Published: January 2014
Non-Hodgkin's lymphoma (NHL) and Hodgkin's lymphoma (HL) are neoplasms arising from cells of the lymphoid lineage. T and B cells originate in the bone marrow, migrate to the thymus or peripheral lymphoid tissues respectively, and develop into highly specialized mediators of the adaptive immune response. Generating and maintaining this dynamic repertoire of cells is a complex error-prone process. Lymphoid cells are at various times susceptible to acquired genetic defects, direct viral infection, chronic stimulation by antigen, and effects of generalized host immunodeficiency—four dynamic factors involved in lymphomagenesis. The heterogeneity of lymphomas, reflecting the complexity of the human immune system, implies that a number of genetic and acquired risk factors play a role in pathogenesis.
Lymphomas are divided into two major groups, NHL and HL, based on a range of pathologic and clinical features. The incorporation of genetic and immunologic characteristics into lymphoma diagnosis is a recent advancement, proposed by the 1994 Revised European-American Classification of Lymphoid Neoplasms (REAL) classification divided lymphomas into B cell neoplasms, T cell neoplasms, and Hodgkin's disease, with clinically relevant categories.1 It also served as the basis for the ensuing World Health Organization (WHO) classification of lymphoid neoplasms. The many WHO subtypes of NHL are made more manageable via grouping into indolent, aggressive, or highly aggressive categories based on their natural history (Box 1). HL encompasses two main categories: classic HL (with four further subgroups) and nodular lymphocyte-predominant HL. Classic HL includes nodular-sclerosis, mixed-cellularity, lymphocyte-rich, and lymphocyte-depleted subgroups. The WHO classification, though complex and continually evolving, establishes a common language for researchers and clinicians and is key to collaborative research aimed at curing lymphoma.
|Follicular lymphoma (grades I and II)|
|Chronic lymphocytic leukemia / small lymphocytic lymphoma|
|Marginal zone: extranodal, mucosa-associated lymphoid tissue (MALT), nodal, splenic|
|Plasma cell myeloma|
|Hairy cell leukemia|
|Diffuse large B cell lymphoma and variants|
|Follicular lymphoma (grade III)|
|Peripheral T cell|
|Anaplastic large cell|
|Precursor B/T lymphoblastic|
*Mantle cell lymphoma may also behave in an indolent fashion.
According to the National Cancer Institute and Centers for Disease Control and Prevention SEER database, lymphoma was diagnosed in about 74,340 people in the United States in 2008, giving an age-adjusted incidence rate of 22.2 per 100,000 per year.2 In general, NHL is increasing in incidence (especially diffuse large B cell lymphoma [DLBCL]), though mortality among those affected with NHL has decreased. HL is much less common than NHL, accounting for about one tenth of all lymphoma cases; its annual incidence is 2.8 per 100,000. The prevalence of lymphomas tends to be much higher than their incidence, given their natural history and availability of effective therapies. For example, the U.S. prevalence of HL was 156,000 (patients with HL or a history of HL) as of January 1, 2005.
As noted, risk factors for the development of lymphoma are not fully understood. Environmental associations with pesticides, agricultural chemicals, and hair dyes have been inconsistently identified. Reports of a protective effect of sun exposure in the development of lymphoma have been conflicting. On the other hand, known risk factors for lymphoma include systemic immunosuppression due to inherited conditions, HIV infection, or medications.
The natural history of a given NHL is reflected in its conceptual grouping (e.g., indolent, aggressive, highly aggressive), although heterogeneity even within specific subtypes is observed. This heterogeneity is due to the broad spectrum of genetic changes, cell-signaling aberrations, and features of the tumor microenvironment that can affect the behavior of an individual lymphoma.
