostat179 with at least one rate in SS 37%; 68% in months; P 185 days); FDA trial244 pretreated with at relief) mTTP 4.9 months, least two lines, one 9.8 months stage IIB with bexarotene or higher responders Romidepsin Pivotal, single-arm, 96 Stage IB–IVA 34%, 38% IIB–IV pruritus mDOR 15 months FDA open-label, phase II245 pretreated relief 43% Mechlorethamine Randomized, 260 IA–IIA Gel vs ointment 58.5% vs 90% of responses FDA and EMA gel observer-blinded, trial 47.7% (non-inferiority) maintained for at mechlorethamine gel least 10 months, no vs ointment181 treatment difference CR, complete response; ECP, extracorporeal photopheresis; mDOR, median duration of response; MF, mycosis fungoides; mPFS, median progression-free survival; mTTP, median time to progression; MTX, methotrexate; NR, not reported; ORR4, overall response lasting for at least 4 months; pcALCL, primary cutaneous anaplastic large-cell lymphoma; PTCLU, peripheral T cell lymphoma unspecified; PUVA, psoralen plus UVA; SS, Sézary syndrome; TSET, total skin electron therapy. for responses in the blood, which still need prospective for CTCL trials. Time to next treatment represents the verification158. interval from the initiation of one treatment to that of Given that, with the exception of curative-indent the next line of therapy and can be used as a surrogate allogeneic HSCT, available treatments lead to incom- for the duration of clinical benefit197. Furthermore, the plete responses and disease palliation, time to next evaluation of QOL could be incorporated as an endpoint treatment has been suggested as a meaningful endpoint in clinical trials. 14 | Article citation ID: (2021) 7:61 www.nature.com/nrdp 0123456789();:  Primer As the tumour microenvironment and the host Circulating neoplastic clones could represent a potential immune response (in particular with new immune- biomarker and a promising new target for therapy198. checkpoint inhibitors) is modulated by the tumour itself and can contribute to the different disease course, future Adverse effects and QOL treatment approaches should target not only the tumour Short-term and long-term toxicity should always be cells but also the tumour microevnironment. Moreover, considered when planning treatment. Of note, SDTs are the finding of multiple neoplastic circulating clones related to local, mostly transient, manageable adverse using whole-exome sequencing in early-stage MF and events199. When choosing a systemic therapy, comorbid- which continuously replenish the skin lesions increas- ities (for example, interferon or liposomal doxorubicin ing their heterogeneity with a tumour seeding mecha- should be avoided in patients with cardiovascular dis- nism, could constitute the rationale for the clinical use eases) and treatment availability should be considered200. of systemic treatments even in the early stage of disease. Patient education and regular follow-up allows the Table 3 | Summary of main new drugs investigated in CTCL Agent Mechanism of action Route of Ongoing trials administration TLR7 and TLR8 agonist ↑ IFNα production by pDCs and Topical Phase II – NCT03292406 (resiquimod) increased TH1 response SGX301 (a synthetic hypericin, Visible light-induced phototherapy Topical Phase III – NCT02448381 ‘FLASH’) with synthetic hypericin Microneedle array – Direct tumour cell killing through Topical Phase I – NCT02192021 doxorubicin (MNA-D) DNA damage Talimogene laherparepvec Oncolysis, induction of host il Phase I – NCT03458117, antitumour immune responses Phase II – NCT02978625 Anti-CD47-ab (TTI-621) ADCP il, iv Phase I – NCT02890368, Phase II – NCT03763149 Anti-KID3DL2-ab (IPH4102, ADCC and ADCP iv Phase II – NCT03902184 latucamab) Recombinant cytotoxic fusion Cell protein synthesis inhibition il Phase III – NTC10871727 protein comprising DT and hIL-2 (E7777) MRG-106 (cobomarsen) Inhibits miR-155 and T cell il, iv Phase I – NCT02580552 (il), proliferation Phase II – NCT03713320 (iv) Anti-PDL1-ab (atezolizumab) Immune-checkpoint inhibition iv