Treatment and prognosis of neuroblastoma

INTRODUCTION — 

The term neuroblastoma is commonly used to refer to a spectrum of neuroblastic tumors (including neuroblastomas, ganglioneuroblastomas, and ganglioneuromas) that arise from primitive sympathetic ganglion cells, and like paraganglioma and pheochromocytomas, have the capacity to synthesize and secrete catecholamines. (See "Clinical presentation and diagnosis of pheochromocytoma" and "Epidemiology, clinical presentation, and diagnosis of paragangliomas".)

Neuroblastomas, which account for 97 percent of all neuroblastic tumors, are clinically heterogeneous, varying in location, histopathologic appearance, and biologic characteristics [1]. They are most remarkable for their broad spectrum of clinical behavior, which can range from spontaneous regression to maturation to a benign ganglioneuroma, or aggressive disease with metastatic dissemination leading to death [2]. Clinical diversity correlates closely with numerous clinical and biologic factors, although its molecular basis remains largely unknown. For example, most infants with disseminated disease have a favorable outcome after treatment with chemotherapy and surgery, while approximately half of children over the age of 18 months who have advanced neuroblastoma die from progressive disease despite intensive multimodality therapy.

The treatment and prognosis of neuroblastoma will be reviewed here and is meant to be an overview for the general oncologist. Due to the rarity of this disease, patients should be managed in a setting where appropriate expertise in the treatment of neuroblastoma is available.

The epidemiology, clinical presentation, and diagnosis of neuroblastoma are presented separately, as is a discussion of neuroblastomas arising in the olfactory epithelium. (See "Epidemiology, pathogenesis, and pathology of neuroblastoma" and "Clinical presentation, diagnosis, and staging evaluation of neuroblastoma" and "Olfactory neuroblastoma (esthesioneuroblastoma)".)

STAGING SYSTEM — 

There is international consensus to use the International Neuroblastoma Risk Group Staging System (INRGSS) to describe the extent of disease for neuroblastoma. This topic is organized using a risk-stratified approach that incorporates the INRGSS (table 1 and figure 1), along with molecular, pathologic, and other clinical characteristics. However, certain sections in this topic define patient risk groups or include studies that use the International Neuroblastoma Staging System (INSS) (table 2), which has been noted where relevant.

Further details of the available staging systems are as follows:

●International Neuroblastoma Staging System – The INSS was adopted in the 1990s and was the first uniform staging system (table 2) [3]. Tumor stage in this system is defined according to resectability and spread to lymph nodes or distant sites.

●INRGSS – In 2009, a revised staging system, the INRGSS, was developed, which incorporated pretreatment imaging parameters rather than findings at resection (table 1) [4,5]. Data suggest that this revised staging system offers improved insights into which patients require intensive treatment (figure 1) [6]. All international cooperative groups, including the Children's Oncology Group (COG), validate and incorporate this staging system into their prospective research.

PROGNOSTIC FACTORS — 

Neuroblastomas are diverse in their clinical behavior. Certain factors influence the biologic behavior of these tumors and are helpful in predicting outcome; some are patient-related (eg, age at the time of diagnosis), while the majority are tumor-related (disease stage, tumor histology, molecular and cytogenetic features).

Stage — The extent of metastatic spread at presentation is the most important factor in determining outcome for patients with neuroblastoma [4,7-9]. Although regional spread to lymph nodes attached or adjacent to the primary tumor does not universally affect outcome, distant metastatic disease (eg, bone marrow involvement) confers a much worse prognosis. (See 'Staging system' above.)

Disease stage at diagnosis is to some extent influenced by age; a greater proportion of patients diagnosed after one year of age have regional or metastatic spread of disease compared with those who present earlier (80 versus 41 percent) [2].

Age — The age at diagnosis is an important prognostic factor in children with neuroblastoma [10-13]. However, predicting outcomes based on age can be complicated by other factors that also impact response to therapy, such as multiple genetic factors and heterogenous pathophysiology [14]. In general, the younger the age at diagnosis, the better the survival rate (with the exception of newborns) [13,15,16]. (See 'Newborns' below.)

Survival based on age — Based on data from the Surveillance, Epidemiology and End Results (SEER) database, overall survival (OS) based on age are as follows:

●The five-year survival rate for all children with neuroblastoma is approximately 80 percent [17].

•Young children beyond the newborn stage who are less than 18 months of age appear to have the best prognosis [13,18]. This age has been incorporated into the treatment regimens as well as the proposed risk group classification schema [4,6]. The significantly better outcome of disseminated disease in infants [19] compared with other age groups is reflected in the special MS (4S) category of disease stage (table 2), which applies only to children less than 18 months of age. (See 'Stage MS (4S) disease' below.)

•Children greater than 18 months of age with metastatic spread are considered to have high-risk disease, with a five-year event-free survival (EFS) of approximately 49 percent and a five-year OS of 59 percent [6].

•For patients diagnosed between ages 5 and 10 years, outcome is inferior with an OS of approximately 49 percent. For patients diagnosed over age 10 years, OS is 46 percent.

•Although adolescent neuroblastoma appears to have a more indolent phenotype, these patients experience a higher number of late relapses and deaths [20]. Even low-stage, histologically favorable, or less aggressive-appearing neuroblastomas fare worse as children get older [13,21]. (See 'Histology' below.)

Newborns — Neonates (ie, newborns younger than two months of age) who have International Neuroblastoma Risk Group Staging System (INRGSS) stage MS (ie, International Neuroblastoma Staging System [INSS] stage 4S) neuroblastoma may present with aggressive disease and are an exception to the general rule that younger age is associated with better outcome. (See 'Survival based on age' above.)

Among this small subset of patients, neuroblastoma metastases in the liver can grow rapidly, with high resulting morbidity and mortality. In addition to liver dysfunction with coagulopathy, these metastases can cause hepatomegaly with subsequent pulmonary compromise, abdominal compartment syndrome, and/or kidney failure [13,22,23]. As an example, the Children's Oncology Group (COG) study ANBL0531 included 49 infants with stage MS (4S) neuroblastoma who required treatment for symptoms, 25 of whom were diagnosed before two months of age [24]. Four of these 25 died due to complications of liver metastases, compared with one of the 24 infants over age two months. For this reason, the COG recommends urgent initiation of chemotherapy in young infants with stage MS (4S) neuroblastoma with evolving hepatomegaly, using a similar chemotherapy approach to those with intermediate-risk disease. (See 'Chemotherapy' below.)

By contrast, newborns who have limited adrenal disease (typically diagnosed by prenatal ultrasound) have a favorable prognosis and frequently do not need treatment. (See 'Infants less than one-year old with localized disease <5 cm' below.)

Stage MS (4S) disease — Stage 4S neuroblastoma per the INSS criteria describes infants below one year of age (table 2) who have resectable primary tumors (stage 1 or 2) and metastases that are limited to the liver, skin, and bone marrow (<10 percent); infants with metastases to cortical bone are excluded from this category. The INRGSS has modified the definition of 4S (called MS) to include patients up to 18 months of age with metastases limited to the liver, skin, and bone marrow (table 1). The prognostic influence of younger age at diagnosis is discussed in detail separately. (See 'Age' above.)

The MS (or 4S) category is an exception to the typically poor prognosis for children with widespread metastases from neuroblastoma [18]. OS for infants in this category is over 90 percent [22]. One contributing factor is that the tumor cells in infants with stage 4S (MS) disease have the capacity to undergo spontaneous regression [25,26].

However, like all stages of neuroblastomas, the tumors that make up stage MS (4S) disease are heterogeneous. For example, newborns less than two months with MS (4S) disease may present with more aggressive disease. Thus, patients require risk-adjusted treatment protocols [23]. (See 'Newborns' above and 'Low-risk disease' below.)

Histology — Histology is an independent prognostic factor, particularly for certain higher stage tumors [4,27]. (See 'Cytogenetics and molecular genetics' below.)