Although some population studies have found a higher risk of lymphoma in first-degree relatives of probands, defining the exact inherited genetic lesions has proved difficult.3 In contrast, genetic abnormalities acquired during early lymphocyte development have been clearly implicated in lymphomagenesis. The most noteworthy include chromosomal translocations during immunoglobulin (IG) gene rearrangement, a complex process that normally provides lymphocytes the diversity of antigen recognition needed for effective host defense. Expression of the viral gene products by host cell machinery contributes to the pathogenesis of HLs and lymphoproliferative disorders in immunosuppressed patients following organ transplant, the most well described being from infection with Epstein-Barr virus (EBV).
Infection by nonviral microbes can also lead to lymphoma, but not by direct infection of lymphocytes. Instead, chronic infection with organisms such as Helicobacter pylori is thought to lead to ongoing antigenic stimulation in lymphoid tissues, creating an environment ripe for selection of a malignant clone. Such stimulation can also follow immune attack on self-antigens, possibly explaining the link between some lymphomas and autoimmune conditions such as rheumatoid arthritis and systemic lupus erythematosis.
Inborn or acquired immunodeficiency is associated with a higher risk of lymphoma. Immunosuppression by HIV dramatically increases the risk for the development of lymphoma which can be reduced by anti-retroviral therapy. The interaction among these etiologic factors, and in particular the cellular interactions among immune and tumor cells (in the tumor microenvironment), are important topics of research in lymphoma pathogenesis and therapy.
Follicular lymphoma (FL) and chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) are the most common indolent lymphomas. Survival from diagnosis of indolent lymphoma is generally measured in years. Although radiation therapy can cure early-stage indolent lymphomas, advanced-stage disease is generally incurable. Despite initial chemosensitivity, such patients tend to face a continual pattern of relapse and treatment-related morbidity until death. Data from Stanford University published in 1984 showed that some asymptomatic patients with advanced FL had no decrement in survival following an initial watch-and-wait approach with spontaneous remission in a minority of cases.4
Based on these factors and the lack of curative therapy, treatment for FL was historically delayed until emergence of disease-related symptoms or organ compromise, and median survival was 8 to 10 years from diagnosis. However, newer treatment approaches using monoclonal antibodies with initial chemotherapy, and autologous stem cell transplantation for patients in relapse, may be prolonging survival and altering the natural history of FL. This has given rise to therapeutic optimism and prompted some to initiate treatment in some groups of patients with newly diagnosed lymphoma who may have been managed expectantly in the past.
In FL, the defining genetic lesion is the translocation between chromosomes 14 and 18 t(14;18), seen in the majority of cases (≥70%). As is typical for lymphomas, this translocation juxtaposes a regulatory sequence next to a normal, intact gene involved in cellular processes. (This contrasts with most leukemias, in which translocations—such as translocation 9;22 in chronic myelogenous leukemia—create a unique fusion gene and protein bearing unique oncogenic properties.) Translocation 14;18 places the BCL2 gene on chromosome 18 under the control of a key regulatory region (the IG heavy chain [IgH] enhancer sequence) on chromosome 14. This results in the overexpression of BCL2, a protein that renders cells resistant to programmed cell death (apoptosis). Affected cells are, in a sense, excessively durable: They defy the usual checks and balances controlling B lymphocyte growth, and they persist in the lymph node to face chronic antigenic stimulation and ongoing mutagenesis processes that can eventually bring about a malignant clone.
The fact that this translocation exists in a large fraction of healthy adults is evidence that further mutagenic events are crucial for lymphomagenesis. Other notable lesions in indolent lymphomas include t(11;14) in mantle cell lymphoma (causing increased cyclin D expression, and thus cell cycle progression), the deletion of chromosome 13q14 in CLL/SLL (a region containing suppressive micro-RNA that normally silences BCL2)5 and the t(11;18) in extranodal marginal zone lymphomas (producing a true fusion gene that also affects apoptosis) (Table 1). It should be noted that mantle cell lymphoma can behave in an indolent or aggressive manner, and current studies favor high-intensity induction chemotherapy for patients requiring treatment in an effort to improve poor outcomes.