Phase II – NCT03357224 Autologous CAR T cells Cell-therapy through CD30-directed iv Phase I – NCT03602157 (ATLCAR.CD30.CCR4/ATLCAR. and CCR4-directed allogenous CAR CD30) T cells Anti-PD1 (sintilimab) and Immune-checkpoint inhibition; iv or oral Phase II – NCT04296786 HDACi (chidamide) histone deacetylase enzyme inhibition AFM13 (CD30/CD16A Redirects NK cells to iv Phase II – NCT04101331 bispecific ab) CD30-expressing tumour cells PI3K inhibitor (duvelisib) PI3K inhibitor: cell cycle regulation, oral, iv, sc Phase I – NCT02783625 with HDACi (romidepsin) apoptosis, DNA repair, senescence, or proteasome inhibitor angiogenesis and cell metabolism (bortezomib) PI3K inhibitor (duvelisib) and Cell cycle regulation, apoptosis, oral, iv Phase I – NCT04652960 anti-PD1-ab (nivolumab) DNA repair, senescence, angiogenesis and cell metabolism; immune-checkpoint inhibition CPI-818 (IL-2-ITKi) Inhibition of TCR signal transduction oral Phase I – NCT03952078 Proteasome inhibitor Induces apoptosis; may inhibit iv Phase II – NCT03487133 (bortezomib) and malignant cell migration by dexamethasone suppressing TGFβ1 and IL-10 through NF-κB inhibition exoIL-12, engineered exosome exoIL-12; analysis of skin punch il Placebo-controlled, therapy biopsies bordering the subcutaneous double-blind; phase I trial injection site of exoIL-12 revealed complete; phase II planned local retention of immunologically detectable IL-12 at the injection site ADCC, antibody-dependent cytotoxicity; ADCP, antibody-dependent cellular phagocytosis; CAR T, chimeric antigen receptor T; HDACi, histone deacetylase inhibitor; il, intralesional; iv, intravenous; NK, natural killer; PI3K, phosphatidylinositol 3-kinase; sc, subcutaneous; TCR, T cell receptor; TH1, type 1 helper; pDCs, plasmacytoid dendritic cells. NATure RevIewS | DISEASE PRIMERS | Article citation ID: (2021) 7:61 15 0123456789();: Primer early recognition of adverse effects (such as central patient populations with a predominance of SS in the hypothyroidism with bexarotene201) and management. mogamulizumab trial and the substantial population of pcALCL in the brentuximab trial might account for the Quality of life differences in QOL improvement observed. CTCL has a multidimensional effect on patient QOL due to cutaneous symptoms and impairment of emo- Outlook tional well-being12. An improvement in HRQOL has Current gaps in research been found in both patients with responsive and stable Compared with other cutaneous malignancies, stable disease, further supporting less aggressive therapies in representative cell cultures for the subtypes of CTCL some cases13. QOL reductions are more frequent and and animal models that would allow the study of key pronounced in patients with MF or SS compared with molecular mechanisms or drive the development of control populations with general dermatology con- therapeutic interventions are lacking. The malignant cell cerns, with largest decrements in social functioning and population in CTCL is difficult to identify as there is physical and emotional health202. Pruritus is one of the no reliable and stable surface marker with the exception major factors contributing to poor HRQOL in patients of CD30 in CD30+ LPDs. Thus, biopsies or blood sam- with CTCL but most commonly in patients with SS203. ples used for translational research may contain variable In addition, disease entity, stage and sociodemographic concentrations of malignant cells. factors were associated with a reduction in QOL in those with advanced disease204. Significant differences were Priorities for the next 5–10 years found between patients with early or advanced MF or Routine treatments for CTCL, such as narrow-band SS by the Skindex-29 tool, which found that patients UVB irradiation, radiotherapy or ECP, have not been with advanced MF or SS are compromised in their daily investigated in large multicenter randomized trials. life routine procedures204. Advanced age, worse disease International groups, such as ISCL or the EORTC stage, alopecia and