Primary cutaneous anaplastic large cell lymphoma

INTRODUCTION — Primary cutaneous anaplastic large cell lymphoma (PC-ALCL) is a CD30+ peripheral T cell lymphoproliferative disorder with no systemic involvement.

The CD30+ family of disorders includes a spectrum of related conditions that range from lymphomatoid papulosis, a benign process that typically regresses spontaneously, to PC-ALCL, which is a generally indolent malignancy that can occasionally behave aggressively; there are also borderline cases that exhibit features of both lymphomatoid papulosis and PC-ALCL.

Most patients with PC-ALCL present with slowly growing solitary or grouped skin nodules, and some patients manifest regional lymph node involvement. The skin lesions of PC-ALCL often regress spontaneously, but they only rarely resolve completely, and cutaneous recurrences are common. By contrast to systemic ALCL, PC-ALCL typically follows an indolent course. PC-ALCL is very treatment-responsive, and it is important to avoid overtreatment.

This topic discusses the evaluation, diagnosis, and management of PC-ALCL.

Related topics include:

●(See "Lymphomatoid papulosis".)

●(See "Clinical manifestations, pathologic features, and diagnosis of systemic anaplastic large cell lymphoma (sALCL)".)

●(See "Initial treatment of systemic anaplastic large cell lymphoma (sALCL)".)

EPIDEMIOLOGY — 

The incidence of PC-ALCL is not well defined.

PC-ALCL represents approximately 8 percent of cases of cutaneous lymphomas [1,2]. The incidence is probably underreported because it is difficult to distinguish PC-ALCL within the larger group of CD30+ cutaneous lymphoproliferative disorders (ie, lymphomatoid papulosis and borderline cases). Analysis of the United States Surveillance, Epidemiology, and End Results (SEER) database reported 157 cases of primary, localized CD30+ cutaneous lymphoproliferative disorder over a 30-year timespan [3]. Median age at diagnosis was 61 years (range 5 to 98 years). There was a slight male predominance (58 percent) and most cases were diagnosed in White individuals (87 percent). Other epidemiologic reports have had similar findings [4,5].

CLINICAL PRESENTATION — 

Classically, PC-ALCL presents as solitary or grouped nodules on the upper half of the body that grow over weeks to months. Nodules are up to several centimeters in diameter and they often ulcerate (picture 1). Nodules regress spontaneously in up to one-half of cases, but only rarely do they resolve completely [6]. Cutaneous patches and plaques may also be present. Multifocal disease is rare at the time of diagnosis (picture 2 and picture 3).

Unlike systemic ALCL, PC-ALCL typically follows an indolent course. Cutaneous relapses are common, but the prognosis remains excellent [7,8]. Extracutaneous spread occurs in approximately 10 percent of cases at relapse [1,2,9].

EVALUATION AND DIAGNOSIS

Clinical evaluation — History, physical examination, and basic laboratory studies should be performed.

●History – Nature and duration of the skin lesions and associated symptoms.

●Physical examination – Complete skin examination and evaluation of lymph nodes, liver, and spleen.

●Laboratory studies – Complete blood count with differential, comprehensive metabolic panel, and lactate dehydrogenase.

Imaging and bone marrow examination are generally not needed unless there is suspicion for involvement of extracutaneous sites. (See 'Staging' below.)

Pathology — An adequate biopsy of a skin lesion is required. Extracutaneous sites suspected of involvement by examination or imaging should be biopsied.

●Skin biopsy – Skin biopsy is evaluated for morphology, growth pattern, immunophenotype, and cytogenetics. Excisional biopsies are preferred to incisional biopsies; if a punch biopsy is taken, the diameter should be at least 4 millimeters.

•Morphology – Cutaneous lesions are characterized by a dense dermal infiltrate that usually does not involve the epidermis.

The dermal infiltrate comprises cohesive sheets of large tumor cells with abundant pale eosinophilic cytoplasm and round, pleomorphic, or horseshoe-shaped nuclei with prominent nucleoli; approximately one-quarter of the malignant cells have an atypical, nonanaplastic appearance. Reactive lymphocytes, histiocytes, eosinophils, and neutrophils are often present in the periphery of the lesion and in ulcerating lesions in association with epidermal hyperplasia [1,2]. (See "Clinical manifestations, pathologic features, and diagnosis of systemic anaplastic large cell lymphoma (sALCL)", section on 'Classical variant'.)

•Immunophenotype – By definition, ≥75 percent of the tumor cells express CD30 [1,2,10]. Most cases express CD4 and have variable loss of CD2, CD3, and CD5. Many cases are positive for cutaneous lymphocyte antigen (CLA, HECA-452), but unlike systemic ALCL (sALCL), they lack expression of epithelial membrane antigen and anaplastic lymphoma kinase (ALK). (See "Clinical manifestations, pathologic features, and diagnosis of systemic anaplastic large cell lymphoma (sALCL)", section on 'Immunophenotype'.)