|Follicular lymphoma||t14:18||BCL2 juxtaposed with IgH regulatory sequence, increasing BCL2 expression||Increased BCL2 expression confers resistance to apoptosis Early step in malignant transformation|
|Chronic lymphocytic leukemia / small lymphocytic lymphoma||Deletion 13q14||Causes loss of suppressive micro-RNA elements||Loss of negative regulation of BCL2 confers resistance to apoptosis
Seen in a favorable prognosis subset
|Mantle cell lymphoma||t11:14||Cyclin D1 juxtaposed with IgH regulatory sequence, increasing cyclin D1 expression||Increased Cyclin D1 expression alters cell cycle control, associated with a high mitotic index
Seen in >70% of cases of MCL
|Extranodal marginal zone lymphoma||t11:18||Fuses an apoptosis-inhibitor gene with a novel gene||Alters signals controlling apoptosis in these cells|
|Burkitt lymphoma||t(8;14)||Myc oncogene juxtaposed with the IgH regulatory sequence, increasing Myc expression||Cell cycle progression|
IgH, immunoglobulin heavy chain; MCL, mantle cell lymphoma.
Survival of patients with aggressive lymphomas is measured in months without treatment, and patients with untreated highly aggressive lymphomas can face even shorter survival (weeks). On the other hand, the curability of a number of these patients is well known, and survival rates are improving for many subtypes with modern treatment regimens. Given their tendency for rapid progression and the availability of effective chemotherapy, aggressive and highly aggressive lymphomas are treated immediately upon diagnosis and at times require urgent hospitalization and tumor lysis precautions.
DLBCL is the most common subtype of aggressive lymphoma. About half of all patients with DLBCL achieve long-term disease-free survival after initial therapy, and relapses after more than 5 years are uncommon. Genetic errors involving BCL6 (a transcription factor), BCL2 (an antiapoptotic protein) and FAS (CD95, a TNF-family receptor), are often linked to the development and behavior of DLBCL.6 However, individual lesions fail to explain the pathologic and clinical heterogeneity of DLBCL. To investigate this variability, investigators have applied molecular tools including gene expression profiling to DLBCL tumor samples, which examines tumor mRNA expression patterns. This approach has identified three groups of patients whose gene-expression patterns suggested a distinct cell of origin of their tumor.7 These groups had significantly different prognoses, and the gene-expression technique provided prognostic information surpassing that gained using clinical risk factors alone. However, the challenges for expression profiling are to isolate true driver mutations in lymphomagenesis and to prove them valuable in guiding clinical decisions (such as selection of initial therapy) via prospective trials.
Aggressive T cell lymphomas are rarer and less well understood than their B cell counterparts, and they generally have a poorer outcome. Only about one in three patients with advanced-stage nodal T cell lymphomas survives this diagnosis at 5 years. An exception is the group of anaplastic T-cell large cell lymphomas patient that overexpress the anaplastic lymphoma kinase–gene (ALK) resulting from a translocation between chromosomes 2 and 5. This tends to occur in younger patients and have a more favorable prognosis. Highly aggressive lymphomas, typified by adult Burkitt's lymphoma and adult B and T cell lymphoblastic leukemia or lymphomas, can require immediate hospitalization and are some of the fastest-growing malignancies known. These diseases are in general highly sensitive to combination chemotherapy, and high cure rates (surpassing 60% in the highest-risk patients) are possible. In contrast to the heterogeneity of DLBCL, the highly aggressive Burkitt lymphoma is defined by the deregulation of a single transcription factor, the myc protein. Myc deregulation is observed in more than 90% of cases to be due to a translocation between chromosome 8 (containing the c-myc gene) and one of various partner chromosomes, most commonly chromosome 14. This results in widespread deregulation of genes involved in cell proliferation.
