Cancer-specific survival patterns in patients with bone metastasis: a registry-based analysis of 13,742 patients

With increasing opportunities for patients with bone metastasis (BM) to benefit from local surgical intervention, accurate survival analysis across different primary cancers remains challenging. Current analytical frameworks commonly rely on single-center, pan-cancer cohorts and provide insufficient integration of cancer-specific characteristics. In this retrospective, multicenter, registry-based cohort study, baseline demographic and clinical characteristics of 13,742 patients with AJCC stage IV or TNM stage M1 metastatic cancer were collected from 42 studies registered in the cBioPortal for Cancer Genomics database. Overall survival (OS) after metastatic diagnosis was the primary outcome. Univariate analyses were performed using Kaplan-Meier methods, log-rank tests, and non-parametric tests. Variables with p < 0.20 were included in multivariable Cox proportional hazards models to examine independent associations with survival. Multiple imputation was applied to address missing data. Among the 25 primary cancers analyzed, approximately half showed observable survival differences between BM and other-site metastasis, with 6 cancers reaching statistical significance. Based on median survival, all cancers could be stratified into 3 distinct survival tiers, ranging from prolonged survival exceeding 15 months to markedly shorter survival of 3-10 months, with multivariable analyses further demonstrating that primary cancer type was the strongest factor associated with survival heterogeneity among BM patients (HR = 1.422-1.758, p < 0.001). Moreover, poorly differentiated or undifferentiated histology was independently associated with worse OS (HR = 1.249, p < 0.001), and age > 60 years was also associated with shorter survival (p < 0.001). No single metastatic site demonstrated a consistent adverse association with survival across cancer types. Overall, BM demonstrates cancer-specific and heterogeneous associations with survival compared with other metastatic sites. All primary cancers could be stratified into 3 groups, representing the most important factor associated with survival differences. Moreover, pathological differentiation was significantly associated with survival among BM patients. Notably, no metastatic site functions as a universal prognostic factor across cancers. Large-scale, multicenter, registry-based analyses provide a valuable framework for cancer-specific survival analysis and for identifying clinically relevant factors that may serve as a reference for risk stratification in surgical decision-making.