Radiotherapy in bone sarcoma: the quest for better treatment option

Abstract

Bone sarcomas are rare tumors representing 0.2% of all cancers. While osteosarcoma and Ewing sarcoma mainly affect children and young adults, chondrosarcoma and chordoma have a preferential incidence in people over the age of 40. Despite this range in populations affected, all bone sarcoma patients require complex transdisciplinary management and share some similarities. The cornerstone of all bone sarcoma treatment is monobloc resection of the tumor with adequate margins in healthy surrounding tissues. Adjuvant chemo- and/or radiotherapy are often included depending on the location of the tumor, quality of resection or presence of metastases. High dose radiotherapy is largely applied to allow better local control in case of incomplete primary tumor resection or for unresectable tumors. With the development of advanced techniques such as proton, carbon ion therapy, radiotherapy is gaining popularity for the treatment of bone sarcomas, enabling the delivery of higher doses of radiation, while sparing surrounding healthy tissues. Nevertheless, bone sarcomas are radioresistant tumors, and some mechanisms involved in this radioresistance have been reported. Hypoxia for instance, can potentially be targeted to improve tumor response to radiotherapy and decrease radiation-induced cellular toxicity. In this review, the benefits and drawbacks of radiotherapy in bone sarcoma will be addressed. Finally, new strategies combining a radiosensitizing agent and radiotherapy and their applicability in bone sarcoma will be presented.

Introduction

Bone sarcoma are rare tumors accounting for 0.2% of all tumors with an incidence in North America and Europe of 0.75 / 100 000 [1]. Bone sarcoma can be classified according to the age of tumor onset. On the one hand, osteosarcoma (OS) and bone Ewing sarcoma (EWS) mostly affect children and young adults, and on the other hand chondrosarcoma (CHS) and chordoma (CD) occur after the age of 40 [1]. The survival rate of adults with bone sarcoma is low, around 50–60% at 5 years and 30% at 10 years, principally because of the indolent nature of these tumors [2,3,4,5]. For localized pediatric bone sarcomas, the 5-year survival rate is around 70% [2,3,4,5] and drops to 30% for pediatric bone sarcoma presenting metastases at diagnosis, which occurs in 20–25% of pediatric bone sarcoma [1, 3].

Notwithstanding the age of tumor onset or histological type of sarcoma, the management of all bone sarcoma patients is based on a transdisciplinary approach where surgery, with complete resection of the primary tumor, remains the cornerstone. Indeed, the quality of resection is an essential prognostic factor for all bone sarcomas. Depending on the location of the tumor and the tumoral invasion of peripheral tissues, surgery can be challenging and is not feasible in all cases. Radiotherapy is frequently used to ensure better local control [5,6,7,8]. In the case of Ewing sarcoma surgical resection and radiotherapy are both standard options for local control. Conversely, radiations are not applied as first-line treatment for resectable osteosarcoma, chordoma and chondrosarcoma, albeit high doses of radiotherapy are used as adjuvant treatment in case of marginal or incomplete resection, and as definitive local treatment for unresectable tumors [9,10,11]. Treatment strategy must be adapted according to tumor location, ease of resection and treatment-associated morbidity, as unnecessary high doses of radiation can trigger serious side effects such as neuropathies and fractures [12,13,14]. Challenging bone sarcoma of the axial skeleton are frequently treated with Intensity Modulated photon Radiation Therapy (IMRT) because of the higher dose applied to the tumor and the sparing of healthy tissues [15]. The development of advanced radiotherapy techniques like carbon-ion, or proton therapy has drastically improved patient care by reducing the exposure of nearby critical organs to radiations and increasing the dose of radiations delivered specifically to the tumor [12,13,14]. Combined proton and photon radiotherapy is also increasingly used for the treatment of sarcoma of the spine and sacrum and seems to improve local control [12, 16]. Excellent clinical results have been observed for sarcomas of the skull and cervical spine treated with proton therapy [13, 17]. Interesting results have also been reported with heavier particles, such as carbon ion. Access to these advanced RT techniques is increasing in developed countries. Hence, radiotherapy is an important component in bone sarcoma management, and in this review, we will discuss the benefits of radiotherapy for bone sarcoma, the mechanisms involved in tumor radioresistance, and the innovative ways to improve radiotherapy efficacy in these tumors.

