Cutaneous T-Cell Lymphoma

The CTCL concept

In 1968, Lutzner and Jordan described the ultrastructural features of the circulating atypical cells in SS. Characteristically, the nuclei of these cells had deep and narrow indentations, giving them a cerebriform appearance. Three years later, similar cells were found in the skin and lymph nodes of patients with MF. In 1975, based on the observation that the neoplastic cells in MF, SS, and related conditions not only had the same morphology but also a common T-cell phenotype, the term “CTCL” was suggested for this group of diseases . Within a short time, this term gained wide acceptance, particularly in the US. The introduction of the CTCL concept can be considered as a landmark in the history of this group of diseases. However, a major disadvantage has been that in many subsequent studies, distinction was no longer made between MF, SS, and other T-cell neoplasms, entities which may vary considerably in clinical presentation and clinical behavior.

The concept of primary cutaneous lymphomas

At about the same time that the CTCL concept was introduced, several European groups started to classify cutaneous lymphomas according to the criteria of the Kiel classification, a system used by hematopathologists for the categorization of nodal lymphomas. It was then determined that many types of CBCL and CTCL (other than classic MF and SS) can present in the skin without any evidence of extracutaneous disease at the time of diagnosis. It appeared that these primary cutaneous lymphomas often have a completely different clinical behavior and prognosis when compared to morphologically similar lymphomas arising within lymph nodes, and therefore require different types of treatment . In addition, differences in the presence of specific chromosomal translocations and in the expression of oncogenes, viral sequences or antigens (e.g. EBV), and adhesion receptors involved in tissue-related lymphocyte homing were found.

Such differences underscored that primary cutaneous lymphomas represent a distinct group and may explain, at least in part, their different clinical behavior. For instance, the observation that the neoplastic T cells in most CTCLs express cutaneous lymphocyte antigen (CLA) and the CC-chemokine receptors 4 (CCR4) and 10 (CCR10) indicates that they are the neoplastic counterparts of normal skin-homing T cells, and this explains why these CTCLs present in the skin. Perhaps most importantly, it appeared that different types of CTCL and CBCL with different clinical behaviors and different therapeutic requirements may have an identical histologic appearance. This implies that histologic features should always be combined with clinical and immunophenotypical data, before a definite diagnosis (classification) is made.

Over the past two decades, such an approach resulted in the delineation of several new types of CTCL and CBCL, and it formed the basis of the European Organization for Research and Treatment of Cancer (EORTC) classification for primary cutaneous lymphomas .

EORTC, WHO and WHO-EORTC classification schemes

The EORTC classification scheme was the first one that was specifically designed for the group of primary cutaneous lymphomas. It contained a limited number of well-defined types of CTCL and CBCL, which together accounted for >95% of all primary cutaneous lymphomas, in addition to some provisional entities, which were not yet fully defined clinically. A distinction was made between cutaneous lymphomas with indolent, intermediate, or aggressive clinical behavior. By including well-defined and recognizable disease entities, this classification provided the clinician with detailed information on staging, preferred mode of treatment, clinical behavior and prognosis, serving as a useful guide to optimal management and treatment.

The third edition of the World Health Organization (WHO) classification for malignant lymphomas adopted the most common types of CTCL from the EORTC classification, but there was controversy with regard to the categorization of some rare types of CTCL . In addition, the WHO classification contained new entities not yet included in the EORTC scheme. In 2004, representatives from both organizations reached an agreement on a new classification scheme for the group of cutaneous lymphomas: the WHO-EORTC classification . Importantly, the fourth edition of the WHO classification for malignant lymphomas, published in 2008 (updated in 2016) and used by pathologists and hematologists worldwide, has adopted the terminology and definitions of the WHO-EORTC classification; therefore, it now includes all the types of CTCL and CBCL included in previous classifications of primary cutaneous lymphomas . The frequency and survival of patients with the different types of CTCL recognized in the WHO-EORTC classification are presented in Table 120.1 . Following a description of practical guidelines for diagnosis, classification and staging, relevant features of the different types of CTCL included in this classification are presented.

Diagnosis

The first step in the evaluation of a patient suspected of having CTCL is to decide if it represents a lymphoma or a benign condition. Skin biopsies – preferably deep punch biopsies 4–6 mm wide or an excisional or incisional biopsy from the most representative skin lesions – should be performed. In the US, broad saucerizations are also done for suspected patch/plaque stage MF.

