This site is intended for health professionals only

Sunday 15 September 2019
Share |

Anti-GD2 mAbs for the treatment of high-risk neuroblastoma

Neuroblastoma is the most common extracranial paediatric solid tumour in Europe. Close to 40% of all neuroblastoma patients are classified as high risk (HR) because the chance of relapse or death within two years of diagnosis is close to 50%. Immunotherapy with anti-GD2 monoclonal antibodies (mAbs) is now considered the only important therapeutic advance in the treatment of HR neuroblastoma in the last decade. The availability and sustainability of these antibodies in the market as well as the cost they represent within an already expensive treatment for neuroblastoma patients are relevant issues affecting parents, physicians, regulatory bodies and insurers. 
 

Neuroblastic tumours

Neuroblastic tumours (NBTs) derive from primordial neural crest cells and are the most common extracranial solid childhood tumours in Europe. Under normal conditions, neural crest cells delaminate and migrate from the dorsal neural tube, and those neuroblastic precursor cells differentiate upon reaching their final embryonic location into tissues and organs that will constitute the sympathetic nervous system. In vitro and in vivo studies have shown that NBTs originate from a block in the process of normal differentiation of these precursor cells.1,2
 
Histologically, NBTs are classified in three categories: neuroblastoma (NB); ganglioneuroblastoma (GNB); and ganglioneuroma (GN). By definition, Schwannian stroma should comprise less than 50% of the tumour tissue to be NB. Undifferentiated NB is composed of neuroblastic cells without identifiable neuropil or rosettes. Tumour cells are small in size and have no discernible cytoplasm. The nuclei are round, have a salt-and-pepper appearance and may contain distinct nucleoli.
 
In these undifferentiated tumours, immunohistochemistry shows a pattern compatible with a presumed immature ganglionic (neuronal) SNS lineage origin. These observations suggest that undifferentiated NB tumours may be locked at an early neuronal differentiation stage, without the capacity to differentiate in response to the factors driving normal sympathetic neuronal differentiation. GNB show well-defined microscopic foci of differentiating neuroblastic cells distributed in a ganglio-neuromatous tissue background. The neuroblastic component of GNB tumours expresses markers reflecting an advanced ganglionic (neuronal) development. 
 
GN are composed predominantly of mature Schwannian stroma and ganglion cells usually surrounded by satellite cells. Mature Schwann cells represent the dominant component of the tumour, characteristically forming multiple fascicles covered with perineurial cells. GN infrequently display neuroendocrine features and usually occurs after the age of four years. Both features suggest that it arises from mature neuronal sympathetic ganglia or adrenal medulla neuronal cells.1,2
 
The incidence of neuroblastoma is 10.2 cases per million children under 15 years of age, and it is the most common cancer diagnosed during the first year of life.3,4 Neuroblastoma is usually diagnosed in very young children; the median age at diagnosis is 17 months.5 The clinical presentation is highly variable, ranging from a mass that causes no symptoms to a primary tumour that causes critical illness as a result of local invasion, widely disseminated disease, or both. Usually the so-called primary tumour arises in the adrenal gland or paravertebral sympathetic ganglia.
 
Two-thirds of neuroblastoma tumours have distant metastases in the bone, bone marrow, lymph nodes, liver or subcutaneous tissue upon diagnosis, whereas lung or central nervous system metastasis are extremely rare.5 One-third of neuroblastoma present as localised with or without involvement of regional lymph nodes. These tumours do not distally metastasise.6
 
There are different ways to stratify neuroblastoma; nonetheless all investigators agree that age, presence or absence of distant metastasis, and MYCN amplification are key factors. Infants under 18 months of age always have a better prognosis compared with older children, not only because neuroblastoma present in young infants more frequently localised but also in the metastatic subgroup. 
 
By contrast, metastatic cases diagnosed beyond six years of age and in adolescents have a more indolent course that eventually will almost always be fatal. Patients with low-risk neuroblastoma have a very good prognosis with a five-year overall survival (OS) of 80–90% treated with minimal or no therapy at all.5 However, patients older than 18 months with metastases and/or patients of any age with MYCN-amplified tumours are considered HR, and their outcome is still poor despite intensive multimodal therapy with a five-year OS <50%.5 Approximately 40% of all neuroblastoma patients are classified as HR, and approximately half of them do not respond to first-line therapy or relapse during the first two years of treatment.
 
