The Association Between Neutrophil-to-Lymphocyte Ratio and Histological Tumor Differentiation in Solid Malignancies: A Systematic Review

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

The manuscript entitled “The Association Between Neutrophil-to-Lymphocyte Ratio and Histological Tumor Differentiation in Solid Malignancies: A Systematic Review” is a systematic review analyzing the association between NLR and tumor grade in 13 articles from the literature.

I appreciate the revisions made to the methods and discussion sections. The methods section is now more transparent and includes all the details necessary to replicate this investigation. Moreover, the discussion has improved.

However, there are still points that the authors should address before the manuscript can be accepted for publication in Diagnostics.

Comments and suggestions:

• The language should be reviewed by a native speaker with experience in academic writing, as many errors and style issues remain (or have even appeared after the revision). For example:

Line 36: “one small colorectal adenocarcinoma study”  the word “small” is ambiguous and could be unclear or misleading. If the authors refer to a study with a small sample size, they should clarify this explicitly, for example: “a small colorectal adenocarcinoma cohort study” or “a study with a small colorectal adenocarcinoma cohort”.

Line 64: “In malignant disease” this could be changed to “in the context of cancer” for greater clarity and precision.

Similar issues should be addressed throughout the manuscript.

• Lines 157–161: Please add the reference number immediately after the tumor type to facilitate access to the cited article (e. g., “breast carcinoma [X]”).
• Why the authors deleted the following subsection from the results section? I found it useful; it could be added as first subsection in the results section with the necessary modifications and corrections.

“In their study, Tang et al. (2017) examined the association between pretreatment NLR and the extent of tumor differentiation in biliary tract cancers (n = 3217). Elevated NLR values were associated with poorer tumor differentiation. However, this association was 175 only marginally statistically significant [23]. Nomelini et al. (2019) investigated the relationship between NLR and the degree of tumor differentiation in ovarian cancer (n = 72). The findings did not demonstrate a statistically significant correlation between NLR values and the histological grade of tumor differentiation [24]. In the study conducted by Arora et al. (2023) investigating this association in breast cancer (n = 73), no statistically significant relationship was found between NLR and tumor differentiation grade [25] In the study by De La Cruz-Ku et al. (2020) evaluating this association in breast cancer (n = 118), patients with lower NLR values exhibited higher histological grades; however, this association was not statistically significant (p = 0.374) [26]. Mano et al. (2015) investigated this association in patients with bladder cancer (n = 107). Elevated NLR values showed a weak positive correlation with higher tumor grades, and this relationship was statistically significant (p = 0.028) [27]. Jadoon et al. analyzed this relationship in a cohort of patients with breast cancer (2050). The median NLR was 2.14 and the cutoff value determined by ROC analysis was 2.5. No statistically significant association was found between NLR and tumor grade (p = 0.694) [28]. Tang et al. investigated this relationship in patients with bladder cancer (n = 302). Patients with high-grade tumors had significantly higher NLR values compared to those with low-grade tumors (median NLR 4.42 vs. 3.42, p < 0.001) [29]. Oh et al. (2016) investigated this association in prostate cancer in a retrospective study including 1106 patients. Higher NLR values were observed in patients with high Gleason score tumors (≥4+3) compared to those with low-grade tumors or negative biopsies, suggesting an association between elevated NLR and more aggressive tumor grades (p < 0.001) [30]. Chan et al. (2018) investigated this relationship in a cohort of 712 patients with soft tissue sarcoma. Elevated NLR values were significantly associated with high-grade tumors and the presence of metastases at diagnosis (p < 0.0001), suggesting a correlation between systemic inflammation and more aggressive tumor characteristics [31]. Chandrasekaran et al. (2022) reported a significant association between elevated NLR values and high tumor grade in renal cell carcinoma in a cohort of 150 patients (p < 0.05) [32]. Nafissi et al. (2025) evaluated the role of multiple peripheral inflammatory indices in predicting tumor grade in breast cancer in a cohort of 114 patients. Elevated NLR values were significantly associated with poorer tumor differentiation (p = 0.027) [33]. Ali et al. (2022) analyzed this relationship in colorectal adenocarcinoma in a cohort of 46 patients. Mean NLR values increased progressively with tumor grade; however, this association was not statistically significant (p = 0.4) [34]. Kawahara et al. (2024) investigated the impact of NLR in patients with pancreatic cancer (n = 461). Elevated NLR values (≥3.2) were significantly associated with poorer histological differentiation (p = 0.002) [35]. Kim et al. (2019) analyzed this relationship in colorectal cancer with synchronous liver metastases (n = 83). Elevated NLR values were significantly associated with poorly differentiated tumors (p = 0.048) [36].”

Author Response

Comments 1: 

The manuscript entitled “The Association Between Neutrophil-to-Lymphocyte Ratio and Histological Tumor Differentiation in Solid Malignancies: A Systematic Review” is a systematic review analyzing the association between NLR and tumor grade in 13 articles from the literature.

 

I appreciate the revisions made to the methods and discussion sections. The methods section is now more transparent and includes all the details necessary to replicate this investigation. Moreover, the discussion has improved.

 

However, there are still points that the authors should address before the manuscript can be accepted for publication in Diagnostics.

Response 1: 

Dear Reviewer,

Thank you very much for your thoughtful comments and constructive suggestions provided during the previous review round. They were extremely valuable and helped us substantially improve the quality of the manuscript. Your recommendations enabled us to improve the overall clarity and scientific presentation of the review. We sincerely appreciate the time and effort you invested in evaluating our work and are grateful for your positive assessment of the revised manuscript.