The balance between neural-type cells and Schwann-type (Schwannian) cells helps to categorize the tumor as neuroblastoma, ganglioneuroblastoma, or ganglioneuroma. In turn, neuroblastomas can have varying degrees of differentiation (undifferentiated, poorly differentiated, or differentiating). These histologic subtypes are described in detail separately. (See "Epidemiology, pathogenesis, and pathology of neuroblastoma", section on 'Pathology'.)

The International Neuroblastoma Pathology Classification (INPC) system, established in 1999 [27-30], relates the histopathologic features of the tumor, other biologic variables, and patient age to clinical behavior [31]. Tumors are classified as favorable or unfavorable based upon the degree of neuroblast differentiation, Schwannian stroma content, the frequency of cell division (the mitosis-karyorrhexis index [MKI]), and age at diagnosis. The INPC is a modification of a previous risk classification scheme, the Shimada system [31]. (See "Epidemiology, pathogenesis, and pathology of neuroblastoma", section on 'International Neuroblastoma Pathology Classification System'.)

In one validation study of the INPC system, five-year EFS was more than three times greater among children with favorable histology tumors compared with unfavorable histology tumors (90 versus 27 percent) [29].

Cytogenetics and molecular genetics — Certain molecular and cytogenetic tumor characteristics correlate with prognosis. These include, but are not limited to, MYCN (N-myc) amplification, deoxyribonucleic acid (DNA) content (ploidy) particularly in infants, and gain or loss of whole or partial chromosomes (ie, segmental chromosomal aberrations [SCAs]). The importance of these features is reflected in their inclusion as prognostic factors in determining risk assignment for treatment (figure 1) [6]. Additional characteristics which portend an inferior outcome and may influence treatment decisions include ALK aberrations [32,33] and telomere maintenance mechanisms [34,35]. (See "Epidemiology, pathogenesis, and pathology of neuroblastoma", section on 'Molecular abnormalities (prognostic impact)' and 'Treatment' below.)

TREATMENT — 

Improvements in outcome for children with neuroblastoma have been the result of cooperative group, multicenter clinical trials, which have integrated combined modality approaches with an understanding of the prognostic factors affecting outcome. Patients should be managed in a setting where appropriate expertise in the treatment of neuroblastoma is available. (See 'Risk stratification' below.)

Risk stratification — The treatment of neuroblastoma is determined based on risk categories (figure 1). Patients are classified into low-, intermediate-, and high-risk categories based on the following characteristics at the time of diagnosis:

●Stage of the disease

●Patient age

●International Neuroblastoma Risk Group Staging System (INRGSS) (table 1)

●Presence or absence of amplification of the MYCN oncogene

●Quantitative DNA content of the tumor (DNA index or ploidy)

●Histologic appearance of the tumor

●Segmental chromosomal aberrations (eg, loss of heterozygosity [LOH] 11q) [36]

Risk categories evolve as newer staging systems are adopted and further knowledge is acquired about molecular and genetic determinants of tumor behavior and prognosis. As an example, an updated Children’s Oncology Group (COG) risk classification scheme incorporating the INRGSS is used to characterize all tumors in ongoing and future COG studies, along with molecular, pathologic, and other clinical characteristics listed above (figure 1) [6]. A previous COG risk classification schema (table 3) was based on the International Neuroblastoma Staging System (INSS) (table 2). (See 'Staging system' above.)

Low-risk disease — Low-risk disease is defined according to the COG risk categorization schema (figure 1). Patients in the low-risk category generally have localized tumors without unfavorable characteristics. This includes low-stage disease (eg, INSS stage 1, 2A, or 2B, (table 2); or INRGSS stage L1 (table 1)), and tumors are MYCN non-amplified, hyperdiploid, and have favorable histology. (See 'Risk stratification' above.)

In general, tumor outcomes for children with low-risk neuroblastoma are excellent. Patients who relapse generally can be salvaged with further surgery or chemotherapy.

Surgery (preferred for most patients with low-risk disease) — Surgery is indicated in most patients with low-risk tumors beyond infancy (eg, children older than one year with INSS stage 1 or 2 disease [L1]) [37-41]. Patients treated with surgery typically do not require adjuvant chemotherapy. Although surgery has been the preferred option for patients with low-risk disease, a subgroup of much younger patients with small tumors (eg, INRGSS stage L1 disease) are typically managed with observation alone. (See 'Subgroups that can be managed with observation alone' below.)

Two- to five-year event-free survival (EFS) rates with surgery alone are greater than 93 percent for children with stage 1 disease; because recurrences can be successfully managed with further surgery or chemotherapy, five-year overall survival (OS) rates are 95 percent [37,38,42].

For most children with asymptomatic stage 2A or 2B disease, the outlook after surgery alone is also excellent. By contrast, children with symptomatic stage 2A or 2B disease are treated similarly to those with intermediate-risk disease and offered initial chemotherapy; among this rare group of patients, surgery is reserved for those who do not respond to chemotherapy. (See 'Intermediate-risk disease' below.)

Chemotherapy was a component of early treatment regimens but has gradually been removed over successive clinical trials without detriment to patient outcomes [37,38]. This was illustrated by a nonrandomized COG clinical trial (P9641) for 915 infants and children INSS stage 2A and 2B disease who underwent surgical resection followed by either observation or chemotherapy [42]. Chemotherapy was reserved for patients with or at risk for symptomatic disease, with less than 50 percent tumor resection at diagnosis, or with unresectable progressive disease after surgery alone. Five-year OS and EFS were similar in patients treated with surgery alone compared with those treated with surgery and chemotherapy (OS 97 versus 98 percent; EFS 89 versus 91 percent).

Subgroups that can be managed with observation alone

Infants less than one-year old with localized disease <5 cm — For infants less than six months of age with small (ie, tumor diameter ≤5 cm), asymptomatic INRGSS stage L1 (table 1) adrenal masses, we recommend observation, forgoing diagnostic surgical biopsy, rather than other interventions (eg, surgery or chemotherapy) at initial diagnosis [43].

●Observation schedule – For these patients, we recommend observation using serial ultrasounds of the adrenal mass with monitoring of urine catecholamines, vanillylmandelic acid (VMA) and homovanillic acid (HVA). These tests may be performed every three weeks for two visits, then every six weeks for two visits, then every 12 weeks for two visits, then every 24 weeks for two visits. If there is increase in tumor size, we recommend returning evaluation frequency to every three weeks. Patients should undergo surgical biopsy and possible resection if there is tumor progression [44]; in this subset of patients, if complete surgical resection can be safely performed, then chemotherapy can be avoided.

With advances in prenatal imaging, adrenal masses may be detected in infants before birth. Similar masses may be incidentally identified in neonates during imaging performed for other indications. These patients represent a favorable cohort [45-47]. While surgery is curative, it can be associated with significant morbidity and mortality, such as tumoral rupture, bleeding, infection, and injury to the great vessels, nerves, or kidneys [48,49]. A number of studies have indicated that expectant observation is safe in newborns with localized neuroblastoma, and that many of these tumors spontaneously regress [45-47,50-53].

In a COG prospective study (ANBL00P2) of 87 infants younger than six months with small adrenal masses, 83 of the 87 were initially observed. Spontaneous reduction in tumor volume was noted in two-thirds of patients, including 27 patients with no residual mass by the end of follow-up [53]. Surgery was avoided in 81 percent of patients with a median follow-up of 3.2 years, and three-year OS was 100 percent. Among the patients who underwent resection, the majority had confirmed stage I disease. A diagnosis other than neuroblastoma was observed in seven patients at the time of surgery; alternative diagnoses included extralobar pulmonary sequestration, adrenal cortical neoplasm, and hematoma.

Some UpToDate experts offer observation to select children greater than six months and less than one year old with asymptomatic, localized, biopsy-proven neuroblastoma (ie, INSS stage 1, 2, or 3 (table 2), INRGSS stage L1 or L2 (table 1)) with both favorable histology and genomics (eg, without MYCN amplification or segmental chromosomal aberrations). These patients may be observed with close monitoring of imaging and tumor markers, such as urine catecholamines. (See 'Histology' above and 'Cytogenetics and molecular genetics' above.)

