Handbook of Cancer Chemotherapy (Lippincott Williams & Wilkins Handbook Series), 8th Ed.

22. Non-Hodgkin Lymphoma

Mark Roschewski and Wyndham H. Wilson


Non-Hodgkin lymphomas (NHLs) are a heterogeneous group of malignant neoplasms in which lymphocytes—either of B-cell, T-cell, or natural killer (NK)–cell origin—have arrested at various stages of differentiation, have acquired the ability to clonally proliferate, and do not undergo apoptosis in a typical fashion. Tremendous variation exists in their molecular profiles, mode of presentation, natural history, and response to therapy. Tumor clonality is established by demonstrating immunoglobulin (Ig) gene rearrangement in B-cells, T-cell receptor rearrangement in T-cells, or more sophisticated methods such as the finding of a reciprocal cytogenetic translocation or molecular rearrangements by fluorescent in situ hybridization or polymerase chain reaction (PCR). The malignant clone most commonly proliferates within the lymphatic system, spleen, and bone marrow, but it can also occur in almost any extranodal site such as the bones, central nervous system (CNS), gastrointestinal tract, and the skin. The inciting events in NHL are multifactorial and result from a combination of alterations in antitumor immunity, changes in the local microenvironment of the tumor, and occasionally antigen selection with certain lymphomas now having established relationships with both viral and bacterial pathogens. Still, although advances in molecular medicine have provided exciting insights into the biology of NHL, the full characterization of lymphomagenesis is not complete, and the precise etiology of most cases of NHL is considered unknown.


A. Epidemiology

NHL is the most common hematologic malignancy in the United States, with an estimated 65,540 cases of NHL in the United States in 2010 (35,380 male and 30,160 female) and approximately 20,210 people dying of the disease. It accounts for 4% to 5% of new cases of cancer as well as 3% to 4% of cancer-related deaths. NHL is the sixth most common cause of cancer in both men and women but shows a male predominance in almost all subtypes. From the early 1970s to the 1990s, the incidence of NHL in the United States had been increasing steadily at a compound rate of about 4% in a fashion that prompted some to describe it as an epidemic, but the incidence since 1991 has been stable for men and slowed to an annual rise of 1.1% for women. Death rates from NHL no longer continue to rise and have actually been decreasing by 3.0% in men and 3.7% in women. The 1-year survival rates for all subtypes of NHL are 80%, but drop to 56% at 10 years. B-cell lymphomas represent about 80% to 85% of all cases, with T-cell lymphomas being represented in the other 15% to 20% of cases and NK-cell lymphomas extremely rare. Even though NHL can affect persons of any age, including children, the incidence of NHL clearly increases steadily from childhood through the age of 80 years. Much of the increase in incidence has been in patients in their sixth and seventh decade of life, and the median age of individuals with NHL has correspondingly risen with the incidence; the median age of patients at the time of diagnosis of NHL is currently between 60 and 65 years of age, which has implications regarding therapeutic decisions.

NHL is also increasing in incidence worldwide, and some geographical distribution exists in certain subtypes of NHL (Table 22.1). Overall, the highest reported incidence rates are in the United States, Europe, and Australia, with the lowest rates reported in Asia. Ethnicity does not seem to directly correlate with risks of developing NHL, but there does seem to be some disparity in outcomes after treatment that is related to a patient's socioeconomic status. For example, shorter survival has been associated with low socioeconomic status among elderly patients with follicular lymphoma, but this relationship has yet to be demonstrated in younger patients with NHL. Whether this disparity in outcome represents poorer access to care, true ethnic differences to therapy, or differences in tumor biology are not yet defined.


B. Risk factors

The increase in diagnoses of NHL in the 1970s to 1990s was due, in part, to the improvement in diagnostic techniques, but another reason was the increase in cases of HIV and AIDS over the same period of time. Many additional factors have also contributed to the incidence in NHL and are listed in Table 22.2. The mode of presentation of NHL may differ based on geography, with extra-nodal presentations occurring in only 15% to 25% of adult cases in the United States but in higher percentages in Europe and the Far East. Familial aggregation of NHL plays only a small role in the rise with a two- to fourfold increased risk for NHL in close relatives of patients with lymphoma or other hematopoietic neoplasms. Lymphomagenesis and the study of its inciting events has best been studied in patients with underlying immunodeficiency states. These conditions can be divided into congenital (or primary) immunodeficiencies and acquired (or secondary) immunodeficiencies (Table 22.3). Common components to all of these disorders are defects in immunoregulation, particularly in T-cell immunity, resulting in unregulated B-cell proliferation in lymphoid tissue, often in association with chronic exposures to antigens such as the Epstein-Barr virus (EBV) genome. Chronic inflammation, immune hyperactivity, and immune deregulation are all elements of autoimmune disorders that predispose patients to lymphoma. Most NHLs that occur in association with immune suppression are B-cell lymphomas, with the exception of the increased risk of T-cell lymphoma seen in ataxia telangiectasia and a small percentage of posttransplant lymphopro-liferative disorders (PTLDs), which are of T-cell origin. Rare, extranodal T-cell lymphomas, such as enteropathy-associated T-cell lymphoma (EATCL), will occur in the setting of celiac sprue, and hepatosplenic T-cell lymphoma (HSTCL) typically occurs in patients with a history of solid organ transplantation, inflammatory bowel disease, systemic lupus, Hodgkin lymphoma, and malarial infection.



In addition to HIV, other viruses have specific clinical associations with subtypes of NHL (Table 22.4). Chronic hepatitis C infection is often associated with an underlying indolent B-cell lymphoma, and Helicobacterpyloriinfection can be demonstrated in over 90% of cases of low grade B-cell lymphoma of mucosa-associated lymphoid tissue (MALT) of the stomach. In some cases of antigen-associated lymphomas, eradication of the infectious agent has resulted in partial or complete remission (CR) of the lymphoma.


The classification systems for lymphomas have changed dramatically and frequently since they were first introduced in the 1950s and have been a source of tremendous confusion to clinicians and controversy among hematopathologists. Even the distinction between “lymphoma” and “leukemia” has been a source of potential confusion as many entities cannot be exclusively categorized. In general, if the site of origin is the bone marrow, the disorder may be classified as a form of lymphocytic leukemia, but when disease is present in both nodes and marrow, the distinction between leukemia and lymphoma is somewhat arbitrary. In view of its clinical diversity, accurate classification of NHL is essential for scientific and clinical purposes.


Ideally, a classification system should identify types of NHL that are scientifically and clinically meaningful as well as those that are relatively homogeneous from a clinical, morphologic, immunologic, and genetic point of view. The systems have evolved along with available technology and scientific discovery from ones that rely heavily on morphologic descriptions to the current working system that incorporates morphology, immunophenotypic characteristics based on both immunohistochemistry and flow cytometry (Tables 22.5 and 22.6), cytogenetic and molecular abnormalities (Tables 22.7 and 22.8), and even clinical variables.

In 1956, Henry Rappaport of the U.S. Armed Forces Institute of Pathology proposed a very simple and reproducible classification system based on the growth pattern of the disease (nodular versus diffuse) as well as the appearance of the predominant cell as well-differentiated, poorly differentiated, undifferentiated, or histiocytic. This was followed in the 1970s by the Lukes-Colins-Lennert classification system that related morphology to lymphocyte lineage by dividing entities into B-cell and T-cell disorders based on their cell surface markers; however, they still did not address clinical concerns and were not uniformly utilized internationally.

In the 1980s, the New Working Formulation defined broad categories of lymphoma based on general clinical prognosis of either low-grade, intermediate-grade, or high-grade in order to assist the clinician in treating the lymphoma. The system, however, did not include information regarding immunophenotype, and therefore was difficult to reproduce and did not foster recognition of new entities.

In 1994, the International Lymphoma Study Group developed a consensus list of diseases that could be recognized by pathologists and that appeared to be distinct clinical entities called the Revised European–American Classification of Lymphoid Neoplasms classification system, which ultimately became the World Health Organization (WHO) classification system: the first international consensus on the classification of hematologic malignancies. The original classification system, published in 2001, defined diseases by four features: morphology, immunophenotype, genetics, and clinical information. The updated 2008 version (Table 22.9) increases the use of both molecular profiles and clinical presentations to define separate clinico-pathologic entities.

