Key points

Introduction

Primary cutaneous T-cell lymphomas (CTCL) represent a heterogeneous group of non–Hodgkin lymphomas (NHLs) that manifest in the skin with no evidence of extracutaneous disease at the time of diagnosis. The exception is MF, the most common type of CTCL, which accounts for more than 50% of primary cutaneous lymphomas. MF is generally associated with an indolent course with most of the patients presenting in early stage of the disease. However, about one-third of patients present with advanced stage (generally considered to be stage IIB and higher) and another 25% progress into higher stage in the course of their disease. SS is the leukemic and most commonly encountered type of aggressive CTCL.

Most patients with early-stage MF respond well to skin-directed therapies with reported long-term remissions. Treatment for patients with advanced disease includes various combinations of skin-directed therapies, biologic response modifiers, histone deacetylase (HDAC) inhibitors, investigational agents, as well as single-agent and/or multiagent chemotherapy regimens. None of these treatment options have been shown to prolong disease-specific survival or overall survival (OS) and often lead to short-term disease control with a median survival ranging from 1.4 to 4.7 years in patients with advanced stages (IIB–IVB) of MF and SS. Borrowing from the paradigm of aggressive lymphomas, hematopoietic stem cell transplant (HSCT) has been explored as a treatment option in patients with advanced-stage MF/SS and other subtypes. The data for using high-dose therapy and autologous HSCT (ASCT) remain disappointing, but the results of allogeneic stem cell transplant are encouraging for the treatment of CTCL. The data series are small, and there is little consensus on conditioning regimens and other aspects of the transplants that are largely driven by institutional preferences. This article discusses the role of allogeneic stem cell transplant in the care of patients with CTCL and presents relevant data to support its use.

Introduction

Primary cutaneous T-cell lymphomas (CTCL) represent a heterogeneous group of non–Hodgkin lymphomas (NHLs) that manifest in the skin with no evidence of extracutaneous disease at the time of diagnosis. The exception is MF, the most common type of CTCL, which accounts for more than 50% of primary cutaneous lymphomas. MF is generally associated with an indolent course with most of the patients presenting in early stage of the disease. However, about one-third of patients present with advanced stage (generally considered to be stage IIB and higher) and another 25% progress into higher stage in the course of their disease. SS is the leukemic and most commonly encountered type of aggressive CTCL.

Most patients with early-stage MF respond well to skin-directed therapies with reported long-term remissions. Treatment for patients with advanced disease includes various combinations of skin-directed therapies, biologic response modifiers, histone deacetylase (HDAC) inhibitors, investigational agents, as well as single-agent and/or multiagent chemotherapy regimens. None of these treatment options have been shown to prolong disease-specific survival or overall survival (OS) and often lead to short-term disease control with a median survival ranging from 1.4 to 4.7 years in patients with advanced stages (IIB–IVB) of MF and SS. Borrowing from the paradigm of aggressive lymphomas, hematopoietic stem cell transplant (HSCT) has been explored as a treatment option in patients with advanced-stage MF/SS and other subtypes. The data for using high-dose therapy and autologous HSCT (ASCT) remain disappointing, but the results of allogeneic stem cell transplant are encouraging for the treatment of CTCL. The data series are small, and there is little consensus on conditioning regimens and other aspects of the transplants that are largely driven by institutional preferences. This article discusses the role of allogeneic stem cell transplant in the care of patients with CTCL and presents relevant data to support its use.

Overview of hematopoietic stem cell transplant

HSCT, formerly known as bone marrow transplant (BMT), is a medical procedure in which multipotent stem cells derived from the bone marrow, peripheral blood, or umbilical cord are infused into a patient for treatment of hematological disorders and malignancies. This procedure requires that the patient’s own hematopoietic and immune function be suppressed enough to accept the infused cells and allow homing of these cells to the marrow spaces and establishment of a donor-derived hematopoietic system in the host. This procedure can be accomplished either by chemotherapy alone or by combination of chemotherapy with radiation therapy called conditioning or preparative regimen given before stem cell infusion. The establishment of a donor-derived hematopoietic system requires some time during which the patient remains pancytopenic and entirely depends on supportive measures to prevent and treat the complication of pancytopenia as well as the conditioning regimen. The stem cells can be derived from the patient’s own hematopoietic system (autologous) or from an HLA-matched donor (allogeneic) who can be a sibling (related) or a matched unrelated donor. Other sources now extend to haploidentical family members and cord blood stem cells and are discussed below. Major indications for stem cell transplant include hematologic malignancies such as leukemia, lymphoma, multiple myeloma, and other myeloproliferative disorders. According to the Center for International Blood and Marrow Transplant Research (CIBMTR) data, approximately 12,000 autologous and 8000 allogeneic transplants were performed in the year 2013 and the numbers are increasing.

