Hodgkin lymphoma (HL) is an uncommon lymphoproliferative malignancy with approximately 9,000 cases diagnosed annually in the United States. The average age at presentation is 32 years with a bimodal incidence curve: one peak occurs in early adulthood before age 25, and the second around age 55. HL is rare in children younger than 5 years of age and only 10% to 15% of cases are diagnosed in children and teenagers. HL is slightly more common in males, first-degree relatives of patients with HL, and in developed nations. Most patients present with disease limited to lymph nodes or to lymph nodes and the spleen. The bone marrow is involved in only 5% of cases. HL generally spreads in a contiguous fashion making the incorporation of radiation therapy (RT) feasible for many patients.

HL is largely a curable disease; up to 95% of cases of early stage HL, and 75% of advanced HL cases are cured with the initial therapy. Patients with advanced disease can be cured with combination chemotherapy, while those with limited disease can be cured either with limited combination chemotherapy and limited RT (combined modality treatment [CMT]), or with chemotherapy alone. While effective and potentially curative, extensive RT as a sole modality for early stage disease has largely been abandoned (see Section IV.B). Despite the gratifying cure rates of initial therapy, about 10% patients fail to respond to primary therapy, and another 20% to 30% of responding patients subsequently relapse after treatment. Salvage chemotherapy followed by autologous stem cell transplantation (ASCT) is the treatment of choice for relapsed or refractory HL, and is potentially curative. Nevertheless, the potential for cure should not lead clinicians and patients to lose sight of the fact that approximately 20% to 25% of patients with HL eventually die of the disease.

Long-term survivors of HL are at risk of developing a range of therapy-related side effects that include cardiac disease and second malignancies such as leukemia, lung cancer, and breast cancer.¹ While deaths from HL level off after 10 to 15 years, deaths from long-term complications continue to increase.²,³,⁴ Over the last 40 years, the search for initial therapy with fewer long-term side effects has been a major research focus, leading, for example, to the abandonment of RT as a sole treatment modality for early stage HL. Ongoing studies are also evaluating response-adapted treatment in an attempt to limit cumulative doses of chemotherapy.

For most cancers, disease-free survival (DFS) is a valuable surrogate marker for overall survival (OS), and thus evaluating DFS is a useful method for choosing optimal initial therapy. However, for HL, the success of salvage therapy means that the treatment options that are associated with superior DFS may not necessarily produce superior OS when the results of salvage therapy are considered, and this makes the selection of initial therapy somewhat more nuanced. In fact, because radiation and chemotherapy have significant and even potentially fatal long-term consequences, DFS may overestimate the value of a specific therapy. Therefore, for each stage of HL, more than one rational therapeutic option may exist.

The presence of Reed-Sternberg (RS) cells (or variants) in a mixed inflammatory background is a key feature of the histopathology of HL. The RS cells of classical HL are of B-cell origin, stain for CD15 and CD30, and are negative for CD45 and CD20. The lack of expression of functional surface immunoglobulin is a hallmark of classical HL. Epstein-Barr virus (EBV) is found in a significant proportion of RS cells (30% to 50% in Western populations and 90% to 100% in developing countries), lending credence to the role of EBV in the pathogenesis of HL via viral mimicry of cellular proliferation and antiapoptotic proteins.

The current World Health Organization classification divides HL into two major groups: Classical HL and Nodular Lymphocyte-predominant HL (NLPHL). Classical HL includes four subtypes: nodular sclerosis HL (approximately 70% of cases), mixed cellularity HL (approximately 20%), lymphocyte-rich HL (less than 5%), and lymphocyte depletion HL (less than 5%). NLPHL (5% of all HL) is a B-cell neoplasm characterized by variant RS cells (“L and H cells” or “popcorn cells”), positive for CD20 and negative for CD30 and CD15. In immunophenotype behavior, NLPHL bears similarities to low-grade non-HL. While advanced NLPHL is generally treated like classical HL, for some situations, a low-grade lymphoma-like approach may be used; for example, radiation of a single site of disease and rituximab for more extensive disease. NLPHL and its management are not the focus of this review.