Gene expression profiling (GEP) is a powerful new technique that allows assessment of the expression of thousands of genes simultaneously on a solid platform. Once this ability to measure gene expression patterns in lymphomas became available, it has become increasingly apparent that the clinical behavior of lymphoid neoplasms is driven by their molecular makeup. The importance of underlying molecular definitions of entities is apparent in the updated WHO 2008 classification system. Newly recognized entities in the WHO 2008 classification system include “grey-zone” lymphomas, which are intermediate between two types of lymphomas. The inclusion of these entities demonstrates that, in some cases, the ability to distinguish between lymphomas requires an understanding of their complete genetic makeup and cannot easily be separated based on clinical data, morphology, immunophenotypic profile, or even cytogenetics alone.





A. Making the diagnosis

One of the most critical, yet often overlooked, aspects surrounding prediction of the prognosis in NHLs is ensuring the accuracy of the pathologic diagnosis. This starts with the appropriateness of the original biopsy procurement procedure, adequacy of the tissue specimen, and the selection of a representative site. In general, fine needle aspiration is not sufficient to accurately classify NHLs given the multiple tests required to accurately classify them. Great preference is placed on excisional lymph node biopsies in order to preserve nodal architecture, while multiple core needle biopsies will suffice in situations where involved nodes are not easily accessible. One should have a low threshold for considering rebiopsy in circumstances where the original biopsy is nondiagnostic or any level of uncertainty remains. Diagnoses made from referring centers are concordant with academic medical centers in almost 95% of cases, but when T-cell lymphomas are suspected, expertise is essential as the discordant rate can be as high as 15%.


B. Ann Arbor staging

1. Recommended workup and imaging.Determining the true extent of disease prior to initiating therapy is important both for determining the treatment plan as well as predicting the likelihood of achieving a complete response to therapy. The staging evaluation of the patient with NHL begins with a history focused on the pace of the disease at presentation, the presence or absence of B symptoms (fevers, chills, drenching night sweats), possible sites of nodal and extranodal involvement, and signs suggestive of a possible underlying immunodeficiency. When performing the physical examination, special care must be given to examining the Waldeyer ring, epitrochlear nodes, and popliteal nodes, which may be difficult to measure on radiographic imaging, as well as examining for sites of extranodal involvement such as the skin and abdomen for signs of hepa-tosplenomegaly. Recommended tests that supplement the history and physical examination at the time of diagnosis are listed in Table 22.10.



The Cotswold modification of the Ann Arbor classification is generally used to stage patients with newly diagnosed NHL and is shown in Table 22.11. Anatomic-based imaging such as contrast-enhanced computed tomography (CT) scans (or magnetic resonance imaging for patients with contrast allergies) are the gold standard for determining which nodal chains are involved with lymphoma. Lymph nodes are considered involved if the long axis is ≥1.5 cm (regardless of short axis) or if the long axis is ≥1.1 cm and the short axis is >1.0 cm. Lymph nodes that are ≤1.0 cm in both axes are considered uninvolved. Determining the stage of lymphomatous involvement based on anatomic imaging alone misses nodes that are involved but not enlarged and can easily miss extranodal sites of disease such as bony involvement which may affect treatment recommendations. Accurate assessment of disease involvement is not only essential at baseline but also when evaluating the response at the end of treatment; functional imaging such as positron emission tomography (PET) scans can greatly aid in this matter.



2. PET/CT imaging. 18Fluorodeoxyglucose (FDG)-PET scans and PET/CT scans exploit the enhanced rate of glucose utilization (both uptake and phosphorylation) seen in many tumor cells as compared to normal surrounding cells (the Warburg effect). Thus, 18FDG-PET scans can provide a semiquantitative measurement of tumor involvement in NHLs, which has proven to have superior sensitivity compared to anatomic imaging alone.

PET/CT scans are not without drawbacks, however, as 18FDG uptake in tumor cells and the subsequent appearance on PET scan is dependent on a number of variables related to the tumor such as blood flow, glucose transporters, tumor cell number, and tumor proliferation, but also related to nontumor variables such as the fat content of the patient, technicalities regarding the procurement of the images, resolution of the scanner, and clinician interpretation. Nonetheless, PET scans and PET/CT scans have essentially become the standard of care in the United States for aiding in the initial evaluation and staging of patients with newly diagnosed aggressive lymphomas. It is estimated that using PET/CT at diagnosis will “upstage” about 15% to 20% of patients with NHL, but the impact of this on treatment options or ultimate outcomes has not yet been elucidated.

C. International Prognostic Index (IPI)

The prognosis of patients with newly diagnosed NHL is clearly related to more than just Ann Arbor staging, and it is standard practice to determine the IPI at diagnosis in patients with intermediate-grade lymphoma such as diffuse large B-cell lymphoma (DLBCL). The IPI score accurately stratifies patients into quartiles with differing probabilities of both complete response as well as 5-year disease-free survival (DFS) in DLBCL based on the age, stage, performance status, number of extranodal sites of disease, and lactate dehydrogenase (LDH) level. For all patients, 5-year survival was 73% for low-risk patients, 51% for low/intermediate-risk patients, 43% for high/intermediate-risk patients, and 26% for high-risk patients (Table 22.12). For patients under the age of 60 years, a slightly modified age-adjusted prognostic system was developed in which the 5-year survival was 83% for low-risk patients, 69% for low/intermediate-risk patients, 46% for high/intermediate-risk patients, and 32% for high-risk patients (Table 22.13).

D. Gene expression profiling (GEP) as prognostic marker

As opposed to clinical variables that may be dominated by extent of disease burden, there has been great interest in attempting to define and predict the clinical behavior of lymphomas based on the expression of genes on tissue microarray. With the advent of GEP, which can simultaneously analyze the overexpression or underex-pression of thousands of genes at one time on an individual's tissue sample, it has become possible to predict outcomes based on the molecular profile of the tissue. The best example of this exists in DLBCL, which can be subdivided into three categorical subtypes: germinal center B-cell (GCB) subtype, activated B-cell (ABC) subtype, and primary mediastinal B-cell lymphoma (PMBL). All three of these subtypes have a widely different prognosis (independent of IPI score), with the GCB subtype responding more favorably to standard chemotherapy regimens than the ABC subtype and the mediastinal B subtype with the best overall prognosis (59%, 30%, and 64% 5-year survivals, respectively).



E. Functional biomarkers

In addition to providing a tool for staging patients, 18FDG-PET scanning also has the potential to serve as an interim functional biomarker during therapy for aggressive lymphomas to predict which patients are responding to therapy and which ones are likely to be refractory. A risk-adapted approach such as that based on interim PET scanning is based on the principle that even though many patients are cured of their lymphoma, a significant minority will be resistant to chemotherapy. Early identification of the patients who will ultimately be failed by their chemotherapy may allow for a change in treatment. Changes suggestive of treatment response occur much earlier in functional imaging than anatomic imaging and this strategy has prompted research into its use. Unfortunately, the promise of this risk-adapted approach is still a research question as the current positive predictive value of 18FDG-PET scanning is suboptimal. Patients who have a negative PET, however, after only one or two cycles of chemotherapy have a much more favorable prognosis than their counterparts.


Indolent lymphoma is a term used to describe the natural history of NHLs that frequently are managed with low-intensity strategies at the time of diagnosis as they may not cause direct symptoms to patients for months or even years. They should not be considered “good lymphomas,” however, because the majority of patients ultimately die of their cancer. Indolent NHLs are generally considered incurable with conventional chemoimmunotherapy strategies. of particular concern is the fact that indolent lymphomas maintain a perpetual risk of transformation into a more aggressive lymphoma; no reliable predictive variables exist to identify the patients at risk for such transformation. The second most frequent subtype of NHL overall (~20%) and prototypical indolent lymphoma specifically is follicular lymphoma (FL). The principles applied to the diagnosis and management of FL can be applied to most other indolent NHLs. Small lymphocytic lymphoma (SLL) is now known to be biologically identical to chronic lymphocytic leukemia (CLL) and is defined by the lack of a leukemic component. CLL/SLL has unique properties which necessitate that it be considered separately from NHL.