The traditional multi-modal therapeutic approach for HR neuroblastoma includes chemotherapy, surgical excision of the primary tumour and radiotherapy. These modalities are most frequently able to drastically reduce the tumour burden in the induction and consolidation phases and can lead to an apparent complete remission of the disease (referred as minimal residual disease (MRD)). 
 
Most cooperative groups would then include high-dose chemotherapy with autologous haemopoietic stem cell rescue as consolidation for HR neuroblastoma patients. HR neuroblastoma patients treated with this so-called standard schema have >50% recurrence rate, which indicates that most therapeutic failures nowadays occur during the stage of MRD.7
 
After a variable period of quiescent (mostly undetectable) disease, many patients relapse, usually with metastatic foci resistant to cytotoxic therapies, and eventually undergo rapid and overwhelming progression. Thus, the major drawback of current therapies is the management of the limited number of cells that escape induction and consolidation therapies.
 
These cells are able to undergo proliferation and/or migration, giving rise to the metastatic recurrence of neuroblastoma. Anti-GD2 immunotherapy is a promising treatment paradigm in this situation. 
 

Immunotherapy

The immune system can be divided into the innate and the adaptive systems, both capable of recognising and eliminating tumour cells. Mechanisms to remove tumour cells are primarily cellular and include CD8+ or effector T lymphocytes and natural killer (NK) cells. The innate immune response includes NK cells and NK-like T lymphocytes, and is responsible for the first non-specific response line, a defence system with mechanisms that operate in a matter of a few hours or days. 
 
In addition to providing direct response to tumours, these cells are important for preparing the adaptive immune response by releasing cytokines that facilitate the activation of T lymphocytes. The adaptive immune response takes several days to develop but is highly specific in its response against antigens and has long-term memory capacity. Once activated, the T lymphocytes develop a very powerful cytotoxic response. These cytotoxic T lymphocytes (CTLs) are capable of killing any cell expressing the antigen bound to major histocompatibility complex (MHC) class A molecules. These cells enable a long-range and memory anti-tumour response. 
 
Neuroblastoma is an ideal malignancy target for immunotherapy because it derives from developing neural crest cells and thus continues to selectively express lineage-specific cell surface markers that are not widely present on mature, non-embryonic tissues. Spontaneous, innate anti-tumour immunity in neuroblastoma has been suspected because some neuroblastomas can spontaneously regress.8 However, an active adaptive immunity against neuroblastoma has been difficult to demonstrate in HR patients. 
 
The large tumour bulk of neuroblastomas and their rapid proliferation overwhelm the immature immune system of the child. Besides, a paucity of somatic mutations makes neuroblastoma poorly immunogenic, and this tumour has developed a sophisticated immunosuppressive microenvironment to ensure that no effective T-cell immunity can develop or become functional.9
 
Immunotherapy has been tested over the last three decades as a potential strategy against MRD in HR neuroblastoma. Most of the clinical experience has focused on mAbs against cell membrane antigens. In 1985, Cheung and colleagues described for the first time four mAbs against, at the time, an unknown glycolipid antigen on the surface of human neuroblastoma cells: GD2.10 Most recent efforts have focused on the discovery of novel cell surface molecules that can be targeted with novel protein-based or cellular immunotherapeutic approaches. One recent example is the identification of the glypican family member 2 (GPC2) as being highly and selectively expressed on most neuroblastomas.11 GPC2 seems to be required for neuroblastoma proliferation and experiments in vitro and in vivo show that it can be targeted with a GPC2-directed antibody–drug conjugate potently cytotoxic to GPC2-expressing neuroblastoma cells.11
 
Disialoganglioside GD2 is a sialic acid-containing glycosphingolipid expressed primarily on the cell surface membrane and plays an important role in the attachment capacity of neuroblastic cells.12 In normal human tissues, GD2 expression is restricted to neurons, skin melanocytes and peripheral pain fibres. GD2 is biosynthesised from precursor ganglioside GD3/GM3 by the b-1,4 N-acetyl-galactosaminyltrasferase (GD2 synthase) and is abundantly expressed in most neuroblastomas regardless of age or stage.13 Two intravenous (IV) anti-GD2 IgG antibodies have been tested extensively in the clinic: chimeric 14.18 (ch14.18) and mouse 3F8.
 
