Search

PURPOSE Desmoplastic small round cell tumor (DSRCT), a rare sarcoma of adolescents/young adults primarily involving the peritoneum, has a long-term survival of < 20% despite aggressive multimodality treatment. B7H3 is expressed on DSRCT cell surface, providing a target for antibody-based immunotherapy. PATIENTS AND METHODS In this phase I study, we evaluated the safety, pharmacokinetics, and biodistribution of intraperitoneal (IP) radioimmunotherapy (RIT) with the anti-B7H3 murine monoclonal antibody 131I-omburtamab in patients with DSRCT or other B7H3-expressing tumors involving the peritoneum. After thyroid blockade, patients received 131I-omburtamab as a single IP injection at escalated activities from 1.11 to 3.33/GBq/m2. A prior tracer dose of IP 74 MBq124I-omburtamab was used for radioimmuno–positron emission tomography imaging. Each injection was followed by IP saline infusion. RESULTS Fifty-two patients (48, three, and one with DSRCT, peritoneal rhabdomyosarcoma, and Ewing sarcoma, respectively) received IP 131I-omburtamab administered on an outpatient basis. Maximum tolerated dose was not reached; there were no dose-limiting toxicities. Major related adverse events were transient: grade 4 neutropenia (n = 2 patients) and thrombocytopenia (n = 1), and grade 1 (10%) and grade 2 (52%) pain lasting < 2 hours related to saline infusion. Hypothyroidism was not observed, and antidrug antibody was elicited in 5%. Mean (± SD) projected peritoneal residence time was 22.4 ± 7.9 hours. Mean projected absorbed doses for 131I-omburtamab based on 124I-omburtamab dosimetry to normal organs were low and well within tolerable limits. More than 80% 131I remained protein bound in blood 66 hours after RIT. On the basis of peritoneal dose and feasibility for outpatient administration, the recommended phase II activity was established at 2.96 GBq/m2. Patients with DSRCT receiving standard whole-abdominal radiotherapy after RIT did not experience unexpected toxicity. CONCLUSION IP RIT 131I-omburtamab was well tolerated with minimal toxicities. Radiation exposure to normal organs was low, making combination therapy with other anticancer therapies feasible.

Neuroblastoma (NB) is a malignant embryonal tumor of the sympathetic nervous system that is most commonly diagnosed in the abdomen, often presenting with signs and symptoms of metastatic spread. Three decades ago, high-risk NB metastatic to bone and bone marrow in children was not curable. Today, with multimodality treatment, 50% of these patients will survive, but most suffer from debilitating treatment-related complications. Novel targeted therapies to improve cure rates while minimizing toxicities are urgently needed. Recent molecular discoveries in oncology have spawned the development of an impressive array of targeted therapies for adult cancers, yet the paucity of recurrent somatic mutations or activated oncogenes in pediatric cancers poses a major challenge to the evolving paradigm of personalized medicine. Although low tumor mutational burden is a major hurdle for immune checkpoint inhibitors, an immature or impaired immune system and inhibitory tumor microenvironment can further complicate the prospects for successful immunotherapy. In this regard, despite the poor immunogenic properties of NB, the success of antibody-based immunotherapy and radioimmunotherapy directed at single targets (eg, GD2 and B7-H3) is both encouraging and surprising, given that most solid tumor antibodies that use Fc-dependent mechanisms or radioimmunotargeting have largely failed. Here, we summarize the current information on the immunologic properties of this tumor, its potential immunotherapeutic targets, and novel antibody-based strategies on the horizon.

e14502 Background: Intravenous administration of anti-ganglioside GD2 monoclonal antibodies (mAbs) is commonly associated with adverse events (AEs) such as pain and hypertension. Naxitamab is an anti-GD2 mAb intended for outpatient use with a short (30-60 minutes) administration time with 60%-70% of patients in clinical studies experiencing grade 3-4 pain or hypotension. To reduce this risk we assessed a novel administration protocol (StU) that modulates the pharmacodynamics (PD) of naxitamab and mitigates infusion-related reactions. Methods: High-Risk Neuroblastoma (HR-NB) patients in complete remission (CR) received naxitamab as consolidation. Naxitamab cycles comprised priming doses of sc GM-CSF for 5 days at 250 μg/m2/day followed by naxitamab + sc GM-CSF for 5 days at 500 μg/m2/day (days 1-5). Standard naxitamab protocol infusion is provided over 60 minutes at 3 mg/kg/day on day 1 and over 30 minutes on days 3 and 5. Infusions in the StU protocol day 1 were initiated at 1 ml/h with doubling of infusion rate every 15 minutes over 75 minutes (16% of the total dose) and completing the infusion to a final cumulative dose of 3 mg/kg over the remaining 45 minutes; total infusion time of 2h (Table). A faster program (30% of the dose administered in 60 minutes and completed with further 30 minutes) was used for days 3 and 5 infusions. Treatment cycles were repeated every 4 weeks for a goal of 5 cycles. Pharmacokinetics was studied by quantifying serum naxitamab concentrations by ELISA. Infusion related adverse events (AEs) were graded according to the CTCAE v 4.0. Results: 42 HR-NB patients were treated during 2021, 19 with the standard, 15 using the StU, and 8 with both, for a total of 159 cycles (77 StU including 20 cycle one), 477 (231 StU) infusions. All pts presented non-serious CTCAE G1-2 AEs and 7 (37%) CTCAE G3/4 AEs (hypertension x1, hypotension x4, pain x3, airway constriction x5) on the standard protocol. Among the 15 pts on the StU protocol, 1 (6.7%) had CTCAE G3 hypertension. Of the 8 pts who received both types of cycles, 5 (62.5%) had CTCAE G3/4 toxicities (laryngospasm x3; pain x2), 3 with the standard regimen and 2 with the StU protocol. When protocols were compared, the standard regimen cycles (22 cycle one) generated 14.6% (12 of 82) G3/4 AEs whereas the StU 3.9% (3 of 77). Using mixed effects logistic regression analysis, an odds ratio of 0.23 with 95% CI=(0.05, 0.96) and p=0.045 is obtained with the StU cycles reducing the chances to develop G3/4 AEs in 77% compared to the standard regimen. The median serum naxitamab levels pre-StU infusion is 11.46 ug/mL and post infusion 100.95 ug/mL, within the same range as the standard protocol. Conclusions: A pharmacodynamics guided protocol of infusion significantly decreased the severe AEs permitting more tolerable infusions of naxitamab.[Table: see text]