However, this observational strategy is investigational, and other experts may alternatively offer standard treatment approaches (eg, surgery or chemotherapy) in these patients. The COG study ANBL1232 (NCT02176967) seeks to expand criteria for patients who may be followed with observation only or after initial biopsy [54].

Data supporting observation for these patients are as follows:

In a prospective, nonrandomized multicenter study, 93 of 340 infants (27 percent) with localized neuroblastoma and without MYCN amplification were observed without either primary resection of gross residual tumor or chemotherapy following the initial diagnosis [55]. The following findings were noted in this study:

●At a median follow-up of 58 months, spontaneous remissions were observed in 44 of 93 patients (47 percent), including 17 complete responses. 35 patients (38 percent) had evidence of local progression or stage 4S, four patients (4 percent) had progression to stage 4, and 10 patients (11 percent) had no change in tumor size.

●Surgery and/or chemotherapy were used to salvage those children who had evidence of progression.

●Tumor regression occurred over an extended time frame, with median times to first evidence and complete regression of 3.3 and 10 months, respectively. The first evidence of regression did not appear until after age one year in 15 of the 44 cases (34 percent).

●The three-year OS and metastasis-free survival rates for those initially managed with observation were 99 and 94 percent, not significantly different from those initially managed with surgery or chemotherapy.

Infants with stage MS (4S) disease without hepatomegaly — Infants with asymptomatic MS (stage 4S) disease without hepatomegaly and with tumors that are MYCN non-amplified, hyperdiploid, and have favorable histology may also be initially observed. In these patients, a high rate of spontaneous regression is seen (up to 70 percent) [26], and treatment can thus be deferred [22,23]. These patients require a tissue biopsy to exclude MYCN amplification and other tumor characteristics that would place them in an intermediate- or high-risk group. This biopsy can be from the primary tumor, bone marrow, or other metastatic sites.

We offer a similar observation schedule to that for newborns with small adrenal masses, using ultrasound images of liver and primary mass. This subset of patients with MS (4S) disease may benefit from a longer period of observation (ie, two to three years from the date of diagnosis). (See 'Infants less than one-year old with localized disease <5 cm' above.)

Patients with tumor growth and/or progressive disease-related symptoms (eg, those at risk for developing neurologic compromise due to spinal cord compression) should be referred for chemotherapy and possible surgery, per the approach for those with intermediate-risk disease. (See 'Chemotherapy' below.)

The COG study ANBL1232 (NCT02176967) seeks to further define criteria for patients up to age 18 months with INRGSS stage MS neuroblastoma (table 1) who may be able to delay or avoid chemotherapy versus those who require urgent treatment due to risk for poor outcome [54].

Intermediate-risk disease — According to the COG risk classification schema, intermediate-risk disease includes patients who do not meet criteria for either low-risk or high-risk disease (figure 1).

Chemotherapy — For children with intermediate-risk neuroblastoma, we suggest moderately intensive multiagent neoadjuvant chemotherapy (eg, with doxorubicin, cyclophosphamide, a platinum drug, and etoposide) with or without surgical resection, rather than surgery alone [12,56-58]. The goal of treatment is to deliver a sufficient duration of chemotherapy (with or without subsequent surgery) to achieve at least a partial response (PR) (at least 50 percent reduction of soft tissue masses) and resolution of metastatic disease [58].

The duration of chemotherapy is typically 6 to 24 weeks and is optimized based on specific tumor histologic and biologic characteristics. As an example, patients who receive longer durations of chemotherapy typically have tumors with unfavorable histology; diploidy (ie, DNA index = 1); or segmental chromosomal aberrations (ie, LOH of 1p or unbalanced LOH of 11q).

The rationale for combined-modality approach to treatment comes from a report from the Children's Cancer Group (CCG) in which children with intermediate-stage disease received five courses of chemotherapy, followed by surgery, another course of chemotherapy, radiation therapy (RT) for gross residual disease, and four more courses of chemotherapy [12]. The four-year EFS for patients with favorable biology was 100 percent, while for infants with at least one unfavorable biologic feature, EFS was 90 percent.

As a result of treatment successes with this regimen, subsequent clinical investigations have decreased chemotherapy intensity and aggressiveness of local control measures, including elimination of RT except in emergent situations, as illustrated by the following studies:

●In a COG trial (A3961) for intermediate-risk neuroblastoma, four cycles of chemotherapy were given for tumors with favorable biologic features, and eight cycles were given for tumors with unfavorable features [56]. Children treated in this way did just as well as in previous studies using more cycles of chemotherapy, with three-year EFS and OS of 88 and 96 percent, respectively.

●In a study performed by the Society of Pediatric Oncology European Neuroblastoma Network (SIOPEN), infants with unresectable tumors received low doses of cyclophosphamide and vincristine until tumors were resectable. This study of 180 infants included 84 children with stage 3 tumors. Again, the reduction in chemotherapy was not detrimental to five-year EFS or OS, which remained 90 and 99 percent, respectively [57].

●The COG trial for intermediate-risk neuroblastoma (ANBL0531) demonstrated excellent survival outcomes with omitting cycles of neoadjuvant chemotherapy and/or surgery and omitting RT [58]. In this single arm phase III trial, 404 patients were stratified and treated based on patient age, stage, genetic, and histologic features. Patients were treated with either two, four, or eight cycles of neoadjuvant chemotherapy. Longer durations of chemotherapy were assigned to patients with unfavorable histology, diploidy (ie, DNA index = 1), or segmental chromosomal aberrations (eg, LOH of 1p or unbalanced LOH of 11q). Patients then underwent subsequent surgery or additional chemotherapy based on tumor response. Surgery was omitted in those with PRs or better, which comprised a majority of patients. Patients who did not achieve the defined treatment endpoints received additional chemotherapy. This study included some patients whose disease would previously have been considered low or high risk, which limits the ability to compare the results to other studies evaluating this approach.

•For the 404 patients enrolled, three-year EFS was 83 percent and OS was 95 percent. Those with localized tumors (stages 2 and 3 patients) had a three-year EFS of 88 percent and OS of 100 percent.

•Infants with stage 4S disease had a three-year OS of 82 percent, largely due to early death from hepatomegaly.

•Infants with stage 4 disease had a three-year EFS and OS of 78 and 91 percent, respectively. Within this group though, those with unfavorable tumor biology fared much worse than those with favorable biology (three-year EFS 67 versus 87 percent), indicating that different treatment strategies may be needed for these patients [58]. (See 'Prognostic factors' above.)

•Three-year EFS for those treated with two, four, or eight cycles of chemotherapy was 87, 87, and 80 percent, respectively; three-year OS was 99, 94, and 87 percent.

Indications for surgery — For intermediate-risk neuroblastoma, initial aggressive surgical resection is discouraged unless the tumor can be removed without threat to adjacent vital structures. In general, patients with L2 tumors should receive chemotherapy prior to potential surgical resection. Preoperative chemotherapy can decrease tumor size and increase ease of resection, or even eliminate the need for resection [58]. Surgery may be indicated in patients who do not achieve a PR or very good partial response (VGPR) with initial chemotherapy, whichever response is mandated based on the patient's overall risk stratification. For such patients, the decision to proceed with surgery rather than additional chemotherapy is based on the likelihood that resection can achieve the desired response without significant risk to vital structures (eg, kidneys, nerves, or vascular structures). Regardless, all patients should undergo initial surgical biopsy to obtain a diagnosis and discern histologic and genomic features that inform duration of initial chemotherapy [56,59-61]. (See "Clinical presentation, diagnosis, and staging evaluation of neuroblastoma", section on 'Biopsy'.)

The feasibility of complete resection is determined by tumor location and mobility, relationship to major nerves and blood vessels, the presence of distant metastases, and patient age. Sacrifice of vital organs to achieve a complete resection of large primary tumors at the time of diagnosis should be avoided. Surgery should be delayed in circumstances when less than 50 percent of the tumor can be safely removed, and chemotherapy should be administered after biopsy, to either achieve a PR or VGPR (based on specific risk stratification) or to allow for a safer surgical procedure to achieve the goal tumor response.