non-university education have an Cutaneous Lymphoma working groups, are develop- incremental effect on QOL, and social deprivation was ing prospective randomized clinical trial programmes frequently observed204. accompanied by a harmonized concept to collect pre- No specific validated QOL questionnaire for CTCL is treatment and on-treatment samples for translational available; therefore, several generic, dermatology-specific research, which should be the focus in the next years. and/or oncology-specific QOL instruments are used in These projects must consider the ultimate goal of this patient population. Whether the existing HRQOL long-term disease control with minimal toxicity and or skin-related QOL instruments effectively meet the distinguish between short-term aggressive tumour specific problems of these conditions is unclear; thus debulking versus maintenance therapy strategies, with there is an urgent need to develop improved tools that special attention on allogeneic HSCT. Finally, the results focus on the specific aspects of the CTCL-related cuta- of these trials will contribute to refined evidence-based neous and extracutaneous problems to identify the most treatment algorithms considering disease subtype, stage, meaningful aspects from a patient’s perspective. tumour load and potential biomarkers. Several clinical studies of denileukin diftitox, oral and topical bexarotene, and ECP alone or with adjuvant Long-term priorities therapy have investigated QOL as a secondary endpoint. The skin is easily accessible for biopsy and biopsy tis- However, early reports on treatment outcome used sue from early and advanced lesions could be used to unvalidated QOL assessment tools allowing only lim- generate a biobanking for CTCL, allowing sophisti- ited conclusions. Further insights in the complexity and cated work-ups. These work-ups could include the use relevance of QOL resulted in the integration of validated of high-throughput molecular biology techniques and QOL tools in the drug development of brentuximab178,205 multichannel imaging procedures such as CyTOF. The and mogamulizumab179,206. Mogamulizumab resulted in generated huge datasets need a careful bioinformatics a significant improvement of QOL compared with vori- analysis. Information from these analyses will provide a nostat using two different assessment tools, reflecting detailed description of the individual tumour microen- the clinical benefits205. vironment, involved signalling networks, and cell com- Superior reductions in cutaneous symptom burden munications and could allow the development of specific (skin itching, burning, pain, irritation and bleeding) but therapeutic interventions. not in emotional or functional domains were associated with an improvement in QOL with brentuximab in a Drug candidates large multicenter trial that included patients with MF and Surface molecules with a selective or at least preferen- pcALCL. There was a more rapid and durable reduction tial expression on the malignant CTCL population are from baseline in symptom burden (symptom domain) potential targets for monoclonal antibodies; however, as with brentuximab compared with physician’s choice most CTCLs have a mature T cell phenotype, it is chal- unrelated to response status. Interestingly, the develop- lenging to find such surface targets. Early CTCL targets, ment of polyneuropathy did not influence the benefits such as CD4, CD25 and CD52, are widely expressed by on QOL. In addition, the histone deacetylase inhibi- T cells and other cell types. CD30 and CCR4 appear tor romidepsin was reported to improve symptoms in preferentially expressed on the malignant T cell popu­ patients with intractable pruritus, thus contributing lations and, thus, the clinical development of the to improved QOL207,208. The differences between the monoclonal antibody mogamulizumab and the fusion 16 | Article citation ID: (2021) 7:61 www.nature.com/nrdp 0123456789();:  Primer Anti-CCR4-ab Mogamulizumab Epidermis CCL17 CCL22 CCR4 TH2 