A. Follicular lymphoma (FL) pathology

FL is characterized by the translocation t(14;18), which places the BCL2 oncogene under the control of the Ig H enhancer, leading to dysregulated and impaired apoptosis. Histologically, two principal cells exist in the normal follicle center (germinal center): the cen-trocyte (small cleaved cell) and the centroblast (large noncleaved cells). In the 2008 WHO classification of FL, a grading system exists by which one can categorize FL into grades I to IIIa-b based on the ratio of centroblasts/centrocytes. Grade 3b FL (sheets of cen-troblasts) is clinically indistinguishable from DLBCL and is commonly treated with algorithms appropriate for intermediate-grade NHLs whereas grades 1 to 3a are considered truly indolent NHLs.

B. Prognostic scoring systems for FL

FL is typically a widespread disease at presentation that frequently involves multiple nodal chains and the bone marrow, even in cases in which the natural history of the tumor growth will be slow. Also, it is common for patients to experience no symptoms attributable to the disease despite the widespread amount of tumor burden. Thus, prognostic scoring models have been developed to help predict the course of disease.

On the basis of multivariate analysis collected on over 4000 patients and ultimately validated on almost 1000 patients, a five-variable prognostic index was constructed: the Follicular Lym-phoma International Prognostic Index (FLIPI; Table 22.14). The scoring system was found to discriminate outcomes better than the IPI, and it does separate patients into three risk groups with variable 5-year overall survival (OS) rates, but it is based on a cumbersome method of determining the number of nodal sites involved (Groupe d'Etude des Lymphomes Folliculaires criteria) and predated the use of the monoclonal anti-CD20 therapy rituximab, which may render its use obsolete. Despite its widespread use, however, the FLIPI does not define for clinicians which patients should be treated at time of diagnosis.


In an attempt to improve on the FLIPI, an international project was conducted on over 1000 patients with newly diagnosed FL; all were treated with rituximab-based regimens. A prognostic score was also determined from this data set of five variables that could also separate patients into three separate risk groups, termed the F2 (Table 22.15). Even though the F2 model has been available for only a short period of time, it has the strengths of using easily obtainable clinical variables and was derived from patients treated with rituximab. Validation of its prognostic value with current treatment strategies will be required.


C. Front-line treatment principles of indolent NHL

Myriad treatment options exist for patients with newly diagnosed untreated FL, but no curative standard therapy has been identified. The life expectancy of FL patients has been prolonged with the use of rituximab, but it is still characterized by being highly responsive to initial therapies with relapse of disease inevitable. Patients typically undergo many different therapies in their lifetimes. The only potentially curative therapy is allogeneic stem cell transplantation, which is only an option for highly selected patients and carries an unacceptably high risk of treatment-related mortality (TRM) in most patients.

“Watchful waiting” or “dynamic observation” is an appropriate strategy in many patients until the disease progression causes undesired symptoms, progressive cytopenias, or threatens the function of an organ. Discussing this strategy with patients who are otherwise fit for therapy is challenging given the fact that many of them have advanced-stage disease and feel uncomfortable with delaying therapy. All available data, however, suggests that early administration of therapy does not ultimately improve outcomes and certainly carries inherent risks that observation does not. If this strategy is employed, then careful attention should be placed on the pace of the disease as well as the identification of signs of histologic transformation, and PET/CT scanning may add supplemental information in this regard.

Rituximab is a chimeric monoclonal antibody against CD20 that has revolutionized the treatment of all B-cell NHLs since its introduction in the early 2000s. The use of rituximab as a single agent in indolent NHLs is an attractive treatment option for infirm patients because it has relatively few toxicities other than infusional ones and can be given to patients with underlying organ dysfunction. The use of rituximab as a single agent at 375 mg/m2for four or eight doses, most often weekly but at various intervals, in FL yields response rates as high as 73% with progression-free survival (PFS) up to 3 years in duration. Extended duration therapy up to 2 years or longer may be an even more effective strategy for some patients and appears to be safe. In general, however, single-agent rituximab is not terribly effective for patients with bulky nodes that are greater than 10 cm and should be reserved for patients who are not candidates for combination chemotherapy.

Most patients with FL are initially treated with one of a handful of standard combination chemotherapy regimens with the addition of rituximab (Table 22.16). All of these regimens produce overall response rates in excess of 90% with approximately 50% complete responses, but can vary in the duration of maintaining that response and have varying toxicities. The use of anthracyclines is not mandatory in patients with untreated FL, but according to the National LymphoCare study, which examined practice patterns in both community- and academic-based institutions, rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone (R-CHOP) remains the most commonly prescribed regimen in the United States with some geographical variation. Other strategies would be to save anthracycline-based therapies for signs of transformation, but no consensus exists. Bendamustine-rituximab is a newly available regimen based on an old chemotherapeutic developed in East Germany in the 1940s that has shown surprisingly high response rates and PFS but with less toxicity in a recent randomized study compared to R-CHOP. Given the fact that this regimen is easily tolerated by patients, it may eventually replace R-CHOP as the preferred regimen in the United States for untreated patients with FL and other indolent lymphomas.


One treatment strategy that has impressive results in phase II studies but remains largely underutilized is the use of the anti-CD20 radioimmunoconjugates such as iodine131-tositumomab (Bexxar) and ibritumomab tiuxetan (Zevalin). These agents, collectively known as radioimmunotherapy, are U.S. Food and Drug Administration (FDA)-approved for use as both the initial therapy of indolent NHL as well as in a consolidative strategy after initial combination chemotherapy. Barriers to its widespread use, however, include cost and reimbursement issues, unfamil-iarity with the conjugates, and the fact that other services (such as nuclear medicine or radiation-oncology) are required to use these agents. Additionally, concerns linger regarding the long-term risks of treatment-related complications such as myelodys-plasia or acute myelogenous leukemia, and long-term safety data are currently lacking.

Because none of the above strategies can claim superiority to other strategies based on controlled data, the list of potential options is long in patients with untreated indolent lymphomas. Many different strategies are employed with different types of patients and there is not currently one standard approach. However, it is probably important to avoid bone marrow toxic agents (fludarabine) in patients who may be candidates for autologous transplantation in the future.

D. Maintenance therapy in indolent NHL

Given that relapse is considered definite after initial therapy for indolent lymphoma, great emphasis has been placed on finding effective strategies at maintenance therapy. Interferon-a was the first agent tried in this setting, but it did not produce significant results in all patients and was difficult to tolerate in many. There have been several attempts at using cancer vaccines in consolidation after chemotherapy with mixed results. Because rituximab has limited long-term toxicities, it has been an attractive agent for this purpose. Its use after initial chemotherapy and now chemoimmunotherapy with the recently reported results of the PRIMA trial has demonstrated a benefit in PFS, but there has been no OS benefit yet. No consensus currently exists on the appropriate duration or schedule of maintenance therapy, however, and toxicities such as hypogammaglobuline-mia and reversible neutropenia have been observed. Until long-term data exists demonstrating a clear OS benefit without unexpected toxicities, this cannot be recommended to all patients, but it appears to be a promising method of prolonging response duration and treatment-free intervals in many patients with indolent lymphomas.

E. Other indolent lymphomas

1. MALT lymphoma.The treatment principles applied to FL do not apply to all indolent lymphomas, and some lymphomas such as extranodal MALT lymphomas (MALTomas) may respond to treatment of underlying associated pathogens. MALTomas can occur at a number of sites including conjunctiva, thyroid, salivary gland, and gastrointestinal tract and are frequently antigen-driven. MALTomas that arise in immunocompetent patients tend to be localized and are associated with a better survival than other low-grade lymphomas. Of particular importance is the fact that many gastric MALTomas associated with H. pylori will respond to antibiotic therapy directed at the bacteria alone and will not require either radiation or chemotherapy.


