Targeting GD2-positive Refractory/Resistant Neuroblastoma and Osteosarcoma with Anti- CD3 x Anti-GD2 Bispecific Antibody Armed T cells

Background: Since treatment of neuroblastoma (NB) with anti-GD2 monoclonal antibodies provides a survival benefit in children with minimal residual disease and our preclinical study shows that anti-CD3 x anti-GD2 bispecific antibody (GD2Bi) armed T cells (GD2BATs) were highly cytotoxic to GD2+ cell lines, we conducted a phase I/II study in recurrent/refractory patients to establish safety and explore the clinical benefit of GD2BATs. Methods: The 3+3 dose escalation study (NCT02173093) phase I involved 9 evaluable patients with NB (n=5), osteosarcoma (OST) (n=3), and desmoplastic small round cell tumors (DSRCT) (n=1) with twice weekly infusions of GD2BATs at 40, 80, or 160 x 106 GD2BATs/kg/infusion with daily interleukin 2 (300,000 IU/m2) and twice weekly granulocyte-macrophage colony stimulating factor (250 μg/m2). Phase II portion of the trial was conducted in patients with NB at the dose 3 level of 160 x 106 GD2BATs/kg/infusion but failed to enroll the planned number of patients. Results: Nine of 12 patients in the phase I completed therapy. There were no dose limiting toxicities (DLTs). All patients developed mild and manageable cytokine release syndrome (CRS) with grade 2-3 fevers/chills, headaches, and occasional hypotension up to 72 hours after GD2BAT infusions. GD2-antibody associated pain was not significant in this study. The median OS for patients in the Phase I and limited Phase II was 18.0 and 31.2 months, respectively, whereas the combined OS was 21.1 months. There was a complete bone marrow response with overall stable disease in one of the phase I patients with NB. Ten of 12 phase II patients were evaluable for response: 1 had partial response. Three additional patients were deemed to have clinical benefit with prolonged stable disease. More than 50% of evaluable patients showed augmented immune responses to GD2+ targets after GD2BATs as measured by interferon-gamma (IFN-γ) EliSpots, Th1 cytokines, and/or chemokines. Conclusions: Our study demonstrated safety of up to 160 x 106 cells/kg/infusion of GD2BATs. Combined with evidence for the development of post treatment endogenous immune responses, this data supports further investigation of GD2 BATs in larger Phase II clinical trials.


Introduction
Despite intensive chemotherapy, autologous stem cell transplant (SCT) 1,2 , and/or anti-GD2 monoclonal antibodies therapy, many children with high-risk disease NB and OST develop recurrent disease 3,4 which is associated with dire outcomes 1,5,6,7 .There have been no new approved drugs for OST since the 1980's 5 .
GD2 is highly expressed on nearly all NB and 80% of OST 8,9 .Antibody-dependent cellular cytotoxicity by anti-GD2 mAbs is effective against minimal residual disease in NB 2,11,12 .We postulated that it would be feasible, safe, and effective to target NB and OST with activated autologous T-cells (ATC) armed with GD2Bi (GD2BATs).Our preclinical studies showed that GD2BATs kill GD2 + tumor targets, proliferate, and induce release of Th 1 type cytokines upon engagement of GD2 + tumor targets 13 .In our previous clinical trials re-directing the non-MHC restricted cytotoxicity of ATC with bispeci c antibody (BiAb) by engaging the TCR and tumor antigen on tumors has provided encouraging clinical and anti-tumor immune responses in breast 14,15 , prostate cancer 16,17 pancreatic cancer 18 , non-Hodgkin's lymphoma 19 , and multiple myeloma 20 .Infusions of BiAb armed ATC (BATs) induced endogenous responses including innate natural killer cell (NK) and cytokine/chemokine responses suggesting in vivo immunization of patients against their own tumor-associated antigens 14,18,21 .
In this study, we tested combination immunotherapy consisting of GD2BATs 13,22 , interleukin-2 (IL-2), and granulocyte-macrophage-colony stimulating factor (GM-CSF) in a phase I/II trial to determine safety, estimate e cacy, and evaluate immune responses.The feasibility and safety pro le reported here together with the evidence of therapy-induced immune responses and observations of clinical activity support further development of the GD2BAT approach in NB.

