BL is the most common NHL of childhood, adolescents, and young adults (CAYA) [23, 24]. The incidence, however, decreases with age, being highest in children <15 years and relatively uncommon in adults >30 years of age. BL is characterized by very distinct epidemiologic differences. Sub-Saharan Africa has an especially unique epidemiology of childhood cancer where endemic BL is the most common pediatric malignancy overall, representing up to one-third to one-half of all pediatric oncologic diagnoses. Sporadic BL, occurring everywhere else in the world, accounts for approximately 40 % of NHL in CAYA outside of sub-Saharan Africa.

BL is characterized on one hand by a high prevalence (around 70 %) of advanced-stage III–IV disease presentation but on the other hand excellent clinical outcomes with event-free survival (EFS) of 85–90 % overall. Most commonly, BL presents with intra-abdominal masses; however masses of the head and neck also occur. The tumor burden in BL is often exceptionally high, with more than half of patients presenting with a significantly elevated lactate dehydrogenase (LDH) level >2 times the upper limit of normal [25]. With this in mind, it is no surprise that BL carries one of the highest risks for the complication of TLS and close monitoring of patients prior to and during induction therapy is vital [18, 19].

The defining characteristic of BL is a translocation of the C–MYC oncogene on chromosome 8 with the immunoglobulin genes on chromosome 14, 22, or 2. Histology classically reveals intermediate-sized cells with round nuclei and scant cytoplasm with lipid vacuoles. BL has one of the highest proliferation rates of any malignancy and usually reveals numerous mitotic figures and apoptotic bodies that can be seen engulfed in scattered macrophages portraying the characteristic “starry-sky” appearance on low power histology [25]. Endemic BL occurs in the holoendemic malaria belt of sub-Saharan Africa and is virtually always associated with Epstein-Barr virus (EBV) infection. On the other hand, sporadic BL is associated with EBV in about 30 % of cases [26].

The immunophenotypic signature of BL is characterized by expression of mature B-cell antigens CD20 and CD19, as well as CD10 and BCL6, which are associated with germinal center derivation. BL expresses the proliferation antigen Ki-67 at rates often >99 % and rarely expresses BCL2 [27].

Landmark gene expression profiling (GEP) studies have recently established an extensive biologic definition of B-NHL, producing a molecular definition of BL that extended the spectrum of the WHO criteria [28]. Additionally, Dave/Staudt et al. established that C–MYC and its target genes as well as a subgroup of germinal center B-cell genes were more highly expressed in BL versus DLBCL [29]. GEP studies in pediatric B-NHL have highlighted similar findings [30].

The major determinants for risk stratification in BL of CAYA are rooted in the original Murphy stage of the clinical presentation. The previous FAB96 study demonstrated the clinical variables associated with a significantly inferior EFS in pediatric mature B-NHL; those include advanced-stage disease, elevated LDH, primary mediastinal involvement, and combined BM and CNS disease [31]. Additionally, one of the most important prognostic indicators to guide treatment decision is the patient’s response to therapy. On the FAB/LMB protocol, failure to achieve at least 20 % reduction in disease burden with the first week of reduction phase chemotherapy is a poor prognostic indicator and requires intensification of the treatment regimen [32]. Additionally, failure to achieve complete remission (CR) by the completion of the first consolidation cycle is also associated with poor long-term survival and is another indication to intensify therapy. Universal guidelines to evaluate treatment response via fludeoxyglucose (FDG)-PET for NHL in CAYA have been published and will serve as the benchmark for further delineating the role of FDG-PET through prospective clinical trials [8, 33].

The prognostic role of BL biology with regard to cytogenetic and minimal-residual disease (MRD) findings is currently being explored. Although specific cytogenetic findings such as deletion 13q, gain of 7q, and complex cytogenetics may be associated with a higher risk for treatment failure, further studies are needed to determine if these patients require and/or benefit from an intensification of the therapeutic approach [34, 35]. Preliminary data on the presence of MRD in B-cell NHL suggested that the presence of MRD on end-of-therapy (EOT) specimens was associated with disease recurrence [36]. Children/adolescents with intermediate-risk B-NHL, treated with FAB/LMB96 modified chemotherapy plus rituximab, had blood and bone marrow specimens from end of induction (EOI) and EOT assessed for MRD [36]. While recurrence rates were similar between the EOI MRD-positive and MRD-negative patients (p = 0.40), EOT MRD results demonstrated one relapse in the MRD-positive group and no recurrences in the MRD-negative group (p = 0.077). More recent data compared the presence of MRD from EOI and end of consolidation, demonstrating that the presence of MRD did not predict relapse and that subsequent therapy actually appeared to eliminate MRD [37]. Further prospective investigations will determine the prognostic role of MRD in BL in CAYA.

