Plaque stage folliculotropic mycosis fungoides: histopathologic features and prognostic factors in a series of 40 patients

Abstract Background Folliculotropic mycosis fungoides (FMF) is a distinct variant of mycosis fungoides. Recent studies recognized indolent and aggressive subgroups of FMF, but there is controversy how patients presenting with plaques should be classified. The present study describes the histopathologic features of 40 FMF plaques. The aim of the study was to identify risk factors for disease progression and poor outcome in this group. Methods Clinical, histopathological, and immunophenotypical data from 40 patients with plaque stage FMF were reviewed and analysed for risk factors for disease progression and survival. Results After a median follow‐up of 80 months, disease progression occurred in 20 of 40 patients. Percentage of atypical cells, cell size, percentage of Ki‐67+ cells, and co‐existent interfollicular epidermotropism, but not the extent of perifollicular infiltrates, were associated with disease progression and reduced survival, while extensive follicular mucinosis was associated with increased survival. Conclusions This study underlines that FMF patients presenting with plaques represent a heterogeneous group and that a subgroup of these patients may have an indolent clinical course. It further shows that histological examination is a valuable tool to differentiate between indolent and aggressive disease.


| INTRODUCTION
Folliculotropic mycosis fungoides (FMF) is recognized as a distinct variant of mycosis fungoides (MF). FMF is histopathologically characterized by the presence of folliculotropic infiltrates consisting of atypical, T-cells, that often spare the epidermis. [1][2][3] Clinically, skin lesions are preferentially located in the head and neck region and are frequently accompanied by local alopecia, and may include (grouped) follicular papules, acneiform and cystic lesions, plaques and tumors and in some patients keratosis pilaris-like skin lesions, which are located preferentially on the extremities or trunk. 1,2,[4][5][6][7][8][9][10][11][12][13] Previous studies emphasized that FMF patients have a worse prognosis when compared to classic MF patients and should be treated more aggressively. 1,2 However, more recent studies reported that not all patients with FMF have an unfavorable prognosis and suggested that distinction should be made between an indolent (early stage) and an aggressive group (advanced stage) of FMF. 5,7,11 Patients presenting with only follicle-based patches, acneiform, or keratosis pilaris-like lesions have early-stage disease (stages IA-IIA), run an indolent clinical course and have an excellent prognosis, while patients presenting with nodules, tumors, erythroderma, and/or extracutaneous disease have advanced FMF (stages IIB-IV) and usually show a much more aggressive clinical course. 5 It is however uncertain how patients presenting with plaques should be categorized. In a study of Hodak et al, all patients with infiltrated plaques were upstaged and considered to have tumor-stage disease (stage IIB). 7 In contrast, studies of our own group showed that a subset of patients presenting with plaques may have a favorable clinical course. Using histopathological criteria, distinction was made between cases with early plaque-stage FMF and advanced plaque-stage FMF. It was found that patients with early plaque-stage FMF had the same indolent clinical behavior as patients with patch-stage FMF, whereas patients with advanced plaque-stage FMF had the same clinical behavior and prognosis as patients with tumor-stage FMF. 5 Histopathologic classification as either early or advanced plaque-stage disease was based on a rough estimate of the extent and cellular composition of the perifollicular infiltrates in a single hematoxylin and eosin (H&E)-stained section, but predefined criteria were lacking. 5 In the present study, clinical, histopathological, and immunophenotypical characteristics of 40 patients with plaque stage FMF at diagnosis were evaluated. The aim of this study was to describe histopathological characteristics of plaque stage FMF and assess parameters associated with disease progression and survival.

| Patients and methods
The database of the Dutch Cutaneous Lymphoma Group was searched for patients with FMF, who met the following criteria: the clinical presence of plaques without concurrent tumors or extracutaneous disease at the time of diagnosis, availability of clinical illustrations, and availability of relevant immunostainings or tissue blocks to perform additional stains. Plaques were defined clinically as palpable, mostly elevated skin lesions with a diameter > 1 cm. 14 In order to study the risk of disease progression or death of lymphoma, only patients with a follow-up period of at least 60 months or death before that time were selected. The total study group included 40 patients with plaque stage FMF. In all cases, the diagnosis of FMF met the criteria of the WHO-EORTC classification and was confirmed by an expert panel of dermatologists and pathologists of the Dutch Cutaneous Lymphoma Group. 15 Clinical and follow-up data were retrieved from the Dutch Cutaneous Lymphoma Registry and from medical records. The date of first diagnostic biopsy was considered the time of diagnosis. Disease progression was defined as the development of skin tumors (stage IIB) or extracutaneous disease (stage IV) or death due to lymphoma. The study was performed in accordance with the Declaration of Helsinki and in accordance with our in-house Biobank protocol, as approved by our Medical Ethics Committee.

