Report: Modeling Treatment Response of NF1-Deleted Sarcoma

An ESUN Study Report
By Rebecca Dodd, PhD and David G. Kirsch, MD, PhD

Neurofibromin 1 (NF1) is a tumor suppressor that functions as a negative regulator of the Ras pathway.¹ NF1 mutations are present in many types of sarcoma, yet its role in the disease and tumor response to treatment is an underexplored area of sarcoma research. The loss of NF1 is well-characterized in malignant peripheral nerve sheath tumors (MPNSTs),² and NF1 mutations have been recently identified in a wide variety of pediatric and adult soft-tissue sarcomas including myxoidfibrosarcoma,³ liposarcoma,³ and rhabdomyosarcoma.^4,5 In addition, patients with Neurofibromatosis Type 1 (NF1), caused by genetic inactivation of the NF1 gene, are at increased risk for developing soft-tissue sarcomas, including MPNST² and myogenic sarcomas such as rhabdomyosarcoma.^6-8 The identification of novel mutations in sarcomas with complex karyotypes provides new opportunities to model human sarcomas and to investigate new therapies for this difficult to treat disease.

We designed this project to bridge boundaries in sarcoma research by using a novel mouse model of NF1-deleted sarcoma to identify new therapies for sarcoma patients. This particular gene mutation occurs in several sarcoma sub-types including rhabdomyosarcoma, Undifferentiated Pleomorphic Sarcoma (UPS), and MPNST. In addition, this project examined sarcomas that occur in children, young adults, and adults. We developed an innovative mouse model of temporally and spatially restricted NF1-deleted sarcoma that reflects the diverse spectrum of NF1-associated sarcomas found in patients.⁹ We generated mice with conditional mutations in both NF1 and the tumor suppressor Ink4a/Arf (NF1^flox/flox; Ink4a/Arf^flox/flox). Under normal conditions in these mice, both NF1 and Ink4a/Arf are expressed at endogenous levels.

However, in the presence of Cre recombinase, a critical part of each tumor suppressor gene is removed, resulting in deletion of both NF1 and Ink4a/Arf. By injecting an adenovirus containing Cre recombinase (Ad-Cre) into the mice, we generated models of MPNST tumors and high-grade myogenic sarcomas. These tumors reflect the histological properties and spectrum of sarcomas found in patients. This model is useful because it facilitates testing therapeutic drug responses, and we have previously used this model to demonstrate that MAPK inhibition slows the growth of primary mouse tumors.⁹ We are now using this model to characterize the biological properties of NF1-deleted sarcoma and as a platform to test therapeutic agents.

Results

Through the support of the Liddy Shriver Sarcoma Initiative, we used our mouse model described above to address unanswered questions in sarcoma therapy. In Aim 1, the model served as a preclinical platform to assess combination chemotherapy that may be beneficial for sarcoma patients. Chemotherapy can play an important role in treating NF1-associated advanced MPNST.¹⁰ Studies have shown that some MPNSTs are responsive to combination chemotherapy regimens, but it is not known if other NF1-deleted sarcomas will have improved response to these combination regimens.¹¹ Clinical trials have attempted to enroll patients to determine if single-agent vs. combination chemotherapy is beneficial for patients with sporadic MPNST, but patient enrollment was low. Therefore, we used our mouse models to compare single-agent vs. combination chemotherapy against NF1-deleted sarcomas.

For this study, we tested if either doxorubicin alone, isfosfamide alone, or a doxorubicin/isfofamide combination treatment was beneficial for MPNSTs. Initial investigations for this investigation included in vivo toxicology dosing studies to establish an acceptable dose in live mice, which was determined to be a maximum of 15mg/kg i.p for Doxorubicin (Dox) and 50mg/kg i.v. for ifosfamide (IFO). Following this, we investigated a multi-agent doxorubicin and ifosfamide combination treatment in tumors generated from MPNST cells that developed in NF1^flox/flox; Ink4a^flox/flox mice. (Figure 1). When tumors reached 200 mm³, mice were treated with a single bolus of either vehicle alone (n=4), doxorubicin alone (Dox n=5; 15mg/kg i.p), ifosfamide alone (IFO; n=5, 50mg/kg i.v.) or Dox+IFO combination (n=6, 10mg/kg i.p. Dox and 50mg/kg IFO i.v.). All treatments were well-tolerated, and no signs of toxicity were observed. Tumor growth was delayed in mice receiving both doxorubicin alone and IFO alone, but the combination treatment resulted in maximal tumor response. Interestingly, there was a range of response to the combination regimen. As seen in Figure 1C, three tumors showed very low fold change in volume at 14 days with Dox/IFO treatment, while three other tumors showed responses similar to single agent treatment. We are currently continuing these studies in primary mouse tumor models using an optimized combination chemotherapy strategy.