2. Cutaneous T-cell lymphomas (CTCL; mycosis fungoides).Another subset of indolent NHLs that deserves special attention are the cutaneous lymphomas. Cutaneous lymphomas include a wide variety of diseases of both B-cell and T-cell origin which present primarily involving the skin and are predominantly indolent. These NHLs are frequently misdiagnosed as other skin disorders such as eczema, and it is not uncommon for patients to be followed for many years before an accurate diagnosis is made. The most common cutaneous lymphoma is CTCL, also known as mycosis fungoides. When there is generalized erythroderma and involvement of the peripheral blood, the syndrome is known as Sezary syndrome. Prognosis in CTCL depends on stage of disease, and special staging systems for CTCL exist (Table 22.17). Clinical stage IA disease is so indolent that it does not impact on normal life expectancy, while prognosis worsens with more advanced stage of disease. The disease may exist as plaques in the skin for many years before progressing to involve skin tumors, lymphadenopa-thy, or visceral disease. Often, this clinical progression occurs in association with a pathologic transformation to a more aggressive lymphoma. If disease is limited to the skin, topical therapy such as topical nitrogen mustard, electron beam radiotherapy, or psoralen (often in conjunction with ultraviolet radiation) may be employed. Combination chemotherapy regimens such as those used for intermediate-grade lymphoma maybe used for disease involving nodes or viscera. Other approaches to this disease include the use of denileukin diftitox, an antibody to CD25, and bexarotene, a novel retinoid X receptor-selective retinoid (see Chapter 14).


A. Diffuse large B-cell lymphoma (DLBCL)

DLBCL is the most common lymphoid neoplasm in adults, representing about 30% of cases diagnosed in the United State annually. In most cases, it is very responsive to combination chemoimmunotherapy regimens, and an approach with curative intent is appropriate in most cases ofnewly diagnosed DLBCL. Still, a significant minority of patients will relapse after initial response to chemotherapy or be primarily refractory. The molecular mechanisms underlying DLBCL have been elegantly defined recently, and we now know that tremendous genetic heterogeneity exists and that the different subsets are accompanied accordingly with a different prognosis. Molecular profiling and DNA microarray studies have demonstrated that the relationship with the stroma (mi-croenvironment) as well as disease subtype as defined by GEP are important in determining the behavior of DLBCLs. Novel agents with rational targets are beginning to enter the clinical arena in the treatment of DLBCL, and exciting treatment breakthroughs are likely to follow.

As previously mentioned, the Lymphoma/Leukemia Molecular Profiling Project has essentially established three subtypes of DLBCL with very different molecular profiles: the GCB subtype, the ABC subtype, and PMBL subtype. Forthcoming from the elucidation that different subtypes of DLBCL exist has been the realization that they are addicted to different signal transduction pathways for their survival. Upfront treatment decisions are not yet routinely altered based on the underlying tissue subtype, but a potentially attainable goal in the future is to define a more “personalized” treatment approach based on the expression of genes unique to individual patients' tumors.

Another emerging concept related to the biology of DLBCL is the fact that certain subsets of DLBCL contain molecular mutations that portend a particularly grave prognosis. It is now recognized that 5% to 10% of DLBCL cases actually possess a rearrangement in the c-myc oncogene, which makes them highly proliferative and less responsive to standard regimens. Two different groups have retrospectively reported their experience with c-myc+ DLBCLs and both found distinctly poor OS in this subset of patients. The 2008 edition of the WHO classification scheme recognizes that some cases of c-myc+ DLBCL may be very difficult to distinguish from Burkitt lymphoma (BL), and a landmark study was able to demonstrate that based on GEP, some cases of DLBCL may actually have a BL-like molecular signature.

In addition to c-myc+ DLBCLs, another subset of DLBCLs with a poor prognosis are those that have a “double-hit” with the BCL2 translocation t(14;18) (as seen in FL) in addition to the c-myc translocation. These tumors are particularly troublesome because they have developed mutations in both proliferation and impaired apoptosis. No consensus exists on the correct approach to patients in these poor-risk subgroups, but it is becoming increasingly important to identify these patients prior to treating them with standard regimens as they all do uniformly poorly.

B. Initial treatment of advanced-stage intermediate-grade NHL

Treatment principles of combining chemotherapeutic agents without overlapping toxicities and giving the maximally tolerated dose have defined the approach to DLBCL since the advent of chemotherapy in the 1960s. Anthracyclines as a class of agents established themselves many years ago as the single most important drug in the treatment of aggressive lymphomas. In the 1980s, single institution phase II studies of dose-intensive regimens such as cy-clophosphamide, doxorubicin, etoposide, bleomycin, vincristine, methotrexate, and prednisone (known as Pro-MACE-CytaBOM); methotrexate, bleomycin, doxorubicin, cyclophosphamide, vin-cristine, and dexamethasone (known as m-BACOD); and metho-trexate plus leucovorin rescue, doxorubicin, cyclophosphamide, vincristine, prednisone, and bleomycin (known as MACOP-B) began to report what appeared to be markedly improved response rates compared to conventional response rates seen with cyclophosphamide, doxorubicin, vincristine, and prednisolone (CHOP) therapy. However, a landmark study in 1993 re-established CHOP therapy as the backbone by which all challengers are to be compared based on OS.

Prior to rituximab, only about 30% to 40% of patients were cured of their DLBCL if found in advanced stages. Almost immediately on its availability, rituximab overcame some of the chemore-sistance of DLBCL (possibly via overcoming BCL2 overexpression) and has become standardly incorporated into every regimen used in this disease with essentially the same result of improving outcomes. Every time chemotherapy with rituximab has been compared to chemotherapy without rituximab, it has been demonstrated superior. Thus, the current standard of care is to administer R-CHOP for six to eight cycles every 3 weeks for patients without contraindication for advanced-stage DLBCL (Table 22.18). Efforts to give the regimen in a “dose-dense” fashion every 14 days with support with growth factors has been advocated by some and show feasibility, but formal comparisons of R-CHOP-14 to R-CHOP-21 disappointingly demonstrated no clear superiority.

In comparison to the bolus administration of chemotherapy in the R-CHOP regimen, the dose-adjusted etoposide, doxorubi-cin, vincristine, cyclophosphamide, prednisone, and rituximab (EPOCH-R) regimen utilizes infusional administration of agents over 96 hours in an effort to affect cells that are not dividing on the first day of administration. Also, the regimen adjusts the dose of the doxorubicin, etoposide, and cyclophosphamide with subsequent cycles using the absolute neutrophil count as a biomarker. The dose-adjusted EPOCH-R regimen is designed to be effective in more proliferative tumors, and response rates in single institution studies and a multi-institutional phase II trial demonstrate response rates that are higher than those seen in R-CHOP; results of a formal phase III comparison between EPOCH-R and CHOP-R are anxiously awaited. Until a regimen such as EPOCH-R demonstrates superiority either in all patients with DLBCL or certain subgroups, R-CHOP will remain the most widely used regimen in DLBCL.


C. Initial treatment of localized intermediate-grade NHL

The contemporary treatment options for patients with early-stage (stage I or II) DLBCL usually employ R-CHOP therapy but utilize fewer cycles and often include consolidative radiation therapy to sites of involvement. Unlike Hodgkin lymphoma, there is almost no role for radiation therapy alone in early-stage DLBCL because the disease does not spread to contiguous nodes as predictably as HL and commonly has disseminated tumor cells that escape detection at diagnosis. Combined modality therapy with brief duration chemotherapy followed by radiation became the standard approach based on a Southwest Oncology Group trial in the late 1990s that demonstrated an OS benefit to this approach, but whether or not rituximab has obviated the need for radiation therapy has not yet been determined. Thus, most patients will be offered a combination of R-CHOP for two to three cycles followed by consolidative radiation to sites of involvement. Reported 5-year OS rates with this approach are in excess of 90%. In patients in whom the risk of radiation may outweigh its benefit in the long term (young women with nonbulky mediastinal masses), it may be reasonable to use R-CHOP for six cycles in lieu of consolidative radiotherapy. The standard therapy for patients with bulky disease (>10 cm), however, remains consolidative radiation therapy.

