The body has different ways of fighting infections and disease. No single way seems perfect for fighting cancer. This research study combines two different ways of fighting disease: antibodies and T cells. Antibodies are molecules that fight infections and protect your body from diseases caused by bacteria and toxic substances. Antibodies work by sticking to those bacteria or substances, which stops them from growing and causing bad effects. T cells are special infection-fighting blood cells that can kill other cells, including tumor cells or cells that are infected. Both antibodies and T cells have been used to treat patients with cancers. They both have shown promise, but neither alone has been enough to cure most patients.

This multicenter study is designed to combine both T cells and antibodies in order to create a more effective treatment. The treatment that is being researched is called autologous T lymphocyte chimeric antigen receptor cells (CAR) cells targeted against the disialoganglioside (GD2) antigen that express Interleukin (IL)-15, and the inducible caspase 9 safety switch (iC9), also known as iC9.GD2.CAR.IL-15 T cells.

In previous studies, it has been shown that when T cells have part of an antibody attached to them, they are better at recognizing and killing cancer cells. The antibody that will be used in this study is called anti-GD2. This antibody floats around in the blood and can detect and stick to cancer cells called neuroblastoma cells because they have a substance on the outside of the cells called GD2. For this study, the anti-GD2 antibody has been changed so that instead of floating freely in the blood, it is now joined to the T cells. However, it is unknown how long the iC9.GD2.CAR.IL-15 T cells last in the body, so their chances of fighting cancer cells are not well known.

To improve the tumor-fighting power of GD2-CAR-T cells, two additional components were added to these cells. The IL-15 gene was added so that the GD2-CAR-T cells can attack tumor cells more effectively. Interleukin-15 (IL-15) is a chemical that cells use to communicate with one another. Other research using IL-15 in combination with CAR-T cells has shown there is an increase in the body's ability to allow the CAR-T cells to survive and grow in the body. The iC9 gene was added as an "off switch" so it can stop the activity of the GD2-CAR-T cells if there are any serious bad side effects. Bad side effects seen previously in patients receiving the GD2 antibody alone include pain. In this study, the "stop switch" can be used to turn off the GD2-CAR-T cells if you experience intense pain that does not respond to normal pain treatments.

The study will enroll a minimum of 10 adult subjects and 10 pediatric subjects; all subjects will undergo lymphodepletion chemotherapy prior to the cell infusion as outlined in the protocol.

All clinical and laboratory data required for determining eligibility must be available in the subject's medical/research record which will serve as the source document.

Because of the nature of iC9.GD2.CAR.IL-15 T cell product preparation, subjects will be assessed for initial study enrollment eligibility (prior to cell procurement) and then will have to meet criteria prior to starting lymphodepletion and prior to T cell infusion.

Inclusion Criteria for the Study:

1. Written HIPAA authorization signed by legal guardian.

2. Adequate performance status as defined by Lansky or Karnofsky performance status of ≥ 60 (Lansky for <16 years of age).

3. Life expectancy ≥12 weeks.

4. Histological confirmation of neuroblastoma or ganglioneuroblastoma at initial diagnosis. Bone marrow samples are acceptable as confirmation of neuroblastoma, confirmation of osteosarcoma at diagnosis

5. High-risk neuroblastoma with persistent/refractory or relapsed disease, defined as:

   1. First or greater relapse of neuroblastoma following completion of aggressive multi-drug frontline therapy.
   2. First episode of progressive neuroblastoma during aggressive multi-drug frontline therapy. Persistent/refractory neuroblastoma as defined by less than a complete response by the revised International Neuroblastoma Response Criteria (INRC) at the conclusion of at least 4 cycles of aggressive multidrug induction chemotherapy on or according to a high-risk neuroblastoma protocol (such as A3973 or ANBL0532).
   3. Patients must be diagnosed with high risk neuroblastoma at initial diagnosis or if non-high risk at time of initial diagnosis must have had evidence of metastatic progression when >18 months of age as defined in the protocol or relapsed or refractory osteosarcoma that is not responsive to standard treatment.

6. Measurable or evaluable disease per Revised INRC for subjects with neuroblastoma or measurable disease by Response Evaluation Criteria In Solid Tumors Criteria (RECIST) v1.1 criteria for subjects with osteosarcoma.

7. Adequate central nervous system function as defined by:

   1. No known Central Nervous System ( CNS) disease
   2. No seizure disorder requiring antiepileptic drug therapy

Exclusion Criteria for the Study Subjects meeting any of the following exclusion criteria will not be able to participate in this study (procurement, lymphodepletion, and cell infusion).

1. Pregnant or breastfeeding (NOTE: breast milk cannot be stored for future use while the mother is being treated on study).
2. Has a known additional malignancy that is active and/or progressive requiring treatment.
3. History of hypersensitivity reactions to murine protein-containing products.
4. History of hypersensitivity to cyclophosphamide or fludarabine.