cell PAMPs DAMPs DC CTCL TLR 7/8 - agonist TH1 Resiquimod cell CTCL treatments Macrophage Imatinib CTL Daratumumab Mesylate CD38 CAR-T Anti-miR-155-ab Cobomarsen LMP2 CTL PDGFR CD38 Brentuximab Anti-CTLA4 vedotin LMP1 CD30 CAR T CD30 Anti-CD47-ab CTLA4 TTI-621 MicroRNA-155 BCR CD47 signalling Tofacitinib JAK PI3K LY294002 Anti-KIR3DL2-ab KIR3DL2 IPH 4102 Tumour WP1066 STAT cell IL-2 and diphtheria toxin IL-2Ra MAPK NF-κB E7777 PIP3 PTEN signalling signalling Bortezomib CD52 Anti-CD52-ab Alemtuzumab Akt PDL1 Anti-PDL1-ab Atezolizumab Rapamycin CD4 Immune-checkpoint inhibition mTOR IL-2 CD25 Temsirolimus Basiliximab PDL1 Anti-CD4 CCR4 PD1 CD56 Oncolytic virotherapy VEGFR T-VEC Mogamulizumab Pembrolizumab IMGN901 Sintilimab CD56 CAR T VEGF Bevacizumab Fig. 6 | Targets for future directions for CTCL therapy. New therapeutic approaches in mycosis fungoides focus on both surface molecules and intracellular targets and involve a variety of agents, including antibodies, small molecules or microRNAs. CAR T, chimeric antigen receptor T; CTCL, cutaneous T cell lymphoma; CTL, cytotoxic T lymphocyte; DC, dendritic cell; DAMPs, damage-associated molecular patterns; MAPK, mitogen-activated protein kinase; PAMPs, pathogen-associated molecular patterns; PI3K, phosphatidylinositol 3-kinase; TH, T helper. Figure adapted with permission from ref.195, Taylor and Francis and ref.247, Elsevier. NATure RevIewS | DISEASE PRIMERS | Article citation ID: (2021) 7:61 17 0123456789();: Primer toxin brentuximab resulted in the approval of these pembrolizumab resulted in an attractive response rate agents (Fig. 6). We cannot yet anticipate how monoclo- of 38% in patients with SS and advanced MF102,214. nal antibodies targeting CD47, which is a glycoprotein Moreover, at a follow-up of 58 weeks, median dura- providing a signal to inhibit phagocytosis and which is tion of response was not reached, suggesting that ubiquitously expressed on the surface of most cells, will lasting responses could be achieved with anti-PD1 impact on the disease manifestation in CTCL. inhibitors102,214. Clinical responses were not associated Another promising surface target is KIR3DL2 with known prognostic/predictive biomarkers, including (CD158k), which belongs to the family of killer PDL1 expression, mutation burden or IFNγ gene expres- immunoglobulin-like receptors. By binding to major sion signature, illustrating the need for high-resolution histocompatibility complex (MHC) class I, it provides translational research. However, as PD1 inhibition an inhibitory signal, the blocking of which results in cell was also reported to induce T cell neoplasms215–217 and death. After extensive preclinical work, IPH4102 is the may also induce the development of CTCL in humans, first-in-class anti-KIR3DL2-antibody, which is in early anti-PD1 approaches should be developed with caution clinical development with some early signals of clinical in these entities. efficacy in SS209. Viruses (wild type or genetically modified) can sub- The phosphatidylinositol 3-kinase (PI3K) pathway is stantially modify the cutaneous microenvironment by a key regulator of proliferation and activation in many inducing innate interferon-driven immune responses, cell types and altered in many malignancies, including switching resident reactive T cells to a TH1 phenotype lymphomas. Early trials are investigating various mole- and recruiting other cell populations such as NK cells, cules, such as tenalisib (RP6530) and duvelisib (IPI-145 M1 polarized macrophages and plasmacytoid den- in NCT02783625), in relapsed/refractory peripheral dritic cells. Since at least some CTCL malignant cell T cell lymphoma and CTCL. Both tenalisib and duv- popu­lations are characterized by interferon signalling elisib are double PI3Kδ and PI3Kγ inhibitors and have deficiencies, viruses may replicate preferentially in the demonstrated reasonable activity210,211, justifying further malignant cells. Again, larger trials need to be initi- trials alone or in combination with