Stem cell sources

Hematopoietic stem cells express properties of multipotency and self-renewal and reside in bone marrow niches supported by cytokines and other microenvironmental factors. Human hematopoietic stem cells express CD34, CD38, CD90, CD133, CD105, CD45, and also c-kit (CD117), the receptor for stem cell factor, and these cells test negative for the markers that are used for the detection of lineage commitment. Historically, stem cell collection was performed in the operating room under general anesthesia using a large trocar to collect bone marrow from the pelvic bones in adults and long bones in children. This procedure has now given way to peripheral blood as a source of stem cells through a process called apheresis. The peripheral blood stem cells can be mobilized into the circulation either by chemotherapy (in case of autologous collections) or by injections of hematopoietic growth factors, that is, granulocyte colony-stimulating factor supplemented by CXR4 inhibitors such as plexiafor, and collected. Most autologous stem cells are cryopreserved in dimethyl sulfoxide before infusion in contrast to allogeneic stem cells that are usually infused fresh on the day of collection. Umbilical cord blood (UCB), which is rich in hematopoietic stem cells, can be cryopreserved and used in an appropriate patient. However, because a cord can yield only small amounts of blood (approximately 50 mL), a single cord can only provide adequate stem cells for a child or small adult. Generally, 2 UCB units need to be combined for adult transplants, and there is now a significant body of data to support the safety and efficacy of this approach.

Donor selection

HLA typing is required to match a donor and recipient for allogeneic stem cell transplants. The major HLA genes fall into 2 categories (Type I and Type II), and serologic and molecular matching is performed on the basis of variability at 6 loci of the HLA gene. A perfect match at these loci is desirable to ensure engraftment and prevent complications of GVHD. Mismatches of the Type I genes (ie, HLA-A, HLA-B, or HLA-C) increase the risk of graft rejection. A mismatch of an HLA Type II gene (ie, HLA-DR, or HLA-DQB1) increases the risk of GVHD. Molecular methods are increasingly being used to increase the accuracy of tissue typing to ensure optimal matching. Sibling donors have a 25% chance of being a match with the recipient. Rarely, patients have a syngeneic donor, that is, a monozygotic twin who is perfectly matched at all HLA loci. Most other patients need to rely on an unrelated HLA-matched donor who may be found through the National Marrow Donor Program and other such worldwide registries. Unfortunately, these registries have marked underrepresentation of specific ethnic and minority racial groups. This underrepresentation has driven the need to look for alternative donors. A haploidentical donor is a partially matched first-degree relative of the patient (child, parent, or sibling). The advantage of haploidentical transplant is immediate and permanent availability of the donor for current and future therapies. Initially, the 3 antigen mismatches in these transplants led to unacceptably high transplant-related toxicity, but with the use of T-cell depletion techniques, improved immunosuppression, and supportive care, haploidentical transplants are increasingly being offered to patients who lack an HLA-matched donor. Umbilical cords provide immunologically naive stem cells and a reduced risk of GVHD and are a relatively accessible source of stem cell transplants. The main disadvantage is the small number of stem cells that may by themselves be inadequate in number for successful engraftment in an adult. Ex vivo expansion, double cord transplants, and combined haploidentical and cord transplants have resulted in improved outcomes for these patients. According to the European Bone Marrow Transplant Registry survey of 2013, as many as 43% of all HSCTs done that year were allogeneic and there was a notable increase in the use of alternate donors.