The ultimate extent of surgical resection necessary for optimal outcomes has been debated [56,59-61]. Data from the COG study ANBL0531 suggest that among patients with localized favorable histology tumors receiving chemotherapy with or without surgery, good EFS and OS can be achieved with an end-of-therapy goal of PR, even if surgical resection and RT are omitted [58].

Avoidance of radiation therapy — Cooperative groups avoid RT for patients with intermediate-risk disease and recommend it only in the setting of emergent life- or organ-threatening complications or disease progression despite surgery and chemotherapy [2,56,58].

High-risk disease — Patients most commonly at the highest risk for disease progression and mortality are generally those who are older than 18 months of age and have either disseminated disease or localized disease with unfavorable markers such as MYCN amplification (figure 1) regardless of age. (See 'Risk stratification' above.)

Multiple genetic and biologic factors contribute to the risk of treatment failure, and tailoring therapy to reflect these factors is an ongoing challenge [14,32,62]. Despite aggressive multimodality therapy, OS rates remain unacceptably low (approximately 50 percent) [9,63], and the improved outcomes have come at a cost of significant early and late toxicity. (See 'Prognosis' below.)

Improved survival outcomes have been achieved using an aggressive multimodality approach that includes chemotherapy, surgical resection, hematopoietic stem-cell transplantation, RT, and immunotherapy [64,65]. Prior to instituting this approach, the long-term survival probability for children with high-risk disease was less than 15 percent. Data from randomized trials have consistently demonstrated improved EFS in patients who received myeloablative chemotherapy with stem cell rescue [66-68], and some of these studies demonstrated an improvement in OS in certain groups of patients [67,68].

Components of treatment — The following briefly summarizes the general components of treatment for high-risk disease, which includes induction therapy, local control (with surgical resection and RT), consolidation with tandem autologous hematopoietic stem cell transplantation, and postconsolidation immunotherapy. Specific treatment details are beyond the scope of this review and treatment should be administered under the guidance of a multidisciplinary team of pediatric oncologists, surgeons, and radiation oncologists with expertise in the treatment of neuroblastoma. Additionally, such treatment protocols may be individualized based on the institution.

Induction

Chemotherapy — Since patients typically present with unresectable disease, the rationale for induction chemotherapy is to reduce primary and metastatic tumor burden and allow attempts at local control. The choice of induction regimen is typically dependent upon institution and location. While observational and randomized studies have compared the impact of various available induction regimens on survival and long-term toxicity, further data are necessary to determine the optimal regimen [69-71]. In the United States, the most widely used induction regimen includes five cycles of intensive chemotherapy with a combination of agents (vincristine, cyclophosphamide, topotecan, doxorubicin, cisplatin, etoposide) [72-74]. Rapid COJEC (a combination of cisplatin [C], vincristine [O], carboplatin [J], etoposide [E], and cyclophosphamide [C]) is used in Europe [75,76].

Investigational induction therapies — For subsets of high-risk patients, the COG phase III clinical trial ANBL1531 (NCT03126916) is also evaluating the safety and impact on EFS of incorporating ALK inhibitors and therapeutic Iobenguane I-131 meta-iodobenzylguanidine (MIBG) therapy with induction chemotherapy.

●ALK inhibition – Approximately 10 to 15 percent of patients with de novo high-risk neuroblastoma have activating mutations of the ALK receptor tyrosine kinase, and preclinical and early-phase trials have demonstrated the safety and efficacy of ALK inhibition [32,77].

In patients with ALK-activating mutations enrolled in COG ANBL1531, an ALK inhibitor is administered after the first cycle of induction chemotherapy until consolidation, and then restarted after the completion of immunotherapy. Of note, close monitoring for respiratory changes is advised when lorlatinib and dinutuximab are administered concurrently. Holding lorlatinib during days of dinutuximab administration should be strongly considered for any patient with a prior history of significant lung dysfunction, prior sinusoidal obstruction syndrome (SOS), or recent/concurrent respiratory infections.

●MIBG therapy – Patients with MIBG avid disease enrolled in COG ANBL1531 are randomly assigned to receive I-131 MIBG therapy between cycles 3 and 4 of induction or to receive standard therapy without I-131 MIBG.

While gamma emitting I-123-MIBG is effective as a highly specific functional imaging modality for neuroblastoma, the higher energy beta and gamma emitting I-131 MIBG is effective in the setting of relapsed disease and works via directed radiation-induced DNA damage [78]. MIBG therapy is effective against bone marrow and bony metastatic disease [79]. This supports a rationale for including MIBG therapy during induction to further decrease metastatic tumor burden during induction chemotherapy. Several phase I and II studies have demonstrated favorable response rates and toxicity profiles with this approach [78,80,81].

●Anti-GD2 monoclonal antibodies during induction – The anti-GD2 monoclonal antibody Hu14.18K322A was investigated as a novel adjunct to induction chemotherapy. In an open label nonrandomized phase II trial (NB2012), 64 patients with high-risk neuroblastoma were treated with Hu14.18K322A coadministered with standard-induction chemotherapy, followed by granulocyte macrophage colony-stimulating factor (GM-CSF) and low-dose interleukin 2 (IL 2). Subsequently, patients received consolidation chemotherapy with busulfan and melphalan (with Hu14.18K322A administered when available, and natural killer cell infusions in some patients) and postconsolidation therapy with Hu14.18K322A, GM-CSF, IL-2, and isotretinoin. After two cycles of induction chemotherapy, PR was seen in 42 patients (67 percent). Additionally, three-year EFS and OS were 74 and 86 percent, respectively, which is higher than what has been in observed in historical controls receiving similar chemotherapy without Hu14.18K322A (between 50 to 66 percent) [73,82].

Dinutuximab, which has an established role in postconsolidation therapy, is also being evaluated as part of induction chemotherapy in a multi-institutional trial (ANBL17P1; NCT03786783). (See 'Postconsolidation therapy' below.)

Local control — As with other aggressive metastatic cancers, local control of the primary tumor plays an important role for high-risk neuroblastoma, and patients are treated with both surgery and radiation (administered after consolidation chemotherapy).

Surgery — The importance of achieving a gross total resection of the primary tumor in patients with disseminated disease is controversial, with some studies [12,63,83-89], but not others [59,90,91], suggesting a better outcome for complete resection. Surgical resection of the tumor should be performed by a pediatric surgeon with experience in resecting extensive, infiltrating tumors. Resection should be performed after several courses of induction chemotherapy, when the tumor is smaller and less invasive.

Radiation therapy — RT to the primary tumor bed is recommended for high-risk neuroblastoma and is administered after consolidation therapy in most treatment protocols. RT is beneficial in preventing local tumor recurrence [92-94]. There is debate about whether the RT field should include lymph nodes adjacent to the primary tumor [91,95]. Similarly, there is discussion about which metastatic lesions need irradiation, and how this impacts local versus overall relapse risk [96,97]. For patients with residual tumor at the primary site after consolidation therapy, data are also conflicting for the efficacy of an additional radiation boost [98,99]. This approach is being investigated in a randomized SIOPEN trial, High-Risk Neuroblastoma Study 2 (HR-NBL2; NCT04221035).

Consolidation — After tumor bulk has been decreased by chemotherapy and surgery, the consolidation phase includes high-dose chemotherapy followed by autologous hematopoietic stem cell transplantation [64,66,72,100,101] and RT. Hematopoietic stem cells are collected using apheresis, usually after the second cycle of induction chemotherapy in most regimens used in North America. Although high-risk patients have bone marrow metastases at the time of diagnosis and may have residual disease in the bone marrow at the time of apheresis, the use of immunomagnetic "purging" of stem cell products [66] did not decrease the risk of recurrence or OS when evaluated in a prospective, randomized trial performed by the COG [102].