Phase I and II studies of murine IgG2 mAb 14G2a, murine 3F8 and human–mouse chimeric mAb ch14.18 showed clinical responses.14 Ch14.18 was constructed by combining the variable regions of original murine IgG3 anti-GD2 mAb 14.18 and the constant regions of human IgG1. The biological activities of the anti-GD2 mAb ch14.18 in vivo have been demonstrated by the capacity of post-infusion sera to mediate complement-dependent cytotoxicity (CDC) and antibody-dependent cellular cytoxicity (ADCC). 
 
Pharmacokinetic and immunological studies showed the differences between the anti-GD2 mAbs; for example, ch14.18 has longer plasma half-life and less immunogenicity when compared to the murine mAb 14G2a.14 The toxicity profile is common among all the clinically tested anti-GD2 antibodies and include difficult to treat neuropathic pain, tachycardia, hypertension, hypotension, fever and rash. Many of these toxicities are dose-dependent, mainly pain, which is dependent on ADCC and CDC after binding to GD2-positive nerve fibres. 
 
The pain associated with anti-GD2 therapy is similar to other neuropathic pain syndromes and is relatively opioid-resistant. Other less common toxicities include hyponatraemia, hypokalaemia, nausea, vomiting, diarrhoea, serum sickness, and changes in pupil reaction to light and accommodation.15 Importantly, studies of a radiolabelled form of murine anti-GD2 mAb 3F8 indicated that it does not cross the intact blood–brain barrier in mice and humans.16 Long-term neurological impact of anti-GD2 therapy is still under assessment/investigation.
 
By activating ADCC to kill NB, anti-GD2 mAbs are most efficient when effector cell populations including NK, granulocytes, and macrophages, are amplified by cytokines. Because NK cells and granulocytes are effectors for ADCC, the cytokines IL-2 and GM-CSF were administered in combination with anti-GD2 mAbs to enhance their activity. GM-CSF has been shown both in vitro and in vivo to enhance anti-tumoural immunity through direct activation of monocytes, macrophages, dendritic cells, and ADCC and indirect T-cell activation via tumour necrosis factor, interferon, and IL-1.17 IL-2 (aldesleukin) causes activation of NK cells, generation of lymphokine-activated killer cells and augments ADCC. Toxicities include pain, fever, capillary leak, O2 requirement due to capillary leak, hypotension, mild reversible increased hepatic transaminases, and infection.18
 

Anti-GD2 antibodies in clinical practice

Targeted immunotherapy using anti-GD2 mAbs is an important clinical advance in the treatment of HR neuroblastoma. In a landmark study published in 2010 by the cooperative American group COG, the addition of the anti-GD2 monoclonal antibody ch14.18 (dinutuximab) combined with cytokines and isotretinoin improved the survival rates compared with isotretinoin alone in the post-consolidation phase.19 As a consequence of this and subsequent studies, in 2015, dinutuximab was approved in Europe and the US for the treatment of HR neuroblastoma, and is now considered part of the standard of care.20
 
There are two anti-GD2 antibodies approved by the European Medicines Agency (EMA): dinutuximab (Unituxin®) and dinutuximab beta (Qarziba®). Other anti-GD2 antibodies are currently being tested in clinical trials.
 

Dinutuximab (Unituxin)

Dinutuximab is a chimeric monoclonal antibody composed of the variable heavy and light-chain region of the murine anti GD2 mAb14.18 and the constant region of human IgG1 heavy-chain and k light-chain. It was initially produced in the murine myeloma cell line SP2/O. On 10 March 2015 and 14 August 2015, the US FDA and the EMA, respectively, approved IV dinutuximab, in combination with GM-CSF, IL-2 and cis-Retinoic acid (CRA), for the treatment of paediatric patients with HR-NB who achieved at least partial response with prior first-line multi agent, multimodality therapy.20 Dinutuximab is now Unituxin®, a registered trademark from United Therapeutics Company (UTC). 
 
