Review Reports - Pelvic Osteosarcoma: Outcomes of Surgically Treated Patients in a Retrospective Single-Center Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study addresses an interesting and relevant topic; however, substantial revisions are required before it can be considered for publication.

Please revise:

Line 148 and 149: Tumor grading was available for 45 patients, with the majority presenting with high-grade G3 lesions (n = 39; 69.6%) and six patients (10.7%) with G4 tumors.

Please revise the tumor grading. Tumor grading must follow the current guidelines (see attachment).

Grade 1: Low‑grade central osteosarcoma and parosteal osteosarcoma
Grade 2: Periosteal osteosarcoma
Grade 3: Conventional osteosarcoma, telangiectatic osteosarcoma, small‑cell osteosarcoma, and high‑grade surface osteosarcoma

Please provide additional details regarding the histologic tumor type, as this information is essential for accurate diagnostic classification.

Please also include the pTNM staging, which is important for prognostic assessment.

Comments for author File: Comments.pdf

Author Response

We thank the reviewer for this important and constructive comment.

Tumor Grading Revision (lines 164 and following)

Tumor grading has been revised according to the current WHO classification of bone tumors. The previous G3/G4 terminology has been corrected to align with contemporary standards. As expected for pelvic osteosarcoma, the vast majority of tumors in our cohort were classified as high-grade (Grade 3). No low-grade (Grade 1 or 2) osteosarcomas were identified. The revised grading has been updated in the Results section accordingly.

Histologic Subtype Clarification (lines 200 and following)

We agree that histologic subtype is relevant for diagnostic precision. Therefore, we have expanded the Results section to provide a detailed breakdown of histologic tumor types, including:

Conventional osteoblastic osteosarcoma
Conventional chondroblastic osteosarcoma
Conventional fibroblastic osteosarcoma
Telangiectatic osteosarcoma

pTNM Staging (lines 206-207)

We appreciate the reviewer’s suggestion to include pTNM staging. However, complete pathologic TNM data were available in only a limited subset of patients (n = 8), primarily due to historical documentation practices and the retrospective design of the study. Given the small number of fully documented TNM cases, a formal subgroup analysis or tabular presentation would not provide statistically meaningful or generalizable results.

We therefore decided not to include a separate pTNM-based analysis, as this could potentially be misleading. Instead, established prognostic variables including tumor size, metastatic status at diagnosis, margin status, and histologic response were analyzed and reported.

We have added a clarifying statement in the manuscript acknowledging the limited availability of complete TNM data as a study limitation.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a retrospective single-center analysis of surgically treated pelvic osteosarcoma over a long study period. The cohort size is respectable for this rare disease and the authors report encouraging survival outcomes. However, there are some methodological and reporting limitations.

Specifically, I have the following comments:

-The median follow-up of 11.5 months is very short for a study reporting 5-year OS and RFS. This raises concerns regarding the robustness of long-term survival estimates. The reported RFS of ~100% at 1, 3, and 5 years is implausible for pelvic osteosarcoma and likely reflects methodological artifact rather than true disease control. The authors must explicitly address how censoring affects survival estimates and consider limiting survival reporting to time points supported by adequate follow-up.

-An R0 rate of 98.2% in pelvic osteosarcoma is “unusually high” compared with most published series. The manuscript does not sufficiently explain pathological margin assessment methodology, use of navigation, cutting guides, intraoperative margin control or whether marginal resections were reclassified. A clearer explanation is required to avoid concerns of selection bias or margin misclassification.

-Only patients treated with curative surgical intent were included. This likely excludes unresectable cases and patients with advanced disease, inflating survival outcomes. The manuscript should explicitly acknowledge this selection bias and clarify how many pelvic osteosarcoma patients were excluded due to non-surgical management.

-Although several prognostic factors are discussed, no comprehensive multivariable Cox regression analysis is presented. Given the conclusions regarding age, metastasis, and margins, this is a noteworthy limitation. If sample size permits, multivariable analysis should be performed or the conclusions should be toned down accordingly.

-MSTS scores were available in only 64% of patients, with retrospective extraction. There is no description of whether assessments were standardized, or whether interobserver variability exists. One-year functional outcome may not reflect long-term recovery after pelvic reconstruction. The limitations of functional outcome assessment should be expanded and more cautiously interpreted.

-The presentation of Enneking subgroups is fragmented and difficult to interpret. A simplified grouping or tabular summary would improve clarity.