D. CNS prophylaxis

Involvement of the CNS at the time of diagnosis of DLBCL is uncommon with incidence rates of only 5%, but at the time of relapse, uncontrolled disease in the CNS is a very poor prognostic sign. Thus, efforts to predict who is at risk to develop CNS disease have focused on patients with an extranodal disease presentation and/or an elevated LDH at diagnosis. By convention, patients with at least two sites of extranodal involvement (including bone marrow) and an elevated LDH as well as those with “sanctuary sites” involved at presentation (testis, nasopharynx, orbit) are offered CNS prophylaxis with intrathecal therapy: usually methotrexate with or without cytarabine. The administration of CNS prophylaxis, however, is very much understudied and no consensus exists. Therefore, significant variation exists in the patients selected as well as the number and total doses of prophylaxis given.

E. Primary mediastinal B-cell lymphoma (PMBL)

As previously mentioned, PMBL has been identified on the basis of DNA microarray profiles to have a unique molecular profile that carries a relatively good prognosis with current treatment strategies. Clinically, it has some unique features, including the fact that it commonly affects persons of young age (classically women) and at the time of initial presentation usually stays within the mediastinum. Bone marrow involvement and nodal chains below the diaphragm are very uncommon, making the Ann Arbor staging system unhelpful in this disease. Thus, all patients with PMBL are treated with advanced-stage disease principles of six to eight cycles of R-CHOP usually followed by mediastinal radiation irrespective of stage. Recently reported data suggests that with the use of dose-adjusted EPOCH-R, mediastinal radiation may be safely omitted, but this finding requires further validation.

F. Response assessment of DLBCL

Of equal importance to finding an effective treatment regimen is the ability to determine its effect at the end of the therapy. An International Working Group has published guidelines that determine the response to therapy based on the bidimensional measurements of involved nodal groups at the time of diagnosis and again at the end of therapy. If all nodes have regressed to no more than 1 cm in size and symptoms of disease have disappeared, the patient is considered in CR. If the sum of the product of diameters has reduced by 50% but not all nodes are no more than 1 cm, then it is considered a partial response. As residual necrotic tissue and inflammatory cells often prevent bulky nodal groups from ever getting to no more than 1 cm, the use of PET/ CT scanning at the end of therapy is highly useful. In patients who present with bulky mediastinal nodes (such as patients with PMBL), the use of PET/CT scans to confirm a complete response is almost mandatory to differentiate between residual tumor cells and dead cells.

G. Mantle cell lymphoma (MCL)

MCL is an uncommon NHL derived from naive small B lymphocytes. These lymphomas have a diffuse pattern; are generally CD5 positive, CD23 negative, positive for cyclin D-1; and are associated with a t(11;14) chromosomal translocation. MCL essentially shares the worst features of both indolent NHLs and aggressive NHLs in that it is practically incurable but usually requires therapy at diagnosis to prevent a rapid downhill course. Current therapies have slightly improved the median survival of MCL from 3 to 4 years to closer to 7 years, but it is still one of the NHLs with the worst prognosis.

The proliferation signature of MCL as defined by GEP has proven to be highly predictive of outcome in that patients with more proliferative variants of MCL have a much worse prognosis than patients with low proliferative variants. In fact, a subset of patients with MCL can often go several years without therapy while enjoying a lack of progression of disease. Clinical, pathologic, and radiographic variables have all been used to attempt to identify which patients can be observed and which need immediate therapy, but it is currently difficult to accurately identify these subsets prospectively. A prognostic score similar to the IPI, termed the MIPI (Table 22.19) has been developed and is currently undergoing validation in numerous prospective studies.

Despite the poor prognosis, MCL is very sensitive to both chemotherapy and radiation with the overall response rate being in excess of 90% with R-CHOP therapy. These responses, however, are usually of brief duration, and relapsed MCL is often unresponsive to conventional chemotherapy. Thus, the most common strategy for otherwise fit individuals with MCL is to undergo treatment with intense regimens with cytarabine-based combinations and to consolidate with high-dose therapy with stem cell rescue in first CR (Table 22.20). It is likely that this approach is not curative, but long-term data are not yet available and it is the current standard approach for fit patients with newly diagnosed MCL.

H. Peripheral T-cell lymphomas (PTCLs)

It has become apparent that PTCLs have a worse prognosis than that of their B-cell intermediate-grade lymphoma counterparts, and as a result, have a relatively poor prognosis except for the ALK+ anaplastic large cell lymphomas (ALCLs). PTCLs are a heterogeneous group of lymphomas that constitute only 5% to 7% of adult NHLs diagnosed in the United States. The International T-cell/NK Lymphoma Project demonstrated that the use of an anthracycline in PTCLs does not correlate to improved outcomes compared to nonanthracycline-based therapies, bringing into question the conventional approach of using CHOP as initial therapy in most cases of PTCL.


Unfortunately, with the exception of CHOP for ALK+ ALCL, an optimal therapy for PTCLs has not yet been defined, and practice patterns vary markedly between institutions. Agents that seem to have activity and are being explored include gemcitabine, alem-tuzumab (a monoclonal antibody that targets CD52), pentostatin, histone deacetylase inhibitors, and the immunomodulatory drug lenalidomide. Patients with PTCL should be encouraged to participate in well-designed clinical trials whenever possible in order to help determine optimal first-line treatment algorithms.

In September 2009, the FDA approved the first agent specifically for the treatment of relapsed/refractory PTCL in the form of the novel antifolate drug pralatrexate. The approval was based on a multicenter phase II trial that demonstrated significant activity of pralatrexate in patients who had been refractory to many previous therapeutic regimens.


I. Rare, extranodal T-cell lymphomas

HSTCL, subcutaneous pannicuiitis-like T-cell lymphoma, and EATCL are rare subtypes of PTCLs that often present with primarily extranodal disease. Despite the fact that these tumors have distinct clinical and pathologic features, they are often diagnosed after significant delay. Also, the use of anthracycline-based combination chemotherapy that is highly effective in B-cell counterparts is not effective in these disorders. The combination of delay in diagnosis, aggressive natural history of these tumors, and ineffective therapies has resulted in a poor prognosis in most cases. In patients who do achieve remissions with initial therapy, consideration should be given to consolidation with allogeneic stem cell transplantation given the dismal prognosis, especially for HSTCL, cutaneous 78 T-cell lymphoma, and the anaplastic variant of EATCL.


A. Lymphoblastic lymphoma (LBL)

LBL is a highly aggressive lymphoma regarded as a variant of acute lymphoblastic leukemia (ALL) that is rapidly fatal without prompt initiation of combination chemotherapy. In adults, it is most commonly of T-cell lineage (85% to 90% of cases), but can also be of B-cell lineage, and demonstrates a male predominance. Tissue specimens are often composed of small- or medium-sized blast cells with scant cytoplasm that are positive for immature markers such as CD99 and TdT. LBL commonly presents with a mediastinal mass (60% to 70% of patients), elevated LDH levels, and B symptoms; bone marrow involvement is present in about 20% of cases. Although all patients with LBL should receive intrathecal therapy as part of their regimen, actual CNS involvement at the time of diagnosis is less than 10%. R-CHOP and other regimens used for intermediate-grade NHLs are not effective in LBL. ALL-like regimens with CNS prophylaxis are required to effectively treat LBL and typically are administered in three phases (induction, consolidation, and maintenance) with prolonged administration lasting up to 24 to 36 months (cases of B-cell lineage do not typically require prolonged maintenance therapy). A major determining factor in treatment outcomes is the underlying cytogenetic and molecular profile of the tumor. Patients with the t(9;22) translocation have a poor prognosis whereas those with t(12;21) and HOX mutations have a better prognosis. Outcomes in adults treated for LBL are inferior to pediatric outcomes, which may reflect disease biology or adherence to treatment protocols.