other approaches. ated given promising data in small patient populations Immunotherapy using monoclonal antibodies inter- treated with talimogene laherparepvec (T-VEC), a rep- fering with checkpoints during T cell activation have licating attenuated herpes simplex virus, replicating dramatically influenced the treatment landscape in measles vaccine virus and a non-replicating adenoviral many malignancies. In CTCL, a disease of proliferating vector carrying a human IFNγ (TG1042)218–220. T cells, interfering with immune checkpoints might be Besides immunotherapy, small molecules, which a double-edged sword212. interfere with signal transduction pathway targets, PD1 is a transmembrane protein, which transduces including Janus kinase-signal transducer and activator of immune-inhibitory signals after binding to its lig- transcription (JAK–STAT), epigenetic modifiers (com- ands PDL1/PDL2 (ref.213). In mouse models, deletion bined approaches to impact on histone acetylation and of PD1 led to enhanced T cell growth and develop- methylation)221 and miRNAs must also be explored195. ment of lymphomas, suggesting a tumour-suppressive Published online xx xx xxxx role in lymphomagenesis112. The anti-PD1 antibody 1. Willemze, R. et al. EORTC classification for primary 9. Ram-Wolff, C., Martin-Garcia, N., Bensussan, A., treatment and follow-up. Ann. Oncol. 29, iv30–iv40 cutaneous lymphomas: a proposal from the cutaneous Bagot, M. & Ortonne, N. Histopathologic diagnosis (2018). lymphoma study group of the European Organization of lymphomatous versus inflammatory erythroderma: 17. Willemze, R. et al. WHO-EORTC classification for for Research and Treatment of Cancer. Blood 90, a morphologic and phenotypic study on 47 skin cutaneous lymphomas. Blood 105, 3768–3785 354–371 (1997). biopsies. Am. J. Dermatopathol. 32, 755–763 (2005). 2. Willemze, R. et al. The 2018 update of the (2010). 18. Scarisbrick, J. J. et al. Cutaneous lymphoma WHO-EORTC classification for primary cutaneous 10. Axelrod, P. I., Lorber, B. & Vonderheid, E. C. Infections international consortium study of outcome in lymphomas. Blood 133, 1703–1714 (2019). complicating mycosis fungoides and Sezary syndrome. advanced stages of mycosis fungoides and sezary This article describes the current EORTC JAMA 267, 1354–1358 (1992). syndrome: effect of specific prognostic markers on classification for primary cutaneous lymphomas. 11. Hodak, E. et al. Should we be imaging lymph nodes survival and development of a prognostic model. 3. Elder D. E., Massi D., Scolyer R. A., Willemze R. at initial diagnosis of early-stage mycosis fungoides? J. Clin. Oncol. 33, 3766–3773 (2015). (eds) WHO Classification of Skin Tumours 4th Edn Results from the PROspective Cutaneous Lymphoma 19. Dobos, G. et al. Epidemiology of cutaneous T-cell (International Agency for Research on Cancer, 2018). International Prognostic Index (PROCLIPI) international lymphomas: a systematic review and meta-analysis 4. Kempf, W. & Mitteldorf, C. Cutaneous T-cell study. Br. J. Dermatol. 184, 524–531 (2020). of 16,953 patients. Cancers 12, 2921 (2020). lymphomas — an update 2021. Hematol. Oncol. 12. Molloy, K. et al. Characteristics associated 20. Geller, S. et al. Outcomes and prognostic factors 39, 46–51 (2021). with significantly worse quality of life in mycosis in African American and black patients with 5. Agar, N. S. et al. Survival outcomes and prognostic fungoides/sezary syndrome from the prospective mycosis fungoides/Sezary syndrome: retrospective factors in mycosis fungoides/Sezary syndrome: cutaneous lymphoma international prognostic index analysis of 157 patients from a referral cancer validation of the revised International Society for (PROCLIPI) study. Br. J. Dermatol. 182, 770–779 center. J. Am. Acad. Dermatol. 83, 430–439 Cutaneous Lymphomas/European Organisation for (2020). (2020). Research and Treatment of Cancer staging proposal. 13. Quaglino, P. et al. Treatment of early-stage mycosis 21. Olsen, E. et al. Revisions to the staging and J. Clin. Oncol. 