Development and validation of a SEER-based auxiliary nomogram for clinical features associated with pulmonary metastasis status in pediatric osteosarcoma

Introduction

Osteosarcoma is the most common primary malignant bone tumor in children and adolescents, although it accounts for less than 0.5% of all cancers (1). It is estimated that approximately 400 children and adolescents in the US receive an osteosarcoma diagnosis annually (2). With chemotherapy for osteosarcoma, the overall survival rate for patients with non-metastatic osteosarcoma has increased from 20% to more than 70% since the 1980s (3). However, about one-fifth of patients with osteosarcoma have metastatic disease at initial diagnosis. The lung is the most common site of metastasis in osteosarcoma and is involved in approximately 80% of patients with metastatic disease (4). Pediatric osteosarcoma has a high mortality rate due to pulmonary metastasis, which occurs in a short period of time (5,6). The 5-year overall survival rate for patients with localized osteosarcoma is approximately 60–70%, whereas survival decreases substantially to approximately 20–30% in patients with metastatic disease, particularly those with pulmonary metastases (7,8). Complete resection of all detectable metastatic lesions, combined with systemic chemotherapy, may improve outcomes in selected patients with pulmonary metastatic osteosarcoma (9-11). Osteosarcoma, like other malignant bone tumors, metastasizes mainly by hematogenous spread resulting in lung metastases, with a major impact on patients’ prognosis. Chest imaging, particularly chest computed tomography (CT), is commonly used to evaluate pulmonary metastasis in osteosarcoma (12). However, imaging-based detection depends on the presence of visible pulmonary lesions and may be affected by nodule size, image quality, and interpretation variability (13). Therefore, identifying clinical features associated with pulmonary metastasis status may help improve risk stratification and guide individualized clinical evaluation. There are still few reports assessing clinical features associated with pulmonary metastasis status in pediatric osteosarcoma.

Nomograms can integrate multiple clinical variables into a visual model to provide individualized estimation and risk stratification; however, their clinical value depends on model performance, validation, and appropriate interpretation (14). In the present study, we analyzed the Surveillance, Epidemiology, and End Results (SEER) data to identify clinical features associated with pulmonary metastasis status recorded at initial diagnosis in pediatric osteosarcoma. We further constructed and validated a SEER-based auxiliary nomogram for clinical feature identification and risk stratification. Importantly, the model should be interpreted as an exploratory clinical association tool rather than a strictly pre-treatment diagnostic prediction model. We present this article in accordance with the TRIPOD reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0342/rc).

Method

Data source and data extraction

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The SEER database is one of the most comprehensive and representative cancer databases of cancer survival and incidence in the US. The SEER Program was established in 1973 to collect and generate population-based cancer statistics (15). The SEER 18 Registries database used in this study covered approximately 28% of the United States population (16). Data were extracted from the “Incidence—SEER Research Data, 18 Registries, Nov 2020 Sub (2000–2018)” database using SEER*Stat software. Primary osteosarcoma cases were identified using the International Classification of Diseases for Oncology, Third Edition (ICD-O-3) primary site codes C40.0–C40.9 and C41.0–C41.9 and histology codes 9180–9186 and 9192–9194. The extracted variables included age, sex, race, tumor grade, T stage, N stage, tumor size, primary site, laterality, chemotherapy, radiotherapy, surgery, pulmonary metastasis status, cancer-specific survival (CSS), survival months, and vital status. The inclusion criteria were: (I) diagnosis between 2004 and 2018; and (II) age ≤18 years at diagnosis. The exclusion criteria were: (I) unknown surgical procedure; (II) non-primary osteosarcoma; (III) unknown tumor size; (IV) unclear primary tumor site; and (V) survival time <1 month. The study design and patient selection flowchart is presented in Figure 1. Patients’ race was divided into three categories: white, black, and other. The patients’ years of diagnosis were between 2004 and 2018. Grade was divided into five categories: I, II, III and IV, unknown. T stage included T1, T2, T3 and T4. N stage was divided into N0 and N1. Primary site included limbs and axial. Chemotherapy and radiation were divided into No/Unknown and Yes. Surgery was classified as no surgery, partial resection, radical excision and amputation.

