Abstract
Objective
Osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle are rare tumor entities, and their clinical features and treatment outcomes have been rarely reported. The present study was performed to evaluate their survival and confirm independent survival predictors.
Methods
Data on patients with osteosarcoma or Ewing sarcoma of the rib, sternum, and clavicle from 1973 to 2016 were retrospectively extracted from the database. Univariate and multivariate Cox regression analyses were used to determine the independent risk factors. Kaplan–Meier survival curves were applied to examine the prognostic difference between the groups.
Results
In total, 475 patients with osteosarcoma or Ewing sarcoma of the rib, sternum, and clavicle were eligible for this study, including 173 (36.4%) with osteosarcoma and 302 (63.6%) with Ewing sarcoma. The 5-year overall survival and cancer-specific survival rates of all patients were 53.6% and 60.8%, respectively. Six independent variables were identified, including age at diagnosis, sex, histological grade, metastatic status, tumor type, and surgery.
Conclusions
Surgical resection is a reliable treatment for osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle. Further research is needed to reconfirm the role of chemotherapy and radiotherapy in survival of these patients.
Introduction
Osteosarcoma and Ewing sarcoma are common primary malignant tumors of bone.1,2 Osteosarcoma is the most common tumor in children and adolescents, and its peak incidence occurs in adolescence. 3 The incidence of Ewing sarcoma is second only to osteosarcoma, accounting for 10% to 15% of malignant bone tumors. 4 Both tumors occur predominantly in children and young adolescents and primarily affect the long bones, especially the distal femur and proximal tibia. The lung is the most common site for metastases. 5 Moreover, the treatment strategies of these two types of sarcomas are basically the same, including preoperative and postoperative intensive systemic chemotherapy and surgical resection.1,6 The clinical features, survival time, and risk factors for survival among patients with osteosarcoma and Ewing sarcoma at common sites have been widely reported. 7 Patients with non-metastatic osteosarcoma and Ewing sarcoma have a good prognosis with a 5-year survival rate of 70% and 60%, respectively.8,9 However, patients with metastasis have a poor prognosis with a 5-year survival rate of <26%.10–13 Many studies have documented the risk factors for survival among patients with osteosarcoma and Ewing sarcoma.14–17 To the best of our knowledge, however, few studies have focused on osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle.
The optimal treatment for osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle remains unknown. Additionally, although radiotherapy has shown a local control effect on both osteosarcoma and Ewing sarcoma, especially for refractory tumors,18,19 its survival benefits in these patients have been seldom reported. This study was performed to investigate the associations between clinical characteristics, treatments, and survival among patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle, based on the Surveillance, Epidemiology, and End Results (SEER) database.
Methods
Patient population
From 1973 to 2016, all patients diagnosed with osteosarcoma or Ewing sarcoma of the rib, sternum, and clavicle in the SEER database were included for analysis. Clinical data from the SEER database were obtained using the case listing session in SEER*Stat version 8.3.8 software. The SEER database (November 2021 submission) collects tumor data from 18 locations, covering 30% of the United States population (www.seer.cancer.gov). This was a retrospective study. The requirements for ethics approval and informed consent were waived by the local ethics committee because the SEER data are publicly available and de-identified. Additionally, we have de-identified all patient details. The reporting of this study conforms to the STROBE guidelines. 20
According to the International Classification of Diseases for Oncology, Third Edition (ICD-O-3), patients with osteosarcoma (ICD-O-3 histologic type: 9180-9187, 9192-9195) and Ewing sarcoma (ICD-O-3 histologic type: 9260) of the rib, sternum, and clavicle (ICD-O-3 site code: C41.3) were identified. Only patients with a histopathological diagnosis were included for analysis. Fourteen patients were excluded because the diagnosis was based only on the clinical presentation or imaging findings. The variables extracted from this cancer database included race, sex, age, tumor grade, tumor stage, histologic type, tumor size, surgical treatment, radiotherapy of primary site, chemotherapy, cause of death, survival in months, and vital status. The main endpoints were overall survival (OS) and cancer-specific survival (CSS), which were defined as the time from diagnosis to death of any cause and death of cancer, respectively.14,21
Statistical analyses
We used SPSS 22.0 software (IBM Corp., Armonk, NY, USA) to perform all statistical analyses and create all graphics. Survival curves generated by the Kaplan–Meier method were applied to intuitively show the prognostic difference between the groups. We first performed univariate Cox regression analysis to screen the risk factors with a P value of <0.05. These potential risk factors were then included in the multivariate Cox regression analysis. We also calculated the hazard ratios and their 95% confidence intervals to determine the impact of each variable on survival. Statistical results were considered significant at a two-sided P value of <0.05.