Because of the pain side effects, dosing is limited. The recommended dosage of dinutuximab is 17.5 mg/m2/day administered IV over 10–20 hours for four consecutive days for maximum of five cycles. The infusion of dinutuximab initiated at a rate of 0.875mg/m2/hour over 30 minutes and the rate can be increased gradually, as tolerated, to the maximum rate of 1.75 mg/m2/hour. Dinutuximab should be diluted with 0.9% sodium chloride for injection prior to infusion. The supplied dinutuximab does not require filtration during preparation and does not need protection from the light during administration. Vials should be stored in the original container tightly closed at 2–8ºC. The solution is stable at room temperature for at least 24 hours when diluted to a concentration between 0.044 and 0.56mg/ml.
 
Dinutuximab, similar to all the anti-GD2 immunotherapies tested, is associated with potentially serious side effects. The most common include neuropathic pain, tachycardia, hypertension, hypotension, fever, and urticaria. In the COG Phase III trial, grade 3 or 4 pain was observed in 52% of patients, more frequent during cycle 1 (37%) and decreasing to 14% during cycle 5.19 The most common site of pain was the abdomen. Grade 3 or 4 hypersensitivity reactions were reported in 25% of patients. Consequently, the US label for dinutuximab contains a warning for the risk of infusion-related reactions and neuropathy.21 Other secondary described effects of dinutuximab include fever, hypokalaemia, hyponatraemia, liver dysfunction, hypotension, diarrhoea, and hypoxia. 
 
Recently acute-onset transverse myelitis as a novel infusion-related toxicity of dinutuximab was described.22 Myelitis rapidly improved when the infusion was stopped and corticosteroids were administered resulting in functional recovery. Importantly, this serious but reversible complication has not been observed or reported with other anti-GD2 antibodies. Overall, the pain side effects remain the major drawback of all anti-GD2 antibodies. Although these side effects are not lethal, the emotional burden to the patients, their parents and the health care professionals, is important. 
 
Furthermore, in rare occasions, analgesics and sedatives used to treat the pain side effects can become life threatening. For such reasons, anti-GD2 immunotherapy should be performed in highly specialised centres. It has been shown that pain control (and safety) becomes more proficient with practice and improvement over the years. 
 
The cost for one 17.5mg single-use vial is $7500 approximated wholesaler acquisition cost (WAC) or manufacturer’s published price to wholesalers. In 2016, the market price was being negotiated in Europe, country by country, when dinutuximab beta (Qarziba®) came into play. The European marketing authorisation of Unituxin was withdrawn in early 2017 and it is no longer licensed in Europe. 
 
Since approval, Unituxin® has been widely used in the USA. Since then, a remarkable advance in the use of dinutuximab has occurred. The COG reported the ANBL1221 study whereby they randomised relapsed neuroblastoma patients to receive irinotecan–temozolomide with either temsirolimus or dinutuximab. 
 
Between 22 Feb 2013 and 23 March 2015, 35 patients were randomised. Eighteen patients were assigned to irinotecan–temozolomide–temsirolimus and 17 to irinotecan–temozolomide–dinutuximab. Of the 18 patients assigned to irinotecan–temozolomide–temsirolimus, one patient achieved a partial response. Of the 17 patients assigned to irinotecan–temozolomide–dinutuximab, nine had objective responses, including four partial responses and five complete responses.23 This study has opened the door for further chemoimmunotherapeutic regimens and the potential use of dinutuximab earlier in the treatment schema of HR neuroblastoma.
 

Dinutuximab beta (Qarziba®)

Dinutuximab beta (also called ch14.18/CHO or APN311) is a mAb directed against GD2 produced in a recombinant Chinese hamster ovary (CHO) cell line. Because the glycosylation pattern varies between ch14.18/CHO and ch14.18/SP2/0 (Unituxin®) due to the different production cell lines, which might influence pharmacokinetics, efficacy and safety of the antibody, the two products are not considered biosimilars. However, no direct clinical comparison between the two products has been conducted. 
 