Production of GD2BATs
Manufacture of GD2Bi and GD2BATs was performed under BB-IND #15827 sponsored by L.G. Lum in the KCI cGMP facility, Detroit, MI and in the UV Center for Human Therapeutics, Charlottesville, VA under FDA Drug Master le File #19248.Clinical-grade naxitamab [humanized 3F8] was chemically heteroconjugated to clinical-grade anti-CD3 mAb (OKT3; Miltenyi, Auburn, CA) to produce GD2Bi.Lymphocytes (derived from PBMCs) were collected from the patients via apheresis, activated with 20 ng/mL of OKT3, expanded using 100 IU/mL of IL-2 and cultured at a concentration of 1 x 10 6 mononuclear cells/mL in RPMI medium supplemented with 2% human serum and 1% L-glutamine for 14-18 days.The ATC were then harvested, armed with 50 ng of GD2Bi per 10 6 ATC to create GD2BATs, dispersed into equal parts and cryopreserved.When ready for release for infusion, quality control assessments were performed on thawed product: included viability, proportion of CD4 + and CD8 + cells, in vitro cytotoxicity against KCNR GD2 + NB cell line (by mixing and incubating patients' PBMCs with targets), testing mycoplasma, bacterial and fungal contamination, and endotoxin.

Human Subjects.
Patients were enrolled in a Phase I/II clinical trial at CHM, UV, and MSK between November 2013 and January 31, 2022.The protocol (Clinicaltrials.govidenti er NCT02173093) was reviewed and approved by the FDA, protocol review committees, and institutional review boards at respective institutions.All guardians/patients signed informed consent before enrollment.Eligible patients were age ≤30 years at the time of enrollment and had recurrent or refractory NB, OST, or desmoplastic small round cell tumors (DSRCT).Only patients with NB were eligible for the Phase II portion.Depending on the tumor type, patients had con rmation of evaluable or measurable disease by bone marrow morphology, computed tomography, magnetic resonance imaging, and/or iodine-123 metaiodobenzylguanidine (MIBG) or positron emission tomography scans.Adequate performance score (Lansky or Karnofsky score ≥ 70) and adequate organ function (total bilirubin ≤ 1.5 times the institutional upper limit of normal (ULN), ALT ≤ 3 times the ULN, creatinine ≤ 1.5 times the ULN, and a cardiac shortening fraction ≥ 28% by echocardiogram) were required.Patients had to have an absolute neutrophil count ≥ 500/mm 3 , absolute lymphocyte count ≥ 500/mm 3 , and unsupported platelet count ≥ 50,000/mm 3 (≥ 20,000/mm 3 with no more than one transfusion per week for NB patients with known bone marrow involvement.A minimum of two weeks from the last chemotherapy and immunotherapy and one week from hematopoietic growth factors were required for enrollment.
Patients who had been previously treated with anti-GD2 antibodies were eligible.Date of data cutoff for this analysis was January 31, 2022.

Clinical trial design
The phase I study followed a standard 3 + 3 dose escalation design with dose levels of 40, 80, and 160 x 10 6 GD2 BATs/kg/infusion twice per week for total dose of 0.32, 0.64 or 1.28 x 10 9 cells/kg in 8 divided doses.Low-dose IL-2 (300,000 IU/m 2 /day) and biweekly subcutaneous GM-CSF (250 µg/m 2 /dose) were administered beginning three days before the rst GD2BAT infusion and ending one week after the last GD2BAT infusion (Fig. 1A).
A maximum of 18 patients were planned for the phase I study and 22 patients were needed for the Phase II cohort to con rm toxicity and estimate clinical e cacy.The primary endpoints were safety (Phase I) and ORR (PR + CR) (Phase II).Secondary endpoints included: OS, immune responses, persistence of infused cell, and early PET response after infusions.In the interval (up to 90 days) between apheresis and initiation of protocol therapy required for GD2BATs production, bridging chemotherapy with a 2-week washout was allowed.If anti-GD2 mAbs were administered during bridging, 6-week washout was required.
Protocol eligibility was re-con rmed with laboratory testing, imaging and bone marrow evaluation as needed prior to commencing protocol therapy.