•Cytogenetics – Most cases show clonal rearrangement of T cell receptor (TCR) genes, although TCR proteins are often not expressed [1,2,10]. Unlike sALCL, genetic rearrangements involving the ALK gene are extremely rare in PC-ALCL [11,12]. Detection of an ALK rearrangement should provoke investigation for sALCL.

Cases of PC-ALCL with DUSP22::IRF4 rearrangement (involving chromosome 6p25.3) typically demonstrate a biphasic histologic pattern, with larger transformed cells infiltrating the dermis and smaller atypical cells infiltrating the epidermis; these cases generally follow an indolent course [13,14]. A case series reported DUSP22 rearrangements in 50 percent of cases of PC-ALCL, but no such rearrangements were seen in other primary cutaneous T cell lymphomas or lymphomatoid papulosis [15]. Abnormalities of PI3 kinase, MAP kinase, and G-protein signal transduction pathways are frequently seen in such cases [16].

●Other sites – Excisional or core needle biopsy should be performed for enlarged lymph nodes or other suspected extracutaneous sites of involvement.

It is important to biopsy enlarged lymph nodes that drain sites of cutaneous disease. Distinguishing lymphadenopathy caused by dermatopathic changes (ie, inflammatory reaction or infection) versus tumor involvement has important management implications, as discussed below. (See 'Localized skin disease' below and 'Multifocal cutaneous disease' below.)

Bone marrow aspirate/biopsy is generally not required unless there is clinical or imaging evidence for extracutaneous involvement or unexplained hematologic abnormalities.

Diagnosis — PC-ALCL should be suspected in patients with ulcerating skin nodules.

Diagnosis requires a skin biopsy that demonstrates sheets of large tumor cells with anaplastic, pleomorphic, or immunoblastic morphology in which >75 percent of cells express CD30 [1,2]. Details of morphology, immunophenotype, and cytogenetics are presented above. (See 'Pathology' above.)

There must be no evidence of systemic lymphoma or mycosis fungoides (MF), although regional lymph node involvement does not exclude a diagnosis of PC-ALCL.

Differential diagnosis — PC-ALCL should be distinguished from benign conditions (eg, lymphomatoid drug reactions, arthropod bites, viral infections), lymphomatoid papulosis, reactive lymphoid hyperplasia, MF, and systemic lymphomas.

●Reactive lymphoid hyperplasia – CD30+ lymphoid infiltrates may develop in response to insect bites or infestations (eg, scabies), viral infections (eg, herpes simplex, varicella-zoster), and certain medications (eg, anticonvulsant medications, amlodipine, carbamazepine, cefuroxime, valsartan) [17]. Nonneoplastic causes of lymphoid hyperplasia are usually associated with a lower percentage of CD30+ cells on biopsy, do not have a waxing and waning clinical course, and do not show clonal TCR rearrangement. (See "Lymphomatoid papulosis", section on 'CD30+ inflammatory and infectious diseases'.)

●Lymphomatoid papulosis – Lymphomatoid papulosis is a recurrent, benign, self-healing condition characterized by a papulonodular skin eruption with proliferation of CD30-expressing atypical T cells. Lesions of lymphomatoid papulosis are usually smaller (<2 cm) than those of PC-ALCL and resolve spontaneously after weeks to months [17-19]. Lesions of lymphomatoid papulosis may leave residual scars and hyperpigmentation [17-19].

There is considerable clinical, histologic, and immunophenotypic overlap between lymphomatoid papulosis and PC-ALCL. A waxing and waning clinical course is consistent with lymphomatoid papulosis, while skin lesions that do not spontaneously regress within eight weeks are more suggestive of PC-ALCL. The term "borderline" is used for cases in which a distinction between PC-ALCL and lymphomatoid papulosis cannot be made, despite careful clinical and pathologic evaluation [1,2]. Particularly difficult to distinguish from PC-ALCL are cases of lymphomatoid papulosis with biphasic histology and rearrangements involving 6p25.3 [20,21]. (See "Lymphomatoid papulosis".)

●Mycosis fungoides – MF and Sézary syndrome (SS; a leukemic variant) are cutaneous T cell lymphomas that originate in skin, but they can also involve lymph nodes, viscera, and blood.

A history of patch or plaque disease or erythroderma is suggestive of MF/SS. Skin biopsies demonstrate small to medium-sized atypical mononuclear cells with cerebriform nuclei, which differ from the large anaplastic tumor cells of PC-ALCL. However, large cell transformation of MF can resemble the tumor cells of PC-ALCL, and CD30 is expressed in up to one-half of cases of transformed MF/SS. MF and PC-ALCL can coexist in the same patient. Details of pathologic features and diagnostic criteria for MF and SS are presented separately. (See "Clinical manifestations, pathologic features, and diagnosis of mycosis fungoides" and "Clinical presentation, pathologic features, and diagnosis of Sézary syndrome".)

