Obesity and Tolerance to Neoadjuvant Chemotherapy in Breast Cancer: A Retrospective Cohort Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Its an original study focusing on the impact of obesity in the clinic of breast cancer. Although its complex, this study shows that some portion of obese patients may require a closer monitoring for the treatment-related toxicities, including peripheral neuropathy. As brought by the authors in this study, reliance on BMI only may not determinant, others biological or co-morbidity parameters should be integrated. I believe that this study may bring more comfort to some patients suffering from obesity and breast cancer. This study may open new avenues for improving the quality of breast cancer treatment and comfort to patients.

Author Response

Reviewer Comment 1:
Its an original study focusing on the impact of obesity in the clinic of breast cancer. Although its complex, this study shows that some portion of obese patients may require a closer monitoring for the treatment-related toxicities, including peripheral neuropathy. As brought by the authors in this study, reliance on BMI only may not determinant, others biological or co-morbidity parameters should be integrated. I believe that this study may bring more comfort to some patients suffering from obesity and breast cancer. This study may open new avenues for improving the quality of breast cancer treatment and comfort to patients.

Response 1:

We sincerely thank the reviewer for the positive and encouraging comments. We truly appreciate the recognition of the originality and clinical relevance of our study, as well as the importance of integrating biological and comorbidity parameters beyond BMI when assessing treatment-related toxicities in patients with obesity and breast cancer. No specific revisions were requested in this comment.

Reviewer 2 Report

Comments and Suggestions for Authors

Please consider referring to "patients with obesity" rather than "obese patients".

Methodology section should include an explanation of clinically significant toxicities variables -- such as ... a specific toxicity needed to be mentioned in the clinician notes as the reason for dose reduction or treatment delay etc.

Were there instances where the specific reason for a change in treatment plan was not listed?

Were any of the specific reasons patient-reported??? 

A limitation is that patient-reported toxicities are more severe and more numerous as compared to clinician observed/reported.

 

 

Author Response

Reviewer 2 Comment 1:

Please consider referring to "patients with obesity" rather than "obese patients".

Response 1:

Thank you for this helpful comment. We agree with the reviewer’s suggestion and have revised the manuscript to use the term “patients with obesity” (OWP) instead of “obese patients” throughout the text.

Reviewer 2 Comment 2:

Methodology section should include an explanation of clinically significant toxicities variables -- such as ... a specific toxicity needed to be mentioned in the clinician notes as the reason for dose reduction or treatment delay etc. Were there instances where the specific reason for a change in treatment plan was not listed?

Response 2:

Thank you for this important remark.  We have now clarified the study methodology in section 2.2 Outcomes (page 4, lines 155-166); we now explicitly state that only adverse events leading to modification of the planned chemotherapy regimen were recorded. Events were classified as toxicity-related when explicitly documented or clearly supported by contemporaneous clinical documentation indicating a treatment-related adverse event. Please also take into consideration that clinician notes include notes taken in pre-scheduled medical outpatient visits as well as notes from emergency visits (if applicable) and also notes from other healthcare professionals such nurses at the Outpatient Chemotherapy Unit.

Reviewer 2 Comment 3:

Were any of the specific reasons patient-reported??? 

Response 3: Toxicity events were identified based on clinician-documented medical records, including scheduled outpatient visits, emergency visits (if applicable) and clinical notes from the outpatient chemotherapy unit. Although patient-reported outcomes were not systematically collected, it is important to note that many clinically relevant toxicities are symptom-driven (e.g., neuropathy, fatigue, nausea, diarrhea, and decline in performance status) and were documented after being reported by patients. Please also take into consideration that in our Early Breast Cancer Clinic, patients are routinely evaluated prior to each chemotherapy cycle, allowing for documentation of new or worsening symptoms that may qualify as adverse events.

Reviewer 2 Comment 4:

A limitation is that patient-reported toxicities are more severe and more numerous as compared to clinician observed/reported.