Toxicity assessment.
Toxicity was assessed using Common Terminology Criteria for Adverse Event (CTCAE) v5.0 and cytokine release syndrome was graded as per the American Society for Transplantation and Cellular Therapy consensus criteria (ASTCT) 23 .Based on prior experience with BATs 14,[18][19][20][21]24 , Grade 3 chills, fever, headache, nausea, emesis, and hypotension lasting < 72 hours were excluded from the de nition of dose-limiting toxicities (DLTs). Fo this study, Grade 3 fever, nausea, or vomiting were de ned as DLTs only if persisting > 72 hours.Patients were seen twice weekly, during their BAT infusion days, for evaluation of toxicities.

Response evaluation.
Imaging and BM testing (in patients with NB) for disease evaluation was performed at enrollment, after bridging therapy (prior to protocol therapy), and 8-12 weeks after the rst GD2BAT infusion.Response assessment in NB was graded using the revised International Neuroblastoma Response Criteria (INRC) 27 .RECIST criteria were used for assessment of radiological response of measurable target lesions in non-NB tumors.

Correlative Studies
GD2BAT persistence studies and immune evaluations.
To assess the persistence of BATs, PBMCs from patients were stained with anti-mouse:IgG2a antibody (present on OKT3) 14 or alternatively, anti-idiotypic antibody A1G4 speci c for hu3F8 on the surface of circulating T cells at 2 and 72 hours post GD2 BAT infusion.
Patient PBMC samples were obtained prior to GD2 BAT infusion # 1 and then 1 month after the last infusion and evaluated for speci c cytotoxic lymphocyte (CTL) activity using interferon-gamma (IFN-γ) enzyme-linked immunosorbent spots (ELISpots) induced by GD2-expressing NB (KCNR) and OST (MG-63) cell lines and natural killer (NK) -targeted cell line K562 13 .

Statistical Analysis.
Immune evaluations included evaluating means, standard deviations, medians, and the distributions of the data to ascertain whether normal theory methods are appropriate.Paired t-test or Wilcoxon signed-ranks test were used for comparative analyses between each pre-and post-immunotherapy (IT) time point for each patient.With a single stage Phase II trial design, it was estimated that a cohort of twenty-two Phase II patients would have 80% power at a signi cance level of 0.05 to detect 25% difference in response rate compared with a historical response rate of 15% or less.Due to slow recruitment, the Phase II portion of the trial only enrolled half of the planned number of patients (12 of planned 22 enrolled and 10 completed therapy) and responses after IT were analyzed descriptively without applying initially planned statistical methods in the study protocol.

Patient characteristics.
Twelve patients (7 NB, 3 OST and 2 DSRCT) were enrolled on the Phase I study between November 2013 and December 2017, and 9 completed therapy.Twelve additional patients with NB were enrolled in the Phase II portion, and 10 completed protocol therapy.Table I summarizes the patient characteristics, diagnoses, prior therapy, product characteristics, and survival of the phase I/II patients.Reasons for not receiving therapy or completing 8 infusions in the Phase I included: failure to grow adequate ATC in 1 NB patient, rapid disease progression before GD2BATs in 1 patient with NB; and rapid disease progression leading to death after the rst 3 infusions of GD2BATs in 1 patient with DSRCT.In addition, 2 patients enrolled into Phase II portion did not undergo T cell collection and did not receive therapy because of rapid disease progression after enrollment.All patients were heavily pretreated.Patients with OST and DSRCT had 2-3 and 1-2 lines of prior therapy, respectively, prior to GD2BATs.All NB patients in the Phase I portion had prior myeloablation followed by autologous stem cell transplant (SCT).Table 2 shows the distribution, disease status, prior therapy, and disease sites.Fourteen patients with NB had prior anti-GD2 antibodies.

