TARGETING NEUROBLASTOMA THROUGH INDIVIDUALIZED THERAPUTICS 

Background:   Although improvements in the past 40 years have led to markedly improved survival rates of approximately 80% overall for pediatric cancers, patients with relapsed,rare and advanced stage tumors still have a very poor prognosis.  Neuroblastoma is the most common extra cranial solid tumor in children, with 700 new diagnoses projected for 2011. It accounts for 7% to 10% of childhood cancers. Whereas the prognosis for infants with neuroblastoma is generally good, currently only 30% of children diagnosed after 12-15 months of age survive despite aggressive multimodal therapies.  Even with high-dose chemotherapy (HDC) followed by hematopoietic stem cell transplantation (HSCT) and maintenance therapy with retinoic acid the 5-year event-free survival remains below 50%. Long-term survival of patients who are treated with conventional therapies following relapse is <5%. As such, neuroblastoma accounts for 15% of all pediatric cancer deaths in the United States. Consequently, the evaluation of new drugs is strongly needed in this disease. Recent evidence has established the genetic heterogeneity of the disease and revealed the existence of several major molecular subsets that collectively may provide prognostic value for future disease management. While the poor prognosis for older neuroblastoma patients underscores the need for new treatment strategies, the elucidation of specific biologic subsets of neuroblastoma suggests a way to improve disease management. The current standard-of-care treatments for relapsed neuroblastoma include a variety of Phase II or Phase I studies that generally have only modest response rates (10%-35%). Even in responding patients, tumors often go on to further rapid relapses and novel strategies to treat this patient population are urgently needed. There are currently few treatment options from which pediatric oncologists can select with any degree of confidence to improve the management of multiply recurrent neuroblastoma patients. The current strategy is to add salvage therapies based on the anecdotal experience of the treating physician, which often leads to drug-related toxicity but may or may not extend life. The identification of agents that target specific molecular pathways associated with the development and/or progression of neoplastic diseases holds promise. Improved approaches that identify in a more rational (data-driven) fashion combinations of existing agents that are likely to be effective should result in a survival benefit in the clinical setting, while avoiding the toxicity associated with agents that are unlikely to be beneficial.

Our proposal outlines an approach by which we can utilize our expanding knowledge of molecular networks and the mechanisms of action of medications in conjunction with standardized biomarker development to create individualized predictions of effective targeted combinations for children with neuroblastoma.

Goals:  Identification of predictive genomic profiles for patients with neuroblastoma of 5 targeted agents from Novartis (BKM120, LBH589, LCL161, LDE225, LEE011). We hypothesize that the in vitro and in vivo response data of these agents in neuroblastoma patient cells and tumor-bearing mice will correlate closely with patient’s genomic profiling predictions based on RNA expression profiles and DNA mutation panels.

Methodology

Neuroblastoma patient cells will be screened by genomic profiling analysis using RNA expression profiling and DNA mutation panels.  We will validate the targeted therapeutic approach in both in vitro and in vivo in cell culture and xenograft mouse models.

Significance

We expect that we will find a close correlation of the in vitro and in vivo data with the patient’s genomic profiling resulting in the creation of molecular signatures of sensitivity to each of the drugs tested. Since each of these are Novartis agents, it will then be possible to translate this into the first targeted Phase I study of testing multiple new agents in one study aimed at pairing the patient with the best medication for them in the best chemotherapeutic combination shown in studies.  As shown previously in targeted clinical trials, this would allow for improvement in response rates and outcomes for our patients.

Project Description  

Goals:  Identification of predictive genomic profiles for patients with neuroblastoma of 5 targeted agents from Novartis (BKM120, LBH589, LCL161, LDE225, LEE011). We hypothesize that the in vitro and in vivo data of these agents in neuroblastoma cells and tumor-bearing mice will correlate closely with patient’s genomic profiling predictions based on RNA expression profiles and DNA mutation panels.

Impact 

In achieving the goals of this research study, we will be ready to offer a first of a kind Phase II study in which novel agents from a single company, Novartis, will be offered in a targeted manner to patients. As medicine advances, we are moving from the era of high dose chemotherapeutics to more selected therapies based on a deeper understanding of a patients’ tumor.  From patients’ cells growing in culture, already profiled, we will identify molecular patterns which predict response. Then, patients enrolling on study will be profiled prior to being assigned a medication arm.  Each targeted agent will be combined with synergistic chemotherapy agents to offer maximum benefit to patients.  