●Systemic lymphomas – Staging evaluation in patients with presumed PC-ALCL should confirm the absence of systemic disease. Some cases of peripheral T cell lymphoma can present with skin lesions. Involvement of lymph nodes does not necessarily indicate systemic disease, since pathologic involvement of loco-regional nodes alone is possible with PC-ALCL, and does not impact prognosis [5].

•sALCL – Expression of ALK or identification of ALK translocation is highly suggestive of cutaneous involvement of sALCL rather than PC-ALCL. (See "Clinical manifestations, pathologic features, and diagnosis of systemic anaplastic large cell lymphoma (sALCL)", section on 'Immunophenotype'.)

•Adult T cell leukemia-lymphoma (ATLL) – ATLL can present with cutaneous lesions that are difficult to distinguish from PC-ALCL. The cells can be anaplastic in appearance, express CD30, and are ALK negative. ATLL is caused by the human retrovirus HTLV-1, and the key distinguishing feature is the presence of HTLV-1 in the malignant cells of ATLL. (See "Clinical manifestations, pathologic features, and diagnosis of adult T cell leukemia-lymphoma".)

•Hodgkin lymphoma (HL) – HL expresses CD30, but it rarely presents with cutaneous involvement. Unlike PC-ALCL, the malignant cells of HL express CD15. (See "Classic Hodgkin lymphoma: Presentation, evaluation, and diagnosis in adults".)

STAGING — 

Staging is based on the clinical evaluation and biopsy of suspected lymph nodes/other extracutaneous sites (as warranted). Staging guidelines are presented by the Tri-Societies (the International Society for Cutaneous Lymphomas, the United States Cutaneous Lymphoma Consortium, and the Cutaneous Lymphoma Task Force of the European Organisation for Research and Treatment of Cancer) [22].

If there is clinical suspicion for extracutaneous disease, contrast-enhanced computed tomography (CT) of the chest, abdomen, and pelvis and/or positron emission tomography/CT should be performed. Suspected extracutaneous involvement must be confirmed by biopsy, since local or regional nodes may be abnormal due to inflammation alone.

We generally reserve bone marrow examination for patients with multifocal tumors, unexplained cytopenias, and those with documented extracutaneous disease [19].

INITIAL MANAGEMENT — 

The initial management of PC-ALCL is guided by the location, extent, and stage of disease.

Most data regarding the management of PC-ALCL come from small series, case reports, and observational studies. Interpretation of the available evidence is further complicated by the fact that patients with relapsing disease are treated with several therapies concurrently, thereby limiting assessment of the efficacy of individual modalities. Nevertheless, it is important to avoid overtreatment of patients with PC-ALCL. Clinicians unfamiliar with these diseases may recommend aggressive systemic chemotherapy, even though such treatments are unnecessary and overly toxic.

Localized skin disease — Patients with localized skin disease (solitary or grouped skin lesions at one site) who have enlarged regional lymph nodes (ie, nodes that drain involved skin sites) should have a lymph node biopsy to distinguish inflammatory or infectious causes of lymphadenopathy from tumor involvement.

The management of localized skin disease is stratified according to nodal involvement.

No nodal involvement — For localized skin disease without regional lymph node involvement, we suggest involved site radiation therapy (ISRT) or surgical excision rather than surgery plus ISRT or observation. Either surgery or ISRT can effectively control localized disease with little morbidity, and there is no demonstrated benefit from adding ISRT after surgical excision.

The choice of surgery versus ISRT is individualized based on the location of the lesions and patient preference. The optimal dose of radiation for PC-ALCL is uncertain, but low-dose radiation therapy (RT; ≤20 gray [Gy]) appears to be as effective as higher doses [23-25].

There is no evidence that combining surgery with ISRT or adding chemotherapy improves outcomes in this setting.

●A study from the Dutch tumor registry of 79 patients with PC-ALCL with no lymph node involvement included nearly one-quarter who had disease at two or more sites [5]. Initial management included RT (47 percent of patients), surgical excision (22 percent), no treatment (because of spontaneous disappearance of skin lesions; 15 percent), and doxorubicin-based multiagent chemotherapy (for some patients with multifocal skin disease; 8 percent). Disease-related five-year survival was 95 percent. Cutaneous lesions subsequently developed at the initial site of disease or at another site in 42 percent of patients, but only 10 percent of patients developed extracutaneous disease.

●A study from the British Columbia tumor registry reported limited-stage disease in 80 percent of 47 patients with PC-ALCL [26]. Treatment included RT alone (19 patients), surgery alone (six patients), RT plus chemotherapy (combined modality treatment [CMT]; nine patients), and chemotherapy alone (three patients). Initial treatment resulted in an 89 percent complete response (CR) and 11 percent partial response (PR). Cutaneous relapse occurred in 40 percent of patients within five years, but only 11 percent of patients experienced systemic lymphoma at first relapse. Five-year rates of overall survival (OS) and disease-specific survival (DSS) were 75 and 86 percent, respectively; systemic lymphoma developed in 14 percent. DSS was similar for patients treated with RT, chemotherapy, and CMT, but time to progression favored RT over chemotherapy or CMT.