Radiotherapy for bone sarcomas

The place of radiotherapy in the treatment of bone sarcoma has evolved with the development of new types of radiations and new ways to deliver these radiations (Table 1). Nevertheless, this evolution raises the question of choosing the best radiotherapeutic approach for the right tumor depending on patient age, tumor locations, histological subtype, tumor grade, previous treatment. The following paragraphs include an overview of the efficacy of conventional radiotherapy and non-conventional radiotherapy in bone sarcoma.

Table 1 Bone sarcomas

Role of radiotherapy in CD and CHS treatment

Over the past few years, more information has become available on the effects of radiotherapy in bone sarcoma patients with unresectable or residual tumors. In this part, we summarize treatment guidelines and present the latest clinical studies evaluating the efficacy of radiotherapy in bone sarcoma (Table 2).

Table 2 Radiotherapy in chordoma and chondrosarcoma

In chondrosarcoma, radiotherapy can be considered for unresectable disease (primary or recurrent), after incomplete surgery and for symptom palliation. High-dose RT is currently recommended for patients with skull base chondrosarcoma and for inoperable, locally advanced, and metastatic high-grade chondrosarcomas with a poor prognosis. For chordoma, en bloc R0 resection is the recommended treatment for primary localized disease when feasible and sequelae are accepted by the patient. If these conditions are not met, RT alone without debulking is an alternative. For skull base and upper cervical tract chordoma, resection with negative margins can rarely be done, and microscopically-positive margins should be the goal of surgery. Adjuvant RT should be considered for skull base and cervical spine chordomas, and for sacral and mobile spine chordomas with R1 resection margins.

A few historical retrospective studies have been conducted to determine whether chordoma and chondrosarcoma patients could benefit from peri-operative radiotherapy. Two major retrospective studies have evaluated the role of radiotherapy in chordoma, comparing surgery alone vs surgery and conventional radiotherapy in 1478 and 5427 chordoma patients, respectively (level of evidence 2b) [9, 19]. Both studies concluded that radiotherapy peri-operatively improves patient local control when surgery with positive margins are performed. High-dose RT is also associated with better outcome [9, 19]. The same observation has be made in a retrospective study of 743 high-grade chondrosarcoma defining radiotherapy as an independent protective factor (level of evidence 2b) [21].

Different advanced radiotherapeutic techniques have been developed in the last few decades (Table 3, see Table 2 [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35]). First, the use of proton therapy is associated with better outcome than conventional radiotherapy in both chordoma and chondrosarcoma [9, 19, 21, 23]. The administration of proton and photon therapy post-operatively tend to be more efficient with a 5-year local control rate of 85.4% in CD, while it does not exceed 74% when combining surgery and photon radiotherapy alone [22]. When radiotherapy is administered as a single treatment (e.g. in unresectable tumors), proton therapy is a better option than conventional radiotherapy for both CHS and CD, resulting in a 5-year overall survival of 75% for CHS and 100% for CD, whereas the 5-year overall survival is only 19.1% for CHS and 34.1% for CD for conventional radiotherapy [23]. In skull base chordoma and chondrosarcoma, which are particularly difficult to handle surgically due to their proximity to vital structures, carbon ion radiotherapy administered peri-operatively has shown promising results with a 5-year local control of 80% and 89% in CD and CHS, respectively [20]. Stereotactic Radiation Therapy (SRT) has also been used in both chordoma and chondrosarcoma, and retrospective studies reported different results, with local control rates varying between 57% at 10 months and 90% at 28 months [28, 29].

Table 3 Radiotherapy principles

Chondrosarcoma and chordoma have a very low incidence, thus international clinical trials uniting bone sarcoma centers worldwide are ongoing to determine the best therapeutic option depending on the type of the tumor, its localization (NCT05033288, NCT01182779) and its resectability (NCT02986516).

Role of radiotherapy in the treatment of Ewing sarcoma and osteosarcoma

Radiotherapy may be considered in osteosarcoma patients with unresectable tumors, primary tumors where surgery would be unacceptably morbid, or as adjuvant treatment of tumors at high risk of local recurrence and with limited option for further surgery. For patients with bone Ewing sarcoma, RT with definitive intent alone should be used instead of surgery if complete surgical excision is not possible and in cases with challenging local sites such as axial or spinal tumors, where surgery will be unacceptably morbid. Adjuvant RT (45–60 Gy) significantly reduces Local Recurrence in patients with large tumors (> 200 ml), poor histological response or inadequate surgical margins and should be recommended in these circumstances [IV, B].