PHAPSE II RELAPSE PREVENTION TRIAL OF DFMO AS A SINGLE AGENT IN PATIENTS WITH HIGH RISK NEUROBLASTOMA IN REMISSION

High risk neuroblastoma remains a challenge in pediatric oncology.  While most patients are able to attain remission, almost 50% will relapse.  Once relapsed, there is currently no curative treatment for these children and relapse is inevitable.  As such, new therapeutic approaches are needed to prevent the children from relapsing, allowing them to survive. Relapsed patients who are able to obtain a second remission are not eligible for relapse therapy since they have no evidence of disease and yet they are likely to relapse within 6 months-1year.  

This study creates a clinical trial for these patients to prevent the relapse as our approach to improve outcome in these patients.  These more aggressive forms of NB respond poorly to hormonal and chemotherapeutic approaches, and therefore,there is a great need for antineoplastic agents with novel mechanisms of action. The MYCN protein up-regulates ornithine decarboxylase (ODC), a gene encoding for the ODC enzyme that is pivotal in polyamine biosynthesis. High polyamine content and elevated ODC activities are commonly found in many tumors including neuroblastoma.  Therefore, suppression of polyamines in cancer cells is an effective means to reduce tumor cell proliferation. Specific polyamine inhibitors such as DFMO have been evaluated in adult clinical cancer trials (patients on drug for 2 years safely) and shown to prevent formation of polyps and colon cancer.  The NMTRC is completing a Phase I study of DFMO in children with neuroblastoma showing safety and tolerability in our patients. 

A CLINICAL TRIAL USING MOLECULAR-GUIDED THERAPY FOR THE TREATMENT OF PATIENTS WITH REFRACTORY OR RECURRENT NEUROBLASTOMA

This is the first study in pediatric oncology evaluating the feasibility of using patient specific genetic information as a predictive tool in personalized therapy. We hypothesize that transcriptome analysis of individual tumor samples can be combined with data concerning molecular pathways and knowledge of drug targets and activity to allow for more rational and individualized selection of potentially active drugs in this population. 

GOAL:ODC/polyamines are critical in cancer growth and therefore present a therapeutic target for the treatment and prevention of recurrence of NB. This study will focus on the use of DFMO in high  risk neuroblastoma patients that are in remission as a strategy to prevent  recurrence and improve overall survival.

• Brief description: Neuroblastoma accounts for 15% of all pediatric cancer deaths in the US with no curative treatment for relapsed patients so there is a clear and urgent need to develop novel strategies to improve patient outcomes. Recent efforts to systematically characterize the molecular basis of neuroblastoma has confirmed biologic differences in this disease. This has revealed major molecular sub-types with varying prognoses, suggesting that different sub-types of neuroblastoma may respond to different therapeutic strategies. This is a pilot study focusing on the patients tumor genetics using a novel method to identify potentially active agents for each patient with relapsed/refractory neuroblastoma. We hypothesize that transcriptome analysis of individual tumor samples can be combined with data concerning molecular pathways and knowledge of drug targets and activity to allow for more rational and individualized selection of potentially active drugs in this population.
• Specific Aims: Aim 1: Determine feasibility of generating genome-wide RNA expression profiles from neuroblastoma cells (derived either from primary tumor specimens or bone marrow samples) and utilizing this data to make real-time treatment decisions. Aim 2: To determine the effectiveness of treatments chosen by this method based on Overall Response Rate (ORR) and Progression-Free Survival (PFS). Aim 3:  To explore in a correlative biologic study the relationship between tumor phenotype and response to therapy.
• Study Design: The current proposal will involve 20 patients with neuroblastoma that are either refractory to, or have relapsed following, current front-line combination therapy. Patients will have a biopsy from which genomic analysis will be performed and a report will be generated based on XB-BIS predictive modeling technology. This will be used to design a therapeutic protocol by a committee of up to eight pediatric oncologists and two pharmacists. Patients will be followed for response to therapy and survival. A correlative biology study will establish xenografts from patient tumor samples for subsequent analysis. Should the method prove feasible and efficacious, formal comparative trial designs will be implemented.
• Impact: This will be the first study in pediatric oncology evaluating the feasibility of using patient specific genetic information as a predictive tool in personalized therapy. It may bring new options to patients with relapsed/refractory neuroblastoma for whom no curative treatment options exist, and so it has potential to extend survival. This method may provide improved ways to make therapeutic decisions for patients with heterogenic diseases such as neuroblastoma.
• Giselle Sholler, M.D. (Lead Investigator)