B. Burkitt lymphoma (BL)

BL is composed of post-germinal center mature B-cells with a highly aggressive presentation and natural history. Morphologically, Burkitt tumor cells are small noncleaved cells arranged in a monotonous pattern and may have a “starry sky” appearance. It is often subdivided into cases that are endemic, sporadic, and associated with underlying HIV. All cases of BL are associated with a c-MYC translocation and most commonly arise from the reciprocal t(8;14) translo-cation although variant translocations of t(2;8) and t(8;22) exist.

In the United States, BL tends to occur in younger patients and is associated with a higher incidence of gastrointestinal disease. R-CHOP is substandard therapy for BL, and high-intensity brief-duration regimens such as the R-Hyper-CVAD regimen (ritux-imab plus hyper-cyclophosphamide, vincristine, doxorubicin, and dexamethasone), or the cyclophosphamide, vincristine, doxorubicin, high-dose methotrexate/ifosfamide, etoposide, high-dose cytarabine (known as CODOX-M/IVAC) regimen (Table 22.21) have been associated with long-term DFS in almost 50% of patients. Patients with disease involving the CNS or bone marrow have a worse prognosis than those patients with limited-stage disease. The role of stem cell transplantation as routine consolidation following initial high-dose therapy or in relapsed disease has not been established by clinical trials.

C. Tumor lysis prophylaxis

Highly aggressive lymphomas such as LBL and BL are composed of rapidly proliferative tumor cells that have the potential to cause spontaneous lysis of tumor cells even before the administration of combination chemotherapy. When tumor cells lyse, they release intracellular cations such as potassium, uric acid, and phosphorus, which can be toxic to renal tubules. This potentially fatal clinical scenario is termed the tumor lysis syndrome and is becoming increasingly common as regimens become more effective (see Chapter 29).



A. Salvage chemotherapy

Up to 50% of patients with DLBCL will ultimately relapse after initial chemoimmunotherapy; the number is higher with MCL and PTCL, meaning that large numbers of patients will ultimately need second-line or “salvage” chemotherapy. Aggressive NHLs such as DLBCL that have relapsed after response to initial chemotherapy are still curable in a significant minority of cases and the duration of response (DOR) can sometimes be used as a surrogate for chemotherapy sensitivity. One of the most important prognostic markers of relapsed NHL is whether it remains chemosensitive or has become resistant to chemotherapy. Tumors that have become chemoresistant or are primarily refractory to initial chemotherapy (defined as never achieving CR or a DOR 6 months) uniformly have a dismal prognosis and are likely incurable.

Despite the importance of salvage chemotherapy, the optimal regimen remains a research question as comparative data is lacking. Typically, a platinum-based regimen (Table 22.22) is administered for two to three cycles (usually with rituximab) and patients are then taken to high-dose therapy with stem cell rescue (autologous transplant) if they demonstrate chemo-sensitivity. The standard practice of taking every patient to autologous transplant in first relapse is based on the PARMA study published in the pre-rituximab era that demonstrated an OS advantage to this approach versus chemotherapy alone, with up to 30% of patients enjoying long-term DFS. In the current era of rituximab being given with initial therapy in almost all cases, however, less is known about the biology of disease relapse and its response to salvage chemotherapy. The recently presented results of the international, multicenter CORAL study suggest that fewer patients (approximately 15%) will be salvaged with autologous transplantation done in second CR, suggesting the need for an alternative approach that utilizes novel agents. In the era of PET/CT scanning, an emerging concept is that patients with residual 18FDG-PET positivity prior to transplant may not benefit from high-dose therapy and should not be offered this approach.


High-dose chemotherapy with stem cell transplantation can also be offered to patients with relapsed low-grade NHL, but since this is not a curative procedure, enthusiasm for this approach is not high. Additionally, the use of high doses of chemotherapy carry a risk of treatment-related myelodysplasia and acute myeloid leukemia, and patients with indolent NHL often get multiple therapies in their lifetime which increases that risk. However, younger patients with indolent NHLs such as FL have been salvaged with autologous transplantation in first CR with an improvement in PFS but not OS being demonstrated.

B. Allogeneic transplantation in relapsed NHL

Myeloablative allogeneic transplantation has traditionally been reserved as a “last ditch” option for select patients with relapsed both indolent and aggressive NHLs that have failed all other therapies. Even though the graft-versus-leukemia effect can be very potent in NHL, the TRM and incidence of at least grade III graft-versus-host disease has been a significant barrier to widespread use of this approach. Nevertheless, it appears that allogeneic transplantation is a potentially curative approach in carefully selected patients with relapsed NHL; with the use of reduced-intensity preparative regimens, more patients are becoming eligible for this high-risk/high-reward therapy.

C. Novel/targeted agents

Numerous agents with novel mechanisms are being tested in relapsed/refractory NHLs in an effort to address this unmet clinical need. Many of these agents are considered “targeted” agents that attempt to selectively inhibit tumor cells in a more selective manner than conventional DNA-damaging agents. Molecular biology has uncovered many pathways important for proliferation and impaired apoptosis, which are the targets of many of these agents. In addition, some novel agents are thought to work by affecting the tumor cell's interactions with the local mi-croenvironment, which may be providing protection from the effects of chemotherapy. Agents that are currently being studied include immunomodulating agents (such as lenalidomide), pro-teasome inhibitors (such as bortezomib), BCL2 inhibitors, his-tone deacetylase inhibitors, mammalian target of rapamycin inhibitors, PI3k/AkT inhibitors, MEK inhibitors, PARP inhibitors, cyclin-dependent kinase inhibitors, Bruton tyrosine kinase inhibitors, and novel monoclonal antibodies targeting both CD20 and CD19. Many of these agents are likely to be important in the management of NHL moving forward.


NHL is very sensitive to the tumoricidal effects of radiation therapy, and it remains an essential treatment modality to improve local control in the contemporary management of patients with NHL. Enthusiasm for widespread use of radiation therapy has been somewhat tempered, however, by concerns about its inherent risks of localized tissue destruction; premature risks of coronary artery and valvular heart disease; secondary malignancies such as carcinomas of the breast, lung, and thyroid; and sarcomas that frequently present 15 to 25 years after patients are cured of their NHL. Additionally, as chemotherapeutic strategies improve and functional imaging modalities such as PET/CT can better define tumor responses in residual nodal masses, questions arise regarding the need for con-solidative radiation.

Current indications for radiation therapy include a potential curative approach in limited-stage indolent NHL such as FL or gastric MALT as well as consolidative therapy to sites of tumor bulk in patients with mediastinal lymphomas and primary CNS lymphoma (PCNSL). Another potential method by which to take advantage of the effects of radiation is to employ the use of “targeted radiation” in the form of anti-CD20 radioimmunoconjugates such as iodine131- tositumomab (Bexxar) or ibritumomab tiuxetan (Zevalin), which are particularly attractive in low-volume disease in patients with indolent lymphomas, either in the relapsed setting or as a consolidative approach after initial chemotherapy. It is important to use caution in patients with significant bone marrow involvement, however.

Currently used modalities for employing radiation therapy may eventually be replaced with advanced modes of high-precision radiotherapy that utilize computer-controlled linear accelerators to deliver precise radiation doses to a malignant tumor or specific areas within the tumor such as intensity modulated radiotherapy and three-dimensional proton beam therapy (PBT). These advancements in the field purport to deliver higher doses and more precise therapy to tumor cells while at the same time minimizing the dose to surrounding normal critical structures. Another potential advantage of using PBT is that clinical data from cancer survivors thus far has not demonstrated an increased risk of secondary malignancies.


A. AIDS-related lymphomas

Among patients with HIV infection, 3% to 6% will develop NHL, and this is considered an AIDS-defining illness. The lymphoid neoplasms that are associated with HIV include DLBCL, PCNSL, primary effusion lymphoma, plasmablastic lymphoma of the oral cavity, and BL (see Table 22.4) and concomitant infection with human herpesvirus 8 and EBV also appear to play a role in pathogenesis. NHL is typically a late manifestation of HIV with most cases occurring when the cluster of differentiation 4 (CD4) cell count is less than 200 mm3 except for BL which can occur with any CD4 count (see Chapter 25for further discussion of the HIV- and AIDS-related lymphomas).