Dinutuximab beta was developed by the Vienna-based biotech company, Apeiron Biologics AG, as a result of a collaborative effort between academic institutions (the SIOPEN group), Austrian private investors, and public and private research initiatives. EUSA Pharma (UK Limited) acquired the exclusive global commercialisation rights of dinutuximab beta from Apeiron Biologics AG in September 2016. On 8 May 2017, the European Commission granted marketing authorisation for ‘dinutuximab beta Apeiron’.
 
The EMA evaluated the clinical efficacy of dinutuximab beta in three studies: APN311-202; 302; and 303. Study 202 was a Phase I/II dose schedule finding study of ch14.18/CHO continuous infusion combined with subcutaneous aldesleukin (IL-2) in patients with primary refractory or relapsed neuroblastoma. Ch14.18/CHO was administered as a ten-day continuous infusion for a total of five cycles, combined with IL-2 and 13-cis RA. A total of 44 patients from Spain, France, Italy, UK, Germany, Israel, and Austria with relapsed/refractory neuroblastoma were enrolled.18
 
At the end of treatment, response was observed in 14/33 patients (42%) with measurable disease. The treatment response was higher in refractory disease (48%; 10/21) than in relapsed disease (33%; 4/12) and the range for the duration of response was from five weeks to three years, median 2.3 years. Study 303 was a ‘Retrospective analysis of data collected during the administration of ch14.18/CHO continuous infusion combined with subcutaneous aldesleukin (IL-2) in patients with HR-NB under a compassionate use program’. 
 
Ch14.18/CHO was given in combination with fixed doses of subcutaneous IL-2, and cis-retinoic acid. Between November 2009 and August 2013, 54 patients were treated. Half of the patients had relapsed NB; 15 (28%) had refractory disease; and 9 (17%) had received first-line neuroblastoma treatment with either complete response or at MRD. At the end of treatment (5–6 cycles) a response (CR+PR) was observed in 12/39 patients (31%) with evidence of disease at baseline while progression occurred in 17/39 patients (44%). In patients with relapse/refractory disease, the response rate was 10/36 (28%). 
 
The 303 Study results were compared with historical controls from a retrospective study performed by Garaventa and colleagues of the Italian Neuroblastoma Registry for patients diagnosed during or after 1999,24 because treatment was deemed comparable to the treatment used prior to immunotherapy in patients included in study APN311-303. Furthermore, comparison was restricted to patients with relapsed neuroblastoma, patients who were ≥1 year of age at initial diagnosis/relapse and who presented with stage 4 at initial diagnosis. 
 
The difference in OS between the two cohorts was highly significant in favour of dinutuximab beta. Finally, APN311-302 (or HR-NBL-1/SIOPEN) evaluated safety and efficacy of ch14.18/CHO from data collected in the high risk neuroblastoma (HRNBL1) study of SIOP-Europe (SIOPEN), a multinational, open-label, randomised, controlled Phase III trial in HR-NB patients conducted by SIOPEN to control minimal residual disease after induction chemotherapy, myeloablative therapy with stem-cell rescue, and radiation therapy. Study APN311-302 evaluated the effect of IL-2 to the ch14.18/CHO + CRA regimen. Patients were randomised to ch14.18/CHO (100mg/m2 per cycle in five daily eight-hour infusions) with or without IL-2 (60 x 106 IU by cycle) and all received 13-cis RA. 
 
Ch14.18/CHO was administered as an eight-hour IV infusion, at a dose of 20 mg/m2/day over five days, every four weeks over five courses. A total of 406 patients from ten countries were enrolled from November 2009 until August 2013. A total of 187 patients received cho14.18 + CRA only (group 1) and 198 cho14.18 + IL-2 and CRA (group 2). Three-year EFS and OS for group 1 was 55.4% and 64.1%; and for group 2, 61.2% and 69.1%, respectively. Historical comparison for the use of dinutuximab beta with patients enrolled in the same SIOPEN protocol but in an earlier phase where these patients did not receive immunotherapy but only CRA showed three-year OS of 59% for the non-immunotherapy group vs 71% for the ch14.18/CHO group, statistically significant. 
 
