Comment on “Primary osseous chest wall sarcomas: Oncological and surgical outcomes in a large single-institution cohort of 100 patients”

Dear editor,

We read with interest the retrospective cohort study by Oteman et al. [1], which provides valuable long-term data on a rare and heterogeneous group of primary osseous chest wall sarcomas. Their conclusion that R0 resection is the strongest modifiable predictor of survival is clinically plausible, but the analysis warrants qualification before this statement is translated into a general prognostic or operative rule.

The non-R0 group combined 13 microscopically positive resections (R1) with four planned macroscopic residual resections (R2). The R2 operations were performed with palliative intent, and progression at the operative site was defined as persistent disease rather than postoperative recurrence. Pooling these fundamentally different clinical states creates substantial confounding by indication and may magnify the apparent association between residual tumour and outcome. In addition, the reported hazard ratios in Table 4 were derived from univariable analyses restricted to the 79 patients with chondrosarcoma, whereas the principal conclusion is framed across all histological subtypes. The conventional ten-events-per-variable rule should not automatically preclude carefully prespecified, parsimonious adjustment or penalised sensitivity analyses, particularly when the aim is to assess confounding rather than build a complex prediction model [2].

A more informative analysis would separate R1 from R2, repeat survival and recurrence analyses after excluding planned R2 resections and the two patients metastatic at presentation, and report histology-stratified absolute risks alongside relative effects. For local recurrence and distant metastasis, cause-specific hazards and cumulative incidence address different clinical questions; explicit alignment of the estimand, risk set, and interpretation is important in the presence of death as a competing event [3]. Patients with macroscopic residual disease should not be treated as conventionally recurrence-free at baseline for an operative-site recurrence analysis.

The reconstruction findings also merit cautious interpretation. Reconstruction was selected according to defect size and location, and the single lung herniation after unsupported resection cannot establish a threshold at which prosthetic support is required. Standardized 30- and 90-day complication reporting, such as the Clavien-Dindo framework [4], together with respiratory and patient-reported functional outcomes, would improve comparison across techniques. A recent larger series demonstrated the feasibility of pairing oncological endpoints with long-term dyspnoea and performance status after chest wall reconstruction [5]. A multicentre registry using these definitions could convert the present descriptive observations into clinically actionable reconstruction guidance.

These refinements would preserve the study's important contribution while distinguishing the established value of complete resection from the strength of evidence supporting margin-specific prognostic and reconstructive recommendations.

CRediT authorship contribution statement

Vatsala Gupta: Resources, Project administration, Methodology. Parag J. Ratnakar: Visualization, Validation, Supervision. Kotina Shridevi: Methodology, Investigation. Gaurav: Writing – review & editing, Visualization, Validation. Dhanya Dedeepya: Funding acquisition, Formal analysis, Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.