●In a multicenter European study, 42 of 47 patients with PC-ALCL had solitary or localized disease [27]. Patients underwent surgical excision (23 patients), RT (nine patients), surgery followed by RT (four patients), or chemotherapy (six patients); four patients received no treatment because of spontaneous regression of the lesion. All but one patient had a CR. The median OS was 42 months, and four patients ultimately died of lymphoma-related causes. Cutaneous relapse occurred in 27 patients, including 12 with nodal relapse.

●A retrospective multicenter study of 56 patients who received RT alone or after surgical excision reported that, with a median follow-up of four years, 98 percent maintained local control and 95 percent had complete clinical CR (cCR) [28]. The median RT dose was 35 Gy (range of 6 to 45 Gy), but CR was seen with doses as low as 6 Gy, and the achievement of cCR was independent of the RT dose.

Involved regional nodes — For patients with localized skin disease plus regional lymph node involvement, we suggest ISRT to the skin lesion(s) and regional nodes.

Treatment and outcomes with ISRT are discussed above. (See 'No nodal involvement' above.)

Multifocal cutaneous disease — The management of multifocal cutaneous disease is informed by the extent of disease, tumor involvement of regional lymph nodes, and presence of systemic symptoms.

Enlarged regional lymph nodes should be biopsied to distinguish enlargement due to dermatopathic changes from nodal involvement by PC-ALCL. (See 'Pathology' above.)

For multifocal cutaneous disease with biopsy-proven regional lymph node involvement, we manage as symptomatic multifocal cutaneous disease, whether the patient is symptomatic. (See 'Symptomatic and/or regional node involvement (symptomatic or asymptomatic)' below.)

Asymptomatic with no regional node involvement — For asymptomatic multifocal PC-ALCL with no regional lymph node involvement, we suggest a period of initial observation rather than immediate surgical excision, ISRT, or systemic therapy. This approach avoids treatment-related adverse effects (AEs) while awaiting possible spontaneous regression.

Spontaneous regression of cutaneous lesions occurs in approximately 10 to 15 percent of patients. Some patients may choose surgical excision or RT for cosmetically disturbing lesions or potentially scarring sites of disease. For large skin lesions, we generally follow the patient for four to six weeks to await spontaneous remission; if the lesion does not resolve spontaneously, it can be excised or treated with ISRT according to patient preference. We treat with brentuximab vedotin if multifocal disease progresses during observation. The likelihood of subsequent cutaneous recurrence or systemic relapse is not affected by surgery, ISRT, or chemotherapy, as discussed above. (See 'Localized skin disease' above.)

●A multicenter Dutch study reported outcomes of 24 patients with initial presentation of multifocal PC-ALCL and 17 patients with relapsed disease [29]. Four patients had spontaneous regression of disease, 21 received RT, 7 received low-dose methotrexate, and 9 were treated with other systemic chemotherapy. Overall response rate (ORR) and CR rate with RT were both 100 percent; corresponding rates were 100 and 78 percent with combination chemotherapy and 57 and 43 percent with methotrexate. Two-thirds of newly diagnosed patients subsequently relapsed. Compared with patients who presented with two to five skin lesions, patients presenting with ≥5 lesions were more likely to develop extracutaneous relapse (56 versus 20 percent) and more often died of lymphoma (44 versus 7 percent).

●Spontaneous regression of lesions occurred in 10 to 15 percent of patients, with recurrent cutaneous lesions in nearly one-half of patients regardless of initial therapy, in studies described above [5,27]. (See 'Localized skin disease' above.)

Symptomatic and/or regional node involvement (symptomatic or asymptomatic) — For multifocal PC-ALCL accompanied by biopsy-proven regional lymph node involvement and/or systemic symptoms (eg, fevers, chills, night sweats, weight loss), we suggest brentuximab vedotin rather than other systemic treatments, based on superior outcomes and acceptable toxicity.

Brentuximab vedotin is an intravenous medication that achieved deeper and longer lasting responses than methotrexate or bexarotene in a randomized trial of cutaneous T cell lymphoma, but it is associated with peripheral neuropathy. Some patients may favor methotrexate because it is an oral medication or because they seek to avoid peripheral neuropathy. Other single agents or combination chemotherapy offer a less favorable balance of efficacy and toxicity. For patients with pre-existent neuropathy or an inadequate response to brentuximab vedotin, ISRT to skin and/or involved nodes is acceptable, but it may be impractical for treating >2 sites of disease.

The administration and toxicity of brentuximab vedotin are discussed below. (See 'Brentuximab vedotin' below.)

