Combination of Zoledronic Acid with Chemotherapy in Mouse Preclinical Models of Ewing's Sarcoma

By Guillaume A. Odri, MD; Nadège Corradini, MD and Françoise Rédini, PhD

Background

Ewing's sarcoma tumors represent the second most common type of bone tumors in children and adolescents, with a peak incidence at 15 years old. In Europe, most patients are treated in the Euro-EWING99 clinical trial with 73% EFS for 3 years for localized tumors. As patients with metastatic tumors outside lungs or pleura or not responding to chemotherapy and have a poor survival rate (EFS and OS < 30%), there is an urgent need to develop adjuvant therapies to improve outcomes. Tumor development in a bone site is associated with a vicious cycle between tumor cell proliferation and bone resorption: tumor cells secrete osteoclast activating factors [such as Receptor activator of NF-kappa B Ligand (RANKL)] which induce the differentiation and activation of osteoclasts, the only cells able to resorb bone. Osteoclast activity allows the release of growth factors stored in the bone matrix which in turn activate tumor cell proliferation. Therefore, targeting molecular or cellular protagonists involved in the bone resorption process represents an interesting therapeutic option for Ewing's sarcoma (EWS). Among inhibitors of bone resorption, Bisphosphonates (BPs) are promising candidates because they have been used with success for many years to treat complications associated with bone metastases, as demonstrated in several randomized clinical trials, in particular for prostate and breast cancers.¹ Beyond their osteoclastic activity, clinical and pre-clinical studies suggest that these molecules also exert a direct antitumor activity, inducing apoptosis in several tumor cell lines including primary bone tumors such as osteosarcoma or chondrosarcoma. Preclinical data obtained in our lab demonstrated that the combination of chemotherapy with zoledronate (N-containing BP) clearly have synergistic effects on inhibition of tumor development, of osteolytic lesions and on the augmentation of animal survival in two animal models of osteosarcoma.^2,3 From these results, the French OS2006 protocol has been established associating zometa^® with conventional therapy in children and adults.

Objectives

Two main objectives were developed in this program:

1. At the primary tumor level: The objectives were to study the potential therapeutic effects of zoledronate (ZOL) alone or combined with chemotherapy in Ewing's sarcoma using complementary in vitro analyses (mechanisms of action: proliferation, apoptosis, cell cycle, resistance, invasion..) and preclinical approaches. Bone and soft tissue models of Ewing's sarcoma were induced by injection of human tumor cells respectively in the tibia medullar cavity or in intra-muscular site in immune-deficient mice. Zoledronate was delivered subcutaneously at the dose of 50-100 microgr/kg, twice/week during 4-5 weeks (corresponding to clinical relevant pediatric dosing regimen in oncology) associated with ifosfamide (15 mg/kg as 1 to 3 courses). Several parameters were analyzed during the treatments by radiography, microCT (bone micro-architecture), histology, and biology (bone remodeling parameters in the serum). The mechanisms of sensitivity/resistance were also studied, thanks to the 10 human Ewing's sarcoma cell lines available.

2. Second, the effect of zoledronic acid was also studied on the invasion and metastasis ability of Ewing's sarcoma cells respectively in vitro and in vivo.

Results

The relevance of testing the therapeutic potential of zoledronic acid in Ewing's sarcoma is based on the existence of complex interactions between tumor cells and bone cells in the bone tumor microenvironment (Figure 1) . As described for osteosarcoma⁴, high amount of RANKL is present in the tumor microenvironment, leading to osteoclast activation. This activation can be direct (direct production of RANKL by tumor cells) or indirect through osteoblasts via the production of soluble factors (cytokines, growth factors) that further activate osteoclasts. When they are activated, mature osteoclasts degrade the bone matrix and allow the release of growth factors stored in the bone matrix, such as Transforming Growth Factor-b (TGF-b), Insulin-like Growth Factor (IGF) and Platelet Derived Growth Fractor (PDGF) that in turn activate tumor cell proliferation.