28, 4730–4739 (2010). fungoides: results from the PROspective Cutaneous classification of mycosis fungoides and Sezary 6. de Coninck, E. C., Kim, Y. H., Varghese, A. & Lymphoma International Study (PROCLIPI study). syndrome: a proposal of the International Society Hoppe, R. T. Clinical characteristics and outcome Br. J. Dermatol. 184, 722–730 (2020). for Cutaneous Lymphomas (ISCL) and the cutaneous of patients with extracutaneous mycosis fungoides. 14. Scarisbrick, J. J. et al. Ethnicity in mycosis fungoides: lymphoma task force of the European Organization J. Clin. Oncol. 19, 779–784 (2001). white patients present at an older age and with more of Research and Treatment of Cancer (EORTC). Blood 7. Kim, Y. H., Liu, H. L., Mraz-Gernhard, S., Varghese, A. advanced disease. Br. J. Dermatol. 180, 1264–1265 110, 1713–1722 (2007). & Hoppe, R. T. Long-term outcome of 525 patients (2019). This article describes the current approach with mycosis fungoides and Sezary syndrome: clinical 15. Scarisbrick, J. J. et al. The PROCLIPI international to staging of mycosis fungoides and Sézary prognostic factors and risk for disease progression. registry of early-stage mycosis fungoides identifies syndrome. Arch. Dermatol. 139, 857–866 (2003). substantial diagnostic delay in most patients. 22. Gallardo, F. et al. Lymphomatoid papulosis associated 8. van Doorn, R. et al. Mycosis fungoides: disease Br. J. Dermatol. 181, 350–357 (2019). with mycosis fungoides: clinicopathological and evolution and prognosis of 309 Dutch patients. 16. Willemze, R. et al. Primary cutaneous lymphomas: molecular studies of 12 cases. Acta Derm. Venereol. Arch. Dermatol. 136, 504–510 (2000). ESMO clinical practice guidelines for diagnosis, 84, 463–468 (2004). 18 | Article citation ID: (2021) 7:61 www.nature.com/nrdp 0123456789();:  Primer 23. Wood, G. S., Crooks, C. F. & Uluer, A. Z. 42. Folkes, A. S. et al. Targeting CD47 as a cancer 65. Choi, J. et al. Genomic landscape of cutaneous T cell Lymphomatoid papulosis and associated cutaneous therapeutic strategy: the cutaneous T-cell lymphoma lymphoma. Nat. Genet. 47, 1011–1019 (2015). lymphoproliferative disorders exhibit a common experience. Curr. Opin. Oncol. 30, 332–337 66. Kiel, M. J. et al. Genomic analyses reveal recurrent clonal origin. J. Invest. Dermatol. 105, 51–55 (1995). (2018). mutations in epigenetic modifiers and the JAK-STAT 24. Goyal, A. et al. Increased risk of second primary 43. Johnson, L. D. S. et al. Targeting CD47 in pathway in Sezary syndrome. Nat. Commun. 6, 8470 hematologic and solid malignancies in patients Sezary syndrome with SIRPalphaFc. Blood Adv. (2015). with mycosis fungoides: a Surveillance, Epidemiology, 3, 1145–1153 (2019). 67. McGirt, L. Y. et al. Whole-genome sequencing reveals and End Results analysis. J. Am. Acad. Dermatol. 83, 44. Aronovich, A. et al. Cancer-associated fibroblasts oncogenic mutations in mycosis fungoides. Blood 126, 404–411 (2020). in mycosis fungoides promote tumor cell migration 508–519 (2015). 25. Scarisbrick, J. J. et al. Prognostic factors, prognostic and drug resistance through CXCL12/CXCR4. 68. Ungewickell, A. et al. Genomic analysis of mycosis indices and staging in mycosis fungoides and Sezary J. Invest. Dermatol. 141, 619–627.e2 (2021). fungoides and Sezary syndrome identifies recurrent syndrome: where are we now? Br. J. Dermatol. 170, 45. Querfeld, C. et al. Primary T cells from cutaneous T-cell alterations in TNFR2. Nat. Genet. 47, 1056–1060 1226–1236 (2014). lymphoma skin explants display an exhausted immune (2015). 26. Quaglino, P. et al. Time course, clinical pathways, checkpoint profile. Cancer Immunol. Res. 6, 900–909 69. Wang, L. et al. Genomic profiling of Sezary syndrome and long-term hazards risk trends of disease (2018). identifies alterations of key T cell signaling and progression in patients with classic mycosis fungoides: 46. Torrealba, M. P. et al. Chronic activation profile differentiation genes. Nat. Genet. 