Results
Clinical features
In total, 475 patients were eligible in our analysis of the SEER data, including 173 (36.4%) patients with osteosarcoma and 302 (63.6%) with Ewing sarcoma. Table 1 summarizes the detailed clinical features of all patients. Most patients were white (90.9%). There were slightly more male patients than female patients, with a male:female ratio of 1.2:1.0. Overall, 227 (47.8%) patients were aged <18 years, 132 (27.8%) were aged 18 to 40 years, and 116 (24.4%) were aged >40 years. Histologically, one-third (33.5%) of the patients had high-grade cancer. About two-thirds (67.6%) of the patients had data regarding tumor size, and they were divided into three groups: ≤5 cm, >5 to 10 cm, and >10 cm. Most of the patients (80.8%) received chemotherapy, more than three-fourths (75.6%) received surgery, and 38.5% received radiotherapy. At the time of data collection, 238 (50.1%) patients were dead. The Kaplan–Meier survival results revealed that the 5-year OS and CSS rates of this population were 53.6% and 60.8%, respectively.
Baseline characteristics of 475 patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle.
Data are presented as n (%) unless otherwise indicated.
OS: overall survival, CSS: cancer-specific survival.
Univariate Cox regression analysis
The results of the univariate survival analysis are shown in Table 2. The univariate analysis showed that race was not a risk factor for survival. Male patients had worse OS and CSS than female patients (Figure 1). Increasing patient age was associated with a poorer prognosis (Figure 2). Patients with Ewing sarcoma had a better prognosis than those with osteosarcoma (Figure 3). Moreover, the tumor grade and metastatic status significantly affected the patients’ prognosis. Tumor size was significantly correlated with CSS but not OS. The beneficial effect of surgery (versus no surgery) on the prognosis was obvious (Figure 4). Radiotherapy was not significantly associated with OS. However, patients who received radiotherapy had worse CSS than those who did not receive radiotherapy. Additionally, chemotherapy did not provide a survival benefit for this population.
Univariate Cox analysis of variables associated with survival of patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle.
OS: overall survival, CSS: cancer-specific survival, HR: hazard ratio, CI: confidence interval. P values were considered significant at <0.05. Values in bold mean are statistically significant.

Kaplan–Meier estimation of (a) overall survival and (b) cancer-specific survival in patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle stratified by sex.

Kaplan–Meier estimation of (a) overall survival and (b) cancer-specific survival in patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle stratified by age.

Kaplan–Meier estimation of (a) overall survival and (b) cancer-specific survival in patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle stratified by tumor type.

Kaplan–Meier estimation of (a) overall survival and (b) cancer-specific survival in patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle stratified by surgery.
Multivariate Cox regression analysis
The results of the multivariate survival analysis are shown in Table 3. Age of >40 years, male sex, osteosarcoma, high grade, distant metastasis, and no surgery were significantly correlated with decreased OS and CSS (P < 0.05). Radiotherapy and tumor size were not identified as independent risk factors for CSS.
Multivariate Cox analysis of variables associated with survival of patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle.
OS: overall survival, CSS: cancer-specific survival, HR: hazard ratio, CI: confidence interval. P values were considered significant at <0.05. Values in bold mean are statistically significant.
Discussion
More than half of patients with osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle were older than 18 years. The prognosis of this population was close to the results of other studies17,22,23 and better than that of patients with spinal bone sarcoma. 24 Multivariate Cox regression showed that age, sex, tumor grade, metastatic status, tumor type, and surgery were independent risk factors for both OS and CSS. This study is the first to present the clinical features of osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle and to identify independent survival predictors based on a public cancer database.