Preparation of cells and cell product characteristics.
Cell products were successfully manufactured in 20 of 21 patients who were apheresed.For patient IT20113, a 6-year-old male with primary refractory metastatic NB, viability of the manufactured product was only 45% (viability release ≥ 70%) and the minimum target cell number was not met (≥ 80% target dose).Cell products for the remaining 20 patients met cell expansion goals and product release criteria.Viability ranged from 73% to 96% (median 82.5%).
Percentages of CD3+, CD8+, and CD4+ T cells and 51 Cr release speci c cytotoxicity against GD2+ NB cell lines are presented in Table 1.The CD8/CD4 ratios in the manufactured products ranged from 4:1 to 1:15. Toxicity.
In the Phase I, 9 patients received a total of 72 infusions of GD2 BATs.One patient with DSRCT quickly deteriorated due to progressive disease and therapy was discontinued after three infusions.Nine patients completed the study, each received 8 GD2 BAT infusions.No patients had their treatment stopped due to adverse effects (Table 3).All patients developed a mild, manageable form of CRS with Grades 1-3 fever, chills, headache, and nausea in 9 of 9 patients.
Occasional Grade 1-2 hypotension was observed only in 3 of 9 patients after 10 of their 27 GD2 BAT infusions.Headaches and fevers lasted for up to 72 hours post infusion.The most common Grade 2 and 3 toxicities (55% each) were fever and chills, starting as early as the end of the infusion, followed by headache and hypotension.Hypotension was mild and transient and did not require vasopressors.Hypoxia was rare (1 patient had 1 episode) and did not require high-ow oxygen.Six out of 9 patients developed Grade 1-2 CRS with Grade 1 episodes after 16 of 75 infusions and Grade 2 CRS episodes after 10 of 75 infusions.Retrospective CRS grading according to ASTCT criteria did not change initial CTCAE v.5.0 grades.Most of the patients developed fever within the rst 12 hours after infusion.Nausea with or without emesis and anorexia occurred in 8 of 9 patients (any Grade) and 5 of 9 patients (any Grade), respectively.
Only three (33%) patients developed Grade 2 lower extremity pain that was managed with activity modi cations and ibuprofen and lasted for up to three days post GD2 BAT infusion.No DLTs were observed, and the MTD was not reached.In the 10 patients treated in the limited Phase II cohort, the toxicities were similar in type and Grade to Phase I patients and there were no DLTs (Table 2).All toxicities resolved to less than Grade 3 within 72 hours.In many cases, the incidence and severity of side effects decreased with subsequent infusions.Although Phase I patients IT20111 and IT20115 did not have objective responses, their clinical courses were remarkable.Patient IT20111 had NB in ltrating his bone marrow with progressive metastatic NB bone and soft tissue disease and was treated in dose level 1. GD2BAT therapy resulted in a complete bone marrow response beginning at 1 week after therapy (Fig 2A) and stable disease in his presacral mass, retroperitoneal lymph nodes and bone metastases that lasted about 6 months.Six months after therapy, he was restarted on topotecan and cyclophosphamide chemotherapy, the same regimen he failed prior to enrolling on study.Despite progressing through this regimen previously, he had a MIBG response and clinical improvement of limping and pain and remained progression free for 2.5 additional years.He progressed 36 months after GD2BAT therapy and died of disease 47 months after therapy.
In dose level 2, Patient IT20115 with recurrent radiation-induced nasopharyngeal osteosarcoma developed an early metabolic PET response with a signi cant decrease in 18 FDG PET uptake when assessed after the 6 th GD2BAT infusion (Fig 2B ).A biopsy of the soft tissue nasal mass showed high numbers of in ltrating CD8+ cells (Fig 2C).The patient had stable disease by RECIST and was started on pazopanib after completing the study.His disease remained stable allowing for tumor resection, which extended his remission for a year.His disease progressed 14 months after GD2BAT therapy, and he died of disease 18 months later.
Phase II Component.
In the phase II group of 10 evaluable patients, one patient (IT00013) had PR by INRC criteria, 5 patients had SD, and 4 patients had PD at the rst evaluation; however, there was additional clinical activity in some of the patients with overall SD including a patient with soft tissue mass response, and a patient with bone marrow response.The median OS for the phase II group was 31.2 months.The patient with PR was observed without additional therapies and progressed 6 months later and all 5 patients with SD were started on additional therapies.
Highlighted Phase II Cases.Patient IT00013 with recurrent NB and multiple soft tissue thoracic and abdominal masses had a PR in all lesions after GD2BAT infusions with corresponding improvement in MIBG scan.He progressed 6 months after therapy, was treated with several other lines of therapy, and is free of disease 1640 days after GD2BAT therapy.Patient IT0031 with recurrent metastatic NB had PD per INRC criteria based on a new single focus of skeletal MIBG uptake, but he also had a PR in the dural intracranial soft tissue mass at the same time.He received several additional lines of therapy and is alive with disease 4 years and 2 months after GD2 BATs.Patient IT00033 with recurrent metastatic NB had SD after GD2BAT therapy with a stable MIBG scan.She had further improvement after restarting her prior salvage chemotherapy of topotecan and cyclophosphamide, followed by irinotecan and temozolomide leading to a negative MIBG scan 1 year after GD2BAT.She is alive with disease 4 years after GD2 BAT.Patient IT 000040 with recurrent metastatic NB had overall INRC SD but cleared her bone marrow disease after GD2BAT therapy.She progressed 9 months later, received additional therapies and is alive with the disease 3 years and 9 months after GD2BAT therapy.
Staining for IgG2a (OKT3 component of the BiAb) was not interpretable due to high background staining so PBMCs were stained with anti-idiotypic antibody A1G4 speci c for hu3F8 in one patient (IT00005) at MSK. Increasing numbers of circulating GD2BATs were detected above baseline (supplemental Fig 2D) after GD2 BAT infusions and up to 2.4% of PBMCs were GD2BATs still detected in the circulation 6 days after the last infusion.