Figure 1 Study design and patient selection flowchart. AUC, area under the receiver operating characteristic curve; DCA, decision curve analysis; ROC, receiver operating characteristic; SEER, Surveillance, Epidemiology, and End Results.

Construction and validation of the nomogram

All screened pediatric osteosarcoma patients were randomly assigned to the training cohort (70%) or validation cohort (30%). Univariate and multivariate logistic regression analyses were used to identify variables associated with pulmonary metastasis status recorded at initial diagnosis. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. A SEER-based auxiliary risk-stratification nomogram was constructed using variables independently associated with pulmonary metastasis in the training cohort. Because pulmonary metastasis in the SEER database is recorded at initial diagnosis, the present model should not be interpreted as predicting future pulmonary metastatic events during follow-up. In addition, treatment-related variables, including surgery and radiotherapy, are generally determined after diagnosis and multidisciplinary evaluation. Therefore, these variables were interpreted as treatment-pattern variables that may reflect disease burden, resectability, tumor extent, and clinical decision-making patterns rather than as causal factors or pre-diagnostic predictors. Calibration curves were used to assess agreement between the estimated and observed probabilities, and the concordance index (C-index) was used to evaluate model discrimination. The area under the receiver operating characteristic (ROC) curve (AUC) was used to assess discriminatory ability.

Clinical utility

Decision curve analysis (DCA) was used to explore the potential clinical utility of the nomogram. Additionally, a risk stratification system was used to group all patients according to their nomogram scores into high-score and low-score groups. We compared the survival and outcomes of patients in different risk groups by log-rank tests and Kaplan-Meier (K-M) curves.

Statistical analysis

Categorical variables were summarized as frequencies and percentages, and continuous variables were summarized as means and standard deviations. Differences between the training and validation cohorts were evaluated using the chi-square test or non-parametric tests, as appropriate. Univariate and multivariate logistic regression analyses were performed to identify variables associated with pulmonary metastasis status at initial diagnosis. K-M curves and log-rank tests were used to compare CSS between groups. All statistical analyses were performed using SPSS version 26.0 and R version 4.1.0. A two-sided P value <0.05 was considered statistically significant.

Results

Demographic baseline characteristics

There were 1,362 pediatric patients with osteosarcoma included according to the inclusion and exclusion criteria. The mean age of the patients was 12.8 years. Among them, 1,015 (74.5%) were white and 210 (15.4%) were black. 764 (56.1%) were male and 598 (43.9%) were female. Grade included 23 (1.69%) I, 44 (3.23%) II, 320 (23.5%) III, 615 (45.2%) IV, and 360 (26.4%) unknown. T stage included 497 (36.5%) T1, 817 (60.0%) T2, 46 (3.38%) T3, 2 (0.15%) T4. 1,332 patients (97.8%) had stage N0. The mean tumor size was 105 mm. The primary site of osteosarcoma in 1,237 (90.8%) children was in the limbs. 1,292 (94.9%) received chemotherapy and only 43 (3.16%) received radiation. There were 79 (5.80%) patients without surgery, 111 (8.15%) patients with partial resection, 885 (65.0%) patients with radical excision, and 287 (21.1%) patients with amputation. The mean survival time was 64.6 months. By 2018, 71 percent of patients were alive. The clinicopathological data of all patients are shown in Table 1. Most baseline characteristics were comparable between the training and validation cohorts. However, the distribution of radiotherapy differed significantly between the two cohorts (P=0.04).

Univariate and multivariate logistic regression analysis

In the univariate logistic regression analysis, age, laterality, T stage, N stage, tumor size, surgery, and radiotherapy were associated with pulmonary metastasis status. The category-specific ORs and 95% CIs are presented in Table 2. These variables were subsequently evaluated in the multivariate logistic regression analysis. Four variables were independently associated with pulmonary metastasis status, including T stage, N stage, surgery, and radiotherapy. Compared with patients with T1 disease, patients with T2 disease had higher odds of pulmonary metastasis (OR =2.194, 95% CI: 1.457–3.305). Patients with N1 disease also had higher odds than those with N0 disease (OR =5.288, 95% CI: 2.155–12.975). The category-specific ORs for surgery and radiotherapy are presented in Table 2. These variables were incorporated into a SEER-based nomogram for identifying clinical features associated with pulmonary metastasis and for auxiliary risk stratification in pediatric osteosarcoma.

