Interferons are a class of TH1 cytokines and function as immune modifiers. Recombinant interferons are a good therapeutic option for patients with mycosis fungoides for all stages of disease including SS. Alpha interferon is the most common formulation available for clinical use and can be given subcutaneously or intramuscularly and has also been used for intralesional injections. The usual dose is three million to ten million units given subcutaneously in various schedules that range from three times a week to daily dosing. Data from studies involving more than 12 patients have reported partial response rates of 17–53 % and complete remission rates of 4–27 %. These studies included varying stages of disease (Olsen 2003), and the responses were higher if this was the first line of systemic therapy (Bunn and Norris 1990). It has limited and reversible dose-dependent side effects that include fever, chills, influenza-like symptoms, myalgias, and arthralgias. More chronic effects include fatigue, anorexia, weight loss, sleep disturbance, and hepatitis. Alpha Interferon can be used alone or can be combined with other treatment modalities to improve response rates and outcomes. These include extracorporeal photopheresis (Olsen et al. 1989), low-dose chemotherapy (Foss et al. 1992, 1994), retinoids (Stadler et al. 1998; Knobler et al. 1991), and phototherapy (Rupoli et al. 1999). While alpha interferon is the most commonly used formulation, there is data using interferon gamma as well in the treatment of MF resulting in prolonged responses (Kaplan et al. 1990).

Interleukins as immune response modifiers that can be used in the treatment of MF. Interleukin-2 and interleukin-12 have been used in clinical trials with good responses, but the excessive toxicity and limited availability make them impractical for general use (Duvic et al. 2006c).

Bortezomib is a dipeptide boronic acid that binds the catalytic site of the 26S proteasome with high affinity and specificity (Bonvini et al. 2007). This agent is used in the treatment of MF/SS but the data supporting its use is limited to one study of ten patients. Bortezomib was administered at a dose of 1.3 mg/m² IV on days 1,4,8, and 11 every 21 days for 6 cycles, and an ORR of 70 % was noted with 1 CR lasting over 12 months (Zinzani et al. 2007). The main side effects associated with its use are myelosuppression, particularly thrombocytopenia, and sensory neuropathy that was seen in 50 % of the patients treated on the MF study. Other effects include diarrhea, asthenia, and headaches. A subcutaneous route of administration is being explored in other diseases and has shown to be just as efficacious but compared to IV infusions is associated with a lower incidence of neuropathy (38 % vs. 53 % P = 0.044) (Moreau et al. 2011). Similar trials are warranted in MF.

Carfilzomib irreversibly binds to and inhibits the chymotrypsin-like activity of the 20S proteasome, an enzyme that degrades unwanted cellular proteins. It is very well tolerated and can be administered for prolonged time periods without significant effects of neuropathy or myelosuppression. Current trials are under way to establish its activity in this group of lymphomas in combination with other agents (ClinicalTrials.gov Identifier: NCT01276717, (Dasmahapatra et al. 2011).

This is a synthetic retinoid that selectively binds to the RXR receptors and is formulated both as an ointment for topical use as well as an oral formulation. Both forms are approved for the treatment of MF/SS both in the USA and Europe for both early-stage disease and advanced disease including SS (Talpur et al. 2002; Breneman et al. 2002; Duvic et al. 2001b; Gniadecki et al. 2007). Response rates and side effect profiles are dose dependent. In phase II/III studies response rates of 54 % were observed at a dose of 300 mg/m² per day and up to 67 % at higher doses (Duvic et al. 2001b) in early-stage MF (stage I–IIA). For advanced-stage disease (IIIB–IVB) the response rate was 48–55 % (Duvic et al. 2001b). The median time to respond was noted to be 8.1 weeks (range 4–16) and 25.7 weeks (2–28), respectively. Main side effects are reversible hyperlipidemia and hypercholesterolemia occurring within 2–4 weeks of initiating therapy that often require therapy with lipid-lowering agents, a decrease in thyroid-stimulating hormone (TSH) resulting in reduced levels of T4, hepatitis, anemia, leucopenia, headache, and dry skin. Bexarotene is contraindicated in pregnancy due to its effect on fetal development (Duvic et al. 2001a, b).