Since prior treatment with topical corticosteroids or PUVA may change the original histology profoundly, biopsies from untreated skin lesions are preferred. Even if an adequate biopsy specimen is obtained, a definite diagnosis is not always possible. First, several types of CTCL, such as MF, are often preceded for years by skin lesions that are neither clinically nor histologically diagnostic. The gradual progression from this prediagnostic phase to overt lymphoma explains why it has always been so difficult to reach a consensus on the minimal histologic criteria needed for an unequivocal diagnosis of MF . Because of the indolent clinical behavior of the disease in such patients, a conservative approach is justified. In most cases, repeated biopsies, when appropriate, will ultimately result in the correct diagnosis.

Secondly, atypical T-cell infiltrates are found not only in CTCL but also in reactive conditions, e.g. lymphomatoid drug eruptions (pseudo-T-cell lymphomas) . Therefore, a definite diagnosis should always be based on a combination of clinical, histologic and, in most cases, immunophenotypical criteria, and it may be supplemented with the results of gene rearrangement analysis .

Immunophenotyping

Immunohistochemical studies on paraffin or frozen sections using antibodies reactive with cell-surface or cytoplasmic molecules are extremely important in the diagnosis and classification of cutaneous lymphomas. Nowadays, antigen retrieval techniques allow immunophenotyping on formalin-fixed, paraffin-embedded tissue sections, eliminating the need for frozen sections. By using a panel of such antibodies (see Ch. 0 ), distinction can be made between neoplasms of T-cell, B-cell, NK-cell, and myeloid or monocytic origin. Within the group of CTCLs, such studies have contributed to the delineation of new subtypes and provided important diagnostic and prognostic criteria.

With respect to the diagnosis of CTCL, demonstration of an aberrant phenotype, i.e. loss of one or more T-cell-associated antigens, such as CD2, CD3, CD4 or CD5, by the neoplastic T cells, can be considered as an important additional criterion in establishing a definite diagnosis of CTCL . However, loss of CD7 expression, which may be caused by chronic T-cell stimulation and is commonly observed in benign dermatoses, is not a reliable marker.

T-cell receptor gene rearrangement analysis

T-cell receptor (TCR) gene rearrangement analysis, utilizing a standardized assay (e.g. BIOMED-2), may provide useful information for the diagnosis and staging of malignant lymphomas . However, caution is warranted in interpreting the results of such analyses. Clonal T-cell populations have been detected not only in skin lesions, lymph nodes, and peripheral blood of patients with CTCL and in cases of chronic dermatitis preceding MF (“clonal dermatitis”; see Ch. 9 ), but also in skin lesions of patients with apparently benign conditions, such as pityriasis lichenoides et varioliformis acuta, lichen planus, lichen sclerosus and some pseudolymphomas .

Demonstration of clonal T-cell populations can therefore not be used as an absolute criterion of malignancy, but should always be considered in conjunction with clinical and histologic features, which together remain the “gold standard”. If the clinical and histologic findings are not consistent, and an aberrant phenotype is not detected, a definite diagnosis of CTCL should not be made. In the future, high-throughput TCR sequencing may be used to enhance detection of T-cell clones in patients with CTCL and for distinguishing CTCL from benign inflammatory diseases .

Classification

Once a diagnosis of CTCL has been made, the type of CTCL should be determined. As noted previously, the histologic diagnosis is not the final diagnosis, which means that CTCL cannot be classified on the basis of histologic criteria alone. For instance, in the case of a skin biopsy showing a dermal infiltrate which is composed predominantly of CD30-positive large anaplastic or pleomorphic T cells, the histologic diagnosis will be (CD30-positive) anaplastic large cell lymphoma (ALCL). However, depending upon the clinical presentation, the definite diagnosis may be: a primary cutaneous ALCL; cutaneous involvement of a systemic ALCL; large cell transformation of MF with expression of CD30; or LyP (in cases of recurrent self-healing papules). Obviously, each of these conditions requires a different approach in terms of staging and treatment.

In Table 120.2 , the differential diagnosis of four histologic categories has been elaborated. This histologic subdivision into epidermotropic CTCL, CTCL with diffuse pleomorphic infiltrates, CD30-positive CTCL, and subcutaneous CTCL is neither complete nor meant as an alternative classification. It is only presented as an illustration of how histology should be combined with clinical and immunophenotypical data to arrive at a definite diagnosis.

Practical guidelines

For those not regularly involved in the diagnosis and classification of CTCL and perhaps overwhelmed by the many new types of CTCL introduced in recent years, a stepwise approach, as presented in Fig. 120.1 , may serve as a practical guide .

• Step one : Based on a combination of clinical, histologic and immunophenotypical criteria, distinction is first made between classic MF, MF variants, and SS on the one hand and CTCL other than these conditions on the other. The rationale for this first step is that dermatologists are familiar with the former set of conditions, which comprise approximately 65% of CTCLs, and these lymphomas require a different clinical approach in terms of staging and therapy compared to the other forms of CTCL.

