Non-Hodgkin lymphoma

Introduction

Childhood non-Hodgkin lymphoma (NHL) is distinguished from adult NHL by differing frequencies of histopathologic types and by the greater frequency of extranodal presentations. With current combination chemotherapy regimens, survival is generally excellent (85 to over 90%) for all patients, including those with disseminated disease, bone marrow involvement, central nervous system (CNS) involvement, and high serum lactate dehydrogenase (LDH).

With improved survival in all subtypes and stages, recent efforts have focused on maintaining excellent event-free survival (EFS) with reduced late toxicity by incorporating more specific targeted therapies for patients with less favorable subgroups of NHL and identifying new prognostic factors.

Incidence and epidemiology

Pathologic classification

The World Health Organization (WHO) classification for lymphoid neoplasms from the International Lymphoma Study Group was revised in 2016 and incorporates histology as well as immunohistochemistry, gene expression profiling, cytogenetic, and molecular and clinical features (the revisions are well summarized in the study of Swerdlow). Differences in pediatric and adult disease become apparent with recognition of new subtypes of lymphoma. This may explain differences in biology that might direct future therapy. Table 21.1 summarizes the most common Pediatric NHL subtypes with specific cytogenetic and genomic markers. Pediatric NHL is mostly (more than 95%) of high grade and includes the following four major subtypes:

• 1.

  B- and T-lymphoblastic lymphoma (LL)

• 2.

  BL

• 3.

  Diffuse large B-cell lymphoma (DLBCL)

• 4.

  ALCL

Many of these high-grade lymphomas disseminate noncontiguously, evolve into a leukemic phase, and involve the CNS. The more common low-grade lymphomas seen in adults, such as follicular and marginal zone, are rare in children. T- and B-LL comprise about 20% of NHL in childhood; T cell is the more common type. Mature B-cell lymphoma includes both BL (19% of cases) and DLBCL and primary mediastinal large B-cell lymphoma (PMBL). Together the latter two comprise approximately 22% of NHL in childhood. Of note, the majority of pediatric DLBCL fall into the germinal center B-cell-like subgroup. In the WHO classification, PMBL is considered distinct from other DLBCL based upon its unique clinical, histologic, and molecular features. Other mature B-cell lymphomas, including pediatric marginal zone lymphoma, pediatric-type follicular lymphoma (PFL), and mucosa-associated lymphoid tissue (MALT) lymphoma as well as rare cutaneous lymphomas are also recognized as distinct entities that rarely occur in childhood. Included in the most recent WHO classification, a subtype of follicular lymphoma, referred to as PFL, has been described and mostly presents as Stage I or II disease. Considered a mature T-cell lymphoma, ALCL accounts for approximately 10% of NHL in childhood. More than 90% are anaplastic lymphoma kinase (ALK) positive in the pediatric age group.

Clinical features

The clinical manifestations of childhood NHL depend primarily on pathological subtype and sites of involvement. Tumors that grow rapidly can cause symptoms based on size and location. Approximately 70% of children present with advanced-stage disease, including extranodal disease with gastrointestinal, bone marrow, and CNS involvement.

Approximately 25% of children with NHL have an anterior mediastinal mass (usually T-LL or PMBL) and present with wheezing, orthopnea, and cough progressing to dyspnea. The majority of these patients are adolescents, and their presentation may manifest as superior vena cava (SVC) syndrome, an oncological emergency discussed in Chapter 31 , Management of Oncologic Emergencies. Patients with large anterior mediastinal masses are at major risk of cardiac or respiratory arrest when laid flat during general anesthesia or deep sedation. A careful workup, including a chest-computed tomography (CT) scan with airway measurements, is essential before attempting any procedures. The least invasive procedure (e.g., biopsy of a peripheral lymph node) should be carried out. If these procedures are not successful in providing a diagnosis, then a CT-guided needle biopsy of the mediastinal mass should be considered. In some clinical situations (e.g., orthopnea or significant airway narrowing), preoperative or preprocedure steroids should be considered for up to 48 hours. The use of steroids has largely replaced localized irradiation in this setting.