10028 Background: Chemoresistant disease is an obstacle for cure of high-risk neuroblastoma (HR-NB). Anti-GD2 monoclonal antibodies (MoAb) dinutuximab and naxitamab in combination with cytokines are FDA-approved to consolidate remission and for chemorefractory osteomedullary HR-NB, but responses in progressive disease (PD) are rare. We investigated the combination of Humanized anti-GD2 MoAb naxitamab (Hu3F8), Irinotecan, Temozolomide and Sargramostim (GMCSF) in a phase II "HITS" protocol against resistant HR-NB (NCT03189706). Noteworthy differences between HITS and COG protocol ANBL 1221 included higher MoAb and temozolomide dosage and overlap of naxitamab with GMCSF. Methods: Patients were treated at Memorial Sloan Kettering (MSK) on protocol and at Hospital Sant Joan de Déu (HJSD) per protocol on compassionate basis. Salient eligibility criteria included evaluable or measurable chemoresistant disease. Prior anti-GD2 MoAb or irinotecan/temozolomide (IT) therapy was permitted. Each cycle, administered 3-5 weeks apart, comprised irinotecan 50 mg/m2/day intravenously (IV) plus temozolomide 150 mg/m2/day IV or orally (days 1-5); naxitamab 2.25 mg/kg/day IV, days 2,4,8 and 10, and GMCSF 250 mg/m2/day subcutaneously, days 6-10. Toxicity was measured by CTCAE v4.0 and responses by International Neuroblastoma Response Criteria. Objective responses (OR) were also noted. The primary endpoint of the phase II trial was complete (CR) and partial response (PR) after 4 cycles with a desirable rate of 40%; type I and II errors of 10% (undesirable=20%). Results: Of 90 heavily prior-treated patients (38 at MSK evaluated on trial, 52 at HJSD), 8 had HR-NB refractory to induction chemotherapy while 82 had up to 6 prior relapses (median=1). 503 cycles (median 5/patient) were administered. Toxicities included myelosuppression and diarrhea expected with IT, pain and hypertension expected with naxitamab, plus febrile neutropenia in 4%. No other >grade 2 unexpected toxicities occurred; treatment was outpatient. Primary endpoint was reached in the phase II trial: INRC response = 30.6%, lower boundary = 20.4%. In the entire cohort, best responses were CR (26%), PR (11%), mixed response (9%), stable disease (27%) and PD (27%). OR were noted in 64%, with soft tissue (48%) and skeletal MIBG uptake (66%). CR in BM was seen in 57%. OR occurred in patients with MYCN-amplified (25%), refractory (100%) and relapsed (61%) HR-NB; and patients who had previously received I/T (64%) or naxitamab (68%). In patients who had previously received dinutuximab/IT, OR rate to HITS was 42% (5/12). Human anti-human antibody did not develop in any patient (n=50). Conclusions: Naxitamab-based chemoimmunotherapy was safe without immunogenicity. It was effective against chemoresistant HR-NB in all disease compartments even in patients with multiple prior relapses, and in patients who previously received anti-GD2 MoAbs and/or IT.