• Step two : The second category to be considered is the group of primary cutaneous CD30-positive lymphoproliferative disorders. It implies that evaluation of skin biopsies from patients with (suspected) CTCL should always include CD30 immunostaining. This group includes cases of cutaneous ALCL (C-ALCL) and LyP, which together represent the second most common group of CTCL (see Table 120.1 ). Patients within this spectrum of disease generally have an excellent prognosis, and most individuals can be managed easily by dermatologists.

• Step three : With the first two steps, approximately 90% of CTCL will be classified correctly. The remaining group (10% of patients) contains rare types of CTCL, including subcutaneous panniculitis-like T-cell lymphoma (SPTCL), extranodal NK/T-cell lymphoma, nasal type, and the broad group of primary cutaneous peripheral T-cell lymphoma (PTCL), not otherwise specified (NOS). From the latter group, primary cutaneous CD8-positive aggressive epidermotropic cytotoxic T-cell lymphoma, primary cutaneous CD4-positive small/medium T-cell lymphoproliferative disorder, and primary cutaneous acral CD8-positive T-cell lymphoma have been separated out as provisional entities (see Table 120.1 ). In the WHO-EORTC classification, the term PTCL, NOS is maintained for remaining cases that do not fit into any of the provisional entities.

  Apart from SPTCL, primary cutaneous CD4-positive small/medium T-cell lymphoproliferative disorder, and primary cutaneous acral CD8-positive T-cell lymphoma, lymphomas included in this third category have an aggressive clinical course, and they should be treated by, or in close collaboration with, a hemato-oncologist. It should be emphasized that because of overlapping clinicopathologic features and highly aberrant phenotypes, distinction between these different types of aggressive CTCL is sometimes very difficult. Distinction between primary and secondary cutaneous involvement is less important than in other types of cutaneous lymphoma in that patients presenting with only skin lesions generally develop extracutaneous disease within a short period of time and have a poor prognosis as well.

Staging

Apart from CTCL, a term preferably used only for primary CTCL, systemic T-cell lymphomas frequently present or relapse in the skin. Adequate staging procedures are required to differentiate between CTCL and these systemic lymphomas secondarily involving the skin. The extent of staging procedures is dependent on the type of (suspected) CTCL, and, in the case of classic MF, on the clinical stage of the disease . In cases of early patch/plaque stage MF, unequivocal cases of LyP, and probably pagetoid reticulosis as well, staging procedures are generally not worthwhile. In cases of suspected SS, adequate staging with special emphasis on assessment of peripheral blood involvement by cytology, immunophenotyping (flow cytometry), and TCR gene rearrangement analysis is essential. In all other types of CTCL, routine hematologic staging, including complete and differential blood cell counts, a serum chemistry panel, computed tomographic (CT) scans and/or positron emission tomography (PET) of the chest and abdomen, and bone marrow sampling, are required .

Definition

Mycosis fungoides (MF) represents the most common type of CTCL and accounts for ~50% of all primary cutaneous lymphomas (see Table 120.1 ). The term MF should be restricted to the classic “Alibert–Bazin” type characterized by the typical evolution of patches, plaques and tumors, or for clinicopathologic variants showing a similar clinical course.

Epidemiology

MF has an incidence of about 0.4 per 100 000 inhabitants per year in the US . MF typically affects older adults (median age at diagnosis: 55–60 years), but it may occur in children and adolescents as well. Men are affected more often than women, with a male-to-female ratio of 1.6–2.0 : 1.

Pathogenesis

The etiology and the pathogenetic mechanisms involved in the development and stepwise progression of MF are largely unknown. Genetic, environmental, and immunologic factors have all been considered.

Clinical features

Characteristically, patients with classic MF progress from patch stage to plaque stage and finally to tumor stage disease, and they have a protracted clinical course over years or even decades. Before a definite diagnosis is made, patients generally have many years of nonspecific eczematous or psoriasiform skin lesions and non-diagnostic biopsies. The median duration from onset of skin lesions to the diagnosis of MF is 4–6 years, but it may vary from several months to more than five decades .

Early patch stage MF is characterized by the presence of variably sized erythematous, finely scaling lesions, which may be mildly pruritic ( Fig. 120.2A ). These early lesions may show variable degrees of atrophy, and a poikilodermatous variant consisting of patches with mottled hyper- and hypopigmentation, atrophy, and telangiectasia has been described (formerly called poikiloderma vasculare atrophicans). Generalized hypopigmented lesions may be observed in patients with darkly pigmented skin ( Fig. 120.3 ), and this is also a common presentation of juvenile-onset MF.

