Epidemiology and nomogram of pediatric and young adulthood osteosarcoma patients with synchronous lung metastasis: A SEER analysis

Abstract

Background Patients with osteosarcoma and synchronous lung metastasis (SLM) have poor survival. This study aimed to explore the epidemiology data and construct a predictive nomogram to identify cases at risk of SLM occurrence among pediatric and young adulthood osteosarcoma patients. Methods All data were extracted from Surveillance, Epidemiology, and End Results 17 registries. The age-standardized incidence rate (ASIR) and annual percentage change was evaluated, and reported for the overall population and by age, gender, race, and primary site. Univariate and multivariate logistic regression analyses were used to identify risk factors associated with SLM occurrence, then significant factors were used to develop the nomogram. The area under the receiver operating characteristic curve (AUC) and calibration curve were used to evaluated the predictive power of the nomogram. Survival analysis was assessed by the Kaplan-Meier method and the log-rank test. Multivariate Cox analysis was used to determine the prognostic factors. Results A total of 278 out of 1965 patients (14.1%) presented with SLM at diagnosis. The ASIR increased significant from 0.46 to 0.66 per 1,000,000 person-years from year 2010 to 2019, with an annual percentage change of 3.5, mainly in patients with age 10–19 years, male and appendicular location. All patients were randomly assigned into train cohort and validation cohort with a spilt of 7:3. In the train cohort, higher tumor grade, bigger tumor size, positive lymph nodes and other site-specific metastases (SSM) were identified as significant risk factors associated with SLM occurrence. Then a nomogram was developed based on the four factors. The AUC and calibration curve in both train and validation cohorts demonstrated that the nomogram had moderate predictive power. The median cancer-specific survival was 25 months. Patients with age 20–39 years, male, positive lymph nodes, other SSM were adverse prognostic factors, while surgery was protective factor. Conclusions This study performed a comprehensive analysis regarding pediatric and young adulthood osteosarcoma patients had SLM. A visual, clinically operable, and easy-to-interpret nomogram model was developed for predicting the risk of SLM, which could be used in clinic and help clinicians make better decisions.