A. Diagnosis. The diagnosis of HL requires excisional biopsy of an involved node and review of the material by a hematopathologist.⁵ Lymph node biopsy is recommended for any patient with lymphadenopathy greater than 1 cm in diameter and persisting for more than 4 weeks. Lymphoma, including HL, may be suspected when the nodes are freely movable and rubbery rather than stony hard. Other clinical features may include B symptoms such as fever, night sweats, and weight loss, and symptoms such as pruritus or alcohol intolerance. However, these clinical features are not specific for HL or for lymphoma in general. When the diagnosis of HL
is made in a patient presenting at an extranodal site or at a nodal site below the diaphragm, the diagnosis should be subjected to greater than usual scrutiny.

B. Staging. Accurate staging of HL is essential for determining an optimal therapy plan. It also provides a baseline to assess the response to treatment.

1. Cotswold staging system. The Cotswold modification of the Ann Arbor Staging System (Table 22.1)⁶ is used for patients with HL. Patients are placed in one of four stages on the basis of anatomic extent of disease and are further classified as to the absence, “A,” or presence, “B,” of systemic symptoms (see below). The subscript E (e.g., II_E) may be used to denote involvement of an extralymphatic site, primarily or, more commonly, to denote direct extension into an organ, such as a large mediastinal mass extending into the lung. Stage III HL is subdivided into two substages, stages III₁ and III₂, on the basis of the extent of intra-abdominal disease. However, as current treatment recommendations are the same for both substages, this distinction is of little clinical relevance.

2. Staging procedures. Before the advent of modern radiographic and nuclear medicine techniques, clinicians made use of the knowledge that HL tends to spread in a contiguous manner, and elegant and detailed descriptions of patterns of disease were made. For instance, it was recognized that because the thoracic duct makes the left supraclavicular area and the abdomen contiguous sites, abdominal disease is found in 40% of patients with
left supraclavicular presentations and in only 8% of patients with right supraclavicular presentations. Although these associations were most useful before modern computed tomography (CT) and positron emission tomography (PET) were available, they are occasionally useful in the interpretation of inconclusive imaging. Procedures used in the staging of HL are as follows:

a. History and physical. As with any patient, the staging of the patient with HL begins with a history and a physical examination. Special attention should be given to symptoms such as bone pain that might signal a specific extranodal site of disease. Unexplained fever >38, drenching sweats, especially at night, and weight loss greater than 10% of body weight over 6 months or less are classified as “B symptoms.” Pruritus is not considered a B symptom, but is not uncommon. Fever in HL can have any pattern. The pattern of days of high fever separated by days without fever, so-called “Pel-Ebstein fever,” has been associated with HL for over a century but is now quite rare as the diagnosis of HL is usually made and effective therapy is initiated earlier in the course of disease. Pain in the sites of HL involvement with alcohol ingestion is a rare finding but may indicate otherwise unsuspected visceral sites of involvement. On physical examination, attention must be paid to all lymph node regions and the spleen. Splenomegaly is seen at presentation in approximately 10% of patients with HL but does not necessarily correlate with pathologic involvement proven by splenectomy or with PET imaging. Performance status should be assessed.

b. Laboratory tests. Complete blood counts (CBCs), erythrocyte sedimentation rate (ESR), serum albumin, and tests of liver and kidney function should be obtained. Hepatic enzymes may be elevated “nonspecifically” in patients with HL and do not necessarily indicate hepatic involvement by HL. A pregnancy test should be obtained for appropriate patients. HIV antibody testing is considered for some patients, as HL has occasionally been the initial manifestation of HIV infection, and patients with HIV-associated HL may have atypical presentations.

c. Chest x-ray and CT scans. CT scans of the neck, chest, abdomen, and pelvis with intravenous (IV) contrast are routinely obtained to identify sites of disease. A posteroanterior (PA) chest x-ray may be used to ascertain whether a mediastinal mass meets the criteria for bulk (“X” designation in the staging system).

d. PET with CT attenuation correction (PET/CT). PET/CT scans should be obtained in all patients. Note that the CT portion of this scan is used for attenuation correction and localization and is of lower quality than the diagnostic and contrast-enhanced CTs recommended above. The use of PET/CT has been shown to yield a more accurate stage than CTs alone.

e. Bone marrow biopsy. The test is rarely positive except in patients who are found to have at least stage III disease by other tests and/or cytopenias. Because stages III and IV are treated identically (see below), and because cytopenias are rare in stage II, marrow biopsy can often be avoided. In addition, PET can be useful in inferring marrow status, with focal marrow positivity often correlating with a positive marrow. It should be noted that a homogeneously positive marrow on PET does not correlate with marrow involvement. In general, the clinician should consider whether pathologic assessment of the marrow will change the treatment approach to decide whether or not to do a marrow biopsy.