These are non-RXR selective and include oral etretinate, arotinoid, acitretin, and isotretinoin (13-cis-retinoic acid). There are no comparative trials but overall response rates based on studies range from 5 to 65 % either as single agents or in combination with PUVA, interferon, or cytotoxic chemotherapy (Burg and Dummer 2000; Zachariae et al. 1982; Stadler et al. 1998; Thomsen et al. 1989; Knobler et al. 1991).

Alemtuzumab is a humanized monoclonal antibody targeting CD52 on the surface of lymphocytes that has activity against many T-cell lymphoproliferative disorders (Piccaluga et al. 2007; Rowan et al. 1998). Alemtuzumab is thought to mediate its effects through antibody-dependent cellular toxicity and activation of complement-dependent and complement-independent cytolysis (Rowan et al. 1998; Dyer et al. 1989). Initial trials in heavily pretreated patients with MF and SS have reported response rates of 55 % with 31 % CRs (Lundin et al. 2003) and a duration of response of less than 12 months. Another small trial of eight patients reported a response rate of only 31 % with a median duration or response lasting 4 months (Kennedy et al. 2003). Main toxicity is hematological and an increase incidence in infections including CMV and EBV reactivation. Alternative dosing and routing schedules have been attempted including subcutaneous administration to reduce the associated toxicity (Bernengo et al. 2007; Zinzani et al. 2005; Querfeld et al. 2009). Alemtuzumab has shown particularly high-response rates of 86–87 % with a CR seen in 37 and 21 % cases, in small studies focused on patients with erythroderma and SS (Lundin et al. 2003; Bernengo et al. 2007; Querfeld et al. 2009), indicating that this may be an effective therapy in otherwise difficult to treat patients with SS.

Zanolimumab (HuMax-CD4) is a humanized monoclonal antibody directed against CD4 expressed universally on helper T cells and blocks the interaction of CD4 receptor and the major histocompatibility complex class II on cells thus preventing the activation of the T cell. It results in cell death via antibody-dependent cellular toxicity (ADCC) but does not induce complement-dependent cytotoxicity (CDC). It results in depletion for CD4-expressing T cells and has been studied in the setting of MF and SS. Two simultaneous phase 2 multicenter trials were conducted in patients with CTCL, one for early-stage disease and the other for advanced disease. Responses were seen in patients with MF and SS, with a median response rate of 56 % and a median duration of response at 81 weeks with more responses noted at the higher dose level. The agent was well tolerated with mild eczema and low-grade infections in spite of effective lowering of the CD4 count in patients (Kim et al. 2007). Further evaluation of this promising agent is warranted.

CCR4 is a chemokine receptor expressed on CD4+ helper T cells and regulatory cells (Tregs) and in varying proportions in T-cell malignancies (Imai and Umezu 1999; Iellem et al. 2001; Ito et al. 2009). KW-0761 is a humanized monoclonal antibody directed against CCR4 that has a defucosylated Fc region that enhances the ADCC due to increased binding affinity to the Fc gamma receptor on cells. A phase 1 study of the antibody given once a week for 4 weeks indicated promising responses in CTCL (Yamamoto et al. 2010).