Construction and validation of the SEER-based auxiliary nomogram

A SEER-based auxiliary nomogram was developed to visualize the clinical features associated with pulmonary metastasis status in pediatric osteosarcoma and to support risk stratification, as shown in Figure 2. Among the variables included in the nomogram, T stage contributed the most to the model, followed by N stage, radiotherapy, and surgery. The model performance was internally evaluated in both training and validation cohorts using calibration curves. Calibration curves suggested agreement between the estimated and observed probabilities in both the training and validation cohorts (Figure 3). In the training and validation cohorts, the C-index was 0.699 (95% CI: 0.656–0.741) and 0.736 (95% CI: 0.675–0.797), respectively, indicating acceptable model performance. The AUC of the training cohort and the validation cohort also suggested acceptable discriminatory ability, with 0.699 (95% CI: 0.656–0.741) and 0.695 (95% CI: 0.652–0.738), respectively in Figure 4.

Figure 2 SEER-based auxiliary nomogram for pulmonary metastasis status in pediatric osteosarcoma. N, node; SEER, Surveillance, Epidemiology, and End Results; T, tumor.

Figure 3 Calibration curves of the nomogram in the training cohort (A) and validation cohort (B).

Figure 4 Receiver operating characteristic curves of the nomogram in the training cohort (A) and validation cohort (B). AUC, area under the receiver operating characteristic curve; CI, confidence interval; ROC, receiver operating characteristic.

Clinical application of nomogram

DCA suggested that the SEER-based auxiliary nomogram may provide greater net benefit than TN staging alone within a range of threshold probabilities (Figure 5). According to the total nomogram score, patients were divided into high-score and low-score groups using a cutoff value of 71.3. The K-M curves showed that patients in the high-score group had worse CSS than those in the low-score group, suggesting that nomogram-based stratification was associated with survival differences (Figure 6). As an exploratory analysis of treatment patterns and survival, we further compared survival outcomes among different surgical subgroups in the high-score and low-score groups. These findings should be interpreted as survival differences associated with treatment selection rather than evidence that surgery predicts pulmonary metastasis at diagnosis (Figure 7).

Figure 5 Decision curves of the SEER-based auxiliary nomogram for pulmonary metastasis status in the training cohort (A) and validation cohort (B). The x-axis is the risk threshold, and the y-axis is the net benefit. The green line represents the treat-none strategy, assuming that no patients have pulmonary metastasis, whereas the brown line represents the treat-all strategy, assuming that all patients have pulmonary metastasis. The nomogram provided a higher net benefit than the treat-all and treat-none strategies across a clinically relevant range of threshold probabilities. N, node; SEER, Surveillance, Epidemiology, and End Results; T, tumor.

Figure 6 Kaplan-Meier curves of CSS for patients in the low-score and high-score groups in the training cohort (A) and validation cohort (B). CSS, cancer-specific survival.

Figure 7 Kaplan-Meier curves of CSS for patients receiving different surgical treatments in the low-score group (A) and high-score group (B). CSS, cancer-specific survival.