The pathophysiology of HL has been even more challenging to elucidate than NHL. HL lacks the presence of a unifying genetic lesion. The origin of the characteristic Reed-Sternberg cell—long debated—has been traced to a precursor B cell using molecular studies. Such cells account for less than 1% of the lymph node cellularity, and although they originate from a B cell, they downregulate the normal B cell genetic program and do not display typical surface markers. The presence of this unusual cell in the appropriate inflammatory background forms the basis for HL diagnosis. Although EBV is commonly found in Reed-Sternberg cells, and infectious mononucleosis increases the relative risk of developing HL, its role in lymphomagenesis is incompletely understood. The natural history of HL resembles that of an aggressive lymphoma, and the disease is uniformly fatal over months if untreated. The disease has a predilection for manifesting with mediastinal disease and spreading in an orderly fashion via contiguous lymph node groups. Nodular lymphocyte–predominant HL is characterized by lymphohistiocytic variants of the Reed-Sternberg cell (known as L&H cells, or popcorn cells for their microscopic appearance). This uncommon Hodgkin's variant tends to manifest as localized disease without involvement of bone marrow or spleen and without B symptoms; overall, it has a more indolent clinical course.
NHL and HL both commonly manifest with painless enlargement of superficial lymph nodes, with HL usually involving supradiaphragmatic sites. Organomegaly may be present due to liver or spleen involvement, sometimes palpable on physical examination or implied by symptoms such as abdominal discomfort or early satiety. Infiltration of bone marrow can result in anemia or thrombocytopenia. Involvement of nonlymphoid tissues (extranodal involvement) is more commonly seen in more-aggressive lymphomas, as is risk of central nervous system (CNS) involvement, which is greatest in highly aggressive forms.
In contrast, indolent NHL is often diagnosed on an incidental basis—for example, after lymph node sampling performed during an unrelated surgery, or because an imaging study performed for a different illness leads to detection and biopsy of an enlarged lymph node. In the majority of cases, the detection of indolent lymphomas is preceded by a variable but long period of undetected, slow growth. For this reason, bone marrow involvement at diagnosis in indolent lymphoma is common.
Signs and symptoms of lymphoma may be local or paraneoplastic in nature. Local tumor growth can cause palpable masses, compression of adjacent structures, lymphatic obstruction and extravascular fluid accumulation (ascites), or infiltration and disruption of normal organs. Tumors secreting a paraprotein (plasma cell myeloma, lymphoplasmacytic lymphoma, some other varieties) can cause increased serum viscosity and related symptoms especially with IgM and IgA classes. Lymphomas can also manifest a broad range of paraneoplastic effects such as impaired immunity, autoimmune phenomena, thrombophilia, or metabolic and endocrine disturbances. Constitutional symptoms can result in activation of the coagulation system, causing thrombosis, or can result in impaired immunity, leading to recurrent infection. The presence or absence of B symptoms related to a cytokine release is considered part of the staging system for lymphomas (Table 2). B symptoms are defined as fevers higher than 38°C, drenching sweats (especially at night), or unexplained weight loss of more than 10% of body weight in the preceding 6 months.
|I||Involvement of a single lymph node region or lymphoid structure|
|II||Involvement of two or more lymph node regions on the same side of the diaphragm|
|III||Involvement of lymph node regions or lymphoid structures on both sides of the diaphragm|
|IV||Involvement of extranodal site(s) beyond that designated as "E"|
|A: No Symptoms|
|B: Unexplained weight loss of >10% of the body weight during the 6 months before staging investigation|
|Unexplained, persistent, or recurrent fever with temperatures >38°C during the previous month|
|Recurrent drenching night sweats during the previous month|
|E: Localized, solitary involvement of extralymphatic tissue, excluding liver and bone marrow|
Routine laboratory studies are used to assess disease-related complications and suitability for therapy. These include a complete blood count with differential and review of the peripheral blood smear (on which circulating lymphoma cells may be observed). A comprehensive metabolic assessment and tests of kidney and liver function are indicated; these can inform the clinician of mass effects or metabolic consequences such as hypercalcemia or tumor lysis syndrome (which can occur before therapy in aggressive subtypes). Lactate dehydrogenase, immunofixation of urine and serum, monoclonal protein quantification, urinalysis, erythrocyte sedimentation rate (in HL), β2 microglobulin in plasma cell disorders, and HIV serology in patients with risk factors are also indicated.