So we decided to analyse the effect of zoledronic acid (ZOL), a bisphosphonate of the third generation, which not only strongly inhibits osteoclast activity, but also influences Ewing's sarcoma cell proliferation in vitro. Preliminary results from the lab evidenced in vivo the major influence of the bone microenvironment on a ZOL-induced inhibitory effect on primary bone tumor development. Indeed, ZOL strongly inhibits tumor progression in the bone in a model induced by intra-tibia injection of Ewing's sarcoma cells (the mice were treated 8 weeks with 100 µg/kg ZOL, twice a week), whereas no effect was observed in a model induced in soft tissue.⁵ The hypothesis is that ZOL is highly concentrated in the bone matrix and is at higher concentration for adjacent tumor cells in bone than when tumors are induced in soft tissue, where ZOL cannot diffuse and concentrate.

A grant from the Liddy Shriver Sarcoma Initiative allowed us to complete this work by studying in depth several aspects of ZOL inhibitory effects on Ewing's sarcoma progression, by complementary in vitro and in vivo approaches. The major aim of this study was to establish proof-of-concept of the beneficial therapeutic effect of ZOL in Ewing's sarcoma before its transfer to the clinics.

1. Effect of zoledronic acid (ZOL) on primary tumor development.

In light of previous promising results observed in bone sites, we wanted to study the effect of ZOL on early Ewing's sarcoma development in bone. Mice were treated or not with 50 or 100 µg/kg ZOL twice or 3 times a week, as early as 1 day after human Ewing's sarcoma A-673 cell injection (106 in the medullar cavity of tibia) up to 18 days. These doses correspond respectively to the paediatric (50 µg/kg, 3x/week) or adult (100 µg/kg, 2x/week) dosing regimen proposed to patients.

At euthanasia, tumor bearing legs and lungs were analyzed by histology. A schedule was developed (Figure 2).

ZOL effect on early tumor development in bone

The invasion of bone tissue by Ewing's sarcoma cells was analysed on tumor + bone tissue slices by Masson's trichrome staining. No differences could be seen on intra-tibia progression up to day 10 after tumor cell injection (Figure 3A: CT and B: ZOL-treated mice).

At day 13-18 after tumor cell injection, the cortical bone was destroyed by Ewing's sarcoma cells in control untreated mice, whereas it was preserved in ZOL-treated animals. The consequence was an early invasion of adjacent soft tissues by tumor cells in control mice (CT, figure 4A).

On the contrary, the tumor remains contained in the medullar cavity in ZOL-treated mice and tumor cells did not invade surrounding soft tissues (figure 4B: ZOL treated mice). However, very few Ewing's sarcoma cells could be seen in the cortical bone of ZOL-treated mice at the end of the experiment (day 35-42, figure 4D). Whereas in control mice, cortical bone destruction was often evidenced as soon as day10 after tumor cell injection (figure 4C).

In conclusion, in untreated mice, we could observe the early presence of tumor cells in the cortical bone (day 7-10) and the very rapid development of tumor volume due to invasion of surrounding soft tissues after cortical bone destruction.

In ZOL treated mice, a tumor development can be observed inside the bone in some cases and Ewing's sarcoma cells could be found in the cortical bone. However, the cortical bone was not degraded (as compared to tibias of untreated mice) and the only way for Ewing's sarcoma cells to invade surrounding soft tissues (seen in rare cases) was through the invasion and destruction of metaphysic bone. On the way out, the bone is slowed down and we can compare this process to a containment mechanism.

In addition, the presence of TRAP positive cells (in red) was compared in both cases, but no differences could be seen at the growth plate level (Figure 5A: CT and B: ZOL).

Direct or indirect ZOL effect in vivo: comparison between models induced by sensitive or resistant Ewing's sarcoma cell lines.

To study this critical aspect in more detail, the effect of ZOL was compared in two models of Ewing's sarcoma: one induced by sensitive cells (A673) and the other by resistant cells (TC-71). Results showed that ZOL significantly inhibits Ewing's sarcoma mean tumor development in sensitive model (Figure 6A: A673 cells), whereas it has no effect in a model induced by the injection of a resistant cell line (Figure 6B: TC-71 cells). The in vivo experiments confirmed the direct effect of ZOL on Ewing's sarcoma cells observed in vitro.

We also compared the effect of different protocols (doses, number and frequency of injections) to see whether we could improve the therapeutic effect of ZOL. Unfortunately, no significant difference could be seen. Therefore, we continued with both dosing regimens: 50 µg/kg, 3 times/wk (human paediatric protocol) or 100 µg/kg, 2x/week (adult protocol).