Response 4: Thank you for this important observation. We acknowledge that patient-reported toxicities may reflect a broader symptom burden. However, in this retrospective study, we focused specifically on events that resulted in treatment modification.  Very important toxicities are symptom-driven and they were considered when they led to a documented change in treatment. We acknowledge that less severe toxicities reported by patients may have been missed in our study and this limitation has now been added to the Discussion section (page 12-13, lines 398-401).

Reviewer 3 Report

Comments and Suggestions for Authors

Major Comments:

1. Study Design:

The study is described as a retrospective cohort analysis with a secondary matched comparison (1:1 age-stratified frequency matching between obese and non-obese patients). However, several aspects require clarification:

• • It is unclear whether the non-obese comparator group was randomly selected before or after assessment of outcomes.
  • The rationale for excluding the remaining non-obese patients from comparative analyses should be explicitly justified.
  • Frequency matching does not eliminate residual confounding; however, unconditional logistic regression was used. The manuscript should clarify whether matching strata were incorporated appropriately in the analysis.
  • A sensitivity analysis using the full cohort (all 487 patients) would strengthen the robustness of findings.

Additionally, although a flow diagram is provided, the description of exclusion criteria could be more precise (e.g., definition of “other malignancies,” handling of oligometastatic disease).

2. Definition and Clinical Meaning of “Global Toxicity”:

The composite endpoint of “clinically significant toxicity” includes heterogeneous events (dose reductions, delays, hospitalization, discontinuation). While pragmatic, this grouping merges outcomes of varying clinical severity.

• • Dose delay of any duration is counted equivalently to hospitalization or permanent discontinuation. This may dilute interpretability.
  • The manuscript should justify why all components were weighted equally.
  • Reporting the absolute frequency of each component alongside the composite is helpful, but interpretation should clearly distinguish between severity levels.

3. Overinterpretation of BMI as a Clinical Tool:

The Discussion appropriately critiques BMI as an imperfect measure of body composition. However, the Conclusion suggests targeted lifestyle interventions based on current findings. While reasonable from a public health standpoint, this study does not test interventions.

The conclusion should be reframed to state that obese patients may benefit from closer monitoring, and that future studies incorporating body composition metrics are warranted.

Minor Comments

1. Consider adding a patient flow diagram to enhance transparency.
2. Clarify whether chemotherapy dosing was strictly based on actual body weight for all patients and whether any empirical dose capping occurred.
3. Define how missing toxicity data were handled.
4. Specify CTCAE version consistently (v6.0 is mentioned; confirm accuracy).
5. Standardize terminology (e.g., “carboplatina” vs “carboplatin”).
6. Clarify whether survival time was calculated from diagnosis or from NAC initiation (both definitions appear in different sections).
7. Consider presenting baseline characteristics of the full cohort alongside the matched cohort for transparency.
8. Minor language editing is recommended for grammar and clarity.

Author Response

Reviewer 3 Comment 1: Study Design: The study is described as a retrospective cohort analysis with a secondary matched comparison (1:1 age-stratified frequency matching between obese and non-obese patients). However, several aspects require clarification:

Reviewer 3 Comment 1.1: It is unclear whether the non-obese comparator group was randomly selected before or after assessment of outcomes.

Response 1.1:

We thank the reviewer for raising this relevant point. We have clarified the study methodology in page 3, lines 119-121. Specifically, we now state explicitly that the non-obese comparator group was randomly selected prior to the assessment of outcomes. The selection process was performed by an individual who was blinded to outcome variables and had access only to age and BMI subgroup information.

Reviewer 3 Comment 1.2: The rationale for excluding the remaining non-obese patients from comparative analyses should be explicitly justified.

Response 1.2:

Thank you for this important comment. Controls were randomly selected from the pool of eligible non-obese individuals using an age-stratified frequency matching procedure to ensure representativeness and comparability of age across groups. Random sampling from the source population preserves internal validity and reduces the risk of selection bias.