47, 1426–1434 a multicenter, retrospective follow-up study from the of circulating CD8+ T cells in Sezary syndrome. (2015). Italian Group of Cutaneous Lymphomas. Cancer 118, Oncotarget 9, 3497–3506 (2018). 70. Woollard, W. J. et al. Candidate driver genes involved 5830–5839 (2012). 47. Yamanaka, K. et al. Decreased T-cell receptor excision in genome maintenance and DNA repair in Sezary 27. Talpur, R. et al. Long-term outcomes of 1,263 patients circles in cutaneous T-cell lymphoma. Clin. Cancer Res. syndrome. Blood 127, 3387–3397 (2016). with mycosis fungoides and Sezary syndrome from 11, 5748–5755 (2005). 71. Waldmann, T. A. & Chen, J. Disorders of the 1982 to 2009. Clin. Cancer Res. 18, 5051–5060 48. Bastidas Torres, A. N. et al. Genomic analysis reveals JAK/STAT pathway in T cell lymphoma pathogenesis: (2012). recurrent deletion of JAK-STAT signaling inhibitors implications for immunotherapy. Annu. Rev. Immunol. 28. Quaglino, P. et al. Global patterns of care in HNRNPK and SOCS1 in mycosis fungoides. Genes 35, 533–550 (2017). advanced stage mycosis fungoides/Sezary syndrome: Chromosomes Cancer 57, 653–664 (2018). 72. Perez, C. et al. Advanced-stage mycosis fungoides: a multicenter retrospective follow-up study from This study combined whole genome sequencing role of the signal transducer and activator of the Cutaneous Lymphoma International Consortium. and RNA sequencing, providing a comprehensive transcription 3, nuclear factor-kappaB and nuclear Ann. Oncol. 28, 2517–2525 (2017). overview of gene fusions, structural alterations factor of activated T cells pathways. Br. J. Dermatol. 29. Benton, E. C. et al. A cutaneous lymphoma and mutations in mycosis fungoides and positions 182, 147–155 (2020). international prognostic index (CLIPi) for mycosis inactivation of HNRNPK and SOCS1 as potential 73. Moyal, L. et al. Oncogenic role of microRNA-155 in fungoides and Sezary syndrome. Eur. J. Cancer 49, driver events in mycosis fungoides development. mycosis fungoides: an in vitro and xenograft mouse 2859–2868 (2013). 49. van Doorn, R. et al. Oncogenomic analysis of mycosis model study. Br. J. Dermatol. 177, 791–800 (2017). 30. Ho, A. W. & Kupper, T. S. T cells and the skin: from fungoides reveals major differences with Sezary 74. Ralfkiaer, U. et al. Diagnostic microRNA profiling protective immunity to inflammatory skin disorders. syndrome. Blood 113, 127–136 (2009). in cutaneous T-cell lymphoma (CTCL). Blood 118, Nat. Rev. Immunol. 19, 490–502 (2019). 50. Laharanne, E. et al. CDKN2A-CDKN2B deletion 5891–5900 (2011). 31. Campbell, J. J., Clark, R. A., Watanabe, R. & defines an aggressive subset of cutaneous T-cell 75. Sandoval, J. et al. MicroRNA expression profiling and Kupper, T. S. Sezary syndrome and mycosis fungoides lymphoma. Mod. Pathol. 23, 547–558 (2010). DNA methylation signature for deregulated microRNA arise from distinct T-cell subsets: a biologic rationale 51. Navas, I. C. et al. p16(INK4a) is selectively silenced in cutaneous T-cell lymphoma. J. Invest. Dermatol. for their distinct clinical behaviors. Blood 116, in the tumoral progression of mycosis fungoides. 135, 1128–1137 (2015). 767–771 (2010). Lab. Invest. 82, 123–132 (2002). 76. van Kester, M. S. et al. miRNA expression profiling This article describes the differences in 52. Nicolae-Cristea, A. R. et al. Diagnostic and prognostic of mycosis fungoides. Mol. Oncol. 5, 273–280 immunophenotype of tumour cells in patients significance of CDKN2A/CDKN2B deletions in patients (2011). with mycosis fungoides or Sézary syndrome and with transformed mycosis fungoides and primary 77. Caumont, C. et al. PLCG1 gene mutations correlation with clinical phenotype. cutaneous CD30-positive lymphoproliferative disease. are uncommon in cutaneous T-cell lymphomas. 32. Horna, P., Moscinski, L. C., Sokol, L. & Shao, H. Br. J. Dermatol. 172, 784–788 (2015). J. Invest. Dermatol. 135, 2334–2337 (2015). Naive/memory T-cell phenotypes in leukemic cutaneous 53. Laharanne, E. et al. Genome-wide analysis 78. Kiessling, M. K. et al. High-throughput mutation T-cell lymphoma: putative cell of origin overlaps disease of cutaneous T-cell lymphomas identifies three profiling of CTCL samples reveals KRAS and NRAS classification. Cytometry B Clin. Cytom. 