Denileukin diftitox (Ontak) is a novel fusion protein consisting of the membrane translocation sequence for the diphtheria toxin and the receptor-binding sequence of the human interleukin-2 that has affinity for the human IL-2 receptor (Williams et al. 1990; Taniguchi and Minami 1993). Initial phase I/II trial confirmed the antitumor activity of denileukin diftitox in patients with CTCL (Saleh et al. 1998). A phase III trial comparing two dose levels of denileukin diftitox at 9 and 18 μg/kg given daily for 5 days every 21 days was conducted in patients with CD25-expressing CTCL and SS. This trial led to the accelerated approval of the agent by the FDA in the USA for the treatment of relapsed and refractory CTCL with more than 25 % expression of CD25. A response rate of 30 % (20 % PRs and 10 % CRs) was reported in the trial. The median duration of response was 6.9 months (2.7–46.1 months) with no difference between the two dose levels. The main side effects were flu-like symptoms, infusional sensitivity reactions, vascular leak syndrome, hypoalbuminemia, and transaminitis with a statistical hint that the side effects may be worse in the higher-dose arm (Olsen et al. 2001). A second placebo-controlled phase III trial was conducted to evaluate the efficacy of the two dose levels of denileukin diftitox, i.e., 9 and 18 μg/kg, compared with placebo in CD25-expressing CTCL and SS patients who had received up to three prior systemic therapies (Prince et al. 2010). One hundred and forty-four patients were enrolled. The ORR was 44 % (34 % PRs, 10 % CRs) with the response rate being higher in the 18μg/kg group, i.e., 49 % vs. 37.8 % in the 9μg/kg group vs. 12 % in the placebo group. Progression-free survival was 124 days better in both dose groups as compared to placebo. There was no difference in the side effect profile at the two doses. This led to the full approval of the agent in 2010 for the treatment of CTCL if there is expression of CD25. The recommended dose of the agent is either 9 or 18μg/kg and is left to the discretion of the treating physician. There is a black box warning in the label for fatal vascular leak syndromes and loss of visual acuity and color vision which may not be reversible. Combination therapies have been evaluated, the most notable being the combination of denileukin diftitox with bexarotene (Foss et al. 2005). The combination was well tolerated with an overall response rate of 67 %. The study also demonstrated that even low doses of bexarotene at 150 mg/day were capable of inducing upregulation of CD25 expression which may have led to the higher response rate.

Brentuximab vedotin is an antibody conjugate consisting of a chimeric monoclonal antibody that targets CD30 (a member of the transmembrane tumor necrosis factor family of proteins) linked to the antimitotic agent monomethyl auristatin E (MMAE). The binding of the agent to CD30 results in internalization of the compound which is then released intracellularly—mitosis is interrupted and the cell undergoes apoptosis. The agent is approved at a dose of 1.8 mg/kg given once every 3 weeks for CD30+ anaplastic large-cell lymphoma including the cutaneous variant of the disease. The response rate using this agent in the relapsed setting is 87 % in ALCL. Side effects are tolerable with the most common being sensory and motor neuropathy. Variable CD30 expression is seen in MF and up to 41 % of the time in the transformed MF (Arulogun et al. 2008). Hence, it is likely that there will be efficacy of this agent in CD30-expressing MF and SS. Early trials are encouraging and a phase III randomized trial is being conducted in CD30-expressing MF patients who need systemic therapy that will compare brentuximab vedotin with standard-dose methotrexate or bexarotene (physician’s choice) in the comparator arm.

Belongs to the class of hydroxamic acids and has both oral and IV formulations that inhibits both class I and II histone deacetylases. The recommended dosing schedule for CTCL is 400 mg orally once a day with dose adjustments recommended for toxicities. In the pivotal phase 2 trial that led to the approval of this agent, the overall response rate was 24 % with a 58 % reduction in pruritus (Olsen et al. 2001; Duvic et al. 2007). Responses were seen across all stages of diseases for stage IIB or higher. The most common toxicities are gastrointestinal, constitutional symptoms, dysgeusia, and hematological especially reversible thrombocytopenia (Mann et al. 2007). Long-term therapy with vorinostat in patients with stable disease or partial responses is feasible with manageable toxicity (Duvic et al. 2009). Combination studies using vorinostat have been conducted with promising results including a phase 1 trial of the combination of bexarotene and vorinostat (Dummer et al. 2008). Case reports of patients receiving vorinostat in addition to their ongoing therapies to improve responses have included combinations with IFN-α, phototherapy, and photopheresis (Geskin 2010).

Romidepsin (FR901228, FK228, depsipeptide) is a potent HDAC inhibitor belonging to the class of cyclic peptides that mainly inhibits HDAC1 and HDAC2 class I enzymes has an intravenous formulation and is approved in the USA for the treatment of relapsed and refractory CTCL after failing at least one prior systemic therapy. The prescribing dose is 14 mg/m² given over a 4-h infusion once a week for 3 weeks followed by a 1-week rest. Two independent phase II trials (Piekarz et al. 2009; Coiffier et al. 2012) have been conducted, and the pooled data from both these trials has shown an ORR of 34 % with a median duration of response of 15 months (Demierre 2009). One of the striking features of romidepsin is the long duration of response that extended beyond 3 years, observed in some patients even after discontinuation of the drug. Main side effects were nausea, asthenia, anorexia, vomiting, and fatigue. The drug needs to be administered with caution in patients with significant preexisting cardiac abnormalities and concomitant medications that prolong QT interval or inhibit CYP3A4. A topical formulation of romidepsin is currently in clinical trials in limited stage CTCL.