The sarcomas initiated by NF1 mutation discussed above model sporadic sarcomas in the general population. It is not known if NF1-deleted sarcomas that arise spontaneously respond differently to treatment than sarcomas in neurofibromatosis patients. These differences could have important clinical implications that could impact patient care. In Aim 2, we generated sarcomas with NF1 +/- stroma to model sarcoma development in neurofibromatosis patients with NF1 haploinsufficieny (+/-). The role of the supporting stroma (endothelial cells, immune cells, fibroblasts, etc) in the development and vascularization of NF1 patient neurofibromas is well-established.^{12, 13} The ultimate goal of this aim was characterization of this model to study the stromal contribution of NF1 haploinsufficient cells to sarcoma formation and for use in future therapeutic studies.

We used Cre-loxP technology to generate tumors in NF1 flox/- ; Ink4a/Arf ^flox/flox and NF1 ^flox/flox ; Ink4a/Arf ^flox/flox paired littermate mice. These genotypes model tumors from patients with neurofibromatosis or tumors from NF1 wild-type patients, respectively. Upon injection of Ad-Cre into the NF1 ^flox/-; Ink4a/Arf ^flox/flox animals, infected cells will lose the wild-type copy of NF1, generating a tumor that is NF1 ^-/-; Ink4a/Arf ^-/-, surrounded by a tumor microenvironment of cells that are haploinsufficient for NF1 and wild-type for Ink4a/Arf (NF1 ^-/+; Ink4a ^+/+). In contrast, Ad-Cre injection into NF1 ^flox/flox ; Ink4a/Arf ^flox/flox mice will result tumor cells that are NF1 ^-/-; Ink4a/Arf ^-/-, surrounded by a tumor stroma of cells that are wild-type for both NF1 and Ink4a/Arf (NF1 ^+/+; Ink4a ^+/+).

We were successful in generating a large cohort of MPNSTs from these mice. We determined that MPNSTs develop with faster kinetics in NF1^flox/-; Ink4a/Arf^flox/flox mice compared to NF1^flox/flox; Ink4a/Arf^flox/flox littermates (Figure 2). Histological analysis by a sarcoma pathologist diagnosed all tumors as MPNST by histopathology, and the tumors were indistinguishable by genotype. Following injection of the mice, we determined that the NF1 status of the stroma had little impact on the level of cell signaling molecules expressed within the tumor, as assessed by immunohistochemistry (IHC) for pERK and pS6. Future studies will examine the phenotypes of cells within the tumor microenviroment between the two genotypes. In addition, we generated a group of cell lines from these tumors for use in future therapeutic screens.

Summary and Conclusion

These studies have developed and applied mouse models of NF1-deleted sarcomas to test systemic therapies. Support from the Liddy Shriver Sarcoma Fund has been instrumental in developing this model so that we can assess chemotherapy responses of sarcomas with NF1 mutation. Future studies will examine the role of combination chemotherapy in primary mouse models of NF1-deleted sarcoma. Additionally, we have demonstrated the feasibility of comparing MPNSTs with differing NF1 status in the stroma. Continued studies will compare the composition of the tumor microenvironment in these models and the impact of the stroma on therapeutic response. Ultimately, we hope that these novel mouse models will serve as tools to uncover new therapeutic avenues for sarcoma patients.

Plan: Modeling Treatment Response of NF1-Deleted Sarcoma

An ESUN Experimental Plan
By David G. Kirsch, MD, PhD and Rebecca Dodd, PhD

Genomic studies of human tumors have successfully identified genes involved in tumor initiation and revealed novel targets for molecular therapy. Recently, analyses of sarcomas have identified novel mutations in a diverse spectrum of human soft tissue sarcomas. One of the most frequently mutated genes described is neurofibromin 1 (NF1), a tumor suppressor that functions as a negative regulator of the Ras pathway. Although the loss of NF1 is well-characterized in malignant peripheral nerve sheath tumors (MPNSTs), NF1 mutations in a wide variety of pediatric and adult soft-tissue sarcomas have only been recently identified in tumors including myxoidfibrosarcoma,¹ liposarcoma,¹ and rhabdomyosarcoma.² In addition, patients with Neurofibromatosis Type 1 (NF1), caused by genetic inactivation of the NF1 gene, are at increased risk for developing soft-tissue sarcomas, including MPNST and myogenic sarcomas such as rhabdomyosarcoma.^3-5 Despite the prevalence of NF1 mutations in soft-tissue sarcomas, the role of NF1 in the development of soft-tissue sarcomas, particularly non-neurogenic tumors, is an unexplored area of sarcoma research. The identification of novel mutations in sarcomas with complex karyotypes provides new opportunities to model human sarcomas and to investigate new therapies for this difficult to treat disease. This study is designed to identify new therapies for sarcoma patients using a novel mouse model of NF1-deleted sarcoma. The proposed project aims to bridge boundaries in sarcoma research by studying mouse models of a particular gene mutation that occurs in several sarcoma sub-types including rhabdomyosarcoma, UPS, and MPNST. In addition, our project will study sarcomas that occur in children, young adults, and adults.

Background

The role of NF1 in soft-tissue sarcomas

The neurofibromin 1 gene (NF1) is a tumor suppressor that functions as a negative regulator of the Ras pathway. Genomic analyses on sarcomas with complex karyotypes have determined that novel somatic mutations in the NF1 gene are common in a wide spectrum of patient sarcomas. Using genomic sequencing and copy number analysis, NF1 mutations were identified in myxofibrosarcoma (10.5%) and pleomorphic liposarcoma (8%).¹ Parallel studies have identified genomic loss of the NF1 locus in 15% of pediatric embryonal rhabdomyosarcomas.² This deletion was mutually exclusive with Ras gene mutations. In addition to identifying NF1 mutations in spontaneous sarcomas in the general population, patients with Neurofibromatosis Type 1 (NF1) are at increased risk for developing aggressive soft-tissue sarcomas. Neurofibromatosis Type 1 effects 1 in 3000 live births and is caused by germline inactivation of a single allele of the NF1 gene.⁶ The most common sarcoma afflicting NF1 patients is malignant peripheral nerve sheath tumors (MPNST), which has a lifetime risk of 8-13% for neurofibromatosis patients.⁶ NF1 patients are also at risk for developing non-neurogenic sarcomas, including rhabdomyosarcoma (RMS) and undifferentiated pleomorphic sarcoma (UPS), with a 3-6% lifetime risk.^3-5

Genetically Engineered Mouse Models

Genetically-engineered mouse models have greatly advanced our understanding of sarcomas⁷ and have served as a preclinical platform for therapies.^8-12 However, a limitation of the current models for NF1-mutant sarcoma is the development of sporadic tumors at multiple sites, making tumor detection and monitoring response to treatment challenging. In addition, although some of these models develop non-neurogenic sarcomas, they do not lend themselves to studying the full spectrum of NF1-mutant sarcomas that develop in patients. A model that temporally and spatially restricts tumor growth would not only provide a system to increase our understanding of NF1-mutant sarcoma, but would also serve as a platform for testing novel therapies.

Generating NF1-deleted sarcoma in the mouse

To study NF1-deleted sarcoma, we have generated an inducible mouse model of NF1-deleted sarcoma that better reflects the diverse spectrum of NF1-associated sarcomas found in patients. We generated mice with conditional mutations in both NF1 and Ink4a/Arf (NF1^flox/flox; Ink4a/Arf^flox/flox). Under normal conditions in these mice, both NF1 and Ink4a/Arf are expressed at endogenous levels. However, in the presence of Cre recombinase, a critical part of each tumor suppressor gene is removed, resulting in deletion of both NF1 and Ink4a/Arf. By injecting an adenovirus containing Cre recombinase (Ad-Cre) into the mice, we generated models of pediatric and adult sarcomas, including MPNST, rhabdomyosarcoma and UPS. We are now using this model to characterize the biological properties of NF1-deleted sarcoma and as a platform to test therapeutic agents.

Controlling gene expression in the mouse

Cre-loxP technology is used to control the expression of genes in specific tissues or at certain timepoints in development. This technique acts like molecular scissors to delete particular genes in the mouse. The gene of interest is “floxed” by surrounding it by a pair of specific DNA sequences called loxP sites. Under normal conditions, floxed genes are expressed at typical levels in the animal. However, addition of Cre recombinase cuts out the DNA surrounded by loxP sites, resulting in deletion of the floxed gene. By controlling the expression of Cre recombinase, scientists can delete genes at specific timepoints or in different tissues in the animal. In our model, we control the time and place of Cre recombinase activity by directly injecting the animal with a virus containing Cre recombinase. Based on the site of injection, we can generate tumors at different sites in the animal.