Although computed tomography (CT) has traditionally been the most universal imaging tool for staging, restaging after treatment, and surveillance, functional imaging with positron emission tomography (PET) has become increasingly utilized. By assessing metabolic activity through uptake of the radioactive tracer 18-fluorodeoxyglucose, PET has more sensitivity for detecting sites of disease not well-imaged by CT, such as bony involvement but also more false positives. In many cases of HL and NHL, interim response to treatment based on PET imaging has been shown to correlative strongly with outcomes. This has led to a number of prospective trials that are investigating the use of functional imaging to risk-stratify patients to different treatments early during their course of therapy.
MRI is indicated in some cases when noncontrast CT is deemed inadequate for patients with renal insufficiency or contrast allergy. PET is being increasingly used, although its sensitivity and specificity (as well as positive predictive value) vary substantially by NHL subtype and timing of test performance. In general, aggressive lymphomas are more PET-avid than are indolent NHLs. Prospective trials incorporating PET at various time points (e.g., during initial staging, early PET during therapy, in restaging after treatment, or in post-treatment surveillance) and for specific lymphoma subtypes are needed to define its proper role in diagnosis and treatment.
Tissue biopsy of affected nodes or tissues is required and fine needle aspiration is inadequate. A core needle biopsy or an excisional biopsy is needed for accurate diagnosis, to determine the lymphoma growth pattern and content of the surrounding tissue. This also helps provide adequate tissue for specialized studies such as immunohistochemistry, fluorescence in-situ hybridization, or polymerase chain reaction. Flow cytometry of the peripheral blood can in the appropriate clinical context provide adequate diagnostic material. Bone marrow aspirate and biopsy is generally required, with few exceptions (such as early-stage HL without B symptoms). Lumbar puncture for evaluation of CNS involvement is often performed in patients at higher risk for central involvement (those with >2 extranodal sites involved, or bone marrow involvement by aggressive lymphoma), sinus or epidural disease or patients who have symptoms or signs suggesting meningeal involvement.
Staging is performed according to the Ann Arbor staging system (see Table 2). Early-stage disease (stage I or II) involves lymph nodes on one side of the diaphragm, and advanced stages involve nodal sites on both sides of the diaphragm (stage III) and/or extranodal sites (including bone marrow, stage IV). CLL/SLL is staged based on the Rai or Benet staging systems, which emphasize sites of spleen involvement and cytopenias but not localization of involved lymph nodes.
The expected clinical behavior of a lymphoma (whether indolent, aggressive, or highly aggressive) is useful in planning management (Table 3). Given the great variability even among disease subtypes, additional clinical and pathologic prognostic information is needed. The presence of disease-related symptoms and patient co-morbidities can also affect the selection and timing of treatment. The effectiveness of treatments are measured by several criteria. One is their response rate in a given disease (determined by the degree of tumor shrinkage by CT criteria as either partial or complete response with PET scanning also being used for this purpose in clinical trials) and the other is the progression-free survival of subjects treated. Effectiveness is measured especially by the overall survival: the percentage of patients alive at a given time (often 5 years) after therapy.
|Category||Indolent Non-Hodgkin's Lymphoma||Aggressive Non-Hodgkin's Lymphoma||Hodgkin's Lymphoma|
|Early stage||Radiation alone may be curative in FL and some other forms||Short-course chemotherapy (CHOP) + radiation for some patients
Chemotherapy (CHOP) with rituximab
|Short-course chemotherapy (ABVD) + radiation for most patients|
Chemotherapy + rituximab for B cell subtypes
|Chemotherapy (CHOP) with rituximab for B cell subtypes for 6-8 cycles
CHOP-based regimen or clinical trial for T cell lymphomas
|Chemotherapy with ABVD for 6-8 cycles, then radiation to persistent bulky masses|
|Relapsed||Different chemotherapy or monoclonal antibody therapy
|Autologous transplant for eligible chemosensitive patients
|Autologous transplant for eligible chemosensitive patients
ABVD, doxorubicin [Adriamcyin], bleomycin, vinblastine, and dacarbazine; CHOP, cyclophosphamide, hydroxydaunorubicin, vincristine (Oncovin), and prednisone.