-Histologic response to neoadjuvant chemotherapy (e.g., necrosis rate) is not reported, despite being a known prognostic factor. The authors should clarify whether this data was unavailable or excluded.

-Radiotherapy is described as “sporadic,” but dose, modality, and timing are not specified. A brief clarification would strengthen the methods section.

-Table 1 combines OS and RFS but does not adequately explain the limitations of RFS interpretation.

Comments on the Quality of English Language

Minor grammatical issues are present (e.g., “sporadicly,” “assesment”).

Language editing is recommended.

Author Response

We thank the reviewer for this important and constructive comment.

Comment 1 – Short median follow-up and implausible RFS

The median follow-up of 11.5 months is very short for a study reporting 5-year OS and RFS. This raises concerns regarding robustness…

Response:

We thank the reviewer for highlighting this important methodological aspect.

We fully agree that the median follow-up of 11.5 months limits the robustness of long-term survival estimation, particularly for recurrence-free survival (RFS). The high RFS estimates at 3 and 5 years primarily reflect substantial right-censoring rather than absence of recurrence.

To address this:

We have explicitly clarified the censoring structure in the Methods and Results sections.
We have limited formal interpretation of survival beyond time points supported by adequate follow-up.
The discussion now explicitly acknowledges that RFS estimates beyond 24 months should be interpreted cautiously.

The relevant sections have been revised accordingly (Methods, Results, Discussion – Limitations).

Comment 2 – Unusually high R0 rate (98.2%)

An R0 rate of 98.2% is unusually high… insufficient explanation of margin assessment…

Response:

We appreciate this critical observation.

The R0 rate was corrected to 96.4%. The high R0 rate reflects strict inclusion criteria: only patients undergoing surgery with curative intent were included. Patients deemed unresectable were not part of this cohort (see response to Comment 3).

We have now expanded the Methods section to clarify:

Margin assessment was performed by specialized musculoskeletal pathologists.
R0 was defined as histologically tumor-free margins with no microscopic residual disease.
Specimens were inked and evaluated using standardized protocols.
Navigation-assisted osteotomies and patient-specific guides were used in selected cases (added clarification).
No marginal resections were reclassified as wide.

We have also added a paragraph in the Discussion acknowledging that the high R0 rate may reflect referral bias and selection of resectable cases.

Comment 3 – Selection bias (only curative surgical cases included)

This likely excludes unresectable cases and inflates survival outcomes…

Response:

We agree with the reviewer.

The study cohort includes only patients treated surgically with curative intent. Patients deemed unresectable or treated palliatively were not included.

We have now:

Explicitly stated this selection criterion in the Methods.
Expanded the Discussion - Limitations section to explicitly acknowledge that this design likely inflates survival estimates relative to population-based data.

Comment 4 – Lack of multivariable Cox regression

No comprehensive multivariable analysis…

Response:

We appreciate this suggestion.

We have therefore:

Added a multivariable Cox analysis.

Comment 5 – MSTS functional assessment limitations

MSTS available in only 64%… retrospective extraction…

Response:

We agree and thank the reviewer.

We have expanded the Methods section to clarify:

MSTS scores were extracted retrospectively from documented outpatient assessments.
Evaluations were performed by treating orthopedic oncologists using the standard MSTS 1993 scoring system.
No formal interobserver reliability analysis was performed.

We have also strengthened the Discussion:

Acknowledging incomplete availability (64%).
Emphasizing that one-year MSTS does not reflect long-term adaptation.
Highlighting potential selection bias toward surviving patients.

Interpretation of functional results has been moderated accordingly.

Comment 6 – Fragmented Enneking subgroup presentation

Presentation is difficult to interpret…

Response:

We thank the reviewer for this suggestion.

We have simplified Enneking subgroup presentation by:

Combination of subgroups.
Added a table for the simplified grouping

Comment 7 – Histologic response to neoadjuvant chemotherapy missing

Necrosis rate not reported…

Response:

We appreciate this observation.

We added the Saltzer-Kuntschik score analysis to the manuscript (Methods, Results, Discussion).

Comment 8 – Radiotherapy details insufficient

Dose, modality, timing not specified…

Response:

We agree.

The Methods section now specifies:

Radiotherapy modality (photon).
Median dose (Gy).
Timing (postoperative).
Indication criteria.

Comment 9 – Table 1 (OS & RFS combined)

Does not adequately explain RFS limitations…

Response:

We thank the reviewer.

Table 1 has been revised:

OS is now presented only.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for revising the manuscript. The updated version addresses the previous concerns and looks good.