Discussion

Osteosarcoma is the most common malignant bone tumor in children and adolescents, although it makes up less than 0.5% of all tumors. Pulmonary metastases are detected at diagnosis in approximately 10–20% of patients with osteosarcoma, and approximately 30–40% of patients who initially present with non-metastatic disease may subsequently develop pulmonary metastases during the disease course (17,18). Survival outcomes for patients with non-metastatic osteosarcoma have improved substantially since the introduction of effective multi-agent chemotherapy in the 1970s. Standard treatment for non-metastatic osteosarcoma includes complete surgical resection combined with neoadjuvant and adjuvant multi-agent chemotherapy, commonly including doxorubicin, cisplatin, high-dose methotrexate, and, in selected protocols, ifosfamide (19). When pediatric osteosarcoma patients develop lung metastases, their long-term survival rate is less than 40% (2). In published studies, complete resection of all known metastases has been proved to be a consistent prerequisite for survival (20-22). Even with combination therapy, metastatic disease still has a poor prognosis, with a majority of cases relapsing (23). Chest CT is the most effective method to detect pulmonary nodules. One limitation of CT scanning is the possibility of motion artifacts due to the patient’s movement or breathing, which can reduce resolution. Breath-holding may be difficult for younger children, and sedation may occasionally be required to obtain diagnostic-quality images (24). The development of multidetector CT scanners, which allow the detection of smaller pulmonary nodules, may improve the sensitivity of the diagnosis of lung metastases, but may reduce the specificity (13).

Identifying clinical features associated with pulmonary metastasis status may help improve risk stratification and guide more individualized clinical evaluation in pediatric osteosarcoma. Kaste et al. reported that tumor size was associated with clinical outcomes in pediatric non-metastatic osteosarcoma. However, our study did not identify a significant association between primary tumor size and pulmonary metastasis status (25). Pastorino et al. analyzed 463 consecutive children, adolescents, and young adults with osteosarcoma who underwent lung metastasectomy, with a median age of 15.9 years (range, 0.2–23.2 years). Multivariable analysis identified the disease-free interval from the primary tumor, the number of pulmonary metastases, and completeness of resection as the most relevant predictors of survival (26). In this study, we investigated clinical variables associated with pulmonary metastasis status in pediatric osteosarcoma using the SEER database. Based on univariate and multivariate logistic regression analyses, T stage, N stage, radiotherapy, and surgery were independently associated with pulmonary metastasis status in the SEER cohort. Therefore, these variables were incorporated into a SEER-based auxiliary nomogram for clinical feature identification and risk stratification. Importantly, radiotherapy and surgery should be interpreted as treatment-pattern variables rather than causal factors or pre-diagnostic predictors. Our results suggested that higher T stage was associated with a higher probability of pulmonary metastasis status recorded at diagnosis in pediatric osteosarcoma. This association may be related to greater local tumor extent and a higher likelihood of hematogenous dissemination. N stage was also associated with pulmonary metastasis status in pediatric osteosarcoma. Regional lymph node involvement in osteosarcoma is rare but may occur when the tumor extends into surrounding soft tissues containing lymphatic vessels, such as the joint capsule or synovium (27). Thampi et al. concluded that osteosarcoma lymphatic metastasis was significantly associated with distant metastasis (28). In the present SEER-based analysis, radiotherapy was statistically associated with pulmonary metastasis in pediatric osteosarcoma. However, this association should not be interpreted as evidence that radiotherapy causes pulmonary metastasis or as indicating that radiotherapy can predict pulmonary metastasis before treatment decisions are made. Radiotherapy is not routinely used for osteosarcoma but may be considered in selected clinical contexts, such as unresectable disease, anatomically complex tumors in which negative margins are difficult to achieve, positive surgical margins, or residual disease (29,30). Therefore, radiotherapy may act as a treatment-pattern variable reflecting advanced disease burden, poor resectability, treatment selection, or indication bias rather than a causal factor for pulmonary metastasis (31,32). Surgery was also associated with pulmonary metastasis status in the SEER cohort. Nevertheless, surgery is a post-diagnostic treatment variable and should not be interpreted as a pre-diagnostic predictor or causal factor for pulmonary metastasis at presentation. Standard treatment strategies for osteosarcoma are generally determined after diagnosis and staging assessment, and usually include surgery combined with systemic chemotherapy in selected patients (1,19). This association may reflect tumor extent, metastatic status, resectability, and selection of local treatment, because surgical decision-making and the feasibility of complete resection are closely related to disease burden and clinical condition (9,10,33). Patients who underwent definitive surgery may have had more favorable baseline disease characteristics, whereas patients who did not undergo surgery may have had more extensive or unresectable disease. This interpretation should be considered cautiously because surgical eligibility is influenced by tumor extent, anatomical involvement, resectability, and multidisciplinary treatment decisions (34). Therefore, the association between surgery and pulmonary metastasis should be interpreted cautiously as a treatment-selection pattern rather than evidence that surgery directly affects the presence of pulmonary metastasis at initial diagnosis. This interpretation is also consistent with the general concern that treatment variables in observational database studies may be affected by confounding by indication, also referred to as treatment selection bias (35).

We analyzed surgical procedures in the high-score and low-score groups. In the low-score group, partial resection had the highest survival rate, followed by radical resection and amputation. Patients who did not have surgery had the lowest survival rates. Surgery had a significantly higher survival rate in the high-score group than no surgery. Most patients underwent radical resection, and patients who underwent radical resection had significantly higher survival rates than the other groups. This study explored clinical features associated with pulmonary metastasis status in pediatric osteosarcoma using the SEER database. These variables were incorporated into a SEER-based auxiliary nomogram for clinical feature identification and risk stratification. After internal validation, the C-index, ROC curves, and calibration curves indicated acceptable model performance. Nevertheless, given the inclusion of treatment-pattern variables, the model should be interpreted as an exploratory clinical association tool rather than a strictly pre-treatment diagnostic prediction model.

This study has several limitations. First, this was a retrospective study based on the SEER database, and selection bias was unavoidable. Second, surgery and radiotherapy are post-diagnostic treatment variables. Although they were statistically associated with pulmonary metastasis status in this SEER-based analysis, they should be interpreted as treatment-pattern variables reflecting disease burden, resectability, and treatment selection rather than as causal or pre-diagnostic predictors. The inclusion of these variables may introduce temporal ambiguity, treatment selection bias, and confounding by indication. In addition, the number of patients who received radiotherapy was small, particularly in the validation cohort, which may have resulted in unstable effect estimates and limited the reliability of radiotherapy as a model variable. Third, SEER lacks detailed information on radiotherapy dose, chemotherapy regimen, surgical margins, number and size of pulmonary nodules, histological response to chemotherapy, imaging findings, and treatment timing. In particular, chemotherapy is recorded as “Yes” or “No/Unknown” in SEER; therefore, the “No/Unknown” category cannot distinguish patients who truly did not receive chemotherapy from those with no evidence of chemotherapy in the available medical records. This coding limitation may partly explain why less than 95% of patients were recorded as receiving chemotherapy in our cohort. Moreover, a small proportion of patients may have had low-grade or special histological subtypes, such as parosteal osteosarcoma, for which chemotherapy may not be routinely administered. Fourth, pulmonary metastases occurring during disease progression were not captured because SEER records pulmonary metastasis status at diagnosis; therefore, the incidence of pulmonary metastasis may have been underestimated. Fifth, this model was internally validated using training and validation cohorts from the same database, and external validation in independent multicenter cohorts is still required. Future prospective studies incorporating imaging, pathological, molecular, and treatment-timing data are needed to validate and refine this model.

Conclusions

This SEER-based study identified clinicopathological and treatment-pattern variables associated with pulmonary metastasis in pediatric osteosarcoma and developed an auxiliary nomogram for risk stratification. The model should be interpreted as an exploratory clinical association tool rather than a strictly pre-treatment diagnostic prediction model. Further prospective studies incorporating imaging, pathological, molecular, and treatment-timing data are needed to validate and refine this model.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0342/rc

