VEGF Signaling Pathway Germline Polymorphisms as Prognostic Pharmacogenetic Biomarkers in Localized High-Grade Osteosarcoma Patients from the GEIS-33 Protocol

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The author designed an interesting study in Precision Oncology and Pharmacogenetic field however, they focused on the “SNPs as Biomarkers” which is completely correct. They included very noticeable clinical characteristics, performed good pharmacological tests, and used nice genotyping assay (TaqMan Assay Technology). I suggested a minor revision to enhance this work for possible acceptance. Here are the comments:
1- Title: I suggest to add “Pharmacogenetic” before “….Biomarker…” in the title which will add more interests and attention and place the paper in the Pharmacogenetic and Precision Oncology category. The authors utilized the pharmacogenetic biomarker in the last paragraph of the Introduction!

2- Abstract: The authors should clearly declare the number of SNPs and their rsIDs.

3- Abstract: The authors should add the genotyping method they used (TaqMan Assay Technology).

4-Abstract: The authors can add Pharmacogenetic and Precision Oncology in the conclusion for advancing their future suggestions.

5- Result: The authors can add a nice table in the sub-section 2.2. before the first line with a brief statement presenting the important info of SNPs in separate columns such as rsID, Gene, Function (Intronic, Synonymous, Promoter, …), Alleles [in a Major>minor format; check Ensembl], Minor Allele Frequencies (MAFs) [they can use a Genome browser such as Ensembl or NCBI Genome browser/SNP], ClinVar data, and PGx Annotation(s) [search your SNP in ClinPGx and report the Variant Annotation(s) -> either significant or not significant].

6-Discussion: The authors are encouraged to add more discussion about Pharmacogenetic biomarkers and Precision Oncology in the discussion, especially in the final paragraphs.

7-Add a better Limitation paragraph incorporating “limits” clearly and keep in mind that also you declared about the little sample size, the number of SNPs can be considered as well as the selection of SNPs based on the higher impacts in their functionalities [Missense > Regulatory > Intronic SNPs]; consequently, the author can conclude that the future studies should focus on the structural and regulatory SNPs of their studied genes in order to decipher the better pharmacokinetic and Pharmacodynamic actions of involving chemo-drugs in Osteosarcoma.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript addresses an understudied topic: the prognostic value of VEGF-pathway germline polymorphisms in localized high-grade osteosarcoma within a prospective multicenter cohort. The clinical need is well-justified, and the dataset is homogeneous. Several interesting associations are reported. However, a thorough revision is required, including toned-down claims, significance reporting, and clearer methods.

-The authors analyzed nine SNPs with several genetic models, yet none of the reported associations would remain significant after the stated Bonferroni correction (p < 0.002). However, the Results and Abstract highlight p-values such as 0.007, 0.02, 0.03 as “significant”. This must be corrected. If you choose not to apply strict Bonferroni correction (e.g., due to LD structure or hypothesis-driven selection), this needs justification and discussion.

-With 69 patients, and as few as 3–6 carriers in minor-allele homozygous groups, hazard ratios become very unstable (e.g., HR = 7.34 with very wide CI). The stated statistical power (74–88%) is based on allelic effect size assumptions that may not be realistic for germline polymorphisms in complex traits. You should present a more realistic power calculation or acknowledge limitations more clearly in the Discussion.

-Survival analysis excludes patients with missing progression or death information, but the number excluded is non-negligible (4 for OS, 6 for RFS). Please describe how this missingness may bias results and whether sensitivity analyses were performed.

-SNP selection appears partially literature-driven, but the rationale is inconsistent. Some SNPs previously linked to osteosarcoma prognosis were not included; others with no functional data were. Provide clearer justification for the final SNP list and state whether any SNPs were excluded due to poor genotyping performance.

-The Discussion acknowledges absence of functional evidence but still over-interprets the association. Either provide stronger mechanistic support or clearly temper claims.

-Patients were divided into three FLT1-combination groups, but the reasoning behind grouping is not well-explained and group sizes are very small. The conclusion that rs7993418 “is more informative” is not statistically well supported and requires either additional explanation or removal.

-The Abstract and Conclusion imply these polymorphisms could “improve therapeutic decision-making,” but the results do not yet justify clinical utility due to: lack of independent validation, modest effect sizes with wide CIs, marginal statistical significance, absence of predictive value for chemotherapy response. Rephrase to avoid overstating applicability.

-The manuscript is generally clear but overly long in the Discussion; consider condensing background sections that reiterate known VEGF biology.

-Please define all abbreviations the first time they appear in the text (OS, RFS etc.)

-Table 1 should clearly mark which p-values are uncorrected.

Comments on the Quality of English Language

moderate editing

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript was improved.