In addition, a 1:1 ratio between patients with obesity and non-obese patients was chosen because it provides high statistical efficiency when the size of the exposed group is fixed, as in our case. Although increasing the number of controls per exposed participant can improve statistical precision, the incremental gain diminishes substantially beyond a 1:1 or 1:2 ratio. Given the adequate size of the non-obese pool and the use of frequency matching on age to improve balance and reduce variance, a 1:1 ratio was considered methodologically appropriate and resource-efficient.

Therefore, there was no formal exclusion procedure of non-obese patients from the comparative analyses. 

Reviewer 3 Comment 1.3: Frequency matching does not eliminate residual confounding; however, unconditional logistic regression was used. The manuscript should clarify whether matching strata were incorporated appropriately in the analysis.

Response 1.3:

We appreciate the reviewer’s insightful methodological comment. As described in Section 1.1, controls were selected using frequency matching on age rather than individual (pairwise or set) matching. Frequency matching ensures balance in the marginal age distribution between exposure groups but does not create matched strata that require conditioning in the analysis.

Conditional logistic regression is appropriate for individually matched designs, in which comparisons are performed within matched pairs or sets. In contrast, for frequency-matched studies, valid effect estimates can be obtained using unconditional logistic regression provided that the matching variable is included in the model.

Accordingly, in our analyses, all regression models incorporated age as a covariate to appropriately account for the frequency-matching structure and to minimize residual confounding.

Reviewer 3 Comment 1.4: A sensitivity analysis using the full cohort (all 487 patients) would strengthen the robustness of findings.

Response 1.4:

As clarified above (Responses 1, 1.1 and 1.2), the study was designed a priori as an age-stratified frequency-matched cohort analysis. Accordingly, data extraction was restricted to the predefined 1:1 sample of patients with obesity and non-obese controls. The remaining eligible non-obese patients were not extracted, and therefore a sensitivity analysis including the full cohort of 487 patients is not feasible.

Reviewer 3 Comment 1.5: Additionally, although a flow diagram is provided, the description of exclusion criteria could be more precise (e.g., definition of “other malignancies,” handling of oligometastatic disease).

Response 1.5:

We thank the reviewer for this pertinent observation. We have revised the Methods section to provide a more precise definition of the exclusion criteria (page3, lines 106–110). We have revised the flow diagram to improve clarity (page 6). Specifically, patients with oligometastatic disease treated with neoadjuvant chemotherapy were excluded (n = 10). Additional exclusion criteria included any prior history of invasive malignancy (non-melanoma skin cancer and in situ carcinomas were included) (n = 29 patients including 17 with breast cancer), treatment primarily at other institutions (n = 7), and pregnancy at the time of diagnosis (n = 1). No additional clinical exclusions were applied.

Reviewer 3 Comment 2: Definition and Clinical Meaning of “Global Toxicity”: The composite endpoint of “clinically significant toxicity” includes heterogeneous events (dose reductions, delays, hospitalization, discontinuation). While pragmatic, this grouping merges outcomes of varying clinical severity. Dose delay of any duration is counted equivalently to hospitalization or permanent discontinuation. This may dilute interpretability. The manuscript should justify why all components were weighted equally.

Response 2: We thank the reviewer for this important comment. We have clarified the study methodology in section 2.2 (page 4, lines 153–166). The composite endpoint of clinically significant toxicity, including hospitalization, treatment delay, dose reduction, or treatment discontinuation, was defined because all these events represent clinically relevant interferences with optimal chemotherapy delivery and dose intensity. Please note that each toxicity component was also analyzed separately.

Reviewer 3 Comment 3: Reporting the absolute frequency of each component alongside the composite is helpful, but interpretation should clearly distinguish between severity levels.

Response 3: Thank you for this important observation. We acknowledge that hospitalization and treatment discontinuation may represent potentially more severe clinical consequences than dose delays or reductions and should therefore be interpreted accordingly. We have revised the Discussion section (page 11, lines 338–341) to emphasize that.