In addition, adjuvant RT should be considered in patients with non-sacral pelvic Ewing Sarcoma regardless of surgical margins, tumor volume or histological response, as this was shown to provide superior local control and survival outcome compared with surgery alone.

Several studies aimed at determining the best use of radiotherapy for EWS patients comparing radiotherapy alone with i) surgery alone, ii) post-operative RT, or iii) polychemotherapy (see Table 4). In a retrospective study (INT0091, INT0154, AEWS0031), radiotherapy alone increased the rate of local relapse compared to surgery alone in EWS patients with localized tumors [31].However, no difference was observed in the overall survival and overall disease control between those two treatments [30]. For patients with extremity and pelvic tumors, surgery clearly improved local control compared to definitive radiotherapy (local relapse rates 3.7% and 3.9% vs 14.8 and 22.4%, respectively) [30]. For other tumor locations, no difference was detected between the different treatment groups. Of note, in this study, patients treated with surgery had favorable prognostic factors such as a younger age or tumors of the extremities, and most of the patients were treated with older techniques of radiotherapy. Another study compared the same treatment options (surgery vs radiotherapy vs combined treatment) in metastatic EWS. The combination of surgery and radiotherapy improved the local control of metastatic tumors compared to surgery or radiotherapy alone (EFS at 3 years: RT: 0.35, surgery: 0.35, combination: 0.56) [31].

Table 4 Radiotherapy in Ewing sarcoma and Osteosarcoma

Stereotactic Body Radiation Therapy (SBRT) (Table 1) uses several radiation beams of various intensities targeting the tumor from different angles and is considered an effective strategy for metastatic EWS and OS [33]. SBRT used to control pulmonary metastases was reported to lead to a 2-year local control of 60% in 13 metastatic patients (IV) [33]. In osteosarcoma, the local control at 5 years was shown to range between 68 and 72% with conventionally fractionated proton RT doses of 68-70 Gy (1.8-2 Gy per day) in a retrospective study including 41 OS unresected or incompletely resected [36]. Carbon ion radiotherapy was effectively used in the treatment of unresectable pediatric osteosarcoma, with a local tumor control of 62% at 5 years in a retrospective study [34]. More recently, Combined Ion Beam Radiotherapy with protons and carbon ions in a multimodal treatment strategy of inoperable osteosarcoma was evaluated. Results showed an overall survival and a progression-free survival of 68% and 45%, respectively (2b). These results are particularly promising in craniofacial osteosarcoma [35]. Recently, a randomized controlled phase III study evaluated the efficacy of carbon ions, photon, and proton therapy in chordoma and chondrosarcoma (except skull-based tumors). This study will be extremely valuable in determining the benefits of using carbon ion radiotherapy as it is a prospective study and it compares the effects to a reference treatment [37].

Even though chemotherapy is a preferred treatment choice, RT plays a primordial role in the treatment of bone sarcomas. The development of new techniques makes RT an approach of interest for the treatment of incompletely or unresectable tumors, for tumors localized near critical structures, and for metastases. These new radiotherapies can lead to a better management of sarcoma patients who have an unfavorable prognosis and limited treatment options. With great advances in the development of targeted therapies, moving on to personalized combination approaches able to enhance the efficacy of radiotherapy, may be a promising strategy. To achieve this goal, a better understanding of radiotherapy mechanisms of action is necessary.

Potential target for combination with radiotherapy in bone sarcomas

Radiotherapy is currently focused on the precise delivery of high doses of radiation within the tumor bulk, sparing surrounding healthy tissues. However, the development of targeted therapy arguably has the potential to enhance radiotherapy efficacy. The possibility to molecularly profile tumors at diagnosis, together with improvements in radiotherapy could potentially pave the way for a more personalized approach to bone sarcoma treatment. Several key molecular pathways could theoretically enhance the therapeutic effect of radiation. In addition, it is important to determine the timing for combining molecular targeted therapy with radiation, as it could greatly affect the outcome depending on which pathway is being inhibited.

To determine which potential pathway could be a promising target in bone sarcomas, it is first necessary to review the radiation process and its consequences at the cellular and molecular levels. This paragraph summarizes, in chronological order, the principal steps and actors involved in the cellular response to radiotherapy (Fig. 1).

Fig. 1

Schematic representation of the major known actors involved in radioresistance of bone sarcomas- when a damage occurs in DNA, ATM and ATR kinases are recruited and activate checkpoint kinases 1 and 2, leading to cell cycle arrest and to the recruitment of diverse effectors of DNA repair, such as the complex MRN PARP, RAD51, NBS1, RAD50. Diverse alterations in cell cycle proteins including p16 and CRIF, and in DNA repair proteins enhance bone sarcoma radioresistance. The accumulation of DNA damage is generally leads to cell death. However, bone sarcoma cells present defects in this pathway too, leading to cell survival after radiotherapy. Created using BioRender.com

Irrespective of the type of radiations used (e.g. X-rays, Proton, carbon ion), ionizing radiation affects all cellular compartments and their main target is DNA. Under ionizing radiations, micro-deposits of energy are generated in the nucleus near DNA. This accumulation of energy destabilizes and causes damage to the DNA structure. Moreover, by ionizing water molecules, a phenomenon known as water radiolysis, radiation triggers the formation of Reactive Oxygen Species (ROS) that lead to further DNA damage. DNA damage caused directly or indirectly by radiations, includes DNA oxidation, loss of a base, single-strand break and double-strand breaks [38, 39]. Double-Strand Breaks (DSB) are considered the most lethal type of lesions and are induced at a higher level by proton rather than photon therapy [40]. Each type of damage is recognized and corrected by specific repair mechanisms, each acting with a different degree of precision and speed (Table 5).

Table 5 Biological differences between photons, protons, and carbon ions

The response of a cancer cell to an ionizing radiation can be divided into several steps, from the recognition of the damage to the induction of cell death. At each step, bone sarcoma cells can have properties allowing them to counteract radiation-induced cell death, representing potential targets for combination therapy (Tables 6 and 7). Most of the studies on the biological effects of radiotherapy in bone sarcoma focus on X-rays or γ-rays, which will be presented in the next paragraph, and since very few studies (only 2 studies) deal with protons or carbon ions these will be presented when necessary.

Table 6 TP53 mutations in bone sarcomas

Table 7 Combination of radiotherapy and pharmacological inhibition of targets in bone sarcoma

DNA damage recognition

DNA damage is first recognized by 2 enzymes: Ataxia Telengiectasie Mutated (ATM) and Ataxia Telengiectasie RAD3-related (ATR). ATM recognizes double-strand breaks, while ATR can detect single-strand breaks and replication fork alterations. After the recognition of a DSB, ATM phosphorylates the histone H2AX (yH2AX), involved in stabilizing DNA extremities and in the recruitment of DNA repair complexes. ATM and ATR also phosphorylate the checkpoint kinases 1 and 2 (CHK1 and CHK2), leading to cell cycle arrest. ATR can phosphorylate many other substrates including Replication Fork components: MCM (MiniChromosome Maintenance) proteins, Rpa (Replication Protein A), polymerase, PCNA (Proliferating Cell Nuclear Antigen), and Claspin (Mrc1) [66, 67]. Cancer cells can resist radiation by increasing their efficiency in DNA repair through the increased expression of proteins involved in DNA damage recognition and repair, including ATM and ATR. A correlation was shown between radioresistance levels and the expression of 7 proteins involved in the DSB DNA repair machinery in 5 sarcoma cell lines, including one OS cell line. ATM, ATR and NBS (Nijmegen breakage syndrome protein 1), proteins involved in DNA damage recognition presented the strongest correlation [68]. In CD, an increased expression of ATM, ATR and yH2AX was observed in 26 patient samples in comparison with surrounding healthy tissue. However, this observation has not been directly correlated to the level of radioresistance of CD [69, 70]. Drugs targeting both ATR and ATM are already approved by the FDA and in clinical trials in other cancers (Bay1895344, NCT03188965; AZD1390, NCT03423628).

Once activated, ATM, CHK1 and CHK2 phosphorylate p53, the most studied tumor suppressing protein. P53 is the protein the most often mutated in all cancers and plays major roles in genomic stability, cell cycle regulation, cell death induction and in radioresistance.

P53 activation

P53 is a transcription factor that is stabilized following radiation and induces transcription of genes associated with cell cycle arrest, apoptosis, and metabolism, thereby functioning as a tumor suppressor [71]. Mutations affecting the normal functions of p53 are found in 80% of OS, 20% of CHS, and 10% of EWS (Table 6) [72]. Typically, the majority of TP53 mutations are missense mutations in its DNA binding domain, preventing TP53 from inducing transcription of its target genes and thus causing the loss of its tumor suppressive function [71]. In OS and CHS TP53 functions can also be altered indirectly through the amplification of Murine Double Minute 2 (MDM2) that results in P53 degradation. Recent results have demonstrated that TP53 mutations are associated with a radioresistant phenotype and poor survival in EWS patients [73]. TP53 is rarely mutated in CD; a whole genome sequencing study conducted on 63 CD samples revealed that only one sample carried a p53 mutation [74]. However, an increased expression of p53 was observed in 9/10 patients presenting relapsed tumors compared to patients with a stabilized disease. Thus, in chordoma overexpression of TP53 is correlated with tumor relapse and is a poor prognostic factor [75, 76]. Other studies are needed to understand the role of p53 in CD radioresistance.