In studies APN311-202 and 303, the most frequent toxicities were fever, pain, vomiting, cough, increased weight, skin reactions, constipation, tachycardia, hypotension, and capillary leak syndrome. For all three studies, two treatment-related deaths occurred due to capillary leak syndrome and acute respiratory distress. Toxicities were generally more frequent in patients who received IL-2 compared with patients who did not receive IL-2; in particular, capillary leak syndrome, platelet abnormalities, hypotension, infections, nausea or vomiting, fever, and pain related to ch14.18/CHO.25 In the continuous infusion studies APN311-202 and -303, motor and sensory neuropathies were reported with an incidence of 9% and 5%, respectively. Regarding pain, the authors report that in the continuous infusion studies, around 90% of the patients experienced pain in cycle 1. The percentage of patients with pain decreased in subsequent treatment cycles, to about 60% in cycle 5. 
 
The Committee for Medicinal Products for Human Use of the EMA recommended the following indication for dinutuximab beta: treatment of HR neuroblastoma in patients aged 12 months and above, who have previously received induction chemotherapy and achieved at least a partial response, followed by myeloablative therapy and stem cell transplantation, as well as patients with history of relapsed or refractory neuroblastoma, with or without residual disease. Before the treatment of relapsed neuroblastoma, any actively progressing disease should be stabilised by other suitable measures. In patients with a history of relapsed/refractory disease and in patients who have not achieved a complete response after first line therapy, dinutuximab beta should be combined with IL-2. These indications are different from those for dinutuximab, which includes only first-line treatment.
 
Dinutuximab beta (Qarziba) is formulated as a concentrate for solution comprising 4.5mg/ml dinutuximab beta formulated with histidine, sucrose, polysorbate 20, hydrochloric acid for adjustment to pH 6.0, and water for injections. To ensure an extractable volume of 4.5ml, the vials are filled with 4.9ml of final product. 
 
A solution of 0.9% sodium chloride containing 1% human albumin for dilution is used for the preparation of the final solution for infusion. The recommended posology consists of five consecutive courses, each course comprising 35 days. The individual dose is 100 mg/m2 per course and two modes of administration are possible, a continuous infusion over ten days at the daily dose of 10mg/m2, or five daily infusions of 20mg/m2 administered over eight hours, on the first five days of each course.
 
When IL-2 is combined with Qarziba, it should be administered as subcutaneous injections of 6x106 IU/ m2/day, for two periods of five consecutive days, resulting in an overall dose of 60x106 IU/m2 per course. The first five-day course should start seven days prior to the first infusion of Qarziba and the second five-day course should start concurrently with Qarziba infusion (days 1 to 5 of each course).
 
Qarziba is available through a global compassionate use program from EUSA. Commercially it was launched in Germany on 15 December 2017. The cost for one single-use vial is 8695€ approximate wholesaler acquisition cost or manufacturer’s published price to wholesalers.
 

Conclusions

Anti-GD2 mAbs for the treatment of HR neuroblastoma have made a very long journey from discovery to clinical practice. Many hurdles have been overcome duting this time; however, several remain. First, commercial availability of Qarziba – the solely available antibody in the EU – was only launched in Germany on 15 December 2017 and its distribution and access is limited in the EU and the whole of Europe. Therefore, most patients today have no easy access to an approved therapy that has shown clinical benefit at least for a subgroup of HR-NB patients. 
 
Second, toxicity from anti-GD2 immunotherapy is high and for drugs such as Qarziba that require long-term infusions, it potentially implies long hospital stays or the use of electronic pumps to ensure continuous administration of the drug during 24 hours, for patients who already have spent long hours admitted in hospitals from prior treatments. A rational approach to select centres to decrease risks and optimise identification of those patients who benefit most from immunotherapy is required in future, as well as investigation for more effective and less toxic anti-GD2-based immunotherapies.
 