96, 234–241 clinically relevant classes. J. Invest. Dermatol. 130, mutations sensitizing tumors toward inhibition of (2019). 1707–1718 (2010). the RAS/RAF/MEK signaling cascade. Blood 117, 33. Roelens, M. et al. Circulating and skin-derived Sezary 54. Vermeer, M. H. et al. Novel and highly recurrent 2433–2440 (2011). cells: clonal but with phenotypic plasticity. Blood 130, chromosomal alterations in Sezary syndrome. 79. Park, J. et al. Genomic analysis of 220 CTCLs identifies 1468–1471 (2017). Cancer Res. 68, 2689–2698 (2008). a novel recurrent gain-of-function alteration in RLTPR 34. Saed, G., Fivenson, D. P., Naidu, Y. & Nickoloff, B. J. This paper highlights the recurrent copy number (p.Q575E). Blood 130, 1430–1440 (2017). Mycosis fungoides exhibits a Th1-type cell-mediated alterations in Sézary syndrome identified using Largest dataset of genomic analysis in cutaneous cytokine profile whereas Sezary syndrome expresses array comparative genomic hybridization. T cell lymphoma; by combining publicly available a Th2-type profile. J. Invest. Dermatol. 103, 29–33 55. Caprini, E. et al. Identification of key regions and sequencing data with their own studies, the authors (1994). genes important in the pathogenesis of Sezary identified putative driver genes that are predicted 35. Vermeer, M. H. et al. CD8+ T cells in cutaneous T-cell syndrome by combining genomic and expression to affect activation pathways, chromatin and lymphoma: expression of cytotoxic proteins, Fas microarrays. Cancer Res. 69, 8438–8446 (2009). immune surveillance. Ligand, and killing inhibitory receptors and their 56. Alexandrov, L. B. et al. The repertoire of mutational 80. van der Fits, L., Out-Luiting, J. J., Tensen, C. P., relationship with clinical behavior. J. Clin. Oncol. 19, signatures in human cancer. Nature 578, 94–101 Zoutman, W. H. & Vermeer, M. H. Exploring the 4322–4329 (2001). (2020). IL-21-STAT3 axis as therapeutic target for Sezary 36. Lindahl, L. M. et al. Antibiotics inhibit tumor and 57. Weed, J. et al. FISH panel for leukemic CTCL. syndrome. J. Invest. Dermatol. 134, 2639–2647 disease activity in cutaneous T-cell lymphoma. Blood J. Invest. Dermatol. 137, 751–753 (2017). (2014). 134, 1072–1083 (2019). 58. Prasad, A. et al. Identification of gene mutations 81. van Kester, M. S. et al. Cucurbitacin I inhibits This article describes a decrease in clinical and fusion genes in patients with Sezary syndrome. Stat3 and induces apoptosis in Sezary cells. symptoms in response to aggressive, transient J. Invest. Dermatol. 136, 1490–1499 (2016). J. Invest. Dermatol. 128, 1691–1695 (2008). antibiotic treatment, further establishing a link 59. Wang, Y. & Bagot, M. Updates in cutaneous 82. Taniguchi, K. & Karin, M. NF-kappaB, inflammation, between Staphylococcus aureus infections and lymphoma: evidence-based guidelines for the immunity and cancer: coming of age. Nat. Rev. Immunol. tumour progression in mycosis fungoides. management of cutaneous lymphoma 2018. 18, 309–324 (2018). 37. Tokura, Y. et al. Cutaneous colonization with Br. J. Dermatol. 180, 443–444 (2019). 83. Chang, L. W. et al. An integrated data resource staphylococci influences the disease activity of Sezary 60. da Silva Almeida, A. C. et al. The mutational landscape for genomic analysis of cutaneous T-cell lymphoma. syndrome: a potential role for bacterial superantigens. of cutaneous T cell lymphoma and Sezary syndrome. J. Invest. Dermatol. 138, 2681–2683 (2018). Br. J. Dermatol. 133, 6–12 (1995). Nat. Genet. 47, 1465–1470 (2015). 84. Zinzani, P. L. et al. Phase II trial of proteasome 38. Blumel, E. et al. Staphylococcal alpha-toxin tilts 61. Borcherding, N. et al. Single-cell profiling of cutaneous inhibitor bortezomib in patients with relapsed or the balance between malignant and non-malignant T-cell lymphoma reveals underlying heterogeneity refractor