2010 Background: The prognosis of diffuse intrinsic pontine glioma (DIPG) is dire with a median overall survival less than one-year. 124I-omburtamab is a radiolabeled monoclonal antibody that targets B7-H3 epitope. We evaluated the safety of administering escalating doses and volumes of 124I-omburtamab via convection-enhanced delivery (CED) in children with DIPG. Methods: MSKCC 11-011 trial is a standard 3+3 phase 1, open-label, dose escalation study in patients with non-progressive DIPG. CED of 124I-omburtamab was performed between 4-14 weeks post-external radiation therapy. Nine dose levels of a single injection of 124I-omburtamab (Y-mAbs Therapeutics, USA) (range 0.25 to 8.0 mCi; and volume of infusion (Vi) from 250 to 8,000 µl) have been evaluated so far. Patients were assessed weekly for 30 days. Results: 46 children were evaluable for primary and secondary endpoints. The median age at enrolment was 6.5 years (range 2-17). Two patients have experienced AEs CTCAE grade 3 that were categorized as dose limiting toxicities (DLTs), which led to inclusion of three more patients at both the 4 and 6 mCi dose levels. Eight patients have reported transient AEs of grade 3 considered related to 124I-omburtamab. The acute grade 3 AEs were generally indicative of nervous system effects due to volume intolerance or radiation injury, and included hemiparesis (n = 3), dysarthria (n = 3), ataxia (n = 3), dysphagia (n = 2), muscular weakness (n = 2) and gait disturbance (n = 1). There were no related AEs CTCAE grade 4 or 5. Estimations of distribution volumes based on T2-weighted imaging were linearly related to volume with a mean volume of distribution/volume of infusion ratio (Vd/Vi) between 3 and 3.5. The mean ratio of lesion-to-whole body absorbed dose was ̃1000. Median overall survival from diagnosis across all cohorts was 14.8 months (n = 46, 95% CI 11.5, 16.8) and the survival rate estimates (with 95% confidence intervals) at 1, 2, 3 and 5 years were 0.63 (0.46;0.76); 0.13 (0.05;0.26); 0.08 (0.02;0.19); and 0.04 (0.00;0.16), respectively. Four patients have survived > 3 years; two remain alive at 46 and 96 months and two have died at 43 and 53 months, both with CNS disease outside of the treatment field and one with extra-CNS metastases. Conclusions: 124I-omburtamab via CED into the brain stem of children with DIPG and previously irradiated provides a possibility for improved treatment of DIPG. A dose of 8mCi and an infusion volume of 8,000 µl is considered safe and may provide a distribution volume large enough to cover tumor volumes up to 20 cm3. The median overall survival of all patients included in the trial appears to be increased with 3-4 months compared to historical control data from consortia trials. A phase 2 trial aiming at investigating the efficacy of radiolabeled omburtamab administered via CED is being planned. Clinical trial information: NCT01502917.

Purpose Immunotherapy is a standard of care for children with high-risk neuroblastoma, where bone marrow (BM) is the predominant metastatic site. Early response markers of minimal residual disease (MRD) in the BM that are also predictive of survival could help individualize patient therapies. Patients and Methods After achieving first remission (n = 163), primary refractory disease (n = 102), or second remission (n = 95), children with stage 4 neuroblastoma received anti-GD2 3F8 antibody immunotherapy. BM MRD before 3F8 treatment and after cycle 2 (postMRD) was measured using a four-marker panel (B4GALNT1, PHOX2B, CCND1, and ISL1) by quantitative reverse transcription polymerase chain reaction. Progression-free survival (PFS) and overall survival (OS) were estimated using the Kaplan-Meier method. Prognostic variables were tested in both univariable and multivariable analyses, and MRD markers were further assessed individually and in combination as binary composite (postMRD: 0 and 1) and as equal sum (postMRDSum: 0 to 4) using the Cox regression models, and their predictive accuracy was determined by the concordance index. Results When BM was evaluated after cycle 2, individual markers were highly predictive of PFS and OS. The prediction accuracy improved when they were combined in postMRDSum. A multivariable model taking into account all the variables significant in the univariable analyses identified postMRDSum to be independently predictive of PFS and OS. When the model for OS also included missing killer immunoglobulin-like receptor ligand, human antimouse antibody response, and the enrollment disease status, the concordance index was 0.704. Conclusion BM MRD after two cycles of immunotherapy was confirmed as an early response marker and a consistent independent predictor of survival.

2008 Background: Diffuse intrinsic pontine glioma (DIPG) represents one of the most deadly central nervous system tumors of childhood with a median survival of less than 12 months. Convection-enhanced delivery (CED) has been recently hypothesized as a means for efficiently distributing therapeutic agents within the brain stem. We conducted this study to evaluate CED in children with DIPG. Methods: We performed a standard phase I dose escalation study in patients with non-progressive DIPG 4 to 14 weeks post-completion of radiation therapy. Seven dose levels of a single injection of 124I-8H9 (Omburtamab) (range 0.25 to 4.0 mCi) were studied. Results: 37 children were treated with 34 evaluable for primary and secondary endpoints. The median age at enrollment was 6.8 years old (range 3.2 - 17.9). There was no dose limiting toxicity (DLT). Among adverse events that were at least possibly related to the treatment, there were no grade 4 or 5 events, and only 4 reversible grade 3 events in 4 patients (2 hemiparesis, 1 skin infection and 1 anxiety). Estimations of distribution volumes based on T2-weighted imaging were dose dependent and ranged from 1.5 to 20.8 cm3, and for dose level 7, 10.5 - 19.0 cm3. The mean volume of distribution/volume of infusion ratio (Vd/Vi) was 3.4 ±1.1, and for dose level 7, 3.5 ± 1.0. The mean lesion absorbed dose was 33.3 ± 25.9 Gy, and for dose level 7, 50.1 ± 22.9 Gy. The mean ratio of lesion-to-whole body absorbed dose was 910. The mean volume of distribution/tumor volume ratio on dose level 7 was 82.5%, but the mean tumor overlap was 40.5%. No death occurred as a result of the treatment. Median survival was 15.3 months (n = 29, 95% CI 12.7 - 17.4). Median follow-up time of the 5 surviving patients is 27.2 months (range 11.5 - 72.4). Overall survival rate at 12 months was 64.7% (22/34, 4 alive), and overall survival rate at 24 months 14.7% (5/34, 3 alive). Conclusions: CED in the brain stem of children with DIPG who were previously irradiated is a safe therapeutic strategy. An infusion volume of 4,000 mcl appears to be a reasonable single dose for a target distribution volume but enhanced tumor coverage is likely needed. There seems to be a survival benefit using this therapeutic strategy and outcomes might be dependent on dosimetry and distribution patterns. Clinical trial information: NCT01502917.