Conditioning regimens

The preparative regimen given before stem cell infusion is called a conditioning regimen and has a 2-fold purpose depending on the type of transplant. In the case of an autologous stem cell transplant, the conditioning regimen consists of chemotherapy with or without radiation and is designed to give a high dose of antitumor therapy for cytotoxic purposes. The hematopoietic system damaged by chemotherapy is then reconstituted with transplanted autologous stem cells. The state of immunosuppression created by the conditioning regimen improves with time, and the patient does not have any GVHD. In allogeneic transplant, conditioning treatment is required for engraftment and prevention of GVHD by suppressing host immunity in addition to antitumor effects. Regimens can be fully MAC for maximal antitumor effects and generally include high-dose cytoxan, busulfan, high-dose etoposide, and/or total body irradiation (TBI). These regimens can have significant immediate toxicity that can contribute to significant regimen-related morbidity and mortality in the immediate posttransplant period. Nonrelapse mortality (NRM) of 22% was reported in a cohort of 60 patients with advanced CTCL (clinical stage IIB–IVB) treated with allogeneic stem cell transplant. In multivariate analysis, MAC was associated with a higher NRM (hazard ratio, 4.5; P = .1) RIC or nonmyeloablative transplants are less cytotoxic and mostly immunosuppressive, resulting in decreased immediate transplant-related toxicity and allowing older and more frail patients to undergo transplant. These regimens consist of purine analogs such as fludarabine and cladribine as well as low-dose TBI and rely on the GVHD effect for their antitumor effects. These regimens are associated with decreased immediate transplant-related toxicity and less-acute GVHD, but there has been no change in the incidence of chronic GVHD with RIC. Use of RIC has extended the use of allogeneic stem cell transplants to patients into their 70s and 80s if they are otherwise in good health. CIBMTR data confirm the increasing use of RIC for allogeneic stem cell transplants and the increasing age of transplant recipients in the United States.

Complications

Several complications are associated with HSCT depending on the type of conditioning, type of transplant, and engraftment phase. The most common cause of transplant-related morbidity and mortality is GVHD, followed by infections. GVHD remains a challenge in the care of patients undergoing allogeneic stem cell transplant. Classically, acute GVHD occurs within the first 100 days of transplant and manifests itself primarily in the skin, gastrointestinal tract, and liver. However, the development of various conditioning regimens and the use of DLI after transplant have changed the time course and presentation for acute and chronic GVHD.

The pathogenesis of acute GVHD is initiated by recipient tissue damage occurring from conditioning regimen leading to release of inflammatory cytokines that lead to the expansion of donor lymphocytes following contact with host and donor antigen-presenting cells that express disparate antigens resulting in alloreactive T cells that induce tissue damage. The incidence of acute GVHD varies from 20% to 70% and is directly related to the degree of mismatch between HLA proteins. The incidence ranges from 35% to 45% in recipients of full-matched sibling donor grafts to 60% to 80% in recipients of one-antigen HLA-mismatched unrelated donor grafts. The same degree of mismatch causes less GVHD using UCB grafts, and the incidence of acute GVHD is lower following the transplant of partially matched UCB units and ranges from 35% to 65%. Chronic GVHD is the major cause of late nonrelapse death following transplant. Older recipient age and a history of acute GVHD are the greatest risk factors for chronic GVHD. Incidence ranges from 60% to 80%, and manifestations of chronic GVHD are protean and often resemble an autoimmune disorder. The National Institutes of Health proposed standard criteria in 2005 for diagnosis, organ scoring, and global assessment of chronic GVHD severity to establish a common platform for subclassifying chronic GVHD.

Prevention strategies using immunosuppressive agents including calcineurin inhibitors, methotrexate, mammalian target of rapamycin inhibitors, or novel agents begin before stem cell infusion and are maintained for some time after allogeneic transplant. Treatment of GVHD is initiated when any sign of GVHD appears and consists of high-dose steroids and other strategies for immunosuppression. Prolonged immunosuppression delays immune reconstitution of the host and increases the risk of serious infections leading to high morbidity and mortality rates. Other complications after stem cell transplant can be related to conditioning regimens causing end-organ toxicity, secondary malignancies, and psychosocial issues related to chronic medical problems. Even with improved supportive care measures, there is a significant morbidity and mortality associated with stem cell transplant, which in the case of allogeneic stem cell transplants can range from 20% to 40% depending on the patient’s comorbidities, donor type, degree of HLA mismatch, underlying disease, and conditioning regimen. Autologous stem cell transplants are associated with a much lower risk of death at less than 5% in most centers, but there is a higher incidence of disease relapse. When determining the eligibility for HSCT, these risks have to be weighed against the risk of death and morbidity posed by the disease.