Reviewer 3 Comment 4: Overinterpretation of BMI as a Clinical Tool: The Discussion appropriately critiques BMI as an imperfect measure of body composition. However, the Conclusion suggests targeted lifestyle interventions based on current findings. While reasonable from a public health standpoint, this study does not test interventions. The conclusion should be reframed to state that obese patients may benefit from closer monitoring, and that future studies incorporating body composition metrics are warranted.

Response Comment 4: We are very grateful for this important comment. We agree with the reviewer’s suggestion and have revised the Conclusion section (page 13, lines 405–409) accordingly. The previous statement regarding targeted lifestyle interventions has been removed from conclusion, and referred only in discussion.

Reviewer 3 Minor Comments:

1. Consider adding a patient flow diagram to enhance transparency.

Response minor comment 1: Thank you for this suggestion. A patient flow diagram was included in the original submission. We have revised Figure 1 (page 6) to improve clarity and ensure consistency with the eligibility criteria described in the Methods section.

2. Clarify whether chemotherapy dosing was strictly based on actual body weight for all patients and whether any empirical dose capping occurred.

Response minor comment 2: Chemotherapy dosing was based on actual body surface area using measured body weight and height, as stated in the Methods (page 3, lines 112–113). We have now clarified that no empirical dose capping was applied.

3. Define how missing toxicity data were handled.

Response minor comment 3: Toxicity events were identified based on clinician-documented medical records, including notes taken in pre-scheduled medical outpatient visits as well as notes from emergency visits (if applicable) and notes from other healthcare professionals such nurses at the Outpatient Chemotherapy Unit. In the absence of documentation of a toxicity-related event, it was assumed that no clinically significant toxicity occurred. No imputation for missing toxicity data was performed. We have clarified the study methodology in section 2.2 Outcomes (page 4, lines 153-159). Please note that all patients were observed before any new chemotherapy cycle; patients starting treatments at other hospitals (where missing data would increase) were excluded from this study.

4. Specify CTCAE version consistently (v6.0 is mentioned; confirm accuracy).

Response minor comment 4: We thank the reviewer for allowing us to clarify this issue. At the time the original clinical records were documented, CTCAE v0.5 was the standard, in which grade 1 was "Asymptomatic" and required clinical or electrodiagnostic confirmation. In routine clinical practice, however, clinicians considered mild symptoms as grade 1 without additional exams. For clarity and consistency with current standards, we retrospectively mapped these data to CTCAE v0.6, in which grade 1 corresponds to “mild symptoms.” Grades 2 and 3 definitions remain unchanged. Please take in consideration that CTCAE v6.0 was published after the time of the original clinical records; therefore, our retrospective mapping does not strictly correspond to the contemporaneous guidelines. However, it reflects clinical reality and ensures comparability with current literature.

5. Standardize terminology (e.g., “carboplatina” vs “carboplatin”).

Response minor comment 5: Thank you for this helpful comment. We have revised the manuscript to use the term “carboplatin”

6. Clarify whether survival time was calculated from diagnosis or from NAC initiation (both definitions appear in different sections).

Response minor comment 6: We thank the reviewer for highlighting this point. We have clarified in the Methods section 2.2.2 that PFS was calculated from initiation of NAC to evaluate treatment-related outcomes, whereas OS was calculated from the date of diagnosis to reflect the overall disease course. This distinction has now been made explicit to avoid ambiguity (page 4, lines 179-181).

7. Consider presenting baseline characteristics of the full cohort alongside the matched cohort for transparency.

Response minor comment 7: We thank the reviewer for this suggestion. Detailed baseline variables were collected only for the selected 1:1 sample. For the full cohort of eligible patients, only age, menopausal status, and BMI were extracted, and these characteristics are presented in Table 1 (page 6).

8. Minor language editing is recommended for grammar and clarity.

Response minor comment 8: We thank the reviewer for this suggestion. The manuscript has undergone careful language revision to improve grammar and clarity.