Discussion
Despite the application of anti-GD2 mAbs, 10-20% of high-risk NB patients never achieve remission and 50-60% of those who get immunotherapy after SCT will relapse 3,4 .After patients with recurrent or refractory NB fail initial treatment and then chemoimmunotherapy, subsequent therapies including MIBG, kinase inhibitors 10 , anti-GD2 antibodies 26 , vaccines, haploidentical SCT 28 have not induced high response rates.Although initially promising, reported clinical trials of CAR T cells targeting GD2 + failed to demonstrate formal objective responses 29,30 , except for the recent work reported from Italian group that demonstrated 63% ORR 31 .OST is responsible for 8.9% of all cancer related deaths in the pediatric age group 32 , however, there have been no new drugs for OST for more than three decades 33 .The combined progression free survival (PFS) from seven Phase II trials in OST was 12% at four months in patients with measurable disease and there were less than 5% objective responses in multiple Phase II trials over the past two decades 34 .
There were no DLTs in over 180 patients treated with BATs targeting HER2+, EGFR+, CD20+, in our Phase I/II trials of BAT therapy in metastatic breast cancer 14,15 , metastatic hormone refractory prostate cancer 16,17 , and metastatic pancreatic cancer 14,[18][19][20][21] .Based on these encouraging results, we showed here that the platform could also be safely applied to target GD2 + tumors.
It was feasible to collect PBMCs by apheresis, expand the T cells, harvest, and arm ATC with GD2Bi, aliquot and cryopreserve multiple doses, and infuse the GD2BATs in pediatric outpatient clinics at MSK, CHM, and UV.Only one patient who received sirolimus as part of his previous treatment regimen failed to produce the required numbers of cells secondary to poor growth and viability.GD2BAT therapy was well tolerated in this highly pretreated pediatric patient population.The bone pain commonly associated with anti-GD2 mAb infusions was not seen except for mild leg pain in 2 patients that did not require narcotics.The MTD was not reached.Infusions were associated with mild CRS-associated symptoms with the most frequent side effects being chills, fever, headache, fatigue, and hypotension, with no ≥ Grade 3 toxicities persisting beyond 72 hours.Since the study was conducted prior to the publication of the ASTCT de nition for CRS, scoring was reported using CTCAE v5.0 and were also retrospectively graded according to the ASTCT criteria 23 and there was no difference for low grades CRS between CTCAE v.5.0 and ASTCT grading systems.All cases of CRS were CTCAE/ASTCT Grade 1 or Grade 2 at most with fever and hypotension that responded to uids.
Although there were no objective radiological responses in the phase I portion, there was clinical activity in 2 patients (IT20111 and IT20115) with NB and OST who had a complete BM response with otherwise stable disease (IT20111) or stable tumor with decrease in FDG PET uptake (IT20115), followed by prolonged OS with additional therapies in both cases.The IT20111 patient's clinical response was also associated with elevated anti-NB CTL, NK activity (K562 stimulated PBMCs EliSpots), and Th 1 cytokine/chemokine levels.In the limited phase II portion, clinical activity was seen in several patients, and immunologic sensitization may have positively affected responses to subsequent therapies.For example, patient IT00013, who was chemorefractory and had multiple soft tissue, thoracic, and abdominal masses developed a radiological PR after GD2BAT therapy, progressed 6 months later, then received several additional lines of therapy resulting in a disease-free remission surviving 1640 days after GD2 BAT.Patient IT00031 had a mixed response to GD2BAT therapy, developing a PR in an intracranial mass but increased MIBG uptake in a skeletal metastasis, and he is alive with disease 4 years and 2 months later having received additional chemotherapy.Patient IT00033 had stable disease by MIBG after GD2BAT therapy, which subsequently responded to irinotecan and temozolomide, to which she was previously refractory.She is alive with the disease 4 years after GD2BAT therapy.Finally, patient IT00084 developed a bone marrow CR associated with overall SD after GD2BAT therapy, progressed 9 months later, but is alive with disease 3 years and 9 months after immunotherapy.These observations may support the concept that GD2BAT infusions "immunosensitize" the tumor to subsequent chemotherapy.A similar immunosensitization response was seen in a patient with metastatic pancreatic cancer whose peritoneal lesion progressed after three anti-CD3 x anti-EGFR BiAb armed ATC (EGFRBATs) infusions but developed a CR 2 months after restarting capecitabin that lasted 53 months 14,18 .
Speci c cellular anti-GD2 CTLs developed in more than half of the patients.Anti-GD2 endogenous CTL activity was elevated in both phase I and II patients.We conclude based on our clinical study that multiple BAT infusions of up to 1.6 x 10 9 BATs/kg in children with NB induce endogenous speci c anti-tumor CTL as was also previously seen in breast, prostate, and pancreatic cancer patients; 14,15,17 and may "sensitize" tumors so that the same prior chemotherapy becomes effective when it is restarted.This latter observation is supported by patients who became responsive after immunotherapy to checkpoint inhibitors to chemotherapy regimens to which they were previously refractory 35 .
This Phase I/II study shows IT with GD2BATs is safe and feasible in resistant/refractory heavily pretreated patients who have failed previous therapy including GD2 mAb therapies.The small extension cohort of 10 patients in the phase II portion not only con rmed safety but also showed a median OS of 31 months which is nearly double the median OS of 16.1 months reported for 469 comparable patients with recurrent NB in multiple trials meta-analysis 6 .Our results provide a strong rationale for conducting a Phase II trial that will rigorously assess the clinical bene ts of GD2BATs in patients with refractory/resistant NB.   * Three Phase I and 2 Phase II patients who were not treated (4) or did not complete therapy (1) were not reported in Table 2.
In phase I patients, observed clinical activity of GD2 BATs was limited to ungraded responses.There were no objective responses according to INRC or RECIST.Six of 9 patients who completed therapy (3/5 NB, 2/3 OST, 1/1 DSRCT) had PD, and 3 patients (2/5 NB, 1/3 OST) had SD.Six patients (one OST and 5 NB) received additional therapy after completing the study.Five of 9 patients (1 OST, 4 NB) survived 1 year post GD2BAT therapy.The median OS survival was 18.0 months for Phase I group (Fig 1B).

