Nutritional Association of Quality of Life Among Colorectal Cancer Survivors in Malaysia: A 6-Month Follow-Up Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript addresses predictors of quality of life (QoL) in colorectal cancer (CRC) survivors in Malaysia, with a six-month follow-up after the end of treatment. It uses validated questionnaires (EORTC QLQ-C30 and QLQ-CR29) and assesses participants' sociodemographic, clinical, and nutritional parameters. The study is relevant because it explores an understudied population and incorporates nutritional status as a predictor of QoL.

However, some methodological and analytical limitations reduce the robustness of the findings, which is why I suggest restructuring the manuscript's content:

• The title is clear, objective, and reflects the study's content well, but "nutritional predictors" could be added, given that nutritional parameters are central to the analysis.
• In the Introduction, the authors contextualize the importance of CRC as a public health problem, justify the relevance of assessing QoL in survivors, and highlight the gap in the literature for Asian populations, especially in Malaysia. However, the literature review is relatively brief and could explore more previous studies on the nutritional impacts on QoL and could more clearly present the study hypothesis and the mechanisms by which nutritional variables could influence QoL.

Regarding the methodology, this is a prospective observational study with a 6-month follow-up, which allows for the analysis of changes after treatment. However, the 6-month period is relatively short to capture variations in QoL, especially in cancer survivors.

The sample consisted of 87 participants from two different hospitals. However, it is noteworthy that the relatively small sample size reduces statistical power, and the calculation of sample size and statistical power is not presented. The authors used multiple linear regression tests to identify predictors. However, there is no mention of checking for multicollinearity. Psychosocial variables, which can be important determinants of QoL, were not considered. Would it be interesting to analyze the stratification of QoL by cancer stage or type of treatment?

Regarding the impact and relevance of the results, the clinical impact of the findings is not discussed in depth; there is no detailed comparison with other Asian studies, and the interaction between nutritional variables and cancer stage could have been explored.

Regarding the discussion, the authors could have further discussed the psychosocial and cultural factors that influence the perception of QoL (even if they were not considered in the data collection). The study mentions high rates of anxiety but does not explore coping strategies or psychological support. There is no mention of differences between men and women, although variables related to sexual function are included.

It would be important to discuss the unaddressed limitations: possible recall bias in data collection via home interviews; lack of multivariate analysis considering different treatments and broader socioeconomic factors (income, education, access to care); and explore more detailed clinical implications for public policy.

Minor comments:

• In the Abstract, you should indicate the sample size to contextualize the results.
• In regression tables, you should include the adjusted R² to indicate the proportion of variance explained.
• Minor grammatical adjustments and standardization of acronyms (e.g., CRC, QoL, HRQoL) throughout the text.

Author Response

Dear Reviewer,

Thank you for your suggestions and comments. Please see attached our point-by-point responses to the Reviewer. 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The topic is important and contextually valuable. The use of Malay-validated EORTC instruments and recruitment across two teaching hospitals strengthens local relevance. However, key aspects of design, reporting, and tables require substantial revision before the study can support “predictor” claims. 

 

Major comments (with exact locations)

1. Design/causal language mismatch (“predictors”). Abstract says patients were evaluated at 6 months post-treatment to determine predictors, yet the paper acknowledges no baseline QoL at diagnosis; limitations reiterate lack of baseline and cross-sectional constraints. Please align title/claims (associations vs predictors) and clarify whether any repeated measures exist. Locations: Abstract; Discussion/Limitations.
2. Ethics timeline and identifiers. Study period is January 2021–July 2022, but USM approval is dated 16 March 2022 (code formatting also varies: USMKK/JEPeM vs USM/JEPeM). Please reconcile coverage for the entire period and harmonize codes. Locations: Study period; IRB statements.
3. Data collection setting ambiguity. Recruitment occurred in hospital clinics, but QoL was assessed via “household interviews.” Specify whether interviews were home visits, telephone, or on-site; describe timing vs treatment completion. Locations: Setting; Instruments/Administration.
4. Table 1 internal inconsistencies and likely errors.
   • “Anterior resection 2 (8.0)”—% is incorrect (2/87 ≈ 2.3%).
   • “Surgical treatment: No 6 (6.9)” but under “Surgical types” there is “None 8 (6.9)”.
   • Immediately after “None 8 (6.9)”, “Yes 68 (78.2) / No 19 (21.8)” appear without a clear header (these seem to be chemotherapy yes/no but are placed under “Surgical types”). Please correct denominators, headings, and percentages. Location: Table 1.
5. Malnutrition claim lacks operational definition. Discussion states “more than half… were malnourished,” yet no malnutrition assessment tool/criteria (e.g., GLIM, PG-SGA, MNA, cut-points on anthropometry) are described in Methods. Add definition, measurement approach, and prevalence table. Locations: Discussion; Methods.
6. Multiple outcomes, limited N, no multiplicity control. You run many regressions across QLQ-C30/CR29 domains with p≤0.05 as the sole criterion; risk of Type I error and unstable estimates is high. Please report model-building strategy, multicollinearity checks (e.g., VIF), residual diagnostics, standardized β, and consider multiplicity adjustment (e.g., FDR). Locations: Statistical Analysis; breadth of outcomes.
7. Tumor site mentioned but not reported. You state tumors were located in the rectosigmoid region, but no tumor-site variable appears in Results tables. Add tumor location to baseline characteristics. Locations: Discussion statement; Table 1 scope.
8. Sex-specific CR29 items and missing-data handling. Table 3 shows “Dyspareuniaa 0.00 (0.00)” and provides sex-specific items (“sexual interest (men/women)”) without Ns or rules for “not applicable.” Clarify scoring for sex-specific items, how NA responses were handled, and provide denominators by sex. Location: Table 3.
9. Inline external link in body text. The FOLFOX URL appears mid-page; move it to References or footnotes. Location: Results page/under Table 2.
10. Terminology and typos to standardize.
    • “GHL/QoL” should be “GHS/QoL.” Location: Table 2 narrative.
    • “QLQC30” instead of “QLQ-C30.” Location: Statistical Analysis lead-in.
    • “first-come, first-serve” → “first-come, first-served.” Location: Sampling method.
    • “Embarassment”, “Dyspareuniaa”. Location: Table 3.
    • Reference list typo “Kr19emer”. Location: References.
11. Scoring citation. Methods say scores were “transformed into 0–100 using a description.” Please cite the official EORTC Scoring Manual and specify which version you followed. Location: Methods—Scoring sentence.
12. Informed consent phrasing. “Written informed consent has been obtained … to publish this paper” reads like a case-report template; rephrase for a cohort study (consent to participate, data use, confidentiality). Location: Informed Consent Statement.