10025 Background: Chemoresistant and relapsed disease are major obstacles to curing high-risk neuroblastoma (HR-NB). Anti-GD2 monoclonal antibody (MoAb) is effective in preventing relapse after remission but responses in relapsed or progressive disease (PD) are rare. We investigated the combination of humanized anti-GD2 MoAb naxitamab, (previously termed Hu3F8), irinotecan, temozolomide and sargramostim (GM-CSF): a pilot HITS protocol against resistant HR-NB now expanded to a phase II study (NCT03189706). Methods: Salient eligibility criteria included evaluable or measurable chemoresistant disease. Prior anti-GD2 MoAb and/or irinotecan/temozolomide (I/T) therapy was permitted. Each cycle comprised of irinotecan 50 mg/m2/day intravenously (IV) plus temozolomide 150 mg/m2/day IV or orally (days 1-5); naxitamab 2.25 mg/kg/day IV over 30 minutes, days 2, 4, 8 and 10 (total 9 mg/kg or 270 mg/m2 per cycle), and GM-CSF 250 mg/m2/day subcutaneously, days 6-10. Toxicity was measured by CTCAE v4.0 and responses by modified International Neuroblastoma Response Criteria. Results: Forty-six (23 enrolled on protocol and 23 on compassionate-use basis) heavily prior-treated patients (median age at enrollment: 6.6 years; median number of prior relapses: 2) have received 175 (median 2; range 1-12) cycles to date. At enrollment, 7 patients had HR-NB refractory to induction chemotherapy while 39 had prior relapse. Toxicities included myelosuppression and diarrhea expected with I/T, and pain and hypertension expected with naxitamab. No other > grade 2 related toxicities occurred; treatment was outpatient. Early responses, assessed after 2 cycles, were documented in 18 (39%) patients and were complete (n = 9), partial (n = 8), and mixed (n = 1); 13 patients had stable disease. Responses were achieved in refractory (3/7;43%) and PD (15/39;38%) subgroups, in patients who had previously received I/T (12/34;35%) and/or anti-GD2 MoAb (14/36;39%), and in soft tissue (6/22; 27%) MIBG-avid skeletal sites (20/36;56%) and on bone marrow histology (9/12; 75%). Conclusions: High-dose naxitamab-based chemoimmunotherapy is safe and effective against chemoresistant HR-NB. This ongoing phase II study may define a broader role for naxitamab which was recently granted breakthrough designation by the FDA. Clinical trial information: NCT03189706.

e13592 Background: Tumors metastasizing to the central nervous system (CNS) are associated with significant mortality. We tested the toxicity and dosimetry of intraventricular 131I-labeled monoclonal antibody 8H9 targeting surface glycoprotein B7-H3 in patients with primary or metastatic CNS tumors. Methods: Tumor B7-H3 expression was assessed by immunohistochemistry. CSF flow was determined by 111Indium-DTPA cisternography. 131 patients received 2 mCi tracer of intra-Ommaya 124I- or 131I-8H9 for nuclear imaging followed by a therapeutic injection (10-80 mCi, dose levels 1-8 in 10 mCi increments for phase I patients; expanded cohort 50 mCi/injection) 131I-8H9. Pharmacokinetics were studied by serial CSF and blood samplings over 48 hours. Dosimetry was based on pharmacokinetics and region of interest analyses on serial PET. Toxicity was defined by the CTCAE v.3.0. 8H9 dosimetry and therapy injections were repeated after 1 month if no serious adverse events or progressive disease ensued. Tumor response was determined by clinical, radiographic, cytologic criteria; overall survival was noted. Results: 57 patients (ages 2 – 54 years, median age 11.7 years) received 158 injections Primary CNS diagnoses included medulloblastoma (n = 23), ependymoma (N = 8), chordoma (n = 1), rhabdoid tumor (n = 1), choroid plexus carcinoma (n = 3), ETMR (n = 3), glioblastoma multiforme (n = 1) , PXA (n = 1); metastatic tumors included sarcoma (n = 9), melanoma (n = 4), retinoblastoma (n = 2), and ovarian carcinoma (n = 1). Injections were well tolerated and routinely administered in the outpatient setting. Rare self-limited adverse events included grade 1 or 2 fever, headache, vomiting; 3 injections were associated with grade 3 toxicities requiring discontinuation of therapy including chemical meningitis (n = 2),and increasing communicating hydrocephalus (n = 1), Although not a dose limiting toxicity, myelosuppression occurred in patients who had received craniospinal radiation and at dose levels 6 and higher (≥60 mCi). 16 patients remain alive including patients with high-risk malignancies including choroid plexus carcinoma, ETMR, recurrent ependymoma and recurrent medulloblastoma. Conclusions: We conclude that intraventricular 131I-8H9 is safe, has favorable dosimetry to CSF, and may have clinical utility in the treatment of primary and metastatic CNS tumors. Clinical trial information: NCT00089245.