At least two other HDAC inhibitors are currently undergoing investigation for efficacy in the treatment of CTCL/MF with indications of activity. These include the Novartis compound LBH589 (panobinostat) and belinostat (Duvic et al. 2008; Ellis et al. 2008; Pohlman et al. 2009).

Nucleoside analogues are antimetabolites that are phosphorylated and incorporated into the growing DNA strand of a dividing cell in the S phase. Some of these agents are also active in MF and SS as described below.

This agent is widely used in the treatment of NHL and CTCL. However, published data to support the use of this agent in CTCL and SS is limited to a few small studies. The agent is given weekly for 3 weeks with a 1 week break at the end of each cycle. Dosing ranges from 1,000 to 1,200 mg/m² given intravenously. Some of the larger studies have included up to 30 patients with advanced-stage disease and have reported responses of up to 70 % with a few CRs as well (Duvic et al. 2006d; Zinzani et al. 2000; Marchi et al. 2005; Sallah et al. 2001). Main side effects are myelosuppression, fever, nausea, vomiting, interstitial pneumonitis, alopecia, and radiation sensitivity. Cases of hyperpigmentation in SS patients have been seen. Rare incidences of hemolytic uremic syndrome have been reported.

Use of single-agent fludarabine has been studied in CTCL and SS. Trials are small but response rates vary from 19 to 30 % with CR of 9 %. Response may be higher in the SS group with a RR of 35 % and CRs of 18 % (Quaglino et al. 2004). Combination chemotherapy with fludarabine has also been evaluated with interferon alpha. 35 patients were treated and 11 % (4/35) reached a CR including 11 patients with SS maintained for 18 months in 3 patients (Foss et al. 1994). Infectious complications remain the major side effect.

Small case series of up to eight patients reported responses in patients with MF including 1 CR (O’Brien et al. 1994; Trautinger et al. 1999). Differing dosing schedules have been employed. The largest series consists of 24 patients with MF/SS treated at 0.1 mg/kg/day by continuous infusion over 5–7 days repeated every 28 days. The RR was 24 % with 3 out of 24 patients reaching a CR (Kong et al. 1997). Based on an analysis of several small series, a RR of 50 % has been reported in SS (Saven et al. 1992; Zaucha et al. 1997; Bouwhuis et al. 2002). Main side effects are immunosuppression and prolonged leucopenia. Occasional constitutional symptoms of fever and nausea can occur.

One of the most widely studied agents in CTCL as activity was established in the first phase I study of this agent (Grever et al. 1983). Several trials have been published that support the use of single-agent pentostatin in the treatment of MF and SS including ECOG and EORTC but there is no consensus on the dose or schedule. These trials are also marked by patient heterogenicity and lack of uniform response criteria. Reported response rates are 31–66 % with higher responses seen in SS (71 %) with reports of CR lasting up to 76 months (Tsimberidou et al. 2004; Cummings et al. 1991; Greiner et al. 1997; Ho et al. 1999; Kurzrock et al. 1999). A combination of pentostatin and interferon produced a response rate of 41 % in a group of 41 patients with 2 CRs in SS patients and 15 PRs (Foss et al. 1992). The most common side effects are myelosuppression, nausea, fever, and elevation of liver enzymes that is transient. Prolonged suppression of CD4 counts can occur putting the patient at risk of potentially life-threatening infections. Neurologic symptoms, pulmonary toxicity, or unexpected nephrotoxicity has been reported.

Two phase I/II studies of forodesine have established its activity in patients with MF/SS using an IV and an oral formulation. Small studies have reported a ORR of up to (Duvic 2007; Duvic et al. 2004, 2006a). Side effects are fatigue, edema, nausea, pruritus, dyspnea, and headaches. Lymphopenia and low CD4 counts have been noticed in patients but opportunistic infections are not common (Duvic et al. 2006b).

Most of the alkylating agents that are used in MF/SS are part of the combination chemotherapy regimens, but single agents used at lower doses have been attempted to enable prolonged use with fewer side effects.

Single-agent use for MF/Ss has been established for the following agents: