1.0 OBJECTIVES AND RATIONALE 
Objectives: Treatment with retroviral vector-mediated gene transfer of the gamma interferon (y IFN) gene into 
neuroblastoma cells is proposed as a means to improve active specific immunotherapy for human neuroblastoma. 
Autologous cells will be employed where available. In patients for which an autologous neuroblastoma cell line is 
not available, single haplotype matched allogeneic neuroblastoma cells will be used. The therapy is proposed for 
patients at high risk of relapse with minimal or no detectable disease following myeloablative therapy and 
autologous bone marrow transplant; and for patients with persistent/progressive disease following conventional 
therapy. The potential increase in specific immunity against immunogenic gene-modified neuroblastoma cells may 
result in an immunologic response resulting in eradication of residual microscopic disease, or a regression/retardation 
of progression of persistent tumors. 
Our long term goal is to improve the outcome of patients with high risk neuroblastoma. Although effective 
induction chemotherapy followed by high dose myeloablative chemotherapy and autologous bone marrow transplant 
(ABMT) achieves complete remission in up to 80% of such patients, over 50% of these develop recurrent fatal 
isease (62,63,71). Biologically based therapy may be able to eliminate the relatively small number of residual 
tumor cells that survive maximal chemotherapy, surgery, and irradiation. The hypothesis of the project is that active 
specific immunotherapy with neuroblastoma cells engineered genetically to evoke an anti-tumor response may have 
an important role in eliminating minimal residual disease, and possibly a role in treatment of progressive disease as 
well. 
To test the hypothesis we will introduce cytokine genes expressing gamma interferon into autologous neuroblastoma 
cells, and test the ability to elicit a specific response in a Phase I Protocol. As not all patients will have autologous 
tumor cells available, we will also investigate whether cytokine producing single HLA locus matched allogeneic 
neuroblastoma cells can serve as an immunogen. For these latter patients we will utilize a neuroblastoma cell bank 
which currently contains approximately 140 different human cell lines to develop a panel of cytokine transduced 
tumor cells. We have initiated HLA typing of these cells to identify cell lines with common (eg A2, Al) subtypes 
suitable for inclusion in this study. 
Neuroblastoma, a tumor of the peripheral nervous system, is the most common extra-cranial solid tumor of 
childhood and has an incidence of 1 per 7000 children under the age of 15 (61). Approximately 60% of children 
with neuroblastoma are at high risk for developing fatal progressive disease despite all available current therapies 
(62,63). The group includes patients with stage IV disease diagnosed after 1 year of age, and patients whose tumors 
have amplification of the N-myc oncogene, regardless of stage of disease or age of diagnosis (63-65). Ability to 
derive an immortalized cell line is itself a strong predictor of high risk (Fig. 12). Current data suggests that high 
risk patients should receive aggressive therapy followed by high dose consolidation therapy (63). With 
myeloablative therapy and ABMT, nearly all patients achieve a complete clinical remission, but the tumor recurs in 
over 60% of patients indicating the need for better therapy (62,63,71). Patients who relapse following induction 
/ABMT generally have survival measuring 4 months w less. Relapses in high risk patients generally occur within 2 
years of completion of therapy. Our investigations are designed to develop gene transfer based therapy that can 
effectively eliminate tumor cells surviving maximal chemothwapy, and which may alter the course of patients who 
have p-ogressive disease despite failure of available therapies. 
Most neuroblastoma cells have reduced expression of Class I and Class II major histocompatibility complex (MHC) 
antigens (1,65,70,75). A potential strategy in immunotherapy involves presentation of tumor antigens in the 
context of MHC class I and class n (2,20,22) (Figure 1). Immunogenicity of tumor cells may be augmented in an 
antigen specific manner by stimulation of MHC molecules. ylFN is known to induce boA class I and class II 
MHC expression (13,15,17,18,28,30,76,78). We have transduced multiple human neuroblastoma cell lines with a 
retroviral based human y-EFN vector (DAHu y-IFN). Our preliminary results demonstrate that y-IFN vector can 
reliably transduce up to 2/3 of available neuroblastoma cell lines (Table IV) and that transduced cells produce high 
levels of y-IFN, and markedly upregulate MHC class I and class II (Table IV, V, Figures 14,15) as well as some 
neuroblastoma associated antigens (NAAg) (Figurel7). 
Anti-tumor response has been enhanced in vitro and in people by expressing a large variety of cloned cytokines in 
tumor cells (22,24-30,74). Anti-tumor effects have been observed with transduction of y-IFN into mouse 
fibrosarcoma, renal cancer, and neuroblastoma cells (28-30). Several cytokines have been shown to augment the 
immune response to neuroblastoma in animal models (28,72). y-IFN secretion by tumor cells is known to decrease 
[90] 
Recombinant DNA Research, Volume 19 