Long-term curative outcomes in pediatric BL have dramatically improved over the past three decades, with EFS rates essentially doubling from the late 1970s to the contemporary era (Fig. 4.6) [38]. Since the turn of the century, clinical trials have focused on establishing risk-stratified therapy to diminish acute and long-term toxicities for patients with favorable prognosis and to intensify regimens for those with higher risk for treatment failure [32, 39–41]. International collaboration in large-scale clinical trials has resulted in the modern day FAB/LMB chemotherapy backbone consisting of cyclophosphamide, high-dose methotrexate (HD MTX), cytarabine, vincristine, doxorubicin, and corticosteroids.

FAB/LMB chemotherapy protocols were established based upon three risk-stratification groups. Low-risk group A patients with fully resected stage I or abdominal stage II disease are treated with two cycles of systemic chemotherapy (without intrathecal), with 4-year EFS rates of 98.3 % [42]. FAB group B intermediate-risk regimens demonstrated equivalent EFS rates of 91 % with reduced total doses of cyclophosphamide and deletion of maintenance cycles [32]. Even patients with group C high-risk disease—defined as having central nervous system (CNS) involvement and/or >25 % bone marrow involvement—achieve long-term EFS rates of 79 % with the FAB/LMB backbone chemotherapy bolstered by the addition of HD MTX (8 g/m² for CNS-positive patients) during induction and cytarabine (12,250 mg/m²) plus etoposide for two cycles of intensification followed by four cycles of maintenance chemotherapy [39]. The Berlin-Frankfurt-Munster (BFM) collaborative group uses a very similar risk-stratification system with four groups instead of three; FAB group B is essentially broken down into two groups in the BFM approach. Similarly, their treatment regimen is nearly the same, with slight variations in chemotherapy dosing, number of cycles, and the addition of ifosfamide in the BFM approach accounting for the nuanced differences (see Table 4.9) [25].

Recent large-scale prospective data from the FAB/LMB96 study demonstrated that adolescent age was not associated with a risk of treatment failure in comparison to children with BL [31]. Historically though, there had previously been a general consensus that adolescents with mature B-cell lymphomas experienced inferior outcomes based upon data from the United States, Germany, and France [38, 43, 44]. However, this data may be clouded by multiple factors including derivation from retrospective analyses spanning decades (CCG and BFM data) and analysis of smaller numbers of patients (SFOP data) and that all three studies occurred prior to the discovery of primary mediastinal B-cell lymphoma as a separate B-NHL diagnostic entity requiring different therapy that tends to occur in adolescents and young adults. It is currently accepted that adolescent age is not prognostically significant in BL.

While adults with BL have always experienced much lower curative rates than pediatric patients, this has been attributed to the intrinsically different treatment approaches utilized. More recently, adult BL clinical trials have incorporated pediatric-style chemotherapy regimens with encouraging success [45]. In general, it is well established that pediatric BL protocols achieve significantly higher curative outcomes for the following reasons: (1) dose intensity is incorporated by utilizing 21 rather than 28-day cycles, (2) HD MTX is uniformly given, and (3) group C patients receive both HD MTX and high-dose cytarabine. Historically, when treated on the same protocol, children and young adults achieve similar outcomes [46, 47]. Figure 4.7 demonstrates survival curves for children and adolescents treated on the FAB/LMB pediatric regimen [31, 45]. Table 4.10 shows a comparison of the most recent survival data for CAYA treated on large-scale prospective clinical trials utilizing FAB/LMB backbone chemotherapy regimens.