Consolidation therapy with hematopoietic stem cell rescue improves EFS but not OS, as demonstrated by a Cochrane review of three randomized trials conducted in 739 children with high-risk neuroblastoma [103]. In this analysis, compared with conventional treatment, high-dose chemotherapy and hematopoietic stem cell transplantation improved EFS (hazard ratio [HR] 0.79, 95% CI 0.70-0.90) and trended towards improved OS in two of the trials, although the results were not statistically significant (HR 0.86, 95% CI 0.73-1.01). There was no difference in secondary malignant disease and treatment-related death between the two treatment groups, although one of the trials showed a significantly higher rate of specific treatment-related toxicities in those treated with stem cell transplantation (eg, kidney effects, interstitial pneumonitis and veno-occlusive disease) compared with those treated with conventional chemotherapy.

Tandem transplants — For patients with high-risk disease, tandem autologous hematopoietic stem-cell transplantation for consolidation has been shown to improve disease outcomes relative to single transplants.

In a multicenter COG study (ANBL0532), 355 patients with high-risk neuroblastoma were randomly assigned after induction therapy to either single transplant with carboplatin-etoposide-melphalan (CEM) or tandem (two) transplants with thiotepa-cyclophosphamide, followed by a modified CEM (TC:CEM) [72]. High-risk disease was defined as a tumor with amplification of the MYCN oncogene occurring in children of any age; localized unresectable disease with unfavorable histology; or metastatic disease occurring in children who were older than 18 months at diagnosis. After a median follow-up of approximately five and a half years, while the tandem transplant group experienced improved three-year EFS compared with those receiving single transplants (61 versus 48 percent), the difference in OS at three years did not reach statistical significance (74 versus 69 percent). For the subset of patients receiving immunotherapy, tandem transplants were associated with improvements in both EFS (73 versus 55 percent) and OS (84 versus 74 percent). Cumulative rates of severe mucosal, infectious, or liver toxicities and regimen-related mortality were similar between arms. In addition, as dose intensity of treatment increases, the need to monitor for late effects including secondary cancers becomes even more important.

Based on these findings, this tandem transplant strategy is incorporated into the COG trial ANBL1531 evaluating various induction strategies for patients with high-risk disease. (See 'Induction' above.)

Investigational consolidation therapies

●Alternative stem cell transplantation techniques (high-dose chemotherapy with melphalan and busulfan) – A randomized study suggests an advantage of high-dose chemotherapy with melphalan and busulfan over carboplatin, etoposide, and melphalan (CEM) [104]. In a European phase III trial, approximately 600 patients with high-risk neuroblastoma that had responded to a multidrug induction regimen were randomly assigned to high-dose chemotherapy with either busulfan and melphalan or CEM, followed by autologous hematopoietic stem cell transplantation. Compared with patients treated with CEM, patients receiving busulfan plus melphalan had an improved three-year EFS (50 versus 38 percent) and five-year OS (54 versus 41 percent). Severe life-threatening toxicities occurred in 4 percent of patients receiving busulfan and melphalan versus 10 percent receiving CEM.

High-dose chemotherapy with melphalan and busulfan followed by one autologous hematopoietic stem cell transplant is being compared with tandem transplant in the COG trial ANBL1531 for patients with high-risk disease. (See 'Induction' above.)

Postconsolidation therapy — Postconsolidation therapy targets minimal residual disease not detected by diagnostic imaging and bone marrow assessment. In this phase of therapy, patients receive immunotherapy with an anti-GD2 monoclonal antibody and cytokine (in North America) and cis-retinoic acid, a differentiating biologic agent. There are multiple formulations of anti-GD2 monoclonal antibodies in use. However, specific agents that have been evaluated in randomized studies include dinutuximab and dinutuximab beta. (See 'Dinutuximab (preferred)' below.)

Dinutuximab (preferred) — Dinutuximab is approved by the US Food and Drug Administration (FDA) and has been evaluated in combination with granulocyte macrophage colony-stimulating factor (GM-CSF), IL-2, and cis-retinoic acid (isotretinoin) [66,105] for the treatment of pediatric patients with high-risk neuroblastoma who achieve at least a PR to prior initial multimodality therapy.

Of note, available upfront trials in the United States and Europe no longer include IL-2 as part of standard consolidation, based on data from trials evaluating its use with dinutuximab beta [106,107].

Dinutuximab (ch14.18) is a chimeric antibody that targets disialoganglioside GD2, a tumor-associated antigen uniformly expressed on neuroblastomas [108,109]. The addition of dinutuximab plus cytokines (GM-CSF and IL-2) was found to have benefit over cis-retinoic acid alone for prevention of recurrence in a randomized trial (ANBL0032) [82]. The immunotherapy approach resulted in superior two-year EFS (66 versus 46 percent) and OS rates (86 versus 75 percent). In extended follow-up of this study after cessation of random treatment assignment, patients treated with dinutuximab plus cytokines also demonstrated durable survival outcomes (five-year EFS and OS of 61 and 72 percent) [110]. Toxicities include serious and potentially life-threatening infusion reactions and severe neuropathic pain.

Dinutuximab beta (alternative) — Dinutuximab beta is a separate anti-GD2 antibody with a longer and slower dosing regimen than dinutuximab [111]. Dinutuximab beta has regulatory approval in Europe but not in the United States.

Dinutuximab beta may be administered without GM-CSF or cis-retinoic acid, although trials have typically included cis-retinoic acid [106]. In the International Society for Paediatric Oncology European Neuroblastoma (SIOPEN) trial, the addition of low-dose IL-2 to long-term infusion dinutuximab beta and cis-retinoic acid increased toxicity without improving outcomes [106,107].

Other postconsolidation therapies

●Addition of chemoimmunotherapy to postconsolidation therapy – In patients with high-risk neuroblastoma, disease recurrence followed by death is a common outcome despite intensive multimodality therapy [2,9]. Improving survival in this group of children is a priority in clinical trials conducted by international cooperative groups. Treatment options under investigation include building upon the success of chemoimmunotherapy in the setting of relapsed/refractory disease (see 'Dinutuximab plus chemotherapy' below) by maximizing the intensity of postconsolidation with the addition of irinotecan and temozolomide. In preliminary results, the ongoing pilot trial ANBL19P1 (NCT04385277) suggests the feasibility of this approach [112].

●Eflornithine – Maintenance therapy with eflornithine is an option after completion of postconsolidation immunotherapy with an anti-GD2 antibody. However, the clinical use of this agent is not standard across all institutions. Therefore, eflornithine may be offered based on institutional guidelines and on an individual patient basis.

Eflornithine (ie, difluoromethylornithine [DFMO], an inhibitor of ornithine decarboxylase) is approved by the FDA as maintenance therapy in patients with high-risk neuroblastoma who achieve at least a PR to prior systemic agents and have completed postconsolidation immunotherapy with an anti-GD2 antibody [113]. Data supporting the use of eflornithine as maintenance therapy in high-risk neuroblastoma are as follows:

•In a phase II trial of 81 patients with high-risk neuroblastoma treated with eflornithine as maintenance therapy after receiving standard induction and consolidation, five-year EFS and OS were 85 and 95 percent, respectively [114,115]. These values were higher than the five-year EFS and OS seen with historical controls (66 and 82 percent, respectively).

•A separate analysis compared 92 patients with high-risk neuroblastoma who completed multimodality treatment ending with dinutuximab followed by two years of maintenance eflornithine as part of one study (NMTRC003B) to 852 matched control patients with high-risk neuroblastoma who also completed multimodality treatment including postconsolidation dinutuximab as part of ANBL0032, but who did not receive eflornithine [116]. At median follow-up of up to 6.1 years, maintenance eflornithine after completion of immunotherapy was associated with improved EFS (HR 0.50, 95% CI, 0.29 to 0.84) and OS (HR, 0.38, 95% CI 0.19 to 0.76); these were further confirmed with propensity score-matched cohorts and sensitivity analyses.