●The phase 3 ALCANZA trial of 131 patients with cutaneous T cell lymphoma (31 with PC-ALCL) reported that brentuximab vedotin achieved superior responses compared with methotrexate or oral bexarotene [30]. Patients were randomly assigned to intravenous brentuximab vedotin once every three weeks (for up to 16 cycles) versus up to 48 weeks treatment with the investigator's choice of oral methotrexate once per week or oral bexarotene once daily. Among 31 patients with PC-ALCL, brentuximab vedotin achieved superior median progression-free survival (PFS; 28 versus 5 months; hazard ratio [HR] 0.25 [95% CI 0.08-0.79]), objective global response sustained for ≥4 months (75 versus 20 percent), CR (31 versus 7 percent), patient-reported symptom relief, and duration of treatment. Two-thirds of patients with PC-ALCL had biopsy-proven regional lymph node involvement. Grade ≥3 AEs were reported in 47 percent in the investigator's choice group and in 41 percent of patients in the brentuximab vedotin group (including 9 percent grade ≥3 peripheral neuropathy); one death was attributed to brentuximab vedotin [30]. One-half of the patients who had peripheral neuropathy with brentuximab vedotin required at least one dose modification (delay, reduction, or dose held) and 14 percent permanently discontinued it.

●The final results of the ALCANZA trial (median follow-up 46 months) confirmed the durable and robust clinical benefits of brentuximab vedotin [31]. Patients with PC-ALCL who received brentuximab vedotin had superior PFS (28 versus 5 months; HR 0.35 [95% CI 0.13-0.95]), superior median time to next treatment (21 versus 7 months), and did not require subsequent antineoplastic treatment within one year (71 versus 31 percent). With the longer follow-up, 86 percent of cases of peripheral neuropathy in patients treated with brentuximab vedotin resolved completely or improved by at least one grade; there were no cases of ongoing grade ≥3 peripheral neuropathy.

By contrast, multiagent chemotherapy, such as CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) can yield high rates of ORR, but responses are generally transient, and toxicity is significant. As an example, in a Dutch Cutaneous Lymphoma Group study, combination chemotherapy in 24 patients who presented with multifocal PC-ALCL was associated with a 100 percent ORR and 78 percent CR, but two-thirds of patients later relapsed [29]. In another study, a review of 53 patients treated with multiagent chemotherapy reported a 92 percent CR, but 62 percent of patients relapsed at a median of four months [19].

Widespread nodal and/or visceral disease — Patients with widespread nodal disease (ie, beyond nodes that drain sites of cutaneous involvement) and/or visceral disease should be treated like systemic ALCL, as discussed separately. (See "Initial treatment of systemic anaplastic large cell lymphoma (sALCL)".)

RESPONSE ASSESSMENT — 

Treatment response is judged by control of cutaneous and nodal sites of involvement and symptom relief.

Outside of a clinical trial, we assess response according to physical examination and the patient's report of symptoms. If there was biopsy-proven extranodal involvement before treatment, we consider repeat imaging studies.

Although formal criteria for judging the response of cutaneous lymphomas in clinical trials have been published [22], it is not necessary to apply these criteria outside of a clinical trial. We routinely use photography to document the extent and bidimensional measurements of skin lesions.

RECURRENT DISEASE — 

The management of recurrent PC-ALCL is individualized according to the sites and extent of relapse, prior treatments, and patient preference.

Nearly all patients with PC-ALCL have a satisfactory response to initial surgery or radiation therapy (RT), but most patients have one or more skin recurrences, and some patients experience nodal or visceral relapse. Many patients receive various treatment modalities over the course of their disease.

Localized cutaneous relapse — For patients with a localized cutaneous relapse of PC-ALCL, we suggest initial observation rather than immediate surgery or RT. This avoids adverse effects (AEs) in the subset of patients who experience spontaneous regression of the lesion.

For symptomatic lesions and for recurrent lesions that do not spontaneously remit (eg, within two months), some patients favor RT or surgical excision, but multiple surgeries and/or episodes of RT can cause significant morbidity and patient inconvenience.

Symptomatic patients may find relief using a topical steroid (eg, super-high potency topical corticosteroids [group 1 (table 1)]) or other topical treatments (eg, topical bexarotene gel). Note that topical bexarotene can cause an intense localized retinoid dermatitis that may require concomitant use of a topical corticosteroid for relief [32,33]. For thicker lesions, we often use intralesional corticosteroids because topical agents may not penetrate adequately, and intralesional injections may avoid the need for systemic corticosteroids and the associated AEs. Case reports and small series have described treatment with imiquimod (an immune response modifier) [34] or intralesional interferon alfa [35]; patients using these agents should be followed in conjunction with a dermatologist who is experienced with their use.

When multiple surgical resections, RT treatments, or other approaches are ineffective or cause excessive morbidity, we manage localized cutaneous relapses like multiple/multifocal recurrences. (See 'Multifocal or multiple cutaneous relapses' below.)