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0342/prf

Funding: This study was supported by grants from the Chongqing Municipal Science and Technology Bureau (No. CSTB2023NSCQ-MSX0424) and the Chongqing Municipal Health Commission Science-Health Collaborative Key Project (No. 2024ZDXM029).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0342/coif). All authors report that this study was supported by grants from the Chongqing Municipal Science and Technology Bureau (No. CSTB2023NSCQ-MSX0424) and the Chongqing Municipal Health Commission Science-Health Collaborative Key Project (No. 2024ZDXM029). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Biermann JS, Adkins DR, Agulnik M, et al. Bone cancer. J Natl Compr Canc Netw 2013;11:688-723. [Crossref] [PubMed]
2. Mirabello L, Troisi RJ, Savage SA. Osteosarcoma incidence and survival rates from 1973 to 2004: data from the Surveillance, Epidemiology, and End Results Program. Cancer 2009;115:1531-43. [Crossref] [PubMed]
3. Gill J, Ahluwalia MK, Geller D, et al. New targets and approaches in osteosarcoma. Pharmacol Ther 2013;137:89-99. [Crossref] [PubMed]
4. Silva JAM, Marchiori E, Amorim VB, et al. CT features of osteosarcoma lung metastasis: a retrospective study of 127 patients. J Bras Pneumol 2023;49:e20220433. [Crossref] [PubMed]
5. Czarnecka AM, Synoradzki K, Firlej W, et al. Molecular Biology of Osteosarcoma. Cancers (Basel) 2020;12:2130. [Crossref] [PubMed]
6. Ottaviani G, Jaffe N. The epidemiology of osteosarcoma. Cancer Treat Res 2009;152:3-13. [Crossref] [PubMed]
7. Harris MA, Hawkins CJ. Recent and Ongoing Research into Metastatic Osteosarcoma Treatments. Int J Mol Sci 2022;23:3817. [Crossref] [PubMed]
8. Odri GA, Tchicaya-Bouanga J, Yoon DJY, et al. Metastatic Progression of Osteosarcomas: A Review of Current Knowledge of Environmental versus Oncogenic Drivers. Cancers (Basel) 2022;14:360. [Crossref] [PubMed]
9. Mettmann VL, Blattmann C, Friedel G, et al. Primary Multi-Systemic Metastases in Osteosarcoma: Presentation, Treatment, and Survival of 83 Patients of the Cooperative Osteosarcoma Study Group. Cancers (Basel) 2024;16:275. [Crossref] [PubMed]
10. Strauss SJ, Frezza AM, Abecassis N, et al. Bone sarcomas: ESMO-EURACAN-GENTURIS-ERN PaedCan Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 2021;32:1520-36. [Crossref] [PubMed]
11. Stork T, Boemans R, Hardes J, et al. Number of metastases and their response to chemotherapy impact survival of patients with isolated lung metastases from bone-derived sarcoma. BMC Cancer 2021;21:375. [Crossref] [PubMed]
12. Chiesa AM, Spinnato P, Miceli M, et al. Radiologic Assessment of Osteosarcoma Lung Metastases: State of the Art and Recent Advances. Cells 2021;10:553. [Crossref] [PubMed]
13. Cho YJ, Kim WS, Choi YH, et al. Computerized texture analysis of pulmonary nodules in pediatric patients with osteosarcoma: Differentiation of pulmonary metastases from non-metastatic nodules. PLoS One 2019;14:e0211969. [Crossref] [PubMed]
14. Efthimiou O, Seo M, Chalkou K, et al. Developing clinical prediction models: a step-by-step guide. BMJ 2024;386:e078276. [Crossref] [PubMed]
15. Friedman S, Negoita S. History of the Surveillance, Epidemiology, and End Results (SEER) Program. J Natl Cancer Inst Monogr 2024;2024:105-9. [Crossref] [PubMed]
16. Lewis DR, Chen HS, Midthune DN, et al. Early estimates of SEER cancer incidence for 2012: Approaches, opportunities, and cautions for obtaining preliminary estimates of cancer incidence. Cancer 2015;121:2053-62. [Crossref] [PubMed]
17. Rasalkar DD, Chu WC, Lee V, et al. Pulmonary metastases in children with osteosarcoma: characteristics and impact on patient survival. Pediatr Radiol 2011;41:227-36. [Crossref] [PubMed]