2533 Background: With the proven success of anti-GD2 monoclonal antibodies in eradicating minimal residual disease in neuroblastoma (NB), exploiting antibody based anti-GD2 in T cell mediated strategies has potential to combat higher disease burden and improve patient outcome. We hypothesized that arming of ex vivo expanded and activated, autologous, blood derived T cells (ATC) with chemically heteroconjugated GD2Bi should redirect them to target NB. In vitro, ATC coated (armed) with 50 ng/106 cells of GD2Bi exhibited specific killing of NB and osteosarcoma (OS) cell lines. Methods: In this phase I study (NCT02173093), patients with GD2-positive tumors received 8, biweekly infusions of GD2BATs + daily low-dose IL-2 and biweekly granulocyte-macrophage colony stimulating factor (GM-CSF). The study followed the standard 3+3 design with dose levels of 40, 80, and 160 x 106 GD2BATs/kg/infusion. Results: Twelve patients (NB = 7, OS = 3, Desmoplastic Small Round Cell Tumor = 2) were enrolled from 11/2013 to 12/2017 and 9 completed therapy. Adequate ATCs could not be grown in one patient and two patients did not complete 8 infusions because of rapid disease progression. Infusions were given in outpatient settings. All patients developed a mild, dose-independent and manageable form of cytokine release syndrome with grades 2-3 fevers/chills, headaches and occasional hypotension for up to 48 hours after infusion. No patients developed significant pain. Maximum tolerated dose was not reached. Evidence of activity was seen in several patients including one patient with OS who had a PET response, one patient with NB who had complete bone marrow response (this patient had remained progression free for 2.5 years after completion of infusions), and another NB patient who had a minor response on MIBG scan. Four patients with NB are currently alive after additional therapies at 12, 14, 18, and 47 months post BAT infusions. Conclusions: Autologous T cells from heavily pretreated patients could be expanded ex vivo to large numbers, armed with GD2Bi, cryopreserved and thawed for safe IV administration up to total dose of 1.28x109/kg. Ongoing phase II arm of the trial will focus on evaluation of clinical activity of GD2BATs in patients with NB. Clinical trial information: NCT02173093.

Purpose Tumors metastasizing to the CNS and leptomeninges (LM) are associated with significant mortality. We tested the toxicity, pharmacokinetics, and dosimetry of intraventricular iodine-131–labeled monoclonal antibody 3F8 (131I-3F8) targeting GD2-positive CNS/LM disease in a phase I clinical trial. Patients and Methods Adequate CSF flow was determined by pretreatment indium-111-DTPA studies. Fifteen patients received a tracer (1 to 2 mCi) and therapeutic injection (10 to 20 mCi) of intra-Ommaya 131I-3F8. 131I-3F8 pharmacokinetics were studied by serial CSF and blood samplings. Dosimetry was based on pharmacokinetics and region of interest (ROI) analyses on whole-body gamma camera scans. Tumor response was determined by clinical, radiographic, and cytologic criteria. Results Total absorbed CSF dose was 1.12 to 13.00 Gy by sampling and 1.00 to 13.70 Gy by ROI data. Average dosimetry ratio (Gy/mCi) of the therapy/tracer administration was 0.88 (± 0.58) and 1.08 (± 0.66) based on CSF pharmacokinetics and ROI analysis, respectively. CSF half-life by sampling was 3 to 12.9 hours. Toxicities included self-limited headache, fever, and vomiting. Dose-limiting toxicity was reached at the 20-mCi dose, when transient elevations in intracranial pressure and chemical meningitis were seen. Three of 13 assessable patients achieved objective radiographic and/or cytologic responses. No late toxicities have been seen in two patients who remain in remission off therapy for more than 3.5 years. Conclusion Intra-Ommaya 131I-3F8 was generally well tolerated; the maximum-tolerated dose was 10 mCi. A high CSF-to-blood ratio was achieved. Tracer studies reliably predicted the therapeutic dose to the CSF. Radioimmunoconjugates targeting GD2 may have clinical utility in the treatment of CNS/LM malignancies.