Questions for the authors (please answer point-by-point, citing sections/tables)

• What was each patient’s interval from end of treatment to QoL assessment? Was “time since treatment” included in models? Locations motivating the question: Abstract timing; Setting.
• Did you assess collinearity (VIF), residuals, and provide adjusted R²? Any multiplicity control? Location: Statistical Analysis.
• Which definition/cut-points classified patients as “malnourished”? Please provide prevalence by category (normal/at-risk/malnourished). Location: Discussion claim.
• Consider adding comorbidity burden, stoma status, current therapy, and tumor site to the models/tables to improve interpretability. Locations: Table 3 indicates stoma-related symptoms; site is not reported.

Promising and locally meaningful study, but the manuscript needs major revision to correct table errors, align design/claims, clarify methods (ethics coverage, data-collection setting, malnutrition criteria), and strengthen statistical reporting.

 

Author Response

We are thankful for the review, which helped us improve the quality of the manuscript. Attached please find our point-by-point responses to the Reviewer.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The contribution is relevant, and the findings are consistent. They demonstrate methodological alignment (STROBE) and statistical transparency. Congratulations.

Author Response

The contribution is relevant, and the findings are consistent. They demonstrate methodological alignment (STROBE) and statistical transparency. Congratulations.

Responses:

Thank you so much for your earlier suggestions. I hope our responses were satisfactory to you.

Best regards,

Amal Mitra

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for your detailed response and for submitting the revised manuscript. We appreciate the improvements you have made. Ethics reporting is now consistent. Baseline tables are cleaner. Tumor site is added. Typographical issues are reduced. Reporting of adjusted R squared has been introduced.

After re-review the manuscript still requires major revision before it can be considered further. Please prepare a point-by-point rebuttal that quotes each comment and your change and include both a clean file and a tracked-changes file.

Key required actions in order of priority

Align design and wording
Use association language throughout. Remove predictor phrasing. Ensure one consistent description of timing across Title Abstract Methods Results and Conclusions. State clearly whether assessments occur about six months after diagnosis or after completion of treatment. If timing varies across participants report summary statistics and consider including time since treatment as a covariate.

Make the data-collection setting fully consistent
Choose one setting for interviews on site in the hospital or at home and use it everywhere. Confirm the mode face to face phone or self-administered and the timing relative to clinic visits.

Define malnutrition using a clinical standard and report categories
Adopt GLIM or another accepted standard and describe operational criteria. Provide a table with categories normal at risk malnutrition. If you intend to treat both undernutrition and excess adiposity under one umbrella change the label to nutritional status categories and justify the choice.

Specify change variables with exact formulas and units
For weight BMI waist and hip define the two time points and give the computation of change for example six-month value minus baseline or percent change. Describe missing-data handling and list the analysis denominator for each variable.

Control multiplicity and report model diagnostics
Given the number of QoL outcomes relative to sample size reduce the number of primary endpoints or control the false discovery rate. Report VIF ranges and residual diagnostics for each model. Keep adjusted R squared and add standardized coefficients where appropriate. Describe variable selection and provide a short sensitivity analysis after multiplicity control.

Harmonize interpretations in Results
Remove statements that mix GHS with specific function scales. Ensure the narrative about top-scoring CR29 domains matches the corresponding table values. Provide denominators by sex and the rule for not applicable responses for sex-specific items.

Cite the official EORTC scoring manual
Replace vague phrasing such as using a description with a citation to the EORTC Scoring Manual and state the version used. Confirm that all QLQ-C30 and QLQ-CR29 scores were transformed to a 0 to 100 scale per manual guidance.

Standardize instrument names
Use QLQ-C30 and QLQ-CR29 consistently in text tables and captions. Use GHS QoL for global health status.

Clarify denominators for surgical variables
In Table 1 remove None from Surgical types or compute those percentages using only patients who underwent surgery. Avoid duplication with the binary surgical treatment variable.

Use TNM staging from an authoritative source
Reference AJCC or UICC for TNM staging. Ensure stage definitions in the text match that source.

Minor edits to incorporate

• Correct multicollinearity spelling
• Remove trailing punctuation in the Table 5 footnote after loss of appetite
• Keep first come first served phrasing consistent

Resubmission checklist

• Point-by-point response letter with page and line references
• Clean manuscript and tracked-changes manuscript
• Updated baseline table with clarified surgical denominators
• New table with GLIM categories and prevalence
• Model-diagnostics table that includes VIF ranges and brief residual checks
• Sensitivity analyses that indicate which associations remain after multiplicity control
• Updated STROBE checklist

Author Response

Dear Reviewer,

Thank you so much for your comments and suggestions. 

Please find here our point-by-point responses.

Sincerely,

Amal Mitra

Author Response File: Author Response.pdf