Reviewer 4 Report

Comments and Suggestions for Authors

General valutation: The manuscript addresses a clinically relevant issue, namely factors influencing tolerance to systemic chemotherapy in breast cancer patients. The focus on real world toxicity outcomes and treatment modifications is timely and of interest for daily clinical practice. The manuscript is overall clearly written and logically structured. That said, some aspects of the discussion would benefit from a broader contextualization, particularly with respect to clinical strategies that have been explored to mitigate chemotherapy-related toxicity and preserve treatment adherence and patient-reported outcomes.

Comment: 
The manuscript appropriately emphasizes the clinical impact of treatment delays, dose reductions, and discontinuations. However, the discussion could be strengthened by briefly framing these findings within the broader literature on supportive or adjunctive interventions aimed at improving chemotherapy tolerance in breast cancer patients. The reported associations are clearly described. Nonetheless, a clearer distinction between non-modifiable patient related factors and potentially modifiable contributors to treatment tolerance would improve the clinical interpretation of the results. Although patient reported outcomes are mentioned, this aspect remains somewhat underdeveloped. Given their increasing relevance in treatment decision making, a short expansion of this section would enhance the translational value of the study.

Minor comments

Literature coverage: 
The reference list is appropriate overall, however, the inclusion of additional clinical studies addressing supportive strategies during adjuvant or neoadjuvant chemotherapy particularly those reporting hematological tolerance and quality of life outcomes could improve completeness. Minor editorial revisions are suggested to ensure consistency in terminology related to toxicity endpoints.

Author Response

Reviewer 4 Comment 1:

General valutation: The manuscript addresses a clinically relevant issue, namely factors influencing tolerance to systemic chemotherapy in breast cancer patients. The focus on real world toxicity outcomes and treatment modifications is timely and of interest for daily clinical practice. The manuscript is overall clearly written and logically structured. That said, some aspects of the discussion would benefit from a broader contextualization, particularly with respect to clinical strategies that have been explored to mitigate chemotherapy-related toxicity and preserve treatment adherence and patient-reported outcomes.

Comment: 
The manuscript appropriately emphasizes the clinical impact of treatment delays, dose reductions, and discontinuations. However, the discussion could be strengthened by briefly framing these findings within the broader literature on supportive or adjunctive interventions aimed at improving chemotherapy tolerance in breast cancer patients. The reported associations are clearly described. Nonetheless, a clearer distinction between non-modifiable patient related factors and potentially modifiable contributors to treatment tolerance would improve the clinical interpretation of the results. Although patient reported outcomes are mentioned, this aspect remains somewhat underdeveloped. Given their increasing relevance in treatment decision making, a short expansion of this section would enhance the translational value of the study.

Response Comment 1: We are very grateful for this important comment. We agree with the reviewer’s suggestion and have revised the Discussion accordingly (page 12, lines 350-352; page 12, lines 381-388). We acknowledge that less severe toxicities reported by patients may have been missed in our study, and this limitation has now been added to the Discussion section (page 12-13, lines 398-401).

Reviewer 4 Minor comments

Literature coverage: The reference list is appropriate overall, however, the inclusion of additional clinical studies addressing supportive strategies during adjuvant or neoadjuvant chemotherapy particularly those reporting hematological tolerance and quality of life outcomes could improve completeness. Minor editorial revisions are suggested to ensure consistency in terminology related to toxicity endpoints.

Response minor comment 1: We thank the reviewer for this helpful suggestion. We have updated the reference list to include additional clinical studies addressing supportive strategies during neoadjuvant chemotherapy (page 12, lines 381-388). We have also made the necessary editorial revisions regarding terminology related to toxicity endpoints throughout the manuscript.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

I would like to thank the authors for their careful and thoughtful revision. They have done a solid job clarifying the methodology, variables, and data structure of this retrospective study. As we all appreciate, collecting and harmonizing retrospective clinical data in a rigorous manner is challenging, and the revisions have substantially improved transparency and readability.

Minor comment:

Please revise Table 1. The column titles appear to be collapsing or merging, which makes the table difficult to read. Formatting should be corrected to ensure clarity before publishing.