Graft-versus-lymphoma effect

The existence of an immunologic GvL reaction associated with allogeneic stem cell transplant is well established. Allogeneic transplant is successful partly because of the GvL effect of the donor graft, independent of the conditioning regimen. Evidence for GvL effects is based on the following criteria: RIC resulting in long-term disease control, increased risk of relapse associated with T-cell depletion, withdrawal of immunosuppression, use of DLIs to eradicate documented disease relapse after allogeneic transplant, and association of disease control with GVHD. A GvL effect against CTCL has been established and is discussed in later sections.

Autologous stem cell transplant for mycosis fungoides and Sézary syndrome

High-dose therapy and ASCT have curative potential in aggressive chemotherapy-sensitive relapsed lymphomas. Using similar approaches, ASCT has been performed in cases of advanced CTCL in order to establish long-term remissions. Conditioning regimens have incorporated high-dose chemotherapy either with or without TBI. Some investigators have reported the use of total skin electron beam (TSEB) therapy to provide improved control of skin disease in patients with CTCL before initiating ASCT and avoid the generalized organ toxicity of TBI. Bigler and colleagues first reported on 6 patients with MF who underwent TBI or TSEB-based regimens followed by ASCT: 5 patients achieved complete remission (CR) but 3 showed relapse in less than 100 days, whereas 2 remained in persistent CR after 1 year. Results demonstrated that the procedure was feasible, and increased transplant-related mortality (TRM) or transplant-related morbidity was not reported for these patients. A few other case reports showed early relapses within the first 100 days. The largest series by Ingen-Housz-Oro et al reported on 10 patients with CTCL that included 1 patient with MF; 8 patients with peripheral T-cell lymphoma (PTCL), unspecified (PTCL, not otherwise specified [NOS]); and 1 patient with subcutaneous panniculitis-like T-cell lymphoma including patients with systemic disease in addition to skin manifestations. Conditioning was TBI based and 7 of 10 patients achieved a CR. However, in 6 patients, the condition relapsed in less than 4 months and in 1, it relapsed at 7 years (PTCL, NOS). Remaining patients experienced fatal progression. Of note, some of the relapses responded to local and biologic therapies.

Another case series was reported by Olavarria and colleagues on 9 patients with MF consisting of 5 patients with stage IIB and 4 patients with stage IVA disease; 8 patients demonstrated a T-cell clone in the peripheral blood. This study was a pilot study of T-cell depletion and ASCT in these patients using double CD34-positive and CD4/CD8-negative selection by immunomagnetic methods. TBI was given to only 2 patients, and the others were conditioned with chemotherapy only. Of the 9 patients, the condition relapsed in 7 patients at a median of 7 months, mostly with limited disease that responded to conventional therapy; 4 patients were analyzed for T-cell receptor rearrangement after transplant and had detectable clone before or at the time of relapse. There is one report of a patient with SS receiving autologous HSCT. This patient was conditioned with the combination of chemotherapy and TBI. He showed relapse after 3 months and died after 15 months of infectious complications. Moreau and colleagues reported successful transplant in 1994 on 4 cases of CTCL other than MF/SS (including 2 patients with a CD30 + lymphoproliferative disorder [CD30 + LPD]) using a TBI-based regimen with CR at 22, 41, 46, 44, and 51 months. This report predates the 2008 World Health Organization classification, and it is unclear from the publication whether patients were diagnosed with primary cutaneous anaplastic large cell lymphoma (PCALCL) or lymphomatoid papulosis or both. No specific data are available for the use of ASCT in large cell transformation (LCT)-MF, although these cases tend to be treated as systemic T-cell lymphomas and are likely included in the series of transplants for systemic diseases.

These limited data show that ASCT is feasible, but in most cases, the responses are not sustained except possibly for CD30 + LPD; however, these entities have already a favorable long-term outcome without the need of aggressive therapies. The duration of remission does not seem to be related to the stage of the disease or the absence of a detectable T-cell clone in the harvest. Although the numbers are very small, the use of TBI did not seem to predict improved outcomes. These early relapses point to the inability of high-dose therapy to control disease pointing to the need for improved targeted therapies. However, some of the relapses seemed to be less aggressive and could be managed with local and biologic therapies for a while. The use of high-dose therapy and ASCT has essentially been abandoned with a belief that this is not a curative approach and that the inherent chemoresistance of these tumors contributes to the low success rate of this approach.