Relapsed or refractory disease — Overall, at least 40 percent of patients with high-risk neuroblastoma will experience disease recurrence after completing standard multimodality therapy [117]. Among these patients, approximately 20 percent will have no/mixed response or progressive disease at the end of induction chemotherapy [72,102]. While most relapses occur within the first two years after treatment completion, late relapse can occur as far out as 10 to 30 years after treatment completion [118]. These findings indicate the need to develop novel therapeutic concepts for neuroblastoma. The safety, toxicities, and clinical efficacy of these approaches are an active area of clinical research.

For patients with relapsed or refractory disease, we refer patients for clinical trials (www.clinicaltrials.gov), as standard treatment options are limited.

For patients who do not select or are without access to clinical trials, standard treatment options available include dinutuximab in combination with irinotecan and temozolomide or I-131 MIBG therapy (where available). However, less than 50 percent of patients will respond to these therapies, highlighting the need for new treatments to improve efficacy.

Dinutuximab plus chemotherapy — The addition of dinutuximab to chemotherapy in patients with relapsed or refractory disease shows high clinical response rates, including those previously receiving dinutuximab therapy [119-121].

In a randomized COG trial ANBL1221 of 36 patients with neuroblastoma in first relapse/progression treated with irinotecan and temozolomide, the addition of dinutuximab improved objective responses rates compared to temsirolimus (53 versus 6 percent) [119]. As a result of these initial data, additional patients were nonrandomly assigned to the dinutuximab arm, and included those who experienced disease relapse after (but not during) previous dinutuximab postconsolidation therapy [120]. Of the 53 total patients who received this therapy, objective responses were seen in 22 patients (42 percent), including 11 complete responses (21 percent), and 22 patients had stable disease (42 percent). However, total duration of response to this regimen could not be assessed because many patients went off study to receive other treatments, such as surgery or high-dose chemotherapy with autologous hematopoietic stem cell transplantation. This information was provided in a subsequent multicenter retrospective study of 146 patients with relapsed high-risk neuroblastoma who received dinutuximab with irinotecan and temozolomide, which showed a median progression-free survival of 13.1 months, with a median duration of response of 15.9 months [121].

MIBG therapy — I-131 MIBG is an effective treatment for relapsed or refractory neuroblastoma. In several phase I and II clinical trials of patients with relapsed or refractory disease who were treated with I-131 MIBG, the mean response rate across these studies was 32 percent [79,81,122]. Other clinical trials are ongoing to evaluate the combination of I-131 MIBG with agents such as dinutuximab and/or vorinostat (NCT03332667, NCT03561259) to maximize both response rates and duration of response.

Other approaches

●Bevacizumab plus chemotherapy – The optimal use of bevacizumab plus chemotherapy (temozolomide as a single-agent or in combination with either irinotecan or topotecan) is unclear for patients with first-relapsed or primary refractory neuroblastoma. In separate randomized trials, bevacizumab plus chemotherapy resulted in lower objective response rates (ORRs; 26 percent) [123] than those seen with dinutuximab plus irinotecan and temozolomide (53 percent) [119]. The combination of bevacizumab plus irinotecan and temozolomide may be an option for patients with relapsed or refractory disease who lack access to dinutuximab or do not tolerate or respond to the combination of chemotherapy and dinutuximab. (See 'Dinutuximab plus chemotherapy' above.)

In a randomized trial (ITCC-SIOPEN BEACON) of 160 patients with relapse or refractory neuroblastoma, at a median follow-up of 1.5 years, the addition of bevacizumab to chemotherapy (ie, irinotecan plus temozolomide, topotecan plus temozolomide, or temozolomide alone) demonstrated a nonstatistically significant trend towards improved ORR (26 versus 18 percent), but failed to improve progression-free survival (PFS) or OS [123]. However, there was a potential PFS benefit for bevacizumab plus irinotecan and temozolomide, over bevacizumab plus temozolomide.

●Naxitamab – The FDA has approved the anti-GD2 monoclonal antibody naxitamab in combination with GM-CSF for select patients with relapsed and refractory neuroblastoma. There are limited published data regarding its efficacy, and it is not yet clear how to best incorporate this drug into clinical practice. Per the FDA label, patients must have disease limited to the bone and/or bone marrow and must have demonstrated at least stable disease with their immediately preceding therapy [113].

In a single-arm study, 34 patients with refractory high-risk neuroblastoma were treated with naxitamab in combination with irinotecan and temozolomide (HITS) [124]. Complete and partial response rates were 29 and 3 percent, respectively.

●Immunotherapy – There is interest in investigating additional immunotherapy approaches in patients with relapsed and refractory neuroblastoma. Such approaches include chimeric antigen receptor (CAR) expressing T cells [125,126], immune checkpoint inhibitors, vaccines [127], and antibodies that modulate the immune microenvironment. (See "Principles of cancer immunotherapy".)

In an open-label phase I/II trial, 27 patients with relapsed or refractory neuroblastoma were treated with GD2-CAR T cells expressing the inducible caspase 9 suicide gene (GD2-CART01) at a dose of 10 x 10⁶ CAR-positive T cells per kg [125]. At median follow-up of 1.7 years, overall responses were seen in 17 patients (63 percent), including nine complete responses (33 percent) and eight PRs (30 percent). Three-year OS and EFS rates were 60 and 36 percent, respectively. The rate of cytokine release syndrome was 74 percent but was mild in most patients. In one patient who developed an altered state of consciousness (later attributed to brain hemorrhage), the caspase 9 suicide gene was successfully activated via administration of rimiducid to rapidly remove GD2-CART01. (See "Cytokine release syndrome (CRS)".)

●Using MYCN-amplification to select therapy – There is interest in using the presence or absence of MYCN-amplification to select therapy in relapsed or refractory neuroblastoma. Further data are necessary prior to incorporating this approach into routine clinical practice.

In an open-label, randomized phase II trial of 129 patients with relapsed or refractory neuroblastoma, at a median follow-up of 72 months, the addition of dasatinib and rapamycin to irinotecan and temozolomide improved PFS (median 11 versus 5 months, HR 0.62, 95% CI 0.42-0.92) [128]. In the entire study population, although OS was higher with the addition of dasatinib and rapamycin to chemotherapy, the difference was not statistically significant (median 20 versus 16 months, HR 0.68, 95% CI 0.45-1.04). Among those with MYCN-amplified tumors, the addition of dasatinib and rapamycin to irinotecan plus temozolomide improved both PFS (median six versus two months, HR 0.45, 95% 0.24-0.84) and OS (median 11 versus 6.5 months, HR 0.51, 95% 0.27-0.96). By contrast, among those without MYCN-amplified tumors, this approach did not show a statistically significant difference in PFS or OS compared with irinotecan and temozolomide alone.

●Integration of cancer "omics" – Various methodologies permit global genomic characterization of neuroblastoma specific epigenetic, metabolic, transcriptional and translational oncogenic mechanisms [129-131]. This includes integration of genomic information retrieved at the single cell level. This work is revealing critical signaling networks, biomarkers, and novel therapeutic targets [132-135].

●ALK inhibition – Molecular alterations of the ALK tyrosine kinase receptor are present in up to 15 percent of patients with de novo high-risk neuroblastoma. Early-phase clinical trials have demonstrated the safety and efficacy of ALK inhibitors such as crizotinib and lorlatinib [77,136]. As an example, in a phase I trial in patients with relapsed or refractory neuroblastoma, single-agent lorlatinib demonstrated a response rate (complete and partial responses) of 13 percent in patients less than 18 years old, and 47 percent in patients greater than 18 years old [136]. Additionally, for patients less than 18 years old, the response rate (complete plus partial responses) of lorlatinib in combination with cyclophosphamide and topotecan was 25 percent. When minor responses were added to the complete and partial responses for patients less than 18 years old, the response rates for lorlatinib as a single agent or in combination with chemotherapy were 30 and 63 percent, respectively.

●Targeting tumor immune microenvironment and heterogeneity – Oncogenic functions in neuroblastoma modulate the tumor immune microenvironment, drive heterogeneity, and promote metastasis [137-141]. In addition, preclinical data suggest that neuroblastoma is a highly complex mixture of adrenergic and mesenchymal subtypes with different tumorigenic, metastatic, and regenerative potential [142-145] and inform ongoing translational research and trials.