Purpose To report on an irinotecan and temozolomide regimen for neuroblastoma (NB). Quality of life and minimizing toxicity were major considerations. Patients and Methods The plan stipulated 5-day courses of irinotecan 50 mg/m2 (1-hour infusion) and temozolomide 150 mg/m2 (oral) every 3 to 4 weeks, with a pretreatment platelet count more than 30,000/μL. Granulocyte colony-stimulating factor was used when the absolute neutrophil count was less than 1,000/μL. Results Forty-nine NB patients received 1 to 15 courses (median, 5). Gastrointestinal and myelosuppressive toxicities were readily managed. Lymphocyte responses to phytohemagglutinin after 2 to 10 courses (median, 3.5) were normal in 10 of 10 patients treated after nonimmunosuppressive therapy, and normalized in five of seven patients first treated less than 2 months after high-dose alkylators. Of 19 patients treated for refractory NB and assessable for response, nine showed evidence of disease regression, including two complete responses and seven objective responses. Of 17 patients treated for progressive disease, three showed evidence of disease regression, including one partial response and two objective responses. Multiple courses entailed no cumulative toxicity and controlled disease for prolonged periods in many patients, including some who were unable to complete prior treatments because of hematologic, infectious, cardiac, or renal problems. Conclusion This regimen has anti-NB activity, spares vital organs, is feasible with poor bone marrow reserve, causes limited immunosuppression, and allows good quality of life.

Purpose: The International Neuroblastoma Response Criteria (INRC) recommend, but do not make mandatory, metaiodobenzylguanidine (MIBG) scans. We present the first report on the effect of MIBG scans on the classification of response to dose-intensive induction therapy. Patients and Methods: After dose-intensive induction and before consolidative therapy, 162 Memorial Sloan-Kettering Cancer Center (MSKCC) patients with high-risk neuroblastoma (NB) had MIBG scans (99 with 131I, 63 with 123I), computed tomography, 99mTc-bone scan, bone marrow (BM) tests, and urine catecholamine measurements. Induction included high-dose cyclophosphamide (140 mg/kg) plus other agents and high-dose cisplatin (200 mg/m2)/etoposide (600 mg/m2). Results: In 90 patients treated with dose-intensive therapy from diagnosis at MSKCC, the use of MIBG scintigraphy increased the incomplete response numbers from 14 (15.5%) to 20 (22%), giving a complete remission/very good partial remission (CR/VGPR) rate of 78%. In 72 patients treated before referral to MSKCC for intensified therapy, MIBG findings changed the response classification of one patient; the CR/VGPR rate was 43%. MIBG scans showed no BM disease in 15 of 38 patients with histologically evident NB in BM but did show uptake consistent with BM involvement in five patients who had no NB observed in BM tests. Conclusion: With the less effective therapy consequent to the intensification of induction only after initial exposure to standard-dose chemotherapy, MIBG scintigraphy merely confirms the findings of other staging modalities for detection of relatively widespread residual NB. However, when dose-intensive therapy is initiated at diagnosis, the reliable achievement of major disease responses makes extensive BM testing and MIBG scintigraphy prerequisites for accurate determination of disease status.

10545 Background: Neuroblastoma metastatic to the central nervous system (CNS NB) is associated with significant mortality (median survival < 6 months, < 10% survival at 36 months). Intraventricular compartmental radioimmunotherapy (cRIT) with radio-iodinated murine IgG1 monoclonal antibody 131I-8H9 targeting tumor cell-surface glycoprotein B7-H3 offers a therapeutic strategy. We analyzed overall survival of patients with CNS NB treated with intraventricular 131I-8H9 cRIT at Memorial Sloan Kettering Cancer Center (MSK) since 2003. Methods: After radiographic and/or pathologic confirmation of CNS NB, and assessment of adequate CSF flow, cRIT eligible patients underwent treatment on an IRB-approved protocol with either temozolomide/irinotecan-based CNS salvage regimen incorporating craniospinal radiation therapy, 131I-8H9 cRIT plus systemic immunotherapy (group 1), or non-regimen therapies with 131I-8H9 cRIT (group 2). cRIT administration involved a 2mCi tracer of 124I- or 131I-8H9 with nuclear imaging and CSF sampling for dosimetry followed by 1 or 2 therapeutic injections up to 70 mCi 131I-8H9. Disease surveillance included serial MR brain/spine, MIBG, CT, and bone marrow evaluation. Data are presented as overall survival after detection of CNS metastasis. Results: 105 patients with CNS NB were evaluated;80 patients (76%) were treated (57 group 1, 23 group 2). Of the 25 patients who were not eligible for cRIT, survival averaged 8.6 months. Of 19 patients with radiographic evidence of disease at the time of cRIT, 7 (36%) demonstrated post cRIT radiographic improvement. At analysis, 45/80 (56%) patients were alive 4.8–152 months (median 58 months) after CNS metastasis, including 36 (45%) at 36 months and 23 (29%) > 60 months. Subgroup analyses of 131I-8H9–treated patients identified age at NB diagnosis (≤18 months), relapse restricted to CNS and group 1 status as factors positively correlated with survival. Conclusions: 76% of patients with CNS NB treated at MSK received 131I-8H9 cRIT, and approximately half completed multimodality CNS salvage regimen with 131I-8H9 cRIT. Despite advanced CNS involvement, over 50% of patients treated with 131I-8H9 cRIT are still alive and nearly 50% have survived at least 36 months. Clinical trial information: NCT00089245.