Molecularly-targeted therapies in NF1-deleted sarcomas

The identification of NF1 mutations in myogenic sarcoma^1,2 provides a new foundation for preclinical studies examining targeted therapies for soft-tissue sarcomas. Because sarcomas in this model develop at a specific site within a specific time window, this model is particularly useful for testing therapeutic drug responses. Additionally, these primary tumors develop from a small population of initiating tumor cells within the native tumor stroma, facilitating the study of endogenous immune cells, cellular growth, and native vasculature in a primary tumor context. To determine the utility of this novel mouse model in examining therapeutic response, we are now testing novel cancer therapies in this pre-clinical model of NF1-deleted sarcomas.

Aims of This Study

The long-term goal of our research is to use these mouse models of NF1-deleted sarcoma to identify therapies that will better serve sarcoma patients. We believe this project has the potential to significantly improve the ability to assess drug response in NF1-deficient sarcomas in the preclinical setting. Specifically, this proposal aims to further develop these mouse models of NF1-deleted sarcoma to identify therapies that will better serve sarcoma patients. In Aim 1, we will demonstrate the ability to use this mouse model as a preclinical platform to identify novel therapies that may be beneficial for patients with NF1-deleted sarcoma. Concurrently, in Aim 2, we will develop a new tumor model to serve as a faithful model of NF1-associated sarcomas in neurofibromatosis patients.

Aim 1: Demonstrate efficacy of the mouse model in therapeutic studies.

We have generated a novel genetically engineered inducible mouse model of NF1-deficient sarcomas. This primary mouse model can be used to study the tumor biology and therapeutic response of sarcomas with NF1 mutation. Using temporally and spatially controlled conditional gene deletion, we have determined that inactivation of NF1 and Ink4a/Arf in the tumor cells is sufficient to initiate both myogenic and MPNST-like sarcomas in mice with a wild-type stroma. We anticipate that this model will be useful as a pre-clinical platform for novel therapies and for studies to examine the interplay between the microenvironment and NF1-deficient sarcoma cells during therapeutic response. An important feature of the NF1-deleted model is the ability of tumors to develop within their native microenvironment, facilitating the study of stromal contribution to therapeutic response, which may be more challenging in cell transplant or xenograft studies.

In this aim, we will use our mouse model described above to address unanswered questions in sarcoma therapy. The model will be used as a preclinical platform to identify novel molecularly-targeted therapies and chemotherapies that may be beneficial for sarcoma patients. By generating sarcomas at a spatially restricted site, we will be able to accurately measure tumor volumes and assess treatment response via caliper measurement. Because the primary tumors develop within a native, immunocompetent tumor microenvironment, we will be able to study the effects of the immune system on therapy.

Chemotherapy can play an important role in treating NF1-associated advanced MPNST.^13,14 Studies have shown that some MPNSTs are responsive to combination chemotherapies regimens,¹³ but it is not known if other NF1-deleted sarcomas will respond better to these regimens. A clinical trial is currently attempting to enroll patients to determine if single-agent vs. combination chemotherapy is beneficial for patients with sporadic MPNST, but patient enrollment is low. Therefore, we will utilize the mouse models to determine if either single-agent or combination chemotherapy is beneficial for NF1-deleted sarcomas in mice. This information can direct care for sarcoma patients with NF1 mutations that may benefit from the addition of new agents to current treatments.

Aim 2: Demonstrate feasibility of deleting NF1 in tumor stroma.

The sarcomas initiated by NF1 mutation discussed above model sporadic sarcomas in the general population. It is not known if NF1-deleted sarcomas that arise spontaneously respond differently to treatment than sarcomas in neurofibromatosis patients. These differences could have important clinical implications that could impact patient care. In this aim, we will generate sarcomas with NF1 +/- stroma to model sarcoma development in neurofibromatosis patients with NF1 haploinsufficieny (+/-). The role of the supporting stroma (endothelial cells, immune cells, fibroblasts, etc) in the development and vascularization of NF1 patient neurofibromas is well-established.^12,15 The ultimate goal of this aim is to study the stromal contribution of NF1 haploinsufficient cells to sarcoma formation. The characterization of this model will provide new opportunities to use this model in future studies.

Summary

There is a critical need to examine the biology of NF1-mutant high-grade sarcomas in mouse models, particularly due to the prevalence of these tumors in the general population. The NF1-deleted primary mouse tumor models discussed above will significantly advance the understanding of NF1 biology in sarcomas by modeling a particular gene mutation that occurs in several sarcoma sub-types, including rhabdomyosarcoma, UPS, and MPNST. In addition, our project will study sarcomas that occur in children, young adults, and adults. This mouse model system has several advantages. First, by spatially and temporally restricting tumor development, therapeutic studies can be easily performed. Second, this model develops a spectrum of sarcomas that are seen in the patient population. Third, tumors develop within their native tumor microenvironment, allowing us to study the stromal contribution to tumor growth and therapeutic response that are not possible by cell transplant or xenograft studies. Taken together, we believe this project will increase our understating and treatment options for sarcomas with NF1-deletion, including MPNST, rhabdomyosarcoma, undifferentiated pleomorphic sarcoma, and others.