Figure 2D Anti-
Figure 2DAnti-GD2 Cytotoxic Activity.Speci c anti-GD2 cytotoxicity mediated by PBMC as measured by IFN-γ EliSpots 1 month after therapy was signi cantly higher than prior toIT in the combined 10 phase II NB patients in responses to KCNR (Fig 3a, P<0.02), MG63 (Fig 3b, p<0.02) whereas responses to K562 were not signi cant.In contrast, post-IT IFN-γ EliSpots were not different for KCNR, MG63 and K562 (NK target) in the 8 phase I patients (Fig S1a-c).If IFN-γ EliSpots responses to KCNR and MG63 by PBMC of the 18 phase I and II patients were analyzed together, the post IT IFN-γ EliSpots were signi cantly higher than the pre-IT levels for KCNR (p<0.03) and MG63 (p<0.04).Changes in NK activity (Fig 3c and 3f) were not apparent in the phase I or II patients (Fig 3c and 3f).It is notable that Patient IT 20111 demonstrated remarkable cellular immune response that was associated with a complete BM response and overall SD.The PBMCs from the patient showed 7, 4.1, and 8.8-fold increases in IFN-γ ELISpot responses to KCNR, MG63, and K562 cell line stimulation, respectively.Serum Cytokine/Chemokine Responses.Cytokines known to be involved in immune responses and CRS were tested pre-IT and post-IT in 18 patients from combined phase I/II cohort and demonstrated signi cantly increased post-IT levels of the IL-12 (p <0.002) over pre-IT levels (Fig 4a) whereas the post IT levels in 10 patients of the phase II group (Fig 4b) had signi cantly increased levels over pre IT levels of IL-12 (p<0.02).Post IT serum levels of MIP1β (Fig 4c) and IL-10 (Fig 4d) from 18 Phase I and II patients were signi cantly higher than pre-IT (p<0.02, and p<0.02, respectively).It is noteworthy that post-IT levels of IL-6(Fig 4e), TNFα (Fig 4f), and IP-10 (Fig 4g) in 18 Phase I/II patients were not elevated with respect to pre IT levels.There were no changes in the pre-IT or post IT results for IFN-γ EliSpot responses of PBMC to KCNR, MG63, and K562 and no signi cant changes in serum cytokine levels for IL-12, TNF-α, IL-10, IL-6, MIP1β, IP-10 in 8 Phase I patients (data not shown).There were no signi cant changes between pre-IT and post-IT levels of TNFα, MIP1β, and IP-10 in the Phase II patients (data not shown).