| Clinical characteristics
The study included 29 males and 11 females (ratio: 2.6) with a median age at diagnosis of 56 years (range: 19-82 years). Distinction was made between patients with disease progression (P-FMF) and patients without disease progression for at least 5 years after diagnosis (NP-FMF). In the P-FMF group, the median time from diagnosis to disease progression was 30 months (range, 3-136 months) and occurred within 5 years after diagnosis in 17 of 20 patients and after 61, 66, and 136 months in the other three patients. The main clinical characteristics of these two groups are presented in Table 1.
The median age at diagnosis of patients in the P-FMF group was higher than in the NP-FMF group (62 vs 51 years, respectively, P = 0.004). No significant differences between the P-FMF group and the NP-FMF group were found in the extent of skin lesions, the predominant involvement of the head and neck region, the presence of pruritus, and the initial type of treatment. The number of patients achieving complete remission upon initial treatment was higher in the NP-FMF group than in the P-FMF group (60% vs 15%, respectively, P = 0.003), but overall response rates were comparable (75% vs 65%, respectively).

| Histopathology
The main histopathological findings in these 40 patients with plaque stage FMF are summarized in Table 2. In 35 of 40 cases, biopsies were obtained from skin lesions in the head and neck area. In two cases of the NP-FMF group and three cases of the P-FMF group, skin lesions on the back (four cases) or arm (one case) were biopsied.
Twenty cases showed sparse (n = 4) or prominent (n = 14) infiltrates confined to the perifollicular area (Figures 1 and 2). Confluent perifollicular and perivascular infiltrates or completely diffuse infiltrates were observed in the other 22 cases, and were more common in the P-FMF group (Figures 3 and 4). , and was much more common in the P-FMF group than in the NP-FMF group (50% vs 11%, respectively P = 0.01)). Epidermotropism was mild in six cases and prominent in another six cases, but was always less pronounced than coexistent folliculotropism. Syringotropism was seen in seven of 40 cases, without any difference between the two subgroups ( Figure 5).

Percentages of atypical cells, including cells with hyper-
chromatic and cerebriform nuclei and blast cells, in the dermal infiltrates were less than 10% in seven cases, 10% to 25% in 12 cases and more than 25% in 21 cases. Percentages above 25% were more frequent in the P-FMF than in the NP-FMF group (75% vs 30%, respectively; P = 0.02). In 33 of 40 cases (83%), these atypical cells were predominantly small to medium-sized ( Figures 2B and 4A). In the other seven cases, all seven belonging to the P-FMF group, they were predominantly medium-sized to

| Prognostic factors
Factors that significantly correlated with both disease progression and reduced survival included the presence of more than 25% atypical cells in dermal infiltrates, the size of neoplastic T-cells, the presence of coexistent interfollicular epidermotropism and presence of more than 10% Ki-67 positive cells. Presence of more than 10% blast cells was associated increased disease progression. In contrast, presence of prominent follicular mucinosis was associated with reduced disease progression and increased survival ( Table 3). The extent of perifollicular infiltrates, which had been used in our previous study as one of the parameters to differentiate between early and advanced plaque stage disease, 5 had no effect on prognosis and disease progression. Other factors that did not correlate with disease progression or survival were extent of folliculotropism, presence or absence of syringotropism,    In a recent study of Hodak et al, extent and depth of perifollicular infiltrates were significantly greater in advanced-stage FMF than in early-stage FMF. In addition, eosinophils and plasma cells in dermal infiltrates as well as pruritus were significantly more common in advanced-stage FMF. 7 However, in that study early-stage FMF included follicle-based patches and acneiform and keratosis pilaris-like lesions that mainly involved trunk and extremities, while advanced-stage FMF did not only include tumors, but also infiltrated plaques that preferentially involved the head and neck region. It should be noted that our study included only plaques and that in 80% of cases biopsies had been taken from the most infiltrated lesions in the head and neck area. This might explain why extent and depth of perifollicular infiltrates as well as the other abovementioned risk factors were not different between the two groups in the present study.
The observation that co-existent interfollicular epidermotropism is a risk factor for disease progression and reduced survival was unexpected and is at present unexplained. In 11 of 12 cases, with co-existent epidermotropism skin biopsies were obtained from characteristic FMF lesions in the face and none of these patients had concurrent skin lesions typical of classic MF.
Previous studies in classic MF showed that low numbers of admixed CD8+ T-cells are associated with disease progression and an inferior prognosis. [17][18][19] In the present study, such an association was not found. Notably, number of reactive CD8+ T-cells infiltrating into the follicular epithelium was very low, similar to the few or absent reactive CD8+ T-cells in the epidermis of early-stage classic MF.
Furthermore, we found that age at diagnosis and lack of complete response upon initial treatment were clinical prognostic factors associated with disease progression in plaque-stage FMF. Age is a well-reported prognostic factor in both classic MF and FMF, while complete response to initial treatment has been reported as a favorable prognostic factor in classic MF. 5,6,[20][21][22][23] In conclusion, the results of the present study show that FMF patients presenting with plaques represent a heterogeneous group and that a subgroup of these patients may have an indolent clinical course. Our results argue against the suggestion that all cases of FMF with plaques should be considered to have tumor-stage disease. They further show that histopathological examination is a valuable tool to differentiate between indolent and aggressive disease.