The initial skin lesions have a predilection for the buttocks and other covered sites of the trunk and limbs. With progression, more infiltrated reddish-brown, scaling plaques develop, which gradually enlarge and may have an annular, polycyclic or typical horseshoe-shaped configuration ( Fig. 120.4A ). It should be stressed that many patients never progress beyond the plaque stage of the disease. However, a minority (~10%) of patients may develop nodules or tumors. These patients with tumor stage MF characteristically show a combination of patches, plaques, and tumors ( Fig. 120.5A ); the latter often show ulceration.

If only skin tumors are present without preceding or concurrent patches or plaques, a diagnosis of MF is highly unlikely and another type of CTCL should be considered. The risk of developing extracutaneous disease correlates with the extent and type of skin lesions. It is exceedingly rare in patients with limited patch/plaque stage disease, relatively uncommon in patients with generalized plaques, and most likely in patients with skin tumors or erythroderma . Extracutaneous dissemination, almost without exception, first involves the regional lymph nodes draining areas of extensive skin involvement. Visceral involvement may develop subsequently and can involve any organ. The bone marrow is rarely involved.

Pathology

Early patch lesions in MF show superficial band-like or lichenoid infiltrates, consisting primarily of lymphocytes. Atypical cells, with small to medium-sized, highly convoluted (cerebriform) and sometimes hyperchromatic nuclei, are few in number and are mostly confined to the epidermis (epidermotropism). These lymphocytes characteristically colonize the basal layer of the epidermis as single cells surrounded by vacuolated halos, often in a linear configuration ( Fig. 120.2B ) .

In plaques , epidermotropism is generally more pronounced ( Fig. 120.4B ). The presence of intraepidermal nests of atypical cells (referred to as Pautrier microabscesses, although the initial description was by Darier) is a highly characteristic feature, but is observed in only a minority of cases. The epidermis may show acanthosis and elongated rete ridges, but spongiosis is generally mild or absent. The dermal infiltrates are also more pronounced, and they may contain a higher number of atypical cells with cerebriform nuclei and occasional blast cells, as well as admixed eosinophils and plasma cells. Rarely, a predominantly interstitial infiltrate that resembles granuloma annulare or morphea is observed (interstitial MF) .

With progression to tumor stage MF, the dermal infiltrates can involve the entire dermis and extend into the subcutaneous tissue. Epidermotropism may no longer be present. The tumor cells increase in number and size, showing variable proportions of small, medium-sized or large cells with cerebriform nuclei, blast cells with prominent nuclei, and intermediate forms ( Fig. 120.5B ). Large cell transformation , defined by the presence of CD30-negative or CD30-positive large cells exceeding 25% of the infiltrate or forming microscopic nodules, may occur and is generally associated with a poor prognosis. However, within this group, CD30-positive patients have a much better prognosis than CD30-negative patients .

Immunophenotype

The neoplastic cells in MF have a mature CD3 ^+, CD4 ⁺ , CD45RO ⁺ , CD8 ⁻ memory T-cell phenotype. In a minority of patients with otherwise classic MF, a CD3 ⁺ , CD4 ⁻ , CD8 ⁺ mature T-cell phenotype or more rarely a γ/δ T-cell phenotype (βF1 ⁻ , TCRγ ⁺ , CD3 ⁺ , CD4 ⁻ , CD8 ⁺ ) may be seen . Immunophenotyping may demonstrate (partial) loss of T-cell antigens in plaque or tumor stage MF, but rarely in early patch lesions, and is therefore of little help in early diagnosis.

Differential diagnosis

Regarding the differential diagnosis of MF, three categories should be considered. The first category contains a diverse group of benign dermatoses, which early MF may resemble clinically, and it includes several types of eczema, psoriasis, superficial fungal infections, and drug reactions. These specific diagnoses can generally be excluded by histologic and other standard dermatologic examinations. This category may also include patients with large plaque parapsoriasis (parapsoriasis en plaque) who show slightly scaly, sometimes atrophic, erythematous patches or plaques, which are commonly located on the trunk and buttocks (see Ch. 9 ). Whereas large plaque parapsoriasis cannot be distinguished clinically from early patch or plaque stage MF, the histologic features are often not consistent with MF. Long-term follow-up studies have documented progression of large plaque parapsoriasis to overt MF in ~10% of cases. However, one prevailing opinion is that large plaque parapsoriasis should be considered a form of MF, rather than a potential precursor of MF, but the author feels there is no consensus. There is even less consensus about whether small plaque parapsoriasis represents MF.