10550 Background: The diagnosis of NB in adulthood is rare and little is known about its biology and clinical course. There is no established therapy for adult NB. Anti-GD2 immunotherapy is now standard in children with HR-NB but its use has not been reported in adults. Methods: After obtaining IRB waiver, records of all patients with adult-onset (≥18 years) NB seen at MSKCC between 1983 and 2015 were reviewed. Overall survival (OS) was tested by log-rank test. Cox-regression was used for multivariate analysis. Results: The subjects were 42 adults (median: 25; range18-71 years); 23 male and 19 female. Five, 1, 1 and 35 patients had INSS stage 1, 2, 3 and 4 disease, respectively. Genetic abnormalities included somatic ATRX (59%) and ALK mutations (43%) but not MYCN-amplification. 16 patients remain alive at a median follow-up of 5.3 years. OS for non-stage 4 patients was superior to stage 4 (median survival 14.6 vs 5.3 years; p < 0.05). However 5/7 patients with < stage 4 NB progressed to stage 4. Among 35 stage 4 patients, 4 achieved complete remission (CR) after induction chemotherapy and surgery, 11 underwent autologous stem cell transplant (ASCT) and 15 received multiple cycles of anti-GD2 antibodies 3F8 or hu3F8 without complications. In univariate analysis, patients ≤ 29 years old (n = 24) at diagnosis, those achieving CR, and those receiving anti-GD2 antibodies had superior OS (p < 0.05 for each). ASCT was not beneficial (p = 0.3 for ASCT vs no ASCT). For stage 4 patients, anti-GD2 immunotherapy was associated with favorable OS in multivariate analysis (95% CI of anti-GD2 antibody: 1.270 to 7.990). Conclusions: Adult-onset stage 4 NB demonstrates a high incidence of somatic mutations and is only partially chemosensitive. However, 3F8/hu3F8-based anti-GD2 immunotherapy appears to improve long-term survival and is well tolerated.

2010 Background: Diffuse intrinsic pontine glioma (DIPG) represents one of the most deadly central nervous system tumors of childhood with a median survival of less than 12 months. Convection-enhanced delivery (CED) has been recently hypothesized as a means for augmenting distribution of therapeutic agents within the brain stem. We conducted this study to evaluate CED in children with DIPG. Methods: We performed a standard 3+3 phase I, open-label, dose escalation study in patients with non-progressive DIPG 4 to 14 weeks post-completion or radiation therapy. Seven dose levels of a single injection of 124I-8H9 (range 0.25 to 4.0 mCi, 250 to 4000 mcl) were studied. Results: 25 children were treated. The average age at enrollment 8 years old (range 3-17). There was no dose limiting toxicity (DLT) and adverse events were limited to grade 1 or 2 (CTCAE v4.0). Estimations of distribution volumes were dose dependent and ranged from 1.5 to 20.1 cm3. The mean volume of distribution/volume of infusion (Vd/Vi) was 3.4 (SD 1.2). The mean lesion absorbed dose was 1527 rad/mCi. The mean tumor coverage on dose level 7 was 107%. Conclusions: CED in the brain stem of children with DIPG who were previously irradiated is a safe therapeutic strategy. Up to 4 mCi of 124I-8H9 was well tolerated. An infusion volume of 4000 mcl appears to be a reasonable single dose for good tumor coverage. PET-based dosimetry validates the conceptual basis for direct drug delivery. Based on our finding CED merits further exploration in early phase clinical trials for children with DIPG. Clinical trial information: NCT01502917.

PURPOSE: Although positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose (18F-FDG) has a major impact on the treatment of adult cancer, the reported experience with extracranial tumors of childhood is limited. We describe a role for PET in patients with neuroblastoma (NB). PATIENTS AND METHODS: In 51 patients with high-risk NB, 92 PET scans were part of a staging evaluation that included iodine-123 or iodine-131 metaiodobenzylguanidine (MIBG) scan, bone scan, computed tomography (and/or magnetic resonance imaging), urine catecholamine measurements, and bone marrow (BM) examinations. The minimum number of tests sufficient to detect NB was determined. RESULTS: Of 40 patients who were not in complete remission, only 1 (2.5%) had NB that would have been missed had a staging evaluation been limited to PET and BM studies, and 13 (32.5%) had NB detected by PET but not by BM and urine tests. PET was equal or superior to MIBG scans for identifying NB in soft tissue and extracranial skeletal structures, for revealing small lesions, and for delineating the extent and localizing sites of disease. In 36 evaluations of 22 patients with NB in soft tissue, PET failed to identify only two long-standing MIBG-negative abdominal masses. PET and MIBG scans showed more skeletal lesions than bone scans, but the normally high physiologic brain uptake of FDG blocked PET visualization of cranial vault lesions. Similar to MIBG, FDG skeletal uptake was diffusely increased with extensive or progressing BM disease but faint or absent with minimal or nonprogressing BM disease. CONCLUSION: In the absence or after resolution of cranial vault lesions, and once the primary tumor is resected, PET and BM tests suffice for monitoring NB patients at high risk for progressive disease in soft tissue and bone/BM.