Primary gastrointestinal involvement occurs in about 30% (usually Burkitt histology), commonly presenting as an abdominal mass with ascites, an “acute abdomen” from an intussusception, or rarely a malnutrition syndrome with colitis symptoms. The majority of children with BL presenting with an ileal–cecal intussusception have limited gastrointestinal involvement that is amenable to complete surgical resection (Murphy Stage II or group A) (please see Table 21.2 ).

In 20–30% of children the head and neck, including Waldeyer’s ring or cervical lymph nodes, are the sites of origin. The remainder of patients have miscellaneous primary sites, including bone, breast, skin, epidural space, or noncervical lymph nodes. Involvement of the bone marrow occurs at diagnosis in 10–30% of patients with BL and LL. Overt CNS involvement at diagnosis is not common but is mostly seen in children with advanced-stage BL and LL. Children who develop BL in endemic areas of the world often have a mass in the head or neck region (especially jaw) in contrast to the abdominal presentation typical of nonendemic BL. Both endemic and sporadic cases of BL have the same chromosomal translocations involving one of the loci encoding immunoglobulin heavy or light chains and c-myc oncogene. The exact role of EBV in the pathogenesis of BL and other malignancies is unknown.

Diagnosis

Tissue is required for diagnosis. For some subtypes of NHL, architecture is critical to diagnosis, such that a fine-needle aspirate or core biopsy may not be sufficient to establish a diagnosis. In some cases, however, diagnostic samples may also be obtained from bone marrow, cerebrospinal fluid (CSF), or pleural/paracentesis fluid. In all cases, samples should be tested by flow cytometry for immunophenotype as well as cytogenetics and molecular assays. Gene expression profiling as well as assays to identify targetable mutations may be important for diagnostic purposes and have an impact on therapy and prognosis.

Recommended laboratory and radiologic testing includes:

• 1.

  Complete blood count with differential.

• 2.

  Electrolytes including uric acid, LDH, calcium, phosphorus and creatinine.

• 3.

  Liver function tests.

• 4.

  Bone marrow aspirate and biopsy*.

• 5.

  Lumbar puncture with CSF cytology, cell count, glucose, protein.

(* these studies can be omitted in subsets of pediatric NHL such as DLCBL and PMBL in which bone amrrow and/or CNS involvement are rare).

Chest radiograph and neck, chest, abdominal, and pelvic CT scans are recommended to define the extent of disease in most instances. Positron emission tomography (PET) scan, or combined PET/CT, may also be performed to help in initial staging as well as to measure tumor response during the course of therapy. Of note, in contrast to HL, the role of PET in initial staging and treatment response has not yet been well established (see Chapter 20 : Hodgkin Lymphoma).

The possibility of inherited or acquired predisposition to NHL should be considered. Several primary immunodeficiencies have been identified as associated with increased incidence of NHL in the pediatric age group. These include ataxia-telangiectasia, common variable immunodeficiency, severe combined immunodeficiency, and Wiskott–Aldrich Syndrome. As well, a growing number of inherited cancer predispositions are being identified. NHL is also associated with some infections, notably EBV and HIV. Finally, some medications, including cytotoxic chemotherapy and immunomodulatory agents, such as infliximab have been associated with increased risk for secondary NHL. In some patients, evaluation for specific infection, including HIV testing, immune function, or inherited cancer predisposition, may be appropriate.

Staging

Staging of NHL requires an investigation to determine the clinical extent of the disease, the degree of organ impairment, and biochemical disturbance present. Correct staging is critically important at the time of diagnosis in NHL due to the high number of patients who present with advanced disease. Prior staging systems for NHL entailed a modification of the Ann Arbor staging system for HL while taking into consideration the common presentations of childhood NHL such as extranodal involvement and metastatic spread to the bone marrow and CNS. The Murphy Staging system has been widely accepted (see Table 21.2 ). A French, American, and British (FAB) childhood NHL risk stratification, used in the FAB Lymphoma Malignant B type (LMB)-96 (LMB) protocol, has also been developed for B-LL, BL, and DLBCL ( Table 21.3 ) and incorporates standard staging along with the impact of dissemination to either the bone marrow or the CNS on prognosis.