2. Mechanisms of ZOL resistance (FPP synthase involvement)

The first step concerned the screening of all Ewing's sarcoma cell lines available in the laboratory for their sensitivity to zoledronic acid. The results showed dose-dependent effects of ZOL on the seven human EWS cell lines studies, but some cell lines were more sensitive than the others (Figure 7).

ZOL inhibits EWS cell proliferation by inducing caspase-dependent apoptosis and blocks the cell cycle in S-G2M phase. The seven human EWs cell lines differ in p53 status, but no correlation could be establish between p53 status and ZOL sensitivity.

Another aspect that we were interested in was the mechanisms of ZOL "resistance" in Ewing's sarcoma. Previous studies from our laboratory evidenced that in osteosarcoma, ZOL resistance was associated with the level of one of the proteases involved in the mevalonate pathway: the farnesyl diphosphate (FPP) synthase.⁶ So we analyzed by RT-qPCR the level of FPP expression in the seven EWS cell lines studied. Unfortunately, no correlation could be evidenced, suggesting that the mechanisms of ZOL resistance are different between osteosarcoma and Ewing's sarcoma.

3. Effect of ZOL on invasion and metastasis dissemination (boyden chambers, lung metastasis)

Another issue that has never been studied is the effect of ZOL on Ewing's sarcoma cell invasion, migration and metastasis ability.

Ewing's sarcoma cells were cultured in Boyden chambers covered with Matrigel. This type of experiment provides the first information about invasion ability of tumor cells. The results clearly demonstrated that ZOL-treated EWS cells lose their ability to invade the Matrigel, the same number of tumor cells being deposited on the upper part of the chamber (figure 8 A and B). The mean number of tumor cells that invade the Matrigel drop down from 130/field for the control untreated cells to 24/field for the ZOL treated cells.

Next we wanted to determine the mechanism involved in the ZOL-induced inhibition of cell invasion. For this purpose, the activity of two major proteinases involved in cell migration and invasion (MMP-2 and MMP-9) was analyzed by the zymography technics. Supernatants of EWS cells treated or not with ZOL were analyzed by gel electrophoresis containing gelatine (denatured collagen) as a substrate. The results showed a decrease mainly of MMP-2 activity in ZOL-treated A673 cells, but also a slight inhibition for MMP-9 (Figure 9).

Complementary analyses were performed in vivo in mice injected (intra-osseous) with EWS cells and treated or not with ZOL (50 µg/kg, 3x/week). Lung metastases were analysed at autopsy macroscopically and by histology (CD99 staining).

Lung metastases were numerous, small and big in the control untreated group (figure 10A), whereas few metastases (but always big) were observed in ZOL treated mice (figure 10B).

In fact, it appears that "big" metastases reflect the initial injection of tumor cells in the bloodstream (at the time of intra-tibia injection), corresponding to "experimental" metastases. Therefore, the fact that very few or even no small metastases are detected in the ZOL-treated group suggests that ZOL inhibits the dissemination of spontaneous metastases from primitive bone tumors. This is the first time that an inhibitory effect of Zoledronic acid was evidenced on metastasis formation whatever the site of metastasis.

This study's results on the effects of ZOL on Ewing's sarcoma cell invasion and metastasis dissemination have been submitted for publication in BMC Cancer.

Conclusions

These results strongly suggest that zoledronic acid is a very promising therapeutic option for primary bone Ewing's sarcoma tumors, and it also inhibits lung metastasis dissemination, one of the negative prognosis factors for Ewing's sarcoma patients.

These results have been presented to clinicians, and the addition of zoledronic acid to conventional chemotherapy is proposed for the next EuroEWING protocol that will begin in France, Great Britain, Italy and potentially other European countries in 2013.

V10N5-6 ESUN Copyright © 2013 Liddy Shriver Sarcoma Initiative.

Targeting Bone Microenvironment as New Therapeutic Strategy in Ewing's Sarcoma

By Guillaume A. Odri, MD; Nadège Corradini, MD and Françoise Rédini, PhD

Abstract

Ewing's sarcoma represents the second cause of malignant bone tumors in children and adolescents. In Europe, most patients are treated in the Euro-EWING99 clinical trial with 73% event free survival (EFS) at three years for localized tumors. Survival rates are poor, however, for patients with metastases outside the lungs or pleura and patients not responding to chemotherapy (EFS and overall survival < 30%).

Tumor development in bones is associated with the existence of a vicious cycle between tumor cell proliferation and bone resorption. Therefore, targeting the bone resorption process represents an interesting therapeutic option for these tumors. The present project aims at combining zoledronic acid, a potent inhibitor of bone resorption, with chemotherapy in preclinical models of Ewing's sarcoma. The impact of these studies will be to propose zoledronic acid as adjuvant therapy for Ewing's sarcoma patients in the next European protocol to help prevent recurrence and metastasis and to improve prognosis for patients with metastatic or unresponsive disease.

Introduction to Ewing's Sarcoma

Ewing's sarcoma was first described in 1921.¹ It is the second most frequent primary bone tumor before 20 years of age, occurring mostly in the second decade of life, with a frequency of 1 per million per year.² The disease is characterized by a rapid tumor growth and extensive bone destruction with periostal reaction (Figure 1).

Ewing's sarcoma features a specific chromosomal translocation between the EWS gene on chromosome 22 and a gene of the erythroblast transformation sequence (ETS) family. This translocation leads to the production of the oncogenic fusion gene EWS-FLI1 (t(11;22)(q24;q12)) in 85% of cases but also to EWS-ERG (t(21;12)(q22;q12)) in 10% of cases and EWS-ETV1,4 or -FEV in the remaining 5% of cases.^3-5 The protein encoded by this fusion gene is an aberrant transcription factor that promotes tumorogenicity in mesenchymal stem cells.⁶ Other tumors have been found to have the same translocations and are now part of the Ewing's sarcoma family of tumors. The presence of this fusion gene is used as a specific diagnostic marker of the Ewing's tumor family thanks to fluorescence in-situ hybridization and RT-PCR.

The Treatments and Their Limits

Before chemotherapy was introduced, Ewing’s sarcoma patients survival was poor, ranging from 3 to 15% at five years.^7,8 Because Ewing’s sarcoma is very sensitive to radiation therapy, it was the preferred treatment in the early years of understanding the disease. However, this led to a poor five-year survival, and surgery was considered the best option when possible. Surgery consisted of primary amputation or radical excision, which led to 20% survival rates. Nevertheless, most of the patients died of metastases which were present at diagnosis (25%) or appeared after the treatment. It became evident that a systemic therapy was needed.

Chemotherapy was added to radiotherapy and surgery 50 years ago. The first trials used single chemotherapeutic agents such as cyclophosphamide, or dactinomycin, and showed encouraging results.^9,10 In 1972, Hustu et al found a 67% two-year disease-free survival after a combination of chemotherapy in 15 treated patients.¹¹ In 1978, Rosen et al. reported a 75% five-year survival for patients receiving multimodal chemotherapy with radiation therapy or surgery.¹² Since then, many controlled trials have been performed to determine the best scheme of chemotherapy with the least toxicity, but survival has remained around 60% for localized tumors, and 20% for metastatic disease. These statistics have not changed over the past 30 years. Prognostic factors have been identified, and the most important are the presence of metastases (patients with bone metastases have a worse prognosis than those with pulmonary metastases), the size of the tumor (<200mL is good prognosis), and the response to chemotherapy (< 10% of viable cells predicts good prognosis). Thus, patients with small localized tumors that respond well to chemotherapy and are removed with wide margins have more than 75% EFS at 10 years, whereas patients with bone metastases have less than 15% EFS at 10 years.^13-15

The Euro-EWING99 Protocol

Since 1999, the still accruing EURO-EWING group (which involves the SFCE, GCF, SIAK, EORTC, and EICESS European groups and the American COG group) tested the clinical benefits of a different chemotherapy combination involving Vincristine, Ifosfamide, Doxorubicin and Etoposide (VIDE). The protocol is composed of 6 sequences of VIDE treatment followed by surgery when possible. The histological response to chemotherapy is then evaluated and patients are divided into 3 arms according to the tumor localization at diagnosis, the volume for unresected tumors and the percentage of residual cells after treatment (Figure 2). The R1 arm includes patients with localized disease with good response to chemotherapy (<10% of residual cells) or with tumor volume below 200ml. The R2 arm involves patients with lung metastases and patients with localized tumor with poor response to chemotherapy or with a tumor volume above 200ml. Finally, the last arm (R3) includes patients with bone, bone marrow or multifocal metastases. For the R1 group, adjuvant chemotherapy is randomized between 7 sequences of a combination of Vincristine and actinomycine with cyclophosphamide (VAC) or ifosfamide (VAI). For the R2 group, adjuvant therapy is randomized between VAI and a stronger chemotherapy composed of busulfan and melphalan. For the R3 group giving the lower survival prognosis, treatment options consist in high dose of melphalan-melphalan, busulfan/melphalan with autologous stem-cell transplantation or, in the worst cases, a Phase II clinical trial. The current survival rate for EURO-EWING patients reaches 80% for localized disease of small volume (R1 arm). Unfortunately, five-year survival for patients with metastases detected at diagnosis remains around 25%, and even around 7% when relapse occurs in the first two years after treatment.

Accordingly, new therapeutic options should be tested, and the inhibition of tumor-associated bone resorption seems to be a promising axis showing encouraging preclinical results in bone tumor pathology.

Biology of Primary Bone Tumors: The Vicious Cycle

Skeletal complications remain the major issue of primary bone tumors. In osteosarcoma and Ewing's sarcoma, as well as in bone metastasis secondary to breast, lung or prostate carcinoma, skeletal lesions lead to bone pain and pathological fractures. Skeletal integrity is controlled by a balance between bone formation and bone resorption, under the control of osteoblasts and osteoclasts, respectively. This equilibrium is altered by tumor cells leading to osteolytic and/or osteoblastic lesions. Moreover, the interaction between tumor cells, tumor-derived humoral factors and bone marrow microenvironment has shown to be essential for bone tumor initiation and promotion.

As evidenced in bone metastases, a vicious cycle occurs during bone tumor development between bone resorption and tumor proliferation.^{16, 17} The tropism of primary and metastatic tumor cells for bone is linked to the presence of growth factors, such as Insulin-like Growth Factor-1 (IGF-1), Transforming Growth Factor-β (TGF-β) or Fibroblast Growth Factor (FGF), in the bone microenvironment itself after their release from the bone matrix during bone resorption.^18-20 All those factors stimulate tumor engraftment and proliferation according to the "seed and soil" hypothesis.²¹ In turn, tumor cells produce pro-osteolytic factors (such as hormones, cytokines and growth factors) directly or indirectly through stromal cell interaction.^22-24 Among these factors, parathyroid hormone related peptide (PTH-rP) promotes osteoclast differentiation, activation and survival through Receptor Activator of NF-Kappa B Ligand (RANKL) production by osteoblasts, leading to final bone resorption. Therefore, the increased bone resorption favours the vicious cycle by inducing the release of growth factors stored in the bone matrix and further enhances tumor proliferation (Figure 3). Increased calcium concentration in the bone microenvironment induced by bone destruction has also been shown to increase tumor growth and PTH-rP production. Among the new investigational approaches to improve therapy in Ewing's sarcoma, bone-specific agents may improve survival and/or quality of life on "continuation" therapy. This would include regimens with fewer short- and long-term side effects and better results for tumors in difficult locations and patients with recurrent disease.

Bisphosphonates, Anti Bone Resorption Drugs

Bisphosphonates (BPs) are an important class of molecules for the treatment of bone diseases with different molecular mechanisms of action. Bisphosphonates (BPs) are structural analogues of pyrophosphate exerting a strong inhibition of bone resorption.²⁵ Two main families can be distinguished: Nitrogen- and non-nitrogen-containing BPs acting on osteoclasts by different molecular mechanisms (Figure 4). In any case, the final common result is the induction of osteoclast apoptosis. BPs act by inhibiting the recruitment, proliferation and differentiation of preosteoclasts or by impeding the resorptive activity of mature osteoclasts.^26-29

For several years nitrogen-containing BPs (N-BPs) have been used with success to treat complications associated with bone metastases, as demonstrated in several randomized clinical trials, in particular trials for prostate and breast cancers. Beyond their osteoclastic activity, clinical and pre-clinical studies suggest that these molecules also exert a direct anti-tumor effect, inducing apoptosis in several tumor cell lines.³⁰ This effect is reinforced by the association with chemotherapeutic agents such as paclitaxel, gemcitabine, doxorubicin or epirubicin. Among all the bisphosphonates tested, zoledronic acid (ZOL), one of the third generation nitrogen-containing BPs, shows the greatest inhibitory effects on both osteoclast activity and tumor cell proliferation.³¹ Quite appropriately, these agents are increasingly used alongside anticancer treatments to prevent skeletal complications and relieve bone pain.

Rationale of the Project

Relevance of Using Bisphosphonates in Primary Bone Tumors Including Ewing's Sarcoma

Considering the "vicious cycle" hypothesis, BPs may have complementary actions on primary bone tumors: an indirect anti-tumor effect by inhibiting osteoclastogenesis and a direct action on tumor cells. Among primary bone tumors, giant cell tumors of bone are characterized by a major osteoclastic component which justifies the clinical use of BPs in this pathology.³² Because osteosarcoma and Ewing's sarcoma cells are metabolically active bone cells,^33,34 studies with BPs and more specifically the third generation nitrogen-containing BPs such as ZOL have shown selective uptake and "poisoning" of these cells.^35-39 We have previously demonstrated that ZOL is able to limit tumor progression in a rat model of osteosarcoma, to prevent tumor relapse as compared to chemotherapy alone and to prevent osteolytic lesions.40 Our results have provided the rationale for the French randomized clinical protocol OS2006 which combines ZOL with conventional therapy for adult and pediatric patients. Since then, other fundamental or preclinical studies have confirmed the beneficial effect of bisphosphonates in osteosarcoma.^36-39,41,42

Preliminary Data

The first preclinical study reporting the therapeutic benefit of using ZOL to treat Ewing's sarcoma showed an inhibitory effect on osteolysis, but no data on survival and tumor volume were given.⁴³ We then performed in vitro experiments to evaluate the effect of ZOL on eight human Ewing's sarcoma cell lines, all expressing the EWS-FLI1 fusion gene but differing in their molecular characteristics by their p53 mutation status and p16/ink4 deletion. Then, in vivo experiments were performed to study the effects of ZOL on tumor progression and animal survival, as a single therapeutic agent or in combination with ifosfamide, a conventional drug used in Ewing's sarcoma clinical protocols. Two different animal models of Ewing's sarcoma were developed reproducing both clinical behaviors of Ewing's sarcoma: soft tissue and bone development (respectively 15% and 85% of total Ewing's sarcomas) (Figure 5A and B, respectively). From these preliminary experiments, several promising results were obtained. Briefly, mice were treated with ZOL and/or ifosfamide, both at doses corresponding to clinical ones administrated in patients. Ewing's sarcoma cell lines showed different sensitivity to ZOL and mafosfamide, but no correlation with their molecular status was evidenced.

In our experimental model of Ewing's sarcoma, ZOL had no effect on soft tissue tumor progression, although it dramatically inhibited Ewing's sarcoma development in bone (Figure 6A). ZOL also prevented the development of bone lesions associated with tumor progression (Figure 6B). When combined with chemotherapy, ZOL exerted synergistic effects in the model developed in soft tissue: its combination with one cycle of ifosfamide resulted in an inhibitory effect on tumor development similar to three cycles of ifosfamide alone. These data have been accepted for publication in Cancer Research. These promising results prompted us to propose a broader scientific program to study the benefits of using ZOL as adjuvant therapy in Ewing's sarcoma tumors and metastases and to study the molecular mechanisms involved in ZOL resistance, which is the purpose of the present project.

Objectives

The objectives of this program are to study the potential therapeutic effects of zoledronate (ZOL) alone or combined with chemotherapy in Ewing's sarcoma using complementary in vitro analyses (mechanisms of action: proliferation, apoptosis, cell cycle, resistance) and preclinical approaches (bone and soft tissue models of primary Ewing's sarcoma induced by injection of human tumor cells respectively in the tibia medullar cavity or in an intra-muscular site in immune-deficient mice). The effects of this therapeutic combination will also be studied in models of bone or pulmonary metastases induced by injection of Ewing's sarcoma cells via intraveinous or intracardiac routes, respectively. Indeed, the patients with high risk of relapse are those who present with metastases at diagnosis or whose tumors are resistant to chemotherapy.