Multifocal or multiple cutaneous relapses — For multiple cutaneous recurrences after surgery and/or RT, and patients with multifocal cutaneous relapses, management is guided by prior therapy:

●We suggest brentuximab vedotin for patients who have none of the following:

•Treatment with brentuximab vedotin within the past six months

•Disease progression while using brentuximab vedotin

•Pre-existent peripheral neuropathy or discontinuation of brentuximab vedotin due to neuropathy or other AEs

This suggestion is based on retreatment with brentuximab vedotin in patients with systemic ALCL [36]. In that setting, retreatment with brentuximab vedotin was associated with an 88 percent overall response rate, including a 63 percent complete response.

●For patients with multifocal or multiple cutaneous relapses who have intolerance to, progression on, or recent (<6 months) treatment with brentuximab vedotin, we treat as for widespread or multiple relapses. (See 'Widespread and/or multiple relapses' below.)

Skin-directed treatments can be added to maximize clinical responses in the skin compartment and provide added efficacy without cumulative toxicities.

Widespread and/or multiple relapses — For widespread cutaneous, nodal, or visceral relapse and for recurrence after brentuximab vedotin, we suggest sequential single agents to prolong disease control with acceptable toxicity.

Beyond methotrexate and bexarotene, other possible single agents include pralatrexate, mogamulizumab, romidepsin, and interferon. No agent has proven to be superior, and the choice of agents and their sequence should be individualized, with consideration of toxicity, comorbidities, and patient preference. We generally begin with better-tolerated systemic therapies and reserve treatments with a higher risk of cumulative toxicity and/or immunosuppression until after exhausting better-tolerated options.

The selection of single agents for treatment of multiple cutaneous or nodal relapses is like that for recurrent advanced-stage mycosis fungoides, as discussed separately. (See "Treatment of advanced-stage (IIB to IV) mycosis fungoides", section on 'Systemic treatments'.)

SYSTEMIC TREATMENTS

Brentuximab vedotin — Brentuximab vedotin is an anti-CD30 monoclonal antibody conjugated with monomethyl auristatin E that has activity against CD30-expressing lymphomas.

●Administration – The typical starting dose of brentuximab vedotin is 1.8 mg/kg (to a maximum dose of 180 mg) given intravenously every three weeks. Treatment can be given indefinitely until intolerance or disease progression.

Brentuximab vedotin is approved by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) for the treatment of cutaneous T cell lymphoma in patients who received ≥1 prior therapy.

●Toxicity – Peripheral neuropathy is the most common dose-limiting adverse effect (AE) [37]. This can generally be managed with dose reduction (eg, to 1.2 mg/kg) or by increasing the dosing interval.

Infusion reactions are uncommon, but anaphylaxis has been reported. (See "Infusion-related reactions to monoclonal antibodies for cancer therapy", section on 'Brentuximab'.)

Progressive multifocal leukoencephalopathy is a rare complication of brentuximab vedotin treatment and typically presents with subacute neurologic deficits, which may include altered mental status, visual symptoms, weakness, ataxia, and seizures. (See "Progressive multifocal leukoencephalopathy (PML): Epidemiology, clinical manifestations, and diagnosis" and "Neurologic complications of cancer treatment with molecularly targeted and biologic agents", section on 'Brentuximab'.)

Outcomes with brentuximab vedotin for PC-ALCL in the phase 3 ALCANZA trial are discussed above. (See 'Symptomatic and/or regional node involvement (symptomatic or asymptomatic)' above.)

Methotrexate — Low-dose oral methotrexate has moderate activity against PC-ALCL, is easily administered, and has a favorable toxicity profile, as discussed above. (See 'Symptomatic and/or regional node involvement (symptomatic or asymptomatic)' above.)

●Administration – A typical starting dose is methotrexate 7.5 mg orally twice per day (total 15 mg) on two days per week. The dose can be adjusted based upon efficacy or tolerance. We increase the dose if there is no clinical improvement within eight weeks of therapy, but we do not typically use doses higher than 25 mg weekly.

Methotrexate is contraindicated in patients with liver disease, kidney disease, and pregnancy or planned pregnancy (for both females and males).

Patients should be screened for viral hepatitis before initiating long-term methotrexate therapy. Those with hepatitis B or hepatitis C infection should be treated with an alternative systemic therapy, receive treatment to eradicate the infection before beginning methotrexate, or receive treatment to suppress viral replication during treatment. Complete blood count, aminotransaminases, and albumin should be monitored periodically.

Patients should take folic acid 1 mg daily on the five days of the week when they are not taking methotrexate. Proton-pump inhibitors can increase methotrexate levels and should be used with caution in this population.

Methotrexate is not approved by the FDA or EMA for the treatment of cutaneous T cell lymphomas, but it has long been used in this setting.

●Duration of therapy – We continue treatment until best response and then decrease the frequency of methotrexate, which can be continued as maintenance therapy for a maximum of three years. In our experience, even once monthly dosing can be effective to maintain remissions.

We limit maintenance to three years to reduce the risk of liver fibrosis. Methotrexate can be reinstituted if there is evidence of recurrence of disease.