C. Organ function testing. Baseline pulmonary function tests with diffusing capacity of carbon monoxide (DLCO) and an assessment of cardiac ejection fraction should be performed in most patients as a baseline, as doxorubicin and bleomycin may cause cardiac and pulmonary toxicity. Fertility preservation should be considered for appropriate patients and referrals to fertility specialists should be made.

D. Prognostic score for advanced HL (International Prognostic Score [IPS]). In 1998, Hasenclever and colleagues created a prognostic model for advanced HL on the basis of a multivariate analysis of patients.⁷ Seven factors were identified as having prognostic value: serum albumin <4 g/dL, hemoglobin <10.5 g/dL, male sex, stage IV disease, age ≥45 years, white blood cell count ≥15,000/cu mm, and lymphocyte count either <600/cu mm or <8% of the white blood cell count. In the paper presenting the model, the prognostic score correlated with both freedom from progression and the OS rate. However, the utility of the IPS may be limited by the fact that most patients had a score of 0 to 3, with only 12% of patients having a score of 4, and only 7% of patients having a score of 5 to 7.

A. General considerations. The goal of HL therapy may be considered twofold: to deliver a therapy with a high cure rate, and to minimize long-term toxicity. The majority of patients can expect to be cured, most with the initial therapy and another significant percentage with salvage therapy. The choice of therapy depends on anatomic stage and, in stages I and II, the presence or absence of unfavorable features. Unfavorable features are age >50 years, the presence of B symptoms or elevated sedimentation rate, large mediastinal mass, and a larger number of nodal sites (>2 or 3).

B. Radiation therapy. RT is remarkably effective for HL, and potentially curative RT was available long before curative chemotherapy was developed. Historically, this led to a bias toward using RT alone or as part of CMT when anatomically feasible. Today, however, there is clear evidence of the long-term side effects of
RT (particularly in large fields or at high dose), including breast cancer, lung cancer, hypothyroidism, coronary artery disease, cardiomyopathy, cardiac valvular disease, and musculoskeletal atrophy.¹,³,⁸,⁹ The cumulative risk of complications may be as high as 15% at 15 years following treatment. In particular, the risk of breast cancer is dramatically elevated in women who have chest irradiation before the age of 30.⁸,¹⁰,¹¹ For these reasons, there is movement toward limiting the dose and fields or toward eliminating RT entirely as part of initial therapy.

C. Chemotherapy. Chemotherapy, alone or in CMT, is now standard therapy for all stages of HL treatment. (Table 22.2)

1. ABVD. The most commonly used chemotherapy for HL is ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine). This
regimen is given every 2 weeks, with two treatments making up one “cycle.” ABVD’s superiority was established by a series of U.S. cooperative group studies in the 1980s and 1990s, showing its superior efficacy and/or lower toxicity than other then-current options.¹² In advanced-stage patients, the failure-free survival rate is as high as 75% to 88%. ABVD is generally well tolerated; however, it does have acute and long-term side effects. Important acute toxicities are neutropenia (34%), nausea/vomiting (13%), and alopecia (31%). Vinca alkaloids may cause peripheral neuropathy and ileus. Important long-term toxicities include cardiotoxicity due to doxorubicin and pulmonary toxicity due to bleomycin. It is recognized that maintenance of dose intensity is important and many experts do not delay the treatment for uncomplicated neutropenia and do not use growth factor support (which has been associated with increased pulmonary toxicity of the regimen).¹³,¹⁴,¹⁵ Although ABVD is likely modestly inferior in complete response (CR) rate and relapse-free survival (RFS) rate to BEACOPP (see below), patients who do not respond to or relapse after ABVD can often be salvaged with alternative chemotherapy and stem cell transplant (see below).¹⁶