PURPOSE: To describe the efficacy of oral etoposide against resistant stage 4 neuroblastoma. PATIENTS AND METHODS: Patients with refractory or recurrent stage 4 neuroblastoma were treated with etoposide 50 mg/m2 taken orally each day, in two or three divided doses, for 21 consecutive days. Treatment could be repeated after a 1-week period. Extent-of-disease studies included imaging with 131-iodine-metaiodobenzylguanidine and extensive bone marrow (BM) sampling. RESULTS: Oral etoposide was used in 20 children between the ages of 2 and 11 years (median, 6 years). Prior treatment included high doses of alkylating agents and a median of 4.5 cycles of etoposide-containing chemotherapy, with cumulative etoposide doses of 1,800 mg/m2 to 3,935 mg/m2 (median, 2,300 mg/m2). Oral etoposide produced antineuroblastoma effects in four of four children with disease refractory to intensive induction treatment; sampling variability could account for resolution (n = 3) or reduction (n = 1) of BM involvement, but improvement in other markers also occurred. Antineuroblastoma effects were also evident in five of five children with asymptomatic relapses after a long chemotherapy-free interval: BM disease resolved and all other disease markers significantly improved in two patients, and disease markers improved or stabilized in three patients on treatment for more than 6 months. In these nine patients, extramedullary toxicity was absent, neutropenia did not occur, transfusional support was not needed, and preliminary data suggested little immunosuppression (phytohemagglutinin responses). Oral etoposide was ineffective in all (11 of 11) patients with rapidly growing tumor masses. CONCLUSION: Given the absence of toxicity to major organs, the minimal myelosuppression or immunosuppression, and the antineoplastic activity in patients with low tumor burdens after high-dose chemotherapy, limited use of low-dose oral etoposide should be considered for inclusion in postinduction consolidative treatment programs aimed at eradicating minimal residual disease.

PURPOSE To assess treatment-related myelodysplasia/leukemia (t-AML) in neuroblastoma patients by a review of the Memorial Sloan-Kettering Cancer Center (MSKCC) data and the literature. PATIENTS AND METHODS We studied 380 previously untreated and treated MSKCC patients. Low-risk patients received no cytotoxic therapy. High-risk patients received the N4, N5, or N6 regimens. Dosing per cycle and cumulative dosing of leukemogenic agents peaked with N6, which included four cycles of cyclophosphamide 4,200 mg/m2 and doxorubicin 75 mg/m2, plus three cycles of cisplatin 200 mg/m2 and etoposide 600 mg/m2. We reviewed the literature. RESULTS t-AML occurred in six MSKCC patients, which included three of 53 patients in whom the only chemotherapy consisted of N6, and three patients treated for relapsed or refractory neuroblastoma; no case of leukemia emerged among the 50 low-risk patients. Four cases were found incidentally in routine follow-up bone marrow tests. The 36-month cumulative incidence of t-AML in the N6 cohort was 7% (95% confidence interval, 0 to 15). Published data parallel the MSKCC experience in that t-AML after neuroblastoma was once rare but has become less so since the mid-1980s, when the intensified use of topoisomerase-II inhibitors and alkylators first gained wide acceptance and produced better response rates and longer survival. CONCLUSION Neuroblastoma itself is not associated with a host susceptibility to leukemia. However, current neuroblastoma treatment programs that use high-dose cyclophosphamide, cisplatin, and topoisomerase-II inhibitors may entail a considerable risk for t-AML. The incidence of t-AML in neuroblastoma patients may be underestimated because treatment and clinical factors can mask its presence. Efforts to devise effective but less leukemogenic treatment for neuroblastoma or to truncate leukemogenic therapy, eg, by exploiting molecular techniques for the early identification of complete remission, are warranted.

PURPOSE To measure the sensitivity of histologic examination in detecting metastatic solid tumor in bone marrow. PATIENTS AND METHODS A total of 145 patients with stage 4 neuroblastoma underwent 840 marrow examinations, each consisting of six sites (four aspirates and two biopsies), from October 1990 to June 1996 at Memorial Sloan-Kettering Cancer Center. Metastasis was detected by either histology (aspirate by Wright-Giemse and biopsy by Hematoxylin-Eosin stains) or immunostaining of aspirates using anti-G(D2) monoclonal antibodies. RESULTS The absence of tumor by histology at a single marrow site was a poor guarantee of the absence of disease. The number of false-negative sites increased as the percent of G(D2)-positive tumor cells in the marrow decreased: zero of six if tumor cell count was > or = 1%, and approximately six of six sites if < or = 0.003%. Sensitivity was comparable between marrow aspirate and biopsy. A lower bound (LB) for the probability of false-negative histology was calculated from the (1) discordance among the six marrow samplings and (2) comparison with immunofluorescence. When disease was extensive (eg, at diagnosis), the LB was 0.13 and 0.3, respectively. After treatment, it increased to 0.37 and 0.8. Examining multiple marrow sites can decrease the LB to 0.15. However, at least three sites have to be negative at relapse, six at diagnosis, and more than 50 during treatment or off-therapy follow-up. The marginal decrease in the LB by additional samplings rapidly diminished to less than 0.05 after two sites. CONCLUSION Except at diagnosis and relapse when gross disease is present, marrow sampling by histology has limited sensitivity. Current practice grossly underestimates the true prevalence of marrow disease.