Methotrexate can be reinstituted if there is a disease recurrence.

●Toxicity – Minor AEs, including nausea, stomach upset, headache, and fatigue, occur in most patients treated with low-dose methotrexate.

Hepatotoxicity, pulmonary fibrosis, and myelosuppression are serious AEs of methotrexate treatment and may infrequently occur with low-dose therapy. (See "Major adverse effects of low-dose methotrexate".)

●Outcomes – Outcomes of treatment with oral methotrexate in the ALCANZA trial are discussed above. (See 'Symptomatic and/or regional node involvement (symptomatic or asymptomatic)' above.)

In another study, treatment with oral methotrexate was associated with a 77 percent overall response rate in 13 patients with PC-ALCL [38]. Most patients responded within four weeks, and the median duration of methotrexate therapy was 39 months.

In a series of patients with PC-ALCL or lymphomatoid papulosis, 5 of 10 patients who were treated with methotrexate (median dose 20 mg per week for ≥39 months) had evidence of early hepatic fibrosis [38].

Other systemic treatments — Other systemic treatments can be used for patients who had intolerance to or disease progression while taking methotrexate and brentuximab vedotin.

Other agents that can be used to treat multiply relapsed or advanced disease are discussed separately. (See "Treatment of advanced-stage (IIB to IV) mycosis fungoides", section on 'Systemic treatments'.)

Bexarotene — Bexarotene is a systemic rexinoid (a synthetic agent that binds to the retinoid "X" receptor) that is widely used to treat mycosis fungoides (MF). We generally reserve bexarotene for patients with PC-ALCL who have a contraindication to methotrexate or have failed methotrexate due to disease progression or toxicity.

●Administration – We generally start bexarotene at 225 mg total daily dose and adjust the dose based on tolerance and response. In most circumstances, we do not exceed 300 mg/m² daily, but doses as high as 400 mg/m² may be used in rare circumstances.

Bexarotene must be used with caution in patients with hypertriglyceridemia, liver dysfunction, or risk factors for pancreatitis. Given the high incidence of hyperlipidemia with bexarotene, we prescribe concomitant fenofibrate. Additional lipid-lowering agents are sometimes necessary. Gemfibrozil should not be used because coadministration results in increased serum levels of bexarotene.

Bexarotene also causes central hypothyroidism, and thyroid replacement must be adjusted based upon free T4 and not thyroid-stimulating hormone (TSH).

Complete blood count, liver function tests, serum lipid levels, and thyroid function (serum free T4) should be monitored during treatment.

Further details of systemic bexarotene administration are presented separately. (See "Treatment of advanced-stage (IIB to IV) mycosis fungoides", section on 'Bexarotene'.)

●Toxicity – The most common AEs are weakness, myalgia, arthralgia, and headaches. Most bexarotene toxicities are reversible after cessation of therapy.

Retinoids have potential hepatotoxic and hyperlipidemic effects, necessitating monitoring. Women of childbearing potential should use two reliable forms of contraception, including at least one nonhormonal form, given the potential teratogenic effects of retinoids.

●Outcomes – Data regarding bexarotene for the treatment of PC-ALCL are limited to case reports and the control arm of the ALCANZA trial, which reported that the response rate to bexarotene was low [30], as discussed above. (See 'Symptomatic and/or regional node involvement (symptomatic or asymptomatic)' above.)

Case reports describe modest activity of bexarotene against PC-ALCL [32,33,39-41].

Interferon — Peginterferon alfa-2a is the only interferon that is available in the United States.

Interferons are parenteral agents that have been used off-label for decades to treat MF/Sézary syndrome, but there are only limited data with systemic treatment for PC-ALCL. We generally reserve interferon therapy for patients who progress on or are intolerant of other treatment options because of its cumbersome method of administration and toxicity.

Administration and toxicity are discussed separately. (See "Treatment of advanced-stage (IIB to IV) mycosis fungoides", section on 'Systemic treatments'.)

Outcomes with interferon are limited to case reports, usually in combination with other agents (eg, bexarotene) [42,43].

ADJUNCTIVE TREATMENTS — 

Adjunctive treatments to manage cutaneous disease-related symptoms and potentially reduce the need for systemic therapies can benefit patients with all stages and phases of PC-ALCL.

Examples include:

●Super-high potency topical corticosteroids, given topically or injected intralesionally (group 1 (table 1)).

●Topical bexarotene gel [32,33] – Topical bexarotene can cause an intense localized retinoid dermatitis that often necessitates concomitant administration of topical corticosteroids.

●Intralesional interferon alfa [35].

●Imiquimod (an immune response modifier) [34].

The use of these agents for the treatment of mycosis fungoides is discussed separately. (See "Treatment of early stage (IA to IIA) mycosis fungoides", section on 'Skin-directed therapies'.)

PROGNOSIS — 

Most cases of PC-ALCL have an excellent prognosis, but relapses are common.