* Management may vary substantially based on individual patient/disease characteristics.
The adverse effects of chemotherapy have been mitigated by improvements in supportive care, including medications for nausea and growth factors to stimulate neutrophil production, but adverse effects remain. Fatigue and cytopenias leading to potentially serious infections are seen with many drugs. Some notable adverse effects of lymphoma drugs include neuropathy (due to vincristine), cardiotoxicity (doxorubicin), hyperglycemia (prednisone), and T cell immunosuppression with viral reactivation (nucleoside analogues including fludarabine). Given these and other toxicities, a thoughtful assessment of the individual patient before treatment is key to maximizing its benefit.
In general, NHL is sensitive to a number of chemotherapeutic agents, as well as monoclonal antibodies, some molecularly targeted small-molecule therapies in development, and radiation. Early-stage indolent lymphoma may be cured with radiotherapy alone, a standard approach for several subtypes. However, only a small fraction of patients with indolent NHL present with stage I or II disease.
Advanced-stage indolent NHL may be managed in a number of ways, often involving the monoclonal antibody rituximab which induces antibody-dependent cell-mediated cytotoxicity by binding to the CD20 receptor on the surface of target cells. It was first approved in the late 1990s for treating relapsed FL, and it is likely responsible in part for the improvement in survival observed in FL patients since then.6 It is safely combined with chemotherapy in many common intravenous regimens, including the rituximab, cyclophosphamide, vincristine, and prednisone (R-CVP) combination for FL and the fludarabine, cyclophosphamide, and rituximab (FCR) combination for CLL/SLL. The addition of the anthracycline doxorubicin to R-CVP (R-CHOP )resulted in a progression-free survival approaching 7 years in one study,8 but no consensus exists regarding its use in the initial treatment of patients with FL.
Recently, interest has grown in using the bendamustine in combination with rituximab (BR) for NHL and FL in particular. Bendamustine is an alkylating agent with properties of a purine analog as well. A randomized study comparing treatments demonstrated that the BR combination is at least as effective as R-CHOP for indolent NHL but associated with significantly fewer side effects.
Rituximab, when given alone in four weekly doses, produces response rates well over 50% in the initial treatment of FL and is well tolerated; its major adverse effects relate to reactions (fever, chills) during infusion. Extended or maintenance rituximab after initial treatment with the drug prolongs progression-free survival in FL, but its benefit in terms of overall survival, or beyond that achieved with retreatment at progression, is to be determined.
Radioimmunoconjugates, which link radionuclides to monoclonal antibodies, deliver a combined antilymphoma effect (antibody plus radiation-induced cell killing) and have shown high response rates in treating relapsed (60% response rate) and untreated (95% response rate) FL.9 Although approved for relapsed and refractory FL, the optimal treatment in this setting is unclear due to the paucity of randomized data showing an overall survival benefit using one therapy instead of another.
Autologous stem cell transplantation (high-dose chemotherapy followed by reinfusion of the patient's own early hematopoetic progenitor cells) also suffers from a lack of randomized prospective studies, especially in the era of rituximab, to inform decision making in this setting of relapsed indolent lymphoma. Allogeneic stem cell transplantation using an HLA-matched donor has been used with variable success, limited by high upfront treatment-related mortality and complications of graft-versus-host disease. Use of a reduced-intensity initial (conditioning) chemotherapy regimen could make stem cell transplantation less toxic while enabling a graft-versus-tumor effect of the donor's immune system to take hold. However, this intensive approach is relevant to few patients and is best performed in the context of a clinical trial when feasible.