If TP53 involvement in radioresistance is quite clear, further molecular studies are needed to precisely determine the underlying mechanisms of p53-driven radioresistance in bone sarcomas in terms of effectors and functions. In addition, although multiple p53 reactivators have been developed, only two drugs have entered clinical trials, APR-246 and COTI-2, currently making p53 hardly targetable.

Cell cycle arrest

Cell cycle regulation is a critical biological function involved in response to radiation. Arresting cell cycle progression is an essential step to enabling the recruitment of DNA repair machinery when DNA damage is caused by radiations. Several major actors of cell cycle regulation are involved in bone sarcoma radioresistance (Fig. 1). The gene Cyclin Dependent Kinase Inhibitor 2A (CDKN2A) encodes the P16 protein that inhibits Cyclin Dependent Kinases 4 and 6 (CDK4/6), inducing cell cycle arrest in G1 phase [77]. CDK4/6 usually bind to cyclin D1 and phosphorylate the tumor suppressor protein Rb1. The phosphorylation of Rb1 prevents its binding to the protein E2F, which in turn activates the transcription of genes allowing entry into the S phase [78].

The CDKN2A locus, is frequently deleted in bone sarcomas [74, 79,80,81]. The absence of p16 allows CDK4/6 activation and entry into the S phase of the cell cycle and could represent an advantage for cancer cells in response to radiation. These alterations could explain their low sensitivity to radiation. Pre-clinical studies refer to the synergistic effect of CDK4/6 inhibitors-radiotherapy combination. For instance, different clinical studies are ongoing in other cancers to determine the efficacy of combining radiation therapy and Palbociclib in breast cancer patients (NCT03691493, NCT03870919) and in locally advanced squamous cell carcinoma (NCT03024489). Further studies need to be done to determine the therapeutic potential of CDK4/6 inhibition in combination with radiotherapy.

Another protein involved in sarcoma radioresistance is CRIF, a protein regulating cell cycle. This protein phosphorylates CDK2, inducing cell cycle arrest and promoting DNA repair [82], a strong expression of CRIF has been detected in OS patient samples. CRIF inhibition by siRNA in both OS cell lines and OS xenografts was shown to increase sensitivity to irradiation, delay DNA damage repair, inactivate G1/S checkpoint, induce mitochondrial dysfunction and tumor regression in vivo [82]. Other strategies aimed at inhibiting cell cycle to reinduce radiosensitivity. In OS, the inhibition of PLK1 [5, 54], WEE1 [83], or PI3K [55] combined to radiotherapy generated cell growth arrest and cell death through mitotic catastrophe. Other studies are urgently needed to decipher the therapeutic potential of cell cycle gene alterations.

Once DNA damage is recognized and the cell cycle is arrested, the next step in cellular response to radiation is DNA repair.

DNA damage repair (DDR)

DNA repair involves a complex machinery and is orchestrated by numerous actors. Here, we will present the major DNA repair actors involved in the response of bone sarcoma to radiation-induced DSBs. For DSB DNA repair, two major pathways are activated: homologous recombination (HR) and non-homologous end joining (NHEJ).

NHEJ occurs during the G1 phase; it binds broken DNA extremities together leading to an increased number of errors. NHEJ initially recognizes DNA damage through a heterodimer Ku70-Ku80. This complex block 5’ DNA extremity and maintains DNA extremities close to each other to allow their binding. This complex also activates the protein 53BP1, which protects DNA extremities from more damage. γH2AX phosphorylation by ATM is also involved in stabilizing DNA extremities. The final steps following assembly of the repair machinery involve binding of DNA extremities by ligases (LIG4, XRCC4, and XLF) [84].