The United States Surveillance, Epidemiology, and End Results (SEER) database reported 5- and 10-year overall survival (OS) rates of 81 and 62 percent, respectively, among 501 cases (2005 to 2016) [44]. Age >60 years and the use of chemotherapy was associated with inferior outcomes. Another report estimated 10-year OS was 90 percent [45].

Recurrences are common, with an estimated five-year failure-free survival rate of 55 percent. Leg involvement portends a worse prognosis than other sites [15]. Patients with disease on the leg have a five-year disease-specific survival of 76 percent, compared with 96 percent for other locations.

Pathologic involvement of regional lymph nodes does not seem to portend an adverse prognosis [8].

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Primary cutaneous lymphoma".)

SUMMARY AND RECOMMENDATIONS

●Description – Primary cutaneous anaplastic large cell lymphoma (PC-ALCL) is a cutaneous CD30+ peripheral T cell lymphoma with no systemic involvement. Males are affected more than females, and most patients present in the sixth decade of life. (See 'Epidemiology' above.)

●Presentation – Typical presentation is slowly enlarging solitary or grouped skin nodules (picture 2) that frequently ulcerate. Many lesions regress spontaneously. (See 'Clinical presentation' above.)

●Pathology – Dense dermal infiltrate of large tumor cells with abundant cytoplasm; round, pleomorphic, or horseshoe-shaped nuclei with prominent nucleoli; ≥75 percent of tumor cells express CD30. (See 'Pathology' above.)

Evaluation and diagnosis – PC-ALCL should be suspected in patients with ulcerating skin nodules.

•Diagnosis requires characteristic pathology on skin biopsy, with staging that excludes systemic disease. (See 'Diagnosis' above.)

•Differential diagnosis includes reactive CD30+ lymphoid infiltrates, lymphomatoid papulosis, mycosis fungoides, and systemic lymphomas. (See 'Differential diagnosis' above.)

●Staging – Clinical evaluation, with biopsy of suspect lymph nodes/extracutaneous sites. (See 'Staging' above.)

●Initial management – Guided by location, extent, and stage of disease. Overtreatment should be avoided. (See 'Initial management' above.)

•Localized skin disease – Solitary or grouped lesions at one site; enlarged regional lymph nodes should be biopsied. (See 'Localized skin disease' above.)

-No nodal involvement – For localized skin disease without nodal involvement, we suggest involved site radiation therapy (ISRT) or surgical excision rather than surgery plus ISRT or observation (algorithm 1) (Grade 2C). (See 'No nodal involvement' above.)

-Tumor-involved regional nodes – For localized skin disease plus tumor-involved regional nodes, we suggest ISRT to skin and nodes rather than surgery (algorithm 1) (Grade 2C). (See 'Involved regional nodes' above.)

•Multifocal skin disease – Guided by extent of disease, systemic symptoms, and regional node involvement:

-Asymptomatic without nodal involvement – For asymptomatic multifocal PC-ALCL without nodal involvement, we suggest initial observation rather than immediate surgery, ISRT, or systemic therapy (algorithm 1) (Grade 2C). (See 'Asymptomatic with no regional node involvement' above.)

-Symptomatic – For multifocal skin disease with systemic symptoms or regional node involvement, we suggest brentuximab vedotin rather than other systemic treatments (algorithm 1) (Grade 2B). (See 'Symptomatic and/or regional node involvement (symptomatic or asymptomatic)' above.)

•Widespread nodal or visceral disease – We treat widespread nodal disease (ie, beyond sites that drain cutaneous involvement) and/or visceral disease like systemic lymphoma. (See 'Widespread nodal and/or visceral disease' above.)

●Response – Assessed by clinical evaluation. (See 'Response assessment' above.)

●Recurrent disease – Guided by sites and the extent of relapse, prior treatments, and patient preference.

•Localized cutaneous relapse – We suggest initial observation rather than immediate surgery or radiation therapy (RT) (Grade 2C). Some patients favor surgery or RT when lesions do not spontaneously regress. (See 'Localized cutaneous relapse' above.)

•Multifocal/multiple cutaneous relapses – Guided by prior therapy (see 'Multifocal or multiple cutaneous relapses' above):

-For patients with no progression on, toxicity-related discontinuation for, or treatment with brentuximab vedotin within ≤6 months, we suggest retreatment with brentuximab vedotin (Grade 2C).

-For others, we treat as widespread and/or multiple relapses (below).

•Widespread and/or multiple relapses – Options include brentuximab vedotin or sequential single agents, guided by prior treatments, toxicity, comorbidities, and patient preference. (See 'Widespread and/or multiple relapses' above.)

●Systemic treatments

•Brentuximab vedotin (see 'Brentuximab vedotin' above)

•Methotrexate (see 'Methotrexate' above)

•Bexarotene (see 'Bexarotene' above)

•Interferon (see 'Interferon' above)

●Prognosis – Most cases have an excellent prognosis, but relapses are common. (See 'Prognosis' above.)