In general, given the lack of demonstrable survival benefit of one treatment over another in relapsed FL and the increasingly rapid pace of drug development in the field, involvement in new therapy clinical trials is an important patient option.
In DLBCL, treatment is based on disease stage and other factors, and chemotherapy is invariably involved. Initiation of treatment is required soon after diagnosis, and tumor lysis precautions may be indicated. The International Prognostic Index (IPI) is a well-studied and reliable method to estimate prognosis in DLBCL. The IPI has been revised to account for improvements in prognosis seen with widespread use of rituximab (R-IPI, Table 4).13
|Risk Factors (1 point each)|
|Age >60 years|
|Eastern Cooperative Oncology Group (ECOG) performance status 2-4|
|Stage III or IV|
|More than one extranodal site of disease|
|Score||Risk Group||4-Year PFS||4-Year OS|
DLBCL, diffuse large B cell lymphoma; OS, overall survival; PFS, progression-free survival; R-CHOP, rituximab, cyclophosphamide, hydroxydaunorubicin, vincristine (Oncovin), and prednisone.
All but some of the earliest-stage cases of DLBCL are treated initially with R-CHOP chemotherapy given every 21 days for six to eight cycles. For some early-stage patients, a strategy involving abbreviated chemotherapy with three cycles followed by local radiation may be adequate. This standard is based on several studies showing that CHOP given in this manner is just as effective as more intensive chemotherapy regimens10 and that the addition of rituximab to three-weekly CHOP improves overall survival.11 The superiority of eight versus six cycles has not been established. Alterations in the schedule of administration, and the addition of other drugs, are strategies under investigation for patients with high-risk DLBCL subtypes, as defined by the IPI or pathologic features. Burkitt's and lymphoblastic lymphomas are treated initially with more-intense regimens, and treatment-related complications and hospitalizations are common. A 60% cure rate may be achieved even in the worse-prognosis subgroup of Burkitt's lymphoma patients. In these patients, and in some patients with DLBCL (such as patients with large cell involvement of the bone marrow, or more than two extranodal sites involved), prophylactic chemotherapy given into the cerebrospinal fluid (intrathecally) via lumbar puncture is used in an effort to reduce the risk of highly morbid CNS relapse.
For patients with DLBCL who relapse but remain sensitive to chemotherapy, an approach using high-dose chemotherapy followed by autologous stem cell transplantation has been proved superior to salvage chemotherapy alone in a landmark prospective, randomized study.12 However, up to half of patients with relapsed DLBCL do not display further sensitivity to chemotherapy. For these patients, or for patients otherwise ineligible for autologous transplantation, the prognosis tends to be poor (10% long-term disease-free survival), and clinical trials of new drugs are an important option.
Mantle cell lymphoma contrasts with DLBCL in that it displays a pattern of continual relapse with a median survival of 2 to 4 years. This has prompted investigation of more-intense upfront chemotherapy regimens which tend to have more toxicity but higher initial remission rates. Recent data suggests that chemotherapy containing alternating cycles of both R-CHOP and cytarabine-containing treatments followed by autologous stem cell transplantation is relatively well tolerated and results in prolonged remission in the majority of patients.
Peripheral T cell lymphoma (PTCL) has historically been treated with CHOP-based regimens in the initial setting, though only one in three patients achieves long-term survival with this approach. The rarity of T cell lymphomas and their diversity has made conducting prospective, well-powered therapeutic trials difficult. Modifications of CHOP or altogether new regimens are being tested in some cooperative clinical trials in the United States. Patients with relapsed PTCL also tend to fare more poorly than those with relapsed DLBCL in most series, including after autologous transplant, making clinical trials of new therapies an important option in this setting.
In contrast to the early 20th century, in which most patients succumbed to HL, about 85% of patients with HL are now cured. In testament to therapeutic successes, the focus of clinical research has shifted toward curing highest-risk groups and preventing late therapy-related complications in patients who are cured.