B. Post-transplant lymphoproliferative disorders (PTLDs)

Abnormal expansions of lymphoid cells (either B-cell or T-cell) in patients who have undergone either solid organ or hematopoietic stem cell transplantation are defined as PTLDs. Given the broad definition, PTLDs are a chnically heterogeneous group of disorders with marked variation in clinical behavior ranging from benign expansions to aggressive and fatal NHL.

Risk factors for the development of PTLD are listed in Table 22.23. EBV plays a pivotal role in the development the majority of PTLDs. In fact, many transplant centers now routinely monitor periodic PCR for EBV and will preemptively treat cases of a rapidly rising viral load of EBV with rituximab.


If the PTLD is limited in extent, one can employ therapy designed at increasing the host response to EBV, such as withdrawing or decreasing immunosuppression, administering interferon, or giving lymphocytes from individuals who have had EBV infection. Such therapy is most effective in limited disease and in patients in whom the lymphocytes are polyclonal.

Radiotherapy or surgery can be effective in localized and accessible disease, but for more advanced disease and for disease in which the lymphocytes are monoclonal, combination chemotherapy with or without rituximab or donor lymphocyte infusions for stem cell transplant patients is often necessary. Unfortunately, response rates are lower for transplant-associated lymphomas than for de novo lymphomas, and long-term survival has been disappointing.

C. Primary central nervous system lymphoma (PCNSL)

PCNSL is defined as a NHL that is confined to the CNS and has no systemic involvement. PCNSL has risen more rapidly than other extranodal sites in part due to the association with HIV, but it has risen in immunocompetent hosts as well. It is of a B-cell origin in almost all cases in the United States and typically of an intermediate-grade histology, such as in DLBCL. For reasons that are not entirely clear, it very infrequently metastasizes outside of the CNS even though very sensitive techniques such as PCR can detect tumor cells in the peripheral blood of a significant minority of patients with PCNSL. The diagnosis can be made with direct tissue biopsy or demonstration of lymphoma cells on examination of the cerebrospinal fluid (CSF). Of note, the presence of a positive PCR for EBV DNA found in CSF is virtually diagnostic of PCNSL. If PCNSL is considered in the differential diagnosis of patients who present with isolated brain lesions, it is important to attempt diagnostic studies prior to the administration of steroids as tumor cells are very sensitive to steroids and the opportunity to accurately characterize the lymphoma may be lost.

Treatment with radiation therapy alone is generally associated with poor OS, and radiotherapy alone is no longer recommended for patients with PCNSL who are able to undergo more intensive therapies. The current standard is a multimodality approach that utilizes combination chemotherapy with agents that cross the blood–brain barrier such as high-dose methotrex-ate and cytarabine (with the addition of rituximab if of B-cell lineage) followed by consolidative whole-brain radiotherapy (WBRT). Such approaches can achieve overall response rates greater than 90%, but 2-year PFS rates are approximately 50% to 55% and relapse is a common clinical problem. Due to concerns of both acute and long-term toxicities on both central and peripheral neurologic function associated with WBRT, some regimens attempt to use chemotherapy alone, but response rates tend to be lower. The overall effect on OS with the omission of WBRT, however, is not yet clear.

D. Testicular lymphomas

Testicular lymphomas represent the most common testicular tumor seen in men over the age of 60 years, with DLBCL being the most common histologic type. Primary testicular lymphoma (PTL) is a rare entity with a distinct natural history compared to nodal DLBCL that frequently affects the contralateral testis, lung pleura, and soft tissues. Recent reports suggest that the majority of these tumors are of the ABC subtype of DLBCL with marked proliferative activity. Therapy consists of a multimodality approach of surgical removal of the primary tumor, systemic chemoimmunotherapy, and radiation of the contralateral testis. Additionally, as PTL originates in an im-munoprivileged site, prophylactic treatment of the CNS is uniformly administered. Despite this approach, PTL has an inferior outcome compared to nodal DLBCL and can be associated with late relapses.

E. NHL in the elderly

NHL in patients over the age of 60 years deserves special mention because the median age at diagnosis of NHL is between 60 and 65 years, the U.S. population is aging, and age consistently is determined to be a prognostic marker in outcomes in NHL despite no known biologic differences in the disease associated with age. Thus, a common clinical problem involves the selection of appropriate therapy for a patient with intermediate-grade NHL over the age of 70 years or older than 60 years with significant comorbid disease. Patients in this category have been underrep-resented in the prospective clinical trials that have defined our standard treatment approaches, with data regarding patients over the age of 80 years being virtually nonexistent. Concerns about the potential toxicities for these patients frequently leads to empiric dose reductions or dose delays that likely compromise the potential for cure. Available data, however, suggest that when CHOP is used in lower doses, remission rates decline and survival is shortened. In the Group d'Etude des Lymphome dAdulte LN 98-5 trial, 399 patients with DLBCL aged 60 to 80 years were treated with R-CHOP versus CHOP. The addition of rituximab was associated with a 15% improvement in CR rate with associated improvements in DFS and OS. In this trial, the most common ≥G3 toxicity was infection, cardiac toxicities ≥G3 were seen in only 8% of patients, and other toxicities were consistent with those expected with the use of CHOP in younger populations. It is likely that, with the use of supportive medications such as granulocyte colony-stimulating factor, many patients over the age of 60 years can be offered standard treatments at standard doses and experience similar toxicities as younger patients with NHL, but identifying patients over the age of 60 years (and especially 80 years) who are at risk for excessive toxicities remains a significant clinical challenge.

Selected Readings

Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature. 2000;403:503–511.

American Cancer Society. Cancer facts & figures, 2010. Atlanta, GA American Cancer Society; 2010.

Anderson JR, Armitage JO, Weisenburger DD. Epidemiology of the non-Hodgkin's lymphomas: distributions of the major subtypes differ by geographic locations. Non-Hodgkin's Lymphoma Classification Project. Ann Oncol.1998;9:717.

Armitage JO. How I treat patients with diffuse large B-cell lymphoma. Blood. 2007; 110:29–36.

Barrans S, Crouch S, Smith A, et al. Rearrangement of MYC is associated with poor prognosis in patients with diffuse large B-cell lymphoma treated in the era of rituximab. J Clin Oncol. 2010;26:1–6.

Cairo MS, Coiffier B, Reiter A, et al. Recommendations for the evaluation of risk and prophylaxis of tumour lysis syndrome (TLS) in adults and children with malignant diseases: an expert TLS panel consensus. Br J Haematol.2010;149:578–586.

Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol. 2007;25:579–586.

Coiffier B, Lepage E, Briere J, et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med. 2002;346:235–242.

Czuczman MS, Grillo-Lopez AJ, White CA, et al. Treatment of patients with low grade B-cell lymphoma with the combination of chimeric anti CD20 monoclonal antibody and CHOP. J Clin Oncol.1999;17:268–276.

Dave SS, Fu K, Wright GW, et al. Molecular diagnosis of Burkitt's lymphoma. N Engl J Med. 2006;354(23):2431–2442.

Dunleavy K, Little RF, Pittaluga S, et al. The role of tumor histogenesis, FDG-PET, and short-course EPOCH with dose-dense rituximab (SC-EPOCH-RR) in HIV-associated diffuse large B-cell lymphoma. Blood. 2010;115:3017–3024.

Federico M, Bellei M, Marcheselli L, et al. Follicular lymphoma International Prognostic Index 2: a new prognostic index for follicular lymphoma developed by the International Follicular Lymphoma Prognostic Factor Project. J Clin Oncol. 2009;27:4555–4562.

Feugier P, Van Hoof A, Sebban C, et al. Long-term results of the R-CHOP study in the treatment of elderly patients with diffuse large B-cell lymphoma: a study by the Groupe d'Etude des Lymphomes de l'Adulte. J Clin Oncol.2005;23(18):4117–4126.

Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin's lymphoma. N Engl J Med. 1993;328:1002–1006.

Friedberg JW, Taylor MD, Cerhan JR, et al. Follicular lymphoma in the United States: first report of the National LymphoCare Study. J Clin Oncol. 2009;27:1202–1208.