Homologous recombination (HR) only takes place in late S and G2 phases of the cell cycle, as this DNA repair mode is based on the use of the sister chromatid to synthesize an identical DNA strand. This reparation system is more precise than NHEJ. Here, The DNA DSB is recognized by the MRN complex composed of 3 proteins (MRE11, RAD50, NBS1), which initiate resection of DNA extremities in collaboration with CTIP. A loop with the sister chromatid is then formed and a DNA polymerase replicates DNA and ligases bind DNA to the strand break [84, 85]. Certain strategies aim at inhibiting DNA repair to induce cell death such as RAD51 inhibition, a recombinase involved in the DDR machinery. In OS and CD, the inhibition of RAD51, combined with radiations lead to a decreased cell proliferation and an increased apoptosis [86, 87]. In CHS and EWS, the PARP1 inhibitor Olaparib in combination with radiations was reported to decrease cell survival and clonogenic capacities [57, 60].

In this system, PARP-1 is rapidly recruited and activated by DNA DSBs. Upon activation, PARP-1 synthesizes a structurally complex polymer composed of ADP-ribose units that facilitate local chromatin relaxation and the recruitment of DNA repair factors [57]. In both CHS and EWS, PARP-1 seems to play a role in radioresistance. In 2 EWS cell lines, the combination of the PARP-1 inhibitor Olaparib and radiation therapy was more effective than radiotherapy or Olaparib alone. This combination induced a 4-fold increase in apoptosis in comparison with both treatments alone and lead to increased and sustained DNA damage in EWS cell lines. Moreover, in in vivo xenografts models of EWS, the combination of Olaparib and radiation therapy stopped tumor progression [57]. In the CHS cell line CH2879, Olaparib enhanced the efficacy of radiation by 1.3-fold for X rays, 1.8-fold for protons and 1.5-fold for carbon ions [60]. In a study of 11 advanced CD, a mutational signature associated with HR deficiency was found in 72.7% of samples, co-occurring with genomic instability. The use of Olaparib led to prolonged survival in a patient with refractory advanced CD [70]. Olaparib is currently being dose escalated in combination with radical (chemo-)radiotherapy regimens for non-small cell lung cancer (NSCLC), breast cancer and head and neck squamous cell carcinoma (HNSCC) in three parallel single institution phase 1 trials (Study protocols of three parallel phase 1 trials combining radical radiotherapy with the PARP inhibitor Olaparib).

After exposure to radiation, cells normally accumulate DNA damage that cannot be repaired fast enough and with enough precision for the cell to reenter the cell cycle. Proteins involved in genomic stability such as p53 then trigger cell death. However, sarcoma cells often lack the proteins supposed to control genomic integrity and present defects in cell death pathways [88].

Cell death

In response to DNA damage, apoptosis can be induced by different ways: i) activation of p53 or ii) accumulation of ROS. TP53 can directly promote cell death after DNA damage or after incomplete repair of DNA damage [89]. This is mediated through the activation of pro-apoptotic proteins, such as Tumor Necrosis factor Receptor superfamily (TNFR), triggering the extrinsic apoptosis pathway [90].

ROS accumulation can also induce cell death through the loss of mitochondrial membrane potential, leading to the release of cytochrome c. Moreover, ROS cause lipid damage, which activates sphyngomyelinase and induces the production and release of ceramide that in turn can activate caspases 1, 3 and 6, leading to cell death [91, 92].

An incomplete DNA repair can also induce a mitotic catastrophe, during which an abnormal chromosomal condensation occurs and the cell enters in mitosis before the end of S and G2 phases of the cell cycle [93].

Few studies have focused on the involvement of cell death defects in the response of bone sarcoma to radiotherapy. In CHS, the anti-apoptotic proteins Bcl-2, Bcl-xL and XIAP were found to be overexpressed in 2 CHS cell lines in comparison with 2 normal chondrocytes cell lines. When the expression of these anti-apoptotic proteins was inhibited by siRNA, a 10-fold increase in radiosensitivity was observed in CHS cell lines [94]. In EWS cell lines, an exposure to 2 to 10 Gy X-rays was reported to increase the expression of the anti-apoptotic protein survivin in a dose-dependent manner. Survivin inhibition by siRNA doubles apoptotic cell death [95, 96]. Several BH3 mimetics are currently used in the clinic, for example Venetoclax is approved for routine clinical practice in chronic lymphocytic leukemia (CLL) and acute myeloid leukemia (AML). To our knowledge, BH3-mimetics have not yet been combined to radiotherapy in patients.

Bone sarcomas arise in a particular environment (i.e. the bone or cartilage) and one of the characteristics of this environment is its hypoxic content that plays a role in resistance to conventional radiotherapy. Other factors of the tumor microenvironment, like the presence of immune cells or the extracellular matrix are likely involved in bone sarcoma radioresistance but studies regarding these are lacking.