A second category includes several benign conditions with histologic features highly suggestive of MF. Examples of these are lymphomatoid contact dermatitis, lymphomatoid drug reactions, and actinic reticuloid. Apart from subtle histologic differences (e.g. the predominance of atypical T cells in the dermal infiltrates rather than in the epidermis), careful evaluation of the clinical features, which are generally not consistent with MF, often results in a correct diagnosis .

The third category includes other types of (epidermotropic) CTCL, which may resemble MF histologically. Diagnostic features of these entities are presented in Table 120.2 .

Staging systems and staging procedures

Staging patients with MF and SS is important, since it determines management and treatment and has prognostic significance. In 2007, a revised clinical staging system for MF and SS was proposed, which is based on the TNM (tumor–node–metastasis) classification system and takes into account the type and extent of skin lesions (T1–4) as well as the presence or absence of nodal (N0–3), visceral (M0–1), and peripheral blood involvement (B0–2) ( Tables 120.3 & 120.4 ) .

Table 120.3

TNMB classification of mycosis fungoides and Sézary syndrome.

TNMB CLASSIFICATION OF MYCOSIS FUNGOIDES AND SÉZARY SYNDROME
T (skin)
T 1	Limited patch/plaque (involving <10% of total skin surface)
T 2	Generalized patch/plaque (involving ≥10% of total skin surface)
T 3	Tumor(s)
T 4	Erythroderma
N (lymph node)
N 0	No enlarged lymph nodes
N 1	Enlarged lymph nodes, histologically uninvolved
N 2	Enlarged lymph nodes, histologically involved (nodal architecture uneffaced)
N 3	Enlarged lymph nodes, histologically involved (nodal architecture [partially] effaced)
M (viscera)
M 0	No visceral involvement
M 1	Visceral involvement
B (blood)
B 0	No circulating atypical (Sézary) cells (or <5% of lymphocytes)
B 1	Low blood tumor burden (≥5% of lymphocytes are Sézary cells, but not B 2 )
B 2	High blood tumor burden (≥1000/mcl Sézary cells + positive clone)

Table 120.4

Clinical staging system for mycosis fungoides and Sézary syndrome.

The shaded boxes highlight the required features for the three subdivisions of stage IV disease.

Open full size image

Evaluation of patients suspected of having MF should include a thorough physical examination with special attention to the type and extent of skin lesions and the presence of palpable lymph nodes, as well as skin biopsies, complete blood counts, and serum chemistries. Enlarged lymph nodes should be biopsied. Histologically, distinction can be made between lymph nodes showing dermatopathic lymphadenopathy without involvement by MF (N1), dermatopathic lymphadenopathy with early MF involvement (N2), and lymph nodes showing effacement of the normal lymph node architecture by neoplastic T cells (N3). The prognostic significance of such a subdivision has been well established.

No further examinations are recommended for patients with stage IA–B disease. CT scans of the chest and abdomen are recommended in patients in whom extracutaneous disease is suspected, but they are less useful in patients with limited patches and/or plaques without lymphadenopathy. Examination of other organs, including the bone marrow, should only be performed if clinically indicated.

Treatment

The choice of an initial treatment in MF depends on the stage of the disease and the general condition and age of the patient . Given the chronic and recurrent nature of MF, treatment should be aimed at improving symptoms while limiting toxicity. Following the traditional “stage-based” approach, skin-directed therapies are preferred in the early stages of MF (stages IA–IIA) and even in patients with limited tumor stage MF (IIB) ( Table 120.5 ). These skin-directed therapies include topical or intralesional corticosteroids, topical cytotoxic agents (e.g. mechlorethamine [nitrogen mustard]), phototherapy, and radiotherapy. The efficacy of skin-directed therapies in MF is explained by the preferential localization of the neoplastic skin-homing T cells to the epidermis and superficial dermis. Systemic multi-agent chemotherapy is not useful in these early stages, since it does not improve survival and is associated with considerable morbidity. In patients with refractory or progressive skin disease, skin-directed therapies can be combined with IFN-α or systemic retinoids. Alternatively, novel agents such as denileukin diftitox or h istone d e ac etylase i nhibitors (HDACi) such as vorinostat and romidepsin can be used, before systemic chemotherapy is considered. Both HDACi and denileukin diftitox have been approved in the US for patients with relapsed and refractory CTCL, but have not been registered for CTCL in Europe. In general, systemic chemotherapy is only indicated in advanced stages when there is nodal or visceral involvement or in patients with rapidly progressive tumors unresponsive to less aggressive therapies.