Geisler CH, Kolstad A, Laurell A, et al. Long-term progression-free survival of mantle cell lymphoma after intensive front-line immunochemotherapy with in vivo-purged stem cell rescue: a nonrandomized phase 2 multicenter study by the Nordic Lymphoma Group. Blood. 2008;112(7):2687–2693.

Gisselbrecht C, Glass B, Mounier N, et al. Salvage regimens with autologous transplantation for relapsed large B-cell lymphoma in the rituximab era. J Clin Oncol. 2010;28:4184–4190.

Hainsworth JD, Litchy S, Burris HA, et al. Rituximab as first-line and maintenance therapy for patients with indolent non-Hodgkin's lymphoma. J Clin Oncol. 2002;20:4261–4267.

Hans CP, Weisenburger DD, Greiner TC, et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood. 2004;103:275–282.

Hoster E, Dreyling M, Klapper W, et al. A new prognostic index (MIPI) for patients with advanced-stage mantle cell lymphoma. Blood. 2008;111:558–565.

Hummel M, Bentink S, Berger H, et al. A biologic definition of Burkitt's lymphoma from transcriptional and genomic profiling. N Engl J Med. 2006;354(23):2419–2430.

International Non-Hodgkin's Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin's lymphoma. N Engl J Med. 1993;329: 987–994.

Kaminski MS, Tuck M, Estes J, et al. 131-I-tositmomab therapy as initial treatment for follicular lymphoma. N Engl J Med. 2005;352:441–449.

Keegan T, McClure L, Foran J, et al. Improvements in survival after follicular lymphoma by race/ethnicity and socioeconomic status: a population-based study. J Clin Oncol. 2009;27:3044–3051.

Kwee TC, Kwee RM, Nievelstein RA. Imaging in staging of malignant lymphoma: a systematic review. Blood. 2008;111:504–516.

Lacasce A, Howard O, Lib S, et al. Modified magrath regimens for adults with Burkitt and Burkitt-like lymphomas: preserved efficacy with decreased toxicity. Leuk Lymphoma. 2004;45(4):761–767.

LaCasce AS, Kho ME, Friedberg JW, et al. Comparison of referring and final pathology for patients with non-Hodgkin's lymphoma in the National Comprehensive Cancer Network. J Clin Oncol. 2008;26:1–6.

Lenz G, Wright G, Dave SS, et al. Stromal gene signatures in large-B-cell lymphomas. N Engl J Med. 2008;359:2313–2323.

Levine AM, Seneviratne L, Espina BM, et al. Evolving characteristics of AIDS related lymphoma. Blood. 2000;96:4084–4090.

Link BK, Martin P, Kaminski MS, et al. Cyclophosphamide, vincristine, and prednisone followed by tositumomab and iodine-131–tositumomab in patients with untreated low-grade follicular lymphoma: eight-year follow-up of a multicenter phase II study. J Clin Oncol. 2010;28:3035–3041.

Marcus R, Imrie K, Belch A, et al. CVP chemotherapy plus rituximab compared with CVP as first-line treatment for advanced follicular lymphoma. Blood. 2005;105:1417–1423.

McMaster ML, Greer, JP, Greco FA, et al. Effective treatment of small non-cleaved cell lymphoma with high intensity, brief duration chemotherapy. J Clin Oncol. 1991;9:941–946.

Miller TP, Dahlberg S, Cassady JR, et al. Chemotherapy alone compared with chemotherapy plus radiotherapy for localized intermediate and high-grade non-Hodgkin's lymphoma. N Engl J Med.1998;339:21–26.

Pfreundschuh M, Trümper L, Österborg A, et al. CHOP-like chemotherapy plus rituximab versus CHOP-like chemotherapy alone in young patients with good-prognosis diffuse large-B-cell lymphoma: a randomised controlled trial by the MabThera International Trial (MInT) Group. Lancet Oncol. 2006;7:379–391.

Prince HM, Whittaker S, Hoppe RT. How I treat mycosis fungoides and Sezary syndrome. Blood. 2009;114:4337–4353.

Robinson KS, Williams ME, van der Jagt RH, et al. Phase II multicenter study of ben-damustine plus rituximab in patients with relapsed indolent B-cell and mantle cell non-Hodgkin's lymphoma. J Clin Oncol. 2008;26:4473–4479.

Romaguera JE, Fayad LE, Feng L, et al. Ten-year follow-up after intense chemoimmuno-therapy with Rituximab-HyperCVAD alternating with Rituximab-high dose metho-trexate/cytarabine (R-MA) and without stem cell transplantation in patients with untreated aggressive mantle cell lymphoma. Br J Haematol. 2010;150(2):200–208.

Roschewski M, Wilson WH. Biology and management of rare primary extranodal T-cell lymphomas. Oncology (Williston Park). 2010;24(1):94–100.

Rosenwald A, Wright G, Chan WC, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large B-cell lymphoma. N Engl J Med. 2002;346:1937–1947.

Rummel MJ, Niederle N, Maschmeyer G, et al. Bendamustine plus rituximab is superior in respect of progression free survival and CR rate when compared to CHOP plus rituximab as first-line treatment of patients with advanced follicular, indolent, and mantle cell lymphomas: final results of a randomized phase III study of the StiL (Study Group Indolent Lymphomas, Germany). Blood. 2009:114:405.

Savage KJ, Johnson NA, Ben-Neriah S, et al. MYC gene rearrangements are associated with a poor prognosis in diffuse large B-cell lymphoma patients treated with R-CHOP chemotherapy. Blood.2009;114:3533–3537.

Seam P, Juweid ME, Cheson BD. The Role of FDG-PET scans in patients with lymphoma. Blood. 2007;110:3507–3516.

Sebban C, Brice P, Delarue R, et al. Impact of rituximab and/or high-dose therapy with autotransplant at time of relapse in patients with follicular lymphoma: a GELA study. J Clin Oncol. 2008;26(21):3614–3620.

Sehn LH, Donaldson J, Chhanabhai M, et al. Introduction of combined CHOP plus rituximab therapy dramatically improved outcome of diffuse large B-cell lymphoma in British Columbia. J Clin Oncol.2005;23:5027–5033.

Solal-Ce'ligny P, Roy P, Colombat P, et al. Follicular lymphoma International Prognostic Index. Blood. 2004;104:1258–1265.

Swerdlow S, Campo E, Harris N, et al. (eds). WHO Classification of tumours of haemato-poeitic and lymphoid tissues. 4th ed. Geneva, Switzerland: WHO Press; 2008.

Thomas DA, Cortes J, O'Brien S, et al. Hyper-CVAD program in Burkitt's type adult acute lymphoblastic lymphoma. J Clin Oncol. 1999;17:2461–2470.

Valasquez W, Cabanillas F, Salvador P, et al. Effective salvage therapy for lymphoma with cisplatin in combination with high-dose Ara-C and dexamethasone (DHAP). Blood. 1988;71:117–122.

Valasquez WS, McLaughlin P, Tucker S, et al. ESHAP—an effective chemotherapy regiment in refractory and relapsing lymphoma: a 4-year follow up study. J Clin Oncol. 1994;12:1169–1176.

van Oers MHJ, Van Glabbeke M, Giurgea L, et al. Rituximab maintenance treatment of relapsed/resistant follicular non-Hodgkin's lymphoma: long-term outcome of the EORTC 20981 phase III randomized Intergroup study. J Clin Oncol. 2010;28:1–7.

Wang M, Burau K, Fang S, et al. Ethnic variations in diagnosis, treatment, socioeconomic status, and survival in a large population-based cohort of elderly patients with non-Hodgkin lymphoma. Cancer.2008;113:3231–3241.

Zinzani PL, Fanti S, Battista G, et al. Predictive role of positron emission tomography (PET) in the outcome of lymphoma patients. Br J Cancer. 2004;91:850–854.

Zinzani PL, Pulsoni A, Perrotti A, et al. Fludarabine plus mitoxantrone with and without rituximab versus CHOP with and without rituximab as frontline treatment for patients with follicular lymphoma./C& Oncol. 2004;22:2654–2661.