Hypoxia

Hypoxia is a common feature of solid tumors, resulting from the imbalance between oxygen availability and consumption, and is defined as one of the most important causes of radiotherapy failure [97]. In bone sarcoma, the presence of hypoxic zones is correlated with tumor relapse, metastases and resistance to treatments [98,99,100,101,102]. These hypoxic zones are also predictive of poor tumor response to conventional radiotherapy. Different mechanisms have been suggested to explain the link between bone sarcoma, radioresistance and hypoxia. Evidence suggests that hypoxia inhibits the indirect effects of radiotherapy driven by the accumulation of ROS, creating more damage in cells which finally undergo cell death. The first mechanism proposed for hypoxia-induced radioresistance is the acceleration of ROS clearance. In a study including 35 OS and 20 EWS samples, it was shown that radiotherapy does not affect oxidative stress levels. However, it is known that radiotherapy induces ROS production which should increase oxidative stress. Hence, if oxidative stress levels remain constant, this implies that ROS clearance in the tumor cells is accelerated. The activation of autophagy and increased antioxidant metabolism are two hypotheses which can explain how sarcoma cells can accelerate ROS clearance. Indeed, it was demonstrated in OS that hypoxia confers cells resistance to radiation through activated autophagy to accelerate the clearance of cellular ROS products [103]. The increased antioxidant metabolism, mediated by the increase in two antioxidant enzymes, namely Aldehyde dehydrogenase (ALDH) 1 and 3, was shown in CD in an in vitro study. In this study, the pharmacological inhibition of the ALDH1 and 3 restored radiosensitivity to CD spheroid models in vitro [65]. Hypoxia-induced conventional radioresistance can potentially be counteracted by the addition of proton therapy, which has a higher efficacy in hypoxic zones (NCT02802969).

Other potential therapeutic targets with pre-clinical efficacy

Inhibition of histone deacetylases or demethylating agents has proven to be effective in combination with radiation, particularly in OS. Indeed, Histone DeACetylase (HDAC) inhibitors in combination with radiation was reported to increase cell death and DNA damages in several OS cell lines [46,47,48,49,50,51]. In CD and CHS, strategies targeting cancer-initiating cells (CIC) have been tested. One study highlighted that the use of disulfiram, an FDA-approved anti-alcoholism drug, complexed with Cu can radiosensitize CHS CIC. Indeed, the addition of DSF/Cu to a CHS cell line and a CD cell line decreased the clonogenicity of cells, while increasing apoptosis. Moreover, in an orthotopic model of CHS, the combination of DSF/Cu and radiation induced a strong decrease in tumor growth [61]. Similar results were obtained in CD, where the inhibition of ALDH1 and 3, proteins overactivated in CIC, radiosensitized 3D culture of CD cell lines [65]. Efforts need to be made to evaluate the potential of other radiosensitizing strategies. To do this, genetic inhibition of targets in combination with radiotherapy have been tested (Table 8).

Table 8 Combination of radiotherapy and genetic inhibition of targets in bone sarcoma

Future directions could also lead to the combination of immunomodulators and radiotherapy. It is now widely accepted that RT can trigger a systemic immune response supporting a strong rationale for the combination of RT and immunotherapy [110]. Radiations induce a series of biological effects including enhancing tumor antigen release and presentation, promoting priming and activation of immune cells, increasing density of tumor-infiltrating lymphocytes, facilitating recognition of tumor cells by T cells [110]. Combination of immunotherapy and radiotherapy has been evaluated in different solid tumors including melanoma, Non-Small Cell Lung Cancer and other solid tumors. The efficiency of Immune Checkpoint Inhibitors as single agents in bone sarcoma patients has been limited [111, 112]. Given the strong systemic anti-tumor immune effect induced by radiotherapy, an interesting rationale could be the combination of radiotherapy and immune checkpoint inhibitors. To our knowledge, no study has been reported in bone sarcoma concerning radiotherapy-induced anti-tumor immunity, or proof of concept of the combination of radiotherapy and immunotherapy so it would be crucial to investigate further pre-clinically the rationale and to determine efficient and precise biomarkers to predict and evaluate response to this kind of treatment.

