Based on newly available preclinical data we changed the CAR construct to a more effective version and will now study that product on a different protocol.

This research study is for patients that have a cancer called Neuroblastoma that has either come back after treatment or did not respond to the standard medicines used to treat it. This study combines two different ways of fighting cancer: antibodies and Natural Killer T cells. Antibodies are types of proteins that protect the body from infectious diseases and possibly cancer. T cells, also called T lymphocytes, are special infection-fighting blood cells that can kill other cells, including cells infected with viruses and tumor cells. Both antibodies and T cells have been used to treat patients with cancers. The investigators have found from previous research that they can put a new gene into T cells that will make them recognize cancer cells and kill them. In a previous clinical trial, the investigators made a gene called a chimeric antigen receptor (CAR), from an antibody called 14g2a that recognizes GD2, a molecule found on almost all neuroblastoma cells (GD2-CAR). They put this gene into the patients' own T cells and gave them back to patients that had neuroblastoma. Nineteen patients were treated on that study and there were no long term side-effects seen after the GD2 T cell infusion. As the investigators have followed the patients over time, they noticed that for those patients with disease at the time of their infusion, the time to progression (the amount of time it takes before their neuroblastoma got worse) was longer in those whom they could find GD2 T cells in the blood for more than 6 weeks after the last T cell infusion. Because of this, the investigators think that if effector cells are able to last longer, they may have a better chance of killing neuroblastoma tumor cells.

Natural Killer T cells are a special subset of innate lymphocytes that can effectively go into tumor tissues of neuroblastoma. Inside the tumor, there are certain white blood cells which help the cancer cells to grow and recover from injury. Natural Killer T-cells can specifically kill these cells. In this study, Natural Killer T cells will be genetically engineered to express GD2-CAR to attack neuroblastoma cells and the white blood cells inside the tumor tissue.

To prepare the neuroblastoma specific Natural Killer T cells (also called GINAKIT cells), about 80 mL (up to 6 tablespoons) of blood will be collected from the patient. For children, the total amount of blood drawn will not be more than 3 ml (less than 1 teaspoon) per 2.2 lbs of body weight.

In case the neuroblastoma specific Natural Killer T cells cannot be expanded from the amount of blood that can safely be drawn from the patient, the investigators may attempt to collect white blood cells from the patient by using a special type of blood donation called leukocyte apheresis. Unlike a regular blood draw or "whole blood" donation in which whole blood is removed from patient's body, the apheresis process connects the patient to a machine that removes a particular component of their blood and then returns the rest of the blood back to them. The white blood cells collected during leukocyte apheresis will be used to make the GINAKIT cells.

Once collected, the white blood cells will be mixed with a special protein to separate the Natural Killer T cells from the white blood cells using a special machine called a CliniMACS Reagent System in the laboratory. This is an investigational device that is not approved by the FDA. Although this device is not approved for use in this country, it has been in use for years and is approved in other countries. The investigators will put the new genes into patient's Natural Killer T cells by mixing them with a retroviral vector (a special virus that can carry a new gene into cells) containing the new gene called iC9-GD2-CD28-OX40. After the new gene has been put into the Natural Killer T cells, the cells will be tested to make sure that they kill GD2-positive neuroblastoma cells and that the cells are killed by the activation of iC9. The GINAKIT cells may be given back to the patient fresh or frozen.

This is a dose escalation study. This means that at the beginning, patients will be started on the lowest dose (1 of 4 different levels) of GINAKIT cells. Once that dose schedule proves safe, the next group of patients will be started at a higher dose. This process will continue until all 4 dose levels are studied. If the side effects are too severe, the dose will be lowered or the infusions will be stopped.

Before getting the GINAKIT cells, patients will receive cyclophosphamide and fludarabine intraveneously for 2 days and then fludarabine alone for one more day. Patients will then have one day of rest with no chemotherapy before receiving the GINAKIT cells.

The treatment will be given by the Center for Cell and Gene Therapy at Texas Children's Hospital. Patients will need to stay in Houston for 4 weeks after the infusion so the investigators can monitor them for side effects. They will have follow-up visits (at weeks 1, 2, 4, 6, and 8; months 3, 6, 9, and 12; twice a year for 4 years and then once a year for the next 10 years - for a total of 15 years) and scheduled disease evaluations after the GINAKIT cell infusion (at week 4 and then as clinically needed). Also, to learn more about the way the GINAKIT cells are working and how long they last in the body, blood will be obtained before the chemotherapy, on the day of the GINAKIT cell infusion (before and at the end of the infusion) and at the follow-up visits (time points listed above).

If the patient experiences no significant side effects during or after the first GINAKIT cell infusion and their cancer remains stable or responds, they may be offered additional doses of the same cell dose in the future.

Procurement Inclusion Criteria:

• High risk neuroblastoma or persistent or relapsed disease
• Life expectancy of at least 12 weeks
• Age ≥1 and ≤18 years old
• Karnofsky/Lansky score of 60% or greater
• Absence of HAMA prior to enrollment (only in patients that have been previously treated with murine antibodies)
• Ability to tolerate leukocyte apheresis
• Informed consent for leukocyte apheresis
• Informed consent and assent (as applicable) obtained from parent/guardian and child.

Procurement Exclusion Criteria:

• Rapidly progressive disease
• History of hypersensitivity to murine protein containing products

Treatment Inclusion Criteria:

• High risk neuroblastoma with persistent or relapsed disease
• Life expectancy of at least 12 weeks
• Age ≥1 and ≤18 years old
• Karnofsky/Lansky score of 60% or greater
• Patients must have an ANC ≥ 500 without the use G-CSF or GM-CSF for at least 48hrs, platelet count ≥ 20,000
• Pulse Ox ≥ 90% on room air
• AST less than 5 times the upper limit of normal
• Bilirubin less than 3 times the upper limit of normal
• Serum creatinine less than 3 times upper limit of normal
• Recovered from the acute toxic effects of all prior chemotherapy (if some effects of chemotherapy are expected to last long term, patient is eligible if meeting other eligibility criteria).
• Absence of human anti-mouse antibodies (HAMA) prior to enrollment for patients who have received prior therapy with murine antibodies
• Patients must have autologous transduced NKTs with greater than or equal to 20% expression of GD2-specific CAR.
• Informed consent and assent (as applicable) obtained from parent/guardian and child.

Treatment Exclusion Criteria:

• Rapidly progressive disease

• Currently receiving other investigational drugs

• History of hypersensitivity to murine protein containing products

• History of cardiomegaly or bilateral pulmonary infiltrates on chest radiograph or CT. However, patients with cardiomegaly on imaging may be enrolled if they have an assessment of cardiac function (i.e., ECHO or MUGA) within 3 weeks of starting protocol therapy that is within normal limits. Additionally, patients with bilateral pulmonary infiltrates on imaging may be enrolled if the lesions are not consistent with active neuroblastoma (i.e., negative on functional imaging with PET or MIBG, or by pathologic assessment).

• Evidence of tumor potentially causing airway obstruction

• Patients who are pregnant, lactating, or unwilling to use birth control

• Patients currently receiving immunosuppressive drugs such as corticosteroids(Patients may receive treatment if treated with corticosteroids with dose of less than 0.5mg/kg/day of prednisone equivalent), tacrolimus or cyclosporine.

• Patients who have previously experienced severe toxicity from cyclophosphamide and fludarabine.

  • All labs must be collected within 10 days prior to initiation of study related treatment (except for verification of GD2 transduction)