MEDICAL COMPLICATIONS ASSOCIATED WITH NEUROBLASTOMA PRESENTATION

Spinal cord compression — Between 5 and 15 percent of children with neuroblastoma present with spinal cord involvement. Spinal cord compression is considered an oncologic emergency. Prompt resolution is important to limit permanent neurologic impairment. (See "Treatment and prognosis of neoplastic epidural spinal cord compression", section on 'Systemic therapy'.)

In patients with neuroblastoma who present with spinal cord compression, most high-volume centers favor initial chemotherapy rather than surgical decompression due to the known risk of late spinal deformities in some children who undergo surgical treatment [146]. Radiation therapy (RT) is rarely used; tumors that progress on chemotherapy are normally treated with surgical decompression.

Neurologic recovery appears to be related to the severity of presenting neurologic deficits [147-149]. Up to 70 percent of patients with symptoms of cord compression at diagnosis may have residual long-term neurologic sequelae. Chemotherapy and laminectomy appear have equivalent outcomes, in terms of short- and long-term symptom relief, as well as overall survival (OS). Each therapeutic modality, however, carries inherent long- and short-term risks and should be determined on an individualized basis [147,150].

Opsoclonus myoclonus — Neuroblastomas associated with the paraneoplastic opsoclonus-myoclonus-ataxia (OMA) syndrome only represent approximately 2 to 3 percent of all newly diagnosed neuroblastomas, and are usually of lower stage and have favorable prognosis for survival [151,152]. However, children with neuroblastoma and opsoclonus-myoclonus syndrome are often left with long-term neurologic deficits (eg, cognitive and motor delays, language deficits, and behavioral issues), indicating the necessity of prompt diagnosis and interventions [151,153-156]. (See "Clinical presentation, diagnosis, and staging evaluation of neuroblastoma", section on 'Paraneoplastic syndromes' and "Opsoclonus-myoclonus-ataxia syndrome".)

Corticosteroids and intravenous immune globulin (IVIg) may improve long-term neurologic outcome for OMA [151,156,157]. Symptoms refractory to these treatments may respond to rituximab immunosuppression, as well as corticotropin for neurologic symptoms [158-160]. In an open label randomized study of 53 patients with OMA and neuroblastoma treated with risk-based chemotherapy and prednisone, the addition of IVIg improved OMA response rates (80 versus 40 percent) [161].

Tumor lysis syndrome — Patients with bulky or advanced stage disease may rarely be classified as being at intermediate risk for tumor lysis syndrome and may be offered prophylaxis against the complications of tumor lysis syndrome [162]. The treatment of tumor lysis syndrome is discussed separately. (See "Tumor lysis syndrome: Prevention and treatment".)

Watery diarrhea, hypokalemia, and hypochlorhydria or achlorhydria (WDHA) syndrome — A small number of patients with neuroblastoma, ganglioneuroblastoma, and ganglioneuroma present with life-threatening secretory diarrhea caused by vasoactive intestinal peptide (VIP) secreted by the tumor cells [163]. Treatment is fluid and electrolyte resuscitation followed by surgical removal of the tumor. (See "Clinical presentation, diagnosis, and management of VIPoma".)

PROGNOSIS — 

The prognosis for patients with neuroblastoma depends upon prognostic factors, risk stratification, extent and site of metastases, and treatment received. (See 'Prognostic factors' above and 'Risk stratification' above and 'Treatment' above.)

●Low-risk disease – Patients with low-risk disease treated either with observation or surgical resection have event-free survival (EFS) rates of greater than 85 percent and overall survival (OS) approaching 100 percent [37,38,42]. (See 'Low-risk disease' above.)

●Intermediate-risk disease – For patients with intermediate-risk disease, long-term survival rates are over 90 percent [24,56]. (See 'Intermediate-risk disease' above.)

●High-risk disease – Children with high-risk disease have long-term survival rates of approximately 50 percent, despite intensive multimodality strategies that include high-dose therapy with autologous hematopoietic stem cell rescue [9,63]. Prior to instituting these aggressive approaches, the long-term survival probability for children with high-risk disease was less than 15 percent. These patients should be encouraged to enroll in treatment protocols evaluating new therapies. (See 'High-risk disease' above.)

LONG-TERM TOXICITIES — 

Neuroblastoma survivors are at increased risk for long-term morbidity and mortality [164-166]. We recommend long-term follow-up that are consistent with childhood cancer survivorship guidelines from the Children's Oncology Group (COG) [167]. These patients should be followed by clinicians attuned to the special needs of this population. The frequency of surveillance varies depending upon extent of disease and level of exposure to radiation and chemotherapy [168]. (See "Overview of cancer survivorship in adolescents and young adults" and "Overview of cancer survivorship care for primary care and oncology providers".)

In addition, there are several web-based patient portals that facilitate long-term access to patient treatment records and empower patients to seek long-term follow-up. Survivorship Care Plans (SCPs) are one important way to improve compliance and can ensure that adult survivors of pediatric cancer get critical follow-up care [169-171]. (See "Assuring quality of care for cancer survivors: The survivorship care plan".)

Data have demonstrated increased risk of morbidity and mortality among childhood cancer survivors of neuroblastoma, which is likely related to disease and associated treatments [164-166]. As an example, the impact of treatment was demonstrated in one series of 954 neuroblastoma patients from the Childhood Cancer Survivor Study (CCSS) diagnosed between 1970 and 1986 who survived for at least five years [164]. This group had a significantly increased risk of overall mortality compared with the general population (standardized mortality ratio 5.6). The primary causes of death were recurrent disease and second malignancy. When compared with a cohort of siblings, the survivors' risk for chronic health conditions was also increased. The 20-year cumulative incidence of chronic health conditions was 41 percent, with most of these conditions involving the neurologic, sensory, endocrine, and musculoskeletal systems.

COG is enrolling patients in the Late Effects After High-Risk Neuroblastoma (LEAHRN) study (ALTE15N2, NCT03057626). The goal of this study is to collect data about the long-term impact of neuroblastoma therapy on growth, organ function, neurocognitive function, and risk for second malignancy [172].

Examples of the mechanisms by which neuroblastoma or its treatment may affect survivors and associated surveillance recommendations include the following:

●Subsequent malignant neoplasms – Data suggest an increased risk of subsequent malignant neoplasms (SMNs) in children with neuroblastoma who are treated with platinum agents, radiation therapy (RT), and radioactive iodine (eg, I-131 MIBG) [173-176]. The 30-year cumulative incidence of SMNs is approximately 3 percent, based on a cohort study of five-year neuroblastoma survivors [176]. SMNs observed during this period included various carcinomas (most commonly bladder and thyroid cancer), sarcoma, hematologic malignancies, melanoma, and peritoneal mesothelioma. Other studies have also reported subsequent myeloid leukemias due to chemotherapy and solid tumors related to RT. Of note, secondary hematopoietic malignancies are more common in the first five years following therapy, while solid tumors predominate later [174]. Patients who received chemotherapy should be followed with complete blood counts, then bone marrow studies if abnormalities arise. Those treated with radiation should be followed with physical examinations focusing on the radiation field. (See "Therapy-related myeloid neoplasms: Epidemiology, causes, evaluation, and diagnosis" and "Overview of cancer survivorship care for primary care and oncology providers", section on 'Risk of subsequent primary cancer'.)

●Cognitive deficits – Cognitive deficits in childhood cancer survivors are commonly attributed to cranial irradiation, either alone or as part of total-body irradiation (TBI), in very young children. Contemporary neuroblastoma treatment regimens rarely include cranial RT. Despite this, a Childhood Cancer Survivor Study (CCSS) has indicated that neuroblastoma survivors have increased risk for psychologic impairment and increased need for special education resources compared with siblings [177]. This risk is present regardless of treatment modality. Neuroblastoma survivors would benefit from formal neurocognitive evaluation after completion of therapy, as well as follow-up of educational achievement through childhood.