Combination of radiotherapy and pharmacological/genetic inhibition of targets in bone sarcoma in clinical trials

Ongoing clinical trials combining drugs with radiotherapy are summarized in Table 9. In CD, 2 clinical trials show promising combinations. These trials evaluated the efficacy of a combination of an anti-brachyury vaccine with radiotherapy. Brachyury is involved in CD tumorigenesis, progression and poor prognosis, and the vaccine targeting brachyury as monotherapy is in phase I. The results of the phase I clinical trial of brachyury vaccine as monotherapy have demonstrated that brachyury vaccine induces a specific immune response. As radiotherapy can induce immunogenic cell death triggering a strong immune response, the combination of brachyury vaccine and radiotherapy could have a synergistic effect. Other studies combining different chemotherapy regimens with radiotherapy are being tested in OS and EWS.

Table 9 Clinical trials combining radiotherapy and FDA approved drugs in bone sarcoma

Other studies are necessary to test the efficacy of specific targeted therapy that could theoretically play a role in the response to radiotherapy. With the development of new radiotherapeutic approaches and their improved efficacy, specific studies deciphering the mechanistic action of these approaches in bone sarcoma would be not only interesting, but welcome to gain further insight into personalized medicine.

Toxicity & limitations

The improved efficacy of new radiotherapy techniques, such as proton beam or carbon ion therapy, offers new therapeutic perspectives in bone sarcoma. However, radiotherapy is still associated with short- and long-term toxicity, as described in Tables 2 and 4. Toxicity depends on the location of the tumor, and children are often particularly vulnerable to radiation-induced late toxicity and to secondary malignancies due to their immature tissue. In a cohort of 222 patients (151 skull-base CD and 71 CHS) treated post-operatively with proton therapy, long-term high grade (> 3) toxicity-free survival was 87%. High-grade late toxicity was characterized by optic neuropathy, temporal lobe necrosis with cerebellum brain parenchyma Grade 3 necrosis, spinal cord necrosis and unilateral hearing loss [113]. In spinal tumors, spinal cord toxicity and insufficiency fractures are the most common radiotherapy-associated side-effects observed [114]. In children pelvic Ewing sarcoma, radiation can cause pelvic pain, premature ovarian deficiency, unequal limb length due to slow bone growth [115]. Aside from radiotherapy toxicity, one major drawback in cancer patient treatment by radiotherapy is the cost and lack of accessibility with only 30 proton therapy centers in Europe.

Conclusion

Bone sarcomas are a group of rare and heterogenous tumors, affecting people of all ages. Surgery is still the mainstay of bone sarcoma patients’ treatment. However, due to the localization of the tumor and the co-morbidity associated with surgery, complete resection is often difficult. Radiotherapy is used in case of incomplete resection or for unresectable tumors.

In the last decades, there has been an improvement in radiotherapy, both in terms of methods of delivery and types of radiation used, leading to more important doses delivered to tumors and less toxicity for surrounding healthy tissue. Currently, retrospective cohorts, case–control studies and systematic reviews are the main studies evaluating the efficacy of radiotherapy in bone sarcoma. Thus high-quality, multicentric randomized controlled trials are desperately needed to precisely determine the benefits of radiotherapy in bone sarcoma. Efforts are ongoing to standardize the treatment in these rare diseases, regroup patients into adapted clinical trials, and improve patient management. A better understanding of the cellular and molecular mechanisms induced by radiotherapy could offer new therapeutic perspectives.

In vitro and in vivo pre-clinical data combining drugs and radiotherapy have shown promising results in bone sarcomas. However, it is important to remember that during the last decade, very few new drugs have been approved for concurrent radiotherapy administration in other cancers where pre-clinical data were also promising. Out of hundreds of clinical trials, only 2 compounds were finally approved for concurrent radiotherapy: the alkylating agent temozolomide and the anti-EGFR antibody cetuximab [116]. This highlights clear gaps between experimental models and the clinical reality that need to be addressed in bone sarcoma research. Efforts need to be made to improve translational research through in vitro and in vivo models to match radiotherapy specificities and challenges, but also through experimental design revision to unveil synergistic combinations. This need is particularly illustrated by the most recent studies showing the strong efficiency of immunotherapy combined to radiotherapy, even in immune desert tumors [117]. The tumor microenvironment plays a primordial role in tumor initiation and progression and a way to improve tumor modeling could be to reproduce the TME, both in vitro and in vivo. This could be of particular interest in CHS and CD, which are considered immune desert tumors, and where radiotherapy could reverse tumor immune desertification. Finally, strategies focusing on the delivery of targeted therapies and radiotherapy may also offer improved approaches in the treatment of bone sarcoma.