Methods

1. In vitro experiments

Initial resistant human Ewing’s sarcoma cell lines (EW-24, TC-32, TC-71, SK-N-MC,) will be analysed by proliferation assay to confirm their resistance (cell viability tests) to zoledronate (1-100 µM, 24 to 72 hours).

The expression of the main enzyme targeted by N-BP, the farnesyl di-phosphate synthase (FPPs), will be analyzed and compared between resistant and sensitive Ewing's sarcoma cell lines, because its level of expression has been correlated with ZOL resistance in osteosarcoma cell lines.⁴⁴

Gene expression profiles will be compared between one sensitive (A673) and one resistant (TC-71) cell line, thanks to collaborative study with Drs O. Delattre and F. Tirode (INSERM U830, Institut Curie, Paris, France).

2. Study of Biological bone markers in Ewing's sarcoma patients

In serum or plasma of patients: the level of bone markers will be analysed by ELISA [osteocalcin, Bone SialoProtein, Bone Alkaline Phosphatase, Tartrate Resistant Acid Phosphatase 5b, Collagen degradation products (CTX)] and cytokines involved in the regulation of bone remodelling (cytokines of the TNF family: OPG, RANK, RANKL, TRAIL; cytokines of the IL-6 family: IL-6, OSM, LIF...).

In biopsies of patients: predictive factors of N-BP sensitivity (FPPsynthase and other genes determined by microarray analysis) will be analyzed.

3. In Vivo Experiments

Complementary models of Ewing's sarcoma will be developed:

1. Using resistant human Ewing's sarcoma cell lines (mainly TC-71, but also EW-24 or TC-32 or SK-N-MC) injected into the medullar cavity of the tibia in order to study the direct and indirect anti-tumor effect of ZOL through bone resorption inhibition.
2. Models of pulmonary or bone metastases induced respectively by intraveinous or intracardiac injection of luciferase-expressing EWS cells.

The effect of ZOL will be tested in these models, alone or combined with chemotherapy (ifosfamide: 15 mg/kg 3 consecutive days/wk, during 3 wks). Several parameters will be studied in these animals, at the clinical (tumor volume, weight), radiography, micro-architecture (micro-scanner analysis), biological (bone remodelling parameters), histology (Masson’s trichrome, von Kossa, TRAP staining), and histomorphology levels.

Details of the protocols: Zoledronate will be delivered subcutaneously at the dose of 100 microgr/kg, twice a week for 4-5 weeks (corresponding to clinical relevant dosing regimen in adult oncology) or at the dose of 50 microgr/kg every two days, 10 times (corresponding to pediatric protocols), associated with ifosfamide (15 mg/kg as 1 to 3 courses) or doxorubicin (2 mg/kg, weekly).

Expected Results

This project will allow us to determine a molecular signature predictive of the resistance of human Ewing's sarcoma cell lines to ZOL. It will also help to understand and compare the relative involvement of the direct and indirect anti-tumoral effect of ZOL in the development of primary bone tumors such as Ewing's sarcoma. Our preliminary data published this year demonstrated the proof-of-concept for the therapeutic interest in using bisphosphonate such as zoledronic acid in Ewing's sarcoma. The proposed program will complete this work, especially for high risk patients (with bone and pulmonary metastases) and for those with tumors resistant to ZOL. The proposed scientific project includes fundamental and preclinical studies which will be undertaken in the research laboratory INSERM U957 directed by Pr Dominique Heymann, located in the faculty of medicine in Nantes (France).

The impact of these studies will be to propose zoledronic acid as adjuvant therapy for Ewing's sarcoma patients in the next European protocol (following the current EuroEWING99) to help prevent recurrence and metastasis and to improve prognosis for patients with metastatic or unresponsive disease. 

V8N3 ESUN Copyright © 2011 Liddy Shriver Sarcoma Initiative.

Grant Funding

The Liddy Shriver Sarcoma Initiative funded this $90,000 grant in June 2011. The study was made possible by generous gifts from the Arlo and Susan Ellison Family, the Jack Langseder 4evRSTRONG Foundation and Strike Out Sarcoma (in memory of Michael Lio).