Baseline CT imaging parameters predicting overall and progression-free survival for patients with pulmonary metastases from soft tissue and bone sarcoma

Eur Radiol. 2026 Jan 3. doi: 10.1007/s00330-025-12289-3. Online ahead of print.

Abstract

Objectives: The lung is the most frequent metastatic site in soft tissue and bone sarcoma, yet CT imaging predictors of survival remain underexplored. Hence, the aim of this study was to evaluate baseline CT imaging predictors for overall survival (OS) and progression-free survival (PFS) in patients with pulmonary metastases from soft tissue and bone sarcoma.

Materials and methods: Consecutive patients treated from 2016 to 2021 were screened; those without lung metastases, systemic therapy, baseline CT imaging, or with prior lung surgery were excluded. Tumor burden (long axis, volume) and number of pulmonary lesions were assessed at baseline CT imaging and tested for OS and PFS. Significant univariate predictors were entered into multivariate models and reported as hazard ratios (HR).

Results: In the 100 included patients, median OS was 22.1 months, and median PFS was 5.8 months. Tumor burden parameters were not predictive of OS nor PFS (HR 0.724-1.354, p = 0.265-0.975). The number of pulmonary lesions was prognostic, with ≥ 5 lesions predicting shorter OS in univariate (HR 1.709, 95% CI: 1.013-2.881, p = 0.033) and multivariate analysis (HR 3.478, 95% CI: 1.509-8.302, p = 0.004). For PFS, ≥ 5 lesions were significant in univariate analysis (HR 1.741, 95% CI: 1.113-2.723, p = 0.011) but not in multivariate analysis (HR 1.530, 95% CI: 0.395-7.391, p = 0.555).

Conclusion: At baseline, CT imaging of patients with soft tissue and bone sarcoma, the number of pulmonary lesions independently predicted OS, whereas tumor burden showed no prognostic value. No imaging parameter independently predicted PFS.

Key points: Question Which baseline CT imaging parameters best predict overall and PFS in patients with pulmonary metastases from soft tissue and bone sarcoma? Findings At baseline, CT imaging, pulmonary lesion count independently predicted OS, while pulmonary tumor burden was not prognostic. No imaging parameter independently predicted PFS. Clinical relevance Baseline CT imaging assessment of the number of pulmonary lesions provides prognostic information and may inform treatment decisions in patients with soft tissue and bone sarcoma.

Keywords: Lung; Sarcoma; Survival analyses; Tomography (X-ray computed); Tumor burden.