Early-stage HL is generally treated with four cycles of chemotherapy using ABVD (doxorubicin [Adriamcyin], bleomycin, vinblastine, and dacarbazine) followed by radiation therapy. The long-term risks of radiation (including cardiac disease and the risk of secondary, in-field solid tumors) may be lower with modern techniques, because substantial radiation-related morbidity and mortality has been seen with extensive exposures to the chest.
Advanced-stage HL is usually treated with six to eight cycles of ABVD, followed by radiation to residual bulky masses. Some international variation in practice exists, with application of more intensive chemotherapy regimens to stage III/IV patients with high-risk disease in some countries. Increased toxicity (ranging from treatment-related infertility to mortality) is seen with more-intensive regimens, and how best to apply increased intensity therapy is a point of debate. The use of PET scanning early during treatment is being investigated in a risk-adapted approach, because early PET negativity (after two cycles of chemotherapy) is associated with much better outcomes than a positive PET scan at that time. The latter group of patients is the focus of a number of prospective studies testing whether an early change in treatment (to a new or more intensive regimen) may be of benefit.
In 2011, brentuximab vedotin became the first drug for the treatment of HL approved by the FDA in decades. This agent is an antibody-drug congugate: it contains amonoclonal antibody targeting CD30 linked to an antimitotic agent. After attaching to the cell surface receptor CD30, brentuximab vedotin internalizes into the target cell where it delivers the active component of the drug. This agent has a very high overall response rate in patients with refractory HL and is undergoing testing as part of front-line therapy.
The impact of rituximab on the treatment of NHL, arguably much greater than was expected during development of the drug, is likely to foreshadow an era of rapid development of anti-lymphoma drugs with new modes of action and attenuated toxicity. An improved understanding of disease biology is providing a range of rational targets. In addition to improvements of monoclonal antibodies (by enhancing the mechanisms of immune killing, linking to radiation or other anti-lymphoma drugs), new compounds designed with specific pathophysiologic features are under development. Some target restoration of apoptosis, for example, in tumors demonstrating BCL2 dependence for survival. Others target the effect of deregulated transcription factors such as BCL6 in subtypes of DLBCL. Others block key signaling pathways (such as temsirolimus, which blocks the AKT/mTOR pathway) involved in protein translation and cell growth.
The proteasome, which degrades ubiquitinated intracellular proteins, is the target of bortezomib, a drug approved for relapsed mantle cell lymphoma. This drug is thought to induce apoptosis via a massive disruption in cells' disposal of proteins, affecting tumor cells more than normal ones.
Immunomodulation of the lymphoma microenvironment is an area of intense investigation. Thalidomide and lenalidomide, thought to act in part by this mechanism, are approved for the treatment of multiple myeloma, and early reports show activity in B cell malignancies as well.
Histone deacetylase inhibitors, which open chromatin structure to restore normal transcription of a number of genes suspected to be silenced in tumor cells, have shown promise in relapsed cutaneous T cell lymphoma (for which one such agent is approved) and in early studies in HL.
A number of drugs targeting signals emanating from the B-cell receptor are under development for the treatment of B-cell lymphomas. These include drugs that target the kinases Btk (ibrutinib), Syk (fostamatinib), and the PI3-kinase (GS1101).
In summary, recognition of the heterogeneity of NHL and HL in their natural history, pathophysiology, and response to treatment is key. The WHO classification system has provided a common language for managing and studying lymphomas. Lymphomas can manifest with local or paraneoplastic effects, and HL and NHL are staged relatively similarly. PET scanning offers promise in some lymphomas, although its clinical value and role remains to be defined by ongoing prospective trials. The success of passive immunotherapy using rituximab for NHL has changed outcomes for indolent and aggressive subtypes of NHL alike. Future therapeutic efforts will exploit newly understood aspects of lymphogenesis and will explore unique maintenance or long-term treatment strategies with drugs that can be tolerated on a chronic basis. Cure of lymphoma may be tantamount to elimination with initial induction therapy, suppression using established, chronically administered, well-tolerated therapies, or both.