18. Marko TA, Diessner BJ, Spector LG. Prevalence of Metastasis at Diagnosis of Osteosarcoma: An International Comparison. Pediatr Blood Cancer 2016;63:1006-11. [Crossref] [PubMed]
19. Anninga JK, Gelderblom H, Fiocco M, et al. Chemotherapeutic adjuvant treatment for osteosarcoma: where do we stand? Eur J Cancer 2011;47:2431-45. [Crossref] [PubMed]
20. Kager L, Zoubek A, Pötschger U, et al. Primary metastatic osteosarcoma: presentation and outcome of patients treated on neoadjuvant Cooperative Osteosarcoma Study Group protocols. J Clin Oncol 2003;21:2011-8. [Crossref] [PubMed]
21. Briccoli A, Rocca M, Salone M, et al. High grade osteosarcoma of the extremities metastatic to the lung: long-term results in 323 patients treated combining surgery and chemotherapy, 1985-2005. Surg Oncol 2010;19:193-9. [Crossref] [PubMed]
22. Salah S, Fayoumi S, Alibraheem A, et al. The influence of pulmonary metastasectomy on survival in osteosarcoma and soft-tissue sarcomas: a retrospective analysis of survival outcomes, hospitalizations and requirements of home oxygen therapy. Interact Cardiovasc Thorac Surg 2013;17:296-302. [Crossref] [PubMed]
23. Huang X, Zhao J, Bai J, et al. Risk and clinicopathological features of osteosarcoma metastasis to the lung: A population-based study. J Bone Oncol 2019;16:100230. [Crossref] [PubMed]
24. Steliga M, Vaporciyan A. Surgical treatment of pulmonary metastases from osteosarcoma in pediatric and adolescent patients. Cancer Treat Res 2009;152:185-201. [Crossref] [PubMed]
25. Kaste SC, Liu T, Billups CA, et al. Tumor size as a predictor of outcome in pediatric non-metastatic osteosarcoma of the extremity. Pediatr Blood Cancer 2004;43:723-8. [Crossref] [PubMed]
26. Pastorino U, Palmerini E, Porcu L, et al. Lung metastasectomy for osteosarcoma in children, adolescents, and young adults: proof of permanent cure. Tumori 2023;109:79-85. [Crossref] [PubMed]
27. Li W, Dong S, Wang B, et al. The Construction and Development of a Clinical Prediction Model to Assess Lymph Node Metastases in Osteosarcoma. Front Public Health 2021;9:813625. [Crossref] [PubMed]
28. Thampi S, Matthay KK, Goldsby R, et al. Adverse impact of regional lymph node involvement in osteosarcoma. Eur J Cancer 2013;49:3471-6. [Crossref] [PubMed]
29. Spałek MJ, Poleszczuk J, Czarnecka AM, et al. Radiotherapy in the Management of Pediatric and Adult Osteosarcomas: A Multi-Institutional Cohort Analysis. Cells 2021;10:366. [Crossref] [PubMed]
30. DeLaney TF, Park L, Goldberg SI, et al. Radiotherapy for local control of osteosarcoma. Int J Radiat Oncol Biol Phys 2005;61:492-8. [Crossref] [PubMed]
31. Huang X, Guo JW, Han F, et al. Establishment of a nomogram for potential prediction of lung metastasis in patients with primary limb bone tumors: a study based on the SEER database. Transl Cancer Res 2024;13:4763-74. [Crossref] [PubMed]
32. Jiang Y, Zhu Y, Ding Y, et al. Nomograms to predict lung metastasis in malignant primary osseous spinal neoplasms and cancer-specific survival in lung metastasis subgroup. Front Oncol 2024;14:1393990. [Crossref] [PubMed]
33. Meazza C, Scanagatta P. Metastatic osteosarcoma: a challenging multidisciplinary treatment. Expert Rev Anticancer Ther 2016;16:543-56. [Crossref] [PubMed]
34. Gerrand C, Amary F, Anwar HA, et al. UK guidelines for the management of bone sarcomas. Br J Cancer 2025;132:32-48. [Crossref] [PubMed]
35. Zawadzki RS, Grill JD, Gillen DL, et al. Frameworks for estimating causal effects in observational settings: comparing confounder adjustment and instrumental variables. BMC Med Res Methodol 2023;23:122. [Crossref] [PubMed]