V10N2 ESUN Copyright © 2013 Liddy Shriver Sarcoma Initiative.

New Grant Funds Research on the NF1 Gene in Sarcomas

An ESUN Announcement

The Liddy Shriver Sarcoma Initiative has announced the funding of a $69,000 grant to fund promising sarcoma research by investigators at Duke University. In the study, David Kirsch, MD, PhD and Rebecca Dodd, PhD will explore the role of neurofibromin 1 (NF1) mutations in the development and therapeutic response of many kinds of sarcoma, including rhabdomyosarcoma, undifferentiated pleomorphic sarcoma, and malignant peripheral nerve sheath tumor.

NF1 mutations are present in many types of sarcoma, yet their role in the disease and its response to treatment is an unexplored area of sarcoma research. The researchers hope to use mouse models of NF1-deleted sarcomas to identify therapies that will better serve sarcoma patients.

A New Mouse Model

According to Dr. Dodd, studying sarcomas in mice is an excellent way to advance research in rare diseases. She explains: "Several [human] trials have attempted to examine if single-agent chemotherapy is better than combination chemotherapy for specific subtypes of sarcoma, but it is difficult to enroll enough patients in these studies. We can conduct a similar study in primary tumors in our mice, hopefully addressing these questions."

A mouse model developed by Dr. Dodd is the key to this particular research study. The new model will isolate tumor growth and allow more effective investigations into the behavior and treatment response of NF-1 deleted sarcomas. She adds, "One of the great features of this mouse model is that the findings can be applied to multiple subtypes of sarcoma, since mutations in the NF1 gene are found across the sarcoma spectrum."

Important Sarcoma-Specific Research and Tools

Dr. Kirsch admits that it can be difficult to get funding for studies like this one. In his experience, grant applications to fund sarcoma research are often considered 'not significant' because sarcomas affect a relatively small number of patients. Dr. Kirsh thinks differently: "I know how much sarcomas impact my patients and their families and how great the need is to develop better sarcoma treatments. Therefore, it is critical that funding be made available specifically for sarcoma research."

This study will not only provide new insights into several types of sarcoma, but it should improve upon the tools that are available to researchers. Dr. Dodd says, "Since sarcomas have such a unique biology, it is vitally important to have sarcoma-specific tools to advance the field of research. One of our goals is to continue improving the available preclinical tools, so that more faithful studies can be done to examine the biology and therapies for sarcomas."

The Funding

This $69,000 grant is co-funded by the Alan B. Slifka Foundation and the Liddy Shriver Sarcoma Initiative, each of which provided $34,500 in support. The Alan B. Slifka Foundation is a private grant-making foundation that is dedicated to making a world safe for difference and healing. The foundation, based in New York, also sponsors novel biomedical research and innovative interventions for Asperger disorder, autism and sarcoma.

The Liddy Shriver Sarcoma Initiative acknowledges a generous donation from the Thumbs Up For Lane Goodwin Childhood Cancer Foundation and donations in memory of Denise Grove, who lost her life to MPNST, and in memory of Brett Reed, Craig Dion, and Michael Cretella, all of whom lost their lives to rhabdomyosarcoma, and in honor of Samara Sheller, who is still under treatment for rhabdomyosarcoma. We are deeply grateful for their help in funding this important research.

From the Investigators

David Kirsch, MD, PhD: As a radiation oncologist at Duke, I work within a multi-disciplinary team to care for patients with sarcomas. I know that for some of our patients, our current treatments provide a cure. However, on a regular basis, I also see that our state-of-the art treatments fail some sarcoma patients. Moreover, for many patients, who are cured of their sarcoma, our current treatment causes short-term toxicity and long-term side effects. Recognizing these challenges in the clinic has motivated me to focus my laboratory on studying sarcoma development, metastasis, understanding how current sarcoma therapy works (or does not work), and identifying novel therapies for sarcoma.

Rebecca Dodd, PhD: I really enjoy working with mouse models to answer questions in sarcoma development and therapy. I believe they are powerful tools in the search to develop new treatments, which is so critical for the sarcoma field at this time. While in the Kirsch lab, I developed the model of NF1-deleted soft-tissue sarcoma that we are using in this study. I am excited to be continuing this research to further develop the model as a preclinical platform for sarcoma therapy. During these studies, I will be conducting the majority of experiments and analysis for the project.

V10N2 ESUN Copyright © 2013 Liddy Shriver Sarcoma Initiative.