In the present study, age of >40 years was an independent predictor of worse survival, which is consistent with the findings of other investigations.23,25,26 Older patients may have a high risk of metastasis or recurrence, and they use chemotherapy less often because of greater toxic side effects. 22 Female sex was an independent determinant of improved survival, and a similar trend has been revealed in other studies.27–29 This can be partially explained by the fact that female patients less frequently presented with aggressive tumors and showed a better response to treatment. 27
The univariate analysis showed that race did not impact patients’ survival, which has been confirmed by many other studies.14,24,26,30 However, Zhou et al. 25 identified race as a significant survival predictor in patients with Ewing sarcoma of bone through a multivariate analysis. Sun et al. 31 reported that white race can independently predict better survival of chondroblastic osteosarcoma. Tumor size was an important survival predictor among patients with osteosarcoma and Ewing sarcoma.17,26,30 Notably, no significant survival difference was observed based on tumor size in this study. This may be a clinical feature of osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle. Our study results are consistent with those of other investigations suggesting that histopathological grade and metastatic status are meaningful survival predictors in patients with osteosarcoma and Ewing sarcoma.28,32–34
Although the standard treatment pattern for osteosarcoma and Ewing sarcoma has been established, treatment strategies for special populations of patients with osteosarcoma or Ewing sarcoma remain unknown, as for the research population in this study. Surgical resection is still a reliable treatment for osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle. Moreover, three-dimensional printed prostheses are increasingly used to repair and reconstruct bone defects in patients with bone sarcoma of the rib, sternum, and clavicle.35–37 Chemotherapy has led to increased survival rates of osteosarcoma and Ewing sarcoma. 38 However, our study revealed no survival benefit of chemotherapy on survival in these patients. Similarly, a recent systematic review and pooled analysis showed no consistent survival benefit for osteosarcoma of the head and neck. 39 Therefore, the sensitivity of osteosarcoma or Ewing sarcoma in uncommon sites to chemotherapy needs to be further studied. Preoperative radiotherapy can induce a higher rate of necrosis and may allow the resection of unresectable lesions in the treatment of osteosarcoma. 40 Similarly, radiotherapy has achieved good local control in Ewing sarcoma. 41 However, our study showed that radiotherapy was not associated with an improved prognosis. Additionally, radiotherapy may worsen the survival of patients with spinal osteosarcoma. 24 This was not a randomized trial, but a retrospective study. Radiotherapy itself may not be the factor that worsens the prognosis. Therefore, clinicians should consider that patients qualified to receive radiotherapy may have more advanced cancer and consequently worse survival.
This study has both limitations and strengths. It was a retrospective study with a survival analysis, which might have generated selection bias (e.g., the ability of a specific patient to receive surgery may have been influenced by the clinical status of the patient or the experience of the surgical team). Second, some variables were not available in the SEER database, such as surgical procedures, surgical margins obtained, and molecular pathological characteristics. Third, because both osteosarcoma and Ewing sarcoma have similar ages at onset, treatment methods, and prognostic patterns, we combined them into a single survival analysis. Nevertheless, the SEER database is a useful tool for studying rare tumor entities such as those in the cancer population of the present study. Additionally, this cancer database collects cancer information from multiple centers and is regularly updated to promote cancer research.
Conclusion
Patients with osteosarcoma of the rib, sternum, and clavicle showed poorer survival than patients with Ewing sarcoma. This study also reaffirms other known prognostic factors for these special tumors. Based on the results of this study, we support surgical treatment of osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle. However, additional chemotherapy and radiotherapy or standard treatment for these patients should be explored further.
Research Data
sj-pdf-1-imr-10.1177_03000605231175763 - Research Data for Prognostic analysis and clinical characteristics of osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle: a SEER-based study
Research Data, sj-pdf-1-imr-10.1177_03000605231175763 for Prognostic analysis and clinical characteristics of osteosarcoma and Ewing sarcoma of the rib, sternum, and clavicle: a SEER-based study by Fangming He, Zhan Wang, Genlian Chen and Huanxin Lu in Journal of International Medical Research
Footnotes
Acknowledgements
We thank the National Foundation and the Provincial Foundation for their support of our work.
Authors’ contributions
Z. Wang and H. Lu contributed to the article framework guidance and revision. F. He and G. Chen contributed to the data acquisition, statistical analysis, and article writing. All authors have read and approved the manuscript.
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors without undue reservation.
Declaration of conflicting interests
The authors declare that they have no competing interests.
Funding
This work was supported by the Zhejiang Provincial Natural Science Foundation (LQ22H160040), National Natural Science Foundation of China (82103499), and China Postdoctoral Science Foundation (2021M692792).
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