1 A
Figures

Figure 2 A
Figure 2 A) Shows a bone marrow response with clearance of neuron-speci c enolase (NSE) positive tumor cells in the bone marrow of a NB patient after 8 infusions of GD2 BATs.B) Shows the metabolic PET changes after 5 infusions of GD2 BAT in an OST patient.C) Flow cytometry staining for CD4+ and CD8+ to evaluate presence of tumor in ltrating lymphocytes in the tumor for Pt # 20115.Most intranasal tumor in ltrating lymphocytes were CD8 + .Cells were gated on CD3 + cells and analyzed for FITC positive CD4 and PE positive CD8 T cells.D) Summarizes the proportion of CD3+A1G4+ cells in the peripheral blood ofpatient # 8 pre-infusion and at various times during and after GD2 BATs infusions., % of CD3 + A1G4 + cells shows the persistence of GD2 BATs in peripheral blood.CD3 + A1G4 + cells were measured by using 2 color FACS gated on lymphocytes using anti-CD3 and A1 G4 anti-idiotypic antibody for hu3F8.

Figure 3 Anti
Figure 3 Anti-GD2 Cytotoxicity (IFN-γ EliSpots).Each panel shows the pre-IT and post-IT IFN-γ EliSpots responses from PBMC after overnight stimulation.Phase II portion: Panel a, b, and c show responses of PBMC from the phase II NB patients to KCNR (p<0.02),MG63 (p<0.02), and K562 (NS).Phase I/II combined: Panels d, e, and f show IFN-γ EliSpots responses of PBMC from all 18 of the phase I/II patients pre-IT and post-IT after overnight exposure to KCNR (p<0.03),MG83 (p<0.04), and K562 (Not signi cant, NS), respectively.

Figure 4 Cytokines
Figure 4 Cytokines/chemokines: Panel a (IL-12) shows signi cant changes between pre-IT and post-IT levels of IL-12 in all 18 Phase I/II patients (p<0.002).Panel b (IL-12) shows a signi cant change between pre-IT and post-IT IL-12 levels for 10 NB in the phase II patients analyzed separately (p<0.02).Panel c (MIP-1β) shows post-IT levels of MIP-1β increased signi cantly over pre-IT levels in 28 phase I/II patients(p<0.02).Panel d (IL-10) shows the levels of IL-10 signi cantly increased from pre-IT to post-IT in all 18 phase I/II patients (p<0.02).Panels e, f, and g show the pre-IT and post-IT serum levels of IL-6, TNF-α, and IP10 (all NS).

Table 2 :
Distribution of Patients, Disease Status, Types of Prior Therapies, and Disease Involvement*

Table 3 :
Incidence and Grade of Toxicities by Dose Level