Strategies of the management of pulmonary metastases in children with Ewing sarcoma- single institution experience of 137 patients.

BACKGROUND

The role of pulmonary metastasectomy in pediatric Ewing sarcoma remains controversial, with significant treatment selection bias complicating the interpretation of retrospective series.

OBJECTIVE

To evaluate treatment outcomes in isolated pulmonary metastases (PM) using propensity score analysis to control for selection bias.

METHODS

Retrospective analysis of 137 patients with pulmonary metastases (85 isolated PM, 52 PM with extrapulmonary metastases) treated between 1998 and 2024. Primary comparison focused on thoracotomy alone versus radiotherapy alone in isolated PM patients with primary tumor surgery (n = 30). Propensity score matching was performed to address selection bias.

RESULTS

Patients with isolated PM had superior overall survival compared to those with extrapulmonary disease (median OS: 111 vs 66 months, p < 0.001). In isolated PM patients with primary tumor surgery, direct comparison between thoracotomy alone and radiotherapy alone revealed superior survival with surgical resection (median OS: 125 vs 38 months, p = 0.009). Stratified analysis by chemotherapy response demonstrated that thoracotomy benefit was primarily evident in patients with poor histologic response (≤10 % viable cells, median OS: 56 vs 27 months, p = 0.017), while patients with good response showed similar outcomes regardless of local treatment approach (p = 0.104). After propensity score matching and multivariable adjustment for age and gender, the apparent benefit of thoracotomy was maintained in direct treatment comparisons, though overall comparison against all non-surgical treatments showed no significant difference (HR 0.75, p = 0.343), highlighting the importance of appropriate treatment comparisons and patient selection.

CONCLUSIONS

Thoracotomy may benefit carefully selected isolated PM patients, particularly those with poor response to chemotherapy. However, significant selection bias complicates the interpretation of retrospective surgical series. The benefit appears limited to specific clinical scenarios rather than representing a general therapeutic advantage. Future studies should focus on developing standardized selection criteria and prospective validation of treatment algorithms.