References

1 Mora J, Gerald M. The origin of neuroblastic tumors: clues for future therapeutics. Expert Rev Mol Diagn 2004;4(3):293–302. 
2 Marshall GM et al. The prenatal origins of cancer. Nat Rev Cancer 2014;14(4):277–89.
3 Ries LAG et al. Cancer incidence and survival among children and adolescents: United States SEER program 1975-1995. Bethesda, MD: National Cancer Institute. 
4 Gatta G et al; RARECARE Working Group. Embryonal cancers in Europe. Eur J Cancer 2012;48(10):1425–33.
5 Maris JM. Recent advances in neuroblastoma. N Engl J Med 2010;362:2202–11.
6 Mora J et al. Survival analysis of clinical, pathologic and genetic features in neuroblastoma presenting as local-regional disease. Cancer 2001;91:435–42.
7 Mora J et al. Gerald. Evolving significance of prognostic markers associated with new treatment strategies in neuroblastoma. Cancer Lett 2003;197:119–24.
8 Brodeur GM, Bagatell R. Mechanisms of neuroblastoma regression. Nat Rev Clin Oncol 2014;11(12):704-13
9 Cheung NK, Dyer MA. Neuroblastoma: Developmental biology, cancer genomics, and immunotherapy. Nat Rev Cancer 2013;13(6): 397–411. 
10 Cheung NK et al. Monoclonal antibodies to a glycolipid antigen on human neuroblastoma cells. Cancer Res 1985;45:2642–9.
11 Bosse KR et al. Identification of GPC2 as an oncoprotein and candidate immunotherapeutic target in high-risk neuroblastoma. Cancer Cell 2017;32(3):295–309.e12
12 Hakomori S, Igarashi Y. Functional role of glycosphingolipids in cell recognition and signaling. J Biochem (Tokyo) 1995;118(6):1091–103.
13 Wu Z et al. Expression of GD2 ganglioside by untreated primary human neuroblastomas. Cancer Res 1986;46:440–3.
14 Modak S, Cheung NK. Disialoganglioside directed immunotherapy of neuroblastoma. Cancer Invest 2007;25:67–77.
15 Simon T et al. Consolidation treatment with chimeric anti-GD2 antibody ch14.18 in children older than 1 year with metastatic neuroblastoma. J Clin Oncol 2004;22:3549–57.
16 Cheung NK et al. Monoclonal antibody based therapy of neuroblastoma. Hematol Oncol Clin. North Am 2001;15(5):853–66.
17 Munn DH, Cheung NK. Antibody-dependent antitumour cytotoxicity by human monocytes cultured with recombinant macrophage colony-stimulating factor. Induction of efficient antibody-mediated antitumour cytotoxicity not detected by isotope release assays. J. Exp Med 1989;170:511–26.
18 Ladenstein R et al. Dose finding study for the use of subcutaneous recombinant interleukin-2 to augment natural killer cell numbers in an outpatient setting for stage 4 neuroblastoma after megatherapy and autologous stem-cell reinfusion. J Clin Oncol 2011;29:441–8.
19 Yu AL et al. Anti-GD2 antibody with GMCSF, interleukin-2, and isotretinoin for neuroblastoma. N Engl J Med 2010;363:1324–34.
20 Mora J. Dinutuximab for the treatment of pediatric patients with high-risk neuroblastoma. Expert Rev Clin Pharmacol 2016;9(5):647–53.
21 United Therapeutics Corp. Unituxin (dinutuximab) injection, for intravenous use: US prescribing information; 2015. 
22 Ding YY et al. Transverse myelitis as an unexpected complication following treatment with dinutuximab in pediatric patients with high-risk neuroblastoma: a case series. Pediatr Blood Cancer. 2018;65(1):doi: 10.1002/pbc.26732.
23 Mody R et al. Irinotecan–temozolomide with temsirolimus or dinutuximab in children with refractory or relapsed neuroblastoma (COG ANBL1221): an open-label, randomised, phase 2 trial. Lancet Oncol. 2017;18(7):946–57.
24 Garaventa A et al. Outcome of children with neuroblastoma after progression or relapse. A retrospective study of the Italian neuroblastoma registry. Eur J Cancer 2009;45(16):2835–42.
25 Lode HN et al. Interleukin-2 adds toxicity but no measurable activity in relapsed/refractory neuroblastoma patients treated with long term infusion of anti-GD2 antibody ch14.18/CHO. Pediatr Blood Cancer 2015;62:S4 Abstract PD-057.