●Cardiac toxicity – Anthracycline chemotherapy (eg, doxorubicin) can cause myocardial dysfunction and impaired myocardial growth. This risk is exacerbated in patients who require RT to fields that include the heart. One study identified cardiac dysfunction in over 20 percent of survivors of high-risk neuroblastoma [178]. Of note, high-risk neuroblastoma protocols have fewer doxorubicin-containing cycles and have incorporated the use of the cardioprotectant agent dexrazoxane, which may decrease future cardiotoxicity [72,102,179,180]. (See "Clinical manifestations, diagnosis, and treatment of anthracycline-induced cardiotoxicity" and "Risk and prevention of anthracycline cardiotoxicity" and "Cancer survivorship: Cardiovascular and respiratory issues".)

●Sensorineural hearing loss – Hearing loss due to platinum chemotherapy is a common complication of treatment for high-risk neuroblastoma, occurring in as many as 82 percent of patients who have undergone high-dose chemotherapy with stem-cell rescue [178,181-184]. A full evaluation of hearing should be performed at the end of therapy, and patients should be followed with periodic audiograms thereafter. (See "Overview of neurologic complications of platinum-based chemotherapy".)

●Chronic kidney disease – A number of neuroblastoma survivors have been noted to have decreased kidney function [183,184]. Platinum drugs (carboplatin, cisplatin) are known for nephrotoxicity. Additional kidney damage may result from RT to the adrenal/abdominal primary tumor beds or from vascular injury at the time of tumor resection. For some patients with high-risk neuroblastoma, nephrectomy is necessary to obtain adequate tumor resection. In these circumstances, patients are left with single kidney physiology, placing them at risk for future kidney failure. (See "Overview of kidney disease in patients with cancer", section on 'Chronic kidney disease in patients with cancer'.)

●Endocrinopathies – Multiple endocrine effects are possible, including impaired linear growth, thyroid dysfunction, and incomplete puberty [178,185,186]. While impaired growth and puberty issues can be related to RT (either abdominal RT or TBI) [178], these late effects are also noted in those patients who have not received TBI. The thyroid dysfunction is associated with radioactive iodine in meta-iodobenzylguanidine (MIBG) therapy or scans [178,186]. (See "Endocrinopathies in cancer survivors and others exposed to cytotoxic therapies during childhood".)

●Scoliosis – Scoliosis may occur in a number of neuroblastoma survivors. Laminectomy, which may be required in the treatment of intraspinal tumor, increases the risk of scoliosis because of asymmetric growth of vertebral bodies [146,147,187]. In addition, RT involving the spine seems to have a dose-associated impact on the risk of developing scoliosis [187]. (See "Scoliosis in the adult".)

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Neuroblastomas".)

SUMMARY AND RECOMMENDATIONS

●Overview – This topic is designed to be an overview for the general oncologist on the management of neuroblastoma. Patients should be managed in a clinical setting where appropriate expertise in the treatment of neuroblastoma is available.

●Clinical presentation – Neuroblastomas are clinically heterogeneous tumors, varying in location, histopathologic appearance, and biologic characteristics. Neuroblastomas have a highly variable natural history, which can range from spontaneous regression, to maturation into benign ganglioneuroma, to aggressive disease with metastatic dissemination leading to death. (See 'Introduction' above.)

●Prognostic factors – Key factors influencing the clinical behavior of neuroblastomas include tumor stage and prognostic factors (eg, patient age at diagnosis; pathologic risk classification; cytogenetics and molecular genetics). (See 'Staging system' above and 'Prognostic factors' above.)

●Treatment approach – The treatment of neuroblastoma is determined based on low-, intermediate-, and high-risk categories based on tumor stage and other clinical characteristics at diagnosis (including age and molecular/pathologic features). An updated risk classification system is available (figure 1) that is based on the International Neuroblastoma Risk Group Staging System (INRGSS) (table 1). A previous risk classification schema (table 3) was based on the International Neuroblastoma Staging System (INSS) (table 2). (See 'Treatment' above and 'Risk stratification' above.)

●Low-risk disease – For most children older than one year with low-risk disease (localized tumors without unfavorable characteristics), surgery alone is indicated. Patients treated with surgery typically do not require adjuvant chemotherapy. (See 'Low-risk disease' above and 'Surgery (preferred for most patients with low-risk disease)' above.)

However, exceptions are made for the following patients with low-risk disease, given the high rate of spontaneous regression seen in such cases. (See 'Subgroups that can be managed with observation alone' above.)

•For infants less than six months of age with small (ie, tumor diameter ≤5 cm), asymptomatic L1 adrenal masses, we recommend observation (Grade 1C), forgoing diagnostic biopsy, and delaying therapy until the point of progression. (See 'Infants less than one-year old with localized disease <5 cm' above.)

Some UpToDate experts also offer observation to select children greater than six months and less than one year old with asymptomatic, localized, biopsy-proven neuroblastoma (ie, INSS stage 1, 2, or 3 (table 2), INRGSS stage L1 or L2 (table 1)) with both favorable histology and genomics (eg, without MYCN amplification or segmental chromosomal aberrations). However, this observational strategy is investigational, and other experts may alternatively offer standard treatment approaches (eg, surgery or chemotherapy) in these patients. (See 'Infants less than one-year old with localized disease <5 cm' above.)

•For infants with MS disease without hepatomegaly with tumors that are MYCN non-amplified, hyperdiploid, and favorable histology, we recommend observation (Grade 1C), delaying therapy until progression. (See 'Infants with stage MS (4S) disease without hepatomegaly' above.)

●Intermediate-risk disease – For children with intermediate-risk disease, we suggest chemotherapy, with or without surgical resection, rather than upfront surgical resection (Grade 2C). Radiation therapy (RT) is rarely indicated. (See 'Intermediate-risk disease' above.)

●High-risk disease – For children with high-risk disease, we recommend multimodality treatment rather than less aggressive treatment (Grade 1B). Multimodality treatment typically include induction chemotherapy, surgical resection, tandem autologous hematopoietic stem cell transplantation, and RT to the tumor bed and, potentially, metastatic sites, followed by biologic/immunologic therapy (eg, dinutuximab). Extended maintenance therapy with eflornithine is an option, but its clinical use is not standard across all institutions. With this multimodality approach, long-term survival exceeds 50 percent, which compares favorably with survival rates of approximately 15 percent, prior to adoption of these strategies. (See 'High-risk disease' above.)

●Relapsed and refractory disease – For patients with relapsed and refractory disease, standard treatment options are limited; we typically refer patients for clinical trials, where available. For patients who do not select or are without access to clinical trials, treatment options include dinutuximab plus irinotecan plus temozolomide, I-131 meta-iodobenzylguanidine (MIBG) therapy (where available), or naxitamab. (See 'Relapsed or refractory disease' above.)

●Disease complications – Neuroblastoma can be associated with certain medical complications including spinal cord compression, opsoclonus myoclonus, and, rarely, tumor lysis syndrome and watery diarrhea, hypokalemia, and hypochlorhydria or achlorhydria (WDHA) syndrome. (See 'Medical complications associated with neuroblastoma presentation' above.)

●Prognosis – The prognosis for patients with neuroblastoma depends upon prognostic factors, risk stratification, extent and site of metastases, and treatment received. In general, the younger the age at diagnosis, the better the survival rate. One exception are newborns less than two months of age with stage MS neuroblastoma, who typically present with aggressive disease. (See 'Prognosis' above and 'Newborns' above.)

●Survivorship – Neuroblastoma survivors are at increased risk for long-term treatment-related toxicities, and treating clinicians should be aware of these potential challenges that face survivors and their families. We recommend long-term follow-up based on childhood cancer survivorship guidelines from the Children's Oncology Group (COG). (See 'Long-term toxicities' above.)

ACKNOWLEDGMENT — 

The UpToDate editorial staff acknowledges Stefanie R Lowas, MD, who contributed to earlier versions of this topic review.