The vast majority of BL cases in CAYA express the CD20 antigen, begging the question of whether adding anti-CD20-targeted immunotherapy to standard chemotherapy will improve outcomes [48]. Adding rituximab to BL chemotherapy has been shown to be safe and effective in the largest prospective trial for adults with BL, with particularly excellent outcomes for AYA [45]. The Children’s Oncology Group (COG) recently published results on a pilot study for mature B-NHL integrating rituximab into the standard FAB/LMB backbone chemotherapy. The study showed that it was safe to combine rituximab with FAB/LMB chemotherapy, and this chemoimmunotherapeutic approach yielded a 95 % probability of EFS at 3 years for 45 patients with advanced-stage III/IV group B intermediate-risk disease [40]. Meanwhile, rituximab combined with group C regimen produced a 3-year EFS of 90 % in 40 patients [41]. The BFM also evaluated the efficacy of rituximab with 37 of 87 newly diagnosed patients with B-NHL (42 %) demonstrating a significant response to a 1-week window of a single dose of rituximab [49]. Current large-scale clinical trials in pediatric BL are focusing on evaluating the combination of rituximab plus FAB/LMB chemoimmunotherapy regimen in a larger cohort of patients [50].

Although the prognosis for patients with newly diagnosed BL is excellent, CAYA with relapsed and/or refractory BL experience dismal outcomes. With reported long-term survival rates ranging from 12 % to 28 %, it is well established that salvaging those patients with relapsed/refractory BL is inordinately difficult (Fig. 4.8) [38, 39, 51]. A reinduction regimen combining ifosfamide, carboplatin, and etoposide (ICE) with rituximab for children with relapsed/refractory B-NHL produced encouraging results; however maintaining long-term remission can only be established utilizing hematopoietic stem cell transplantation (HSCT) [52, 53]. While the data for HSCT in relapsed/refractory BL is not robust, it has been demonstrated repeatedly that long-term survival utilizing high-dose chemotherapy followed by autologous HSCT is <30 % [54, 55]. The combination of autologous HSCT followed in tandem by reduced-intensity allogeneic HSCT to maintain long-term remission has shown promise in small numbers of CAYA with relapsed/refractory BL; five of eight patients achieved long-term CR (1.9–8.8 years) utilizing this approach [56]. There seems to be a potential graft-versus-lymphoma (GVL) effect in BL and it is vital to further determine the potential role of allogeneic HSCT in CAYA with relapsed/refractory BL.

Certainly long-term survival in patients with relapsed or refractory BL in the modern era is meager. In the absence of established therapies with curative potential, novel approaches to therapy are desperately needed. Therapeutic strategies that are currently under investigation include agents targeting specific molecular pathways and novel cellular and humoral immunotherapies. Promising targeted cell-signaling pathway inhibitors include ibrutinib, a Bruton’s tyrosine kinase (BTK) inhibitor, and idelalisib, a phosphoinositide-3-kinase inhibitor [57, 58]. Chimeric antigen receptor-based cellular immunotherapies are under evaluation in clinical trials for CAYA based upon exciting results from preclinical data as well as success in treating acute lymphoblastic leukemia (ALL) [59, 60]. Finally, novel humoral immunotherapies targeting both CD19 (blinatumomab and SAR3419) and CD20 (obinutuzumab and ofatumumab) are also being evaluated in clinical trials. As a novel anti-CD20 monoclonal antibody that has demonstrated superior results in comparison with rituximab in both preclinical and clinical trials, obinutuzumab is currently being evaluated in a clinical trial for CAYA with relapsed/refractory mature B-NHL in combination with the reinduction ICE chemotherapy platform (NCT02393157) [61–65]. With a number of novel therapeutic agents on the horizon, there is hope to not only improve the chances for salvaging patient with relapsed/refractory disease but to also improve overall outcomes by incorporating targeted therapies with chemotherapy in up-front regimens.

Depending on the country in pediatric practice, DLBCL is included either in studies designed for BL (LMB and BFM studies) [44, 66] or in studies designed for large cell NHL in general (Pediatric Oncology Group [POG] studies) [67, 68]. In the LMB89 study, DLBCL represented 10 % of all registered patients. As with BL, they are treated according to initial resection and stage. The EFS is similar to that of BL (Fig. 4.9), but it should be noted that the proportion of patients with advanced stages is lower than in BL. However, by stage, EFS is not significantly different [69]. In the BFM90 study, the EFS of DLBCL is also similar to that of BL. The criticism of such an approach is that too much CNS-directed therapy is given in DLBCL, in which the risk of CNS disease is lower than in BL.