Comments 2:

The language should be reviewed by a native speaker with experience in academic writing, as many errors and style issues remain (or have even appeared after the revision). For example:

Line 36: “one small colorectal adenocarcinoma study”  the word “small” is ambiguous and could be unclear or misleading. If the authors refer to a study with a small sample size, they should clarify this explicitly, for example: “a small colorectal adenocarcinoma cohort study” or “a study with a small colorectal adenocarcinoma cohort”.

 

Line 64: “In malignant disease” this could be changed to “in the context of cancer” for greater clarity and precision.

 

Similar issues should be addressed throughout the manuscript.

Response 2: 
Thank you for this helpful observation. The specific expressions indicated in Lines 36 and 64 have been revised accordingly. In addition, the manuscript has undergone a further language review, and several expressions throughout the text have been rephrased to improve clarity, precision, and overall academic English. All language-related revisions introduced during the present round are highlighted in green.

Comments 3:

Lines 157–161: Please add the reference number immediately after the tumor type to facilitate access to the cited article (e. g., “breast carcinoma [X]”).

Response 3: We agree with this helpful suggestion. Reference numbers have now been added immediately after the corresponding tumor types throughout the manuscript.

Comments 4: 

Why the authors deleted the following subsection from the results section? I found it useful; it could be added as first subsection in the results section with the necessary modifications and corrections.

“In their study, Tang et al. (2017) examined the association between pretreatment NLR and the extent of tumor differentiation in biliary tract cancers (n = 3217). Elevated NLR values were associated with poorer tumor differentiation. However, this association was 175 only marginally statistically significant [23]. Nomelini et al. (2019) investigated the relationship between NLR and the degree of tumor differentiation in ovarian cancer (n = 72). The findings did not demonstrate a statistically significant correlation between NLR values and the histological grade of tumor differentiation [24]. In the study conducted by Arora et al. (2023) investigating this association in breast cancer (n = 73), no statistically significant relationship was found between NLR and tumor differentiation grade [25] In the study by De La Cruz-Ku et al. (2020) evaluating this association in breast cancer (n = 118), patients with lower NLR values exhibited higher histological grades; however, this association was not statistically significant (p = 0.374) [26]. Mano et al. (2015) investigated this association in patients with bladder cancer (n = 107). Elevated NLR values showed a weak positive correlation with higher tumor grades, and this relationship was statistically significant (p = 0.028) [27]. Jadoon et al. analyzed this relationship in a cohort of patients with breast cancer (2050). The median NLR was 2.14 and the cutoff value determined by ROC analysis was 2.5. No statistically significant association was found between NLR and tumor grade (p = 0.694) [28]. Tang et al. investigated this relationship in patients with bladder cancer (n = 302). Patients with high-grade tumors had significantly higher NLR values compared to those with low-grade tumors (median NLR 4.42 vs. 3.42, p < 0.001) [29]. Oh et al. (2016) investigated this association in prostate cancer in a retrospective study including 1106 patients. Higher NLR values were observed in patients with high Gleason score tumors (≥4+3) compared to those with low-grade tumors or negative biopsies, suggesting an association between elevated NLR and more aggressive tumor grades (p < 0.001) [30]. Chan et al. (2018) investigated this relationship in a cohort of 712 patients with soft tissue sarcoma. Elevated NLR values were significantly associated with high-grade tumors and the presence of metastases at diagnosis (p < 0.0001), suggesting a correlation between systemic inflammation and more aggressive tumor characteristics [31]. Chandrasekaran et al. (2022) reported a significant association between elevated NLR values and high tumor grade in renal cell carcinoma in a cohort of 150 patients (p < 0.05) [32]. Nafissi et al. (2025) evaluated the role of multiple peripheral inflammatory indices in predicting tumor grade in breast cancer in a cohort of 114 patients. Elevated NLR values were significantly associated with poorer tumor differentiation (p = 0.027) [33]. Ali et al. (2022) analyzed this relationship in colorectal adenocarcinoma in a cohort of 46 patients. Mean NLR values increased progressively with tumor grade; however, this association was not statistically significant (p = 0.4) [34]. Kawahara et al. (2024) investigated the impact of NLR in patients with pancreatic cancer (n = 461). Elevated NLR values (≥3.2) were significantly associated with poorer histological differentiation (p = 0.002) [35]. Kim et al. (2019) analyzed this relationship in colorectal cancer with synchronous liver metastases (n = 83). Elevated NLR values were significantly associated with poorly differentiated tumors (p = 0.048) [36].”

Response 4:
We removed this subsection during the previous revision in response to another reviewer’s concern that the information presented in the Results section was repetitive, as the same study-level data were already summarized in the corresponding tables. However, we agree that this introductory overview improves the structure and readability of the Results section. We have therefore reinstated it as the first subsection of the Results, with the necessary revisions and corrections to reduce redundancy.

Reviewer 2 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The work has undergone significant changes and now looks excellent. For the ideal version, several corrections are required.

Required corrections before final publication:

1. Problem with reference 10.

Das, S.; Dey, M.K.; Devireddy, R.; Gartia, M.R. Biomarkers in Cancer Detection, Diagnosis, and Prognosis. Sensors 2023, 364 24, 37, doi:10.3390/s24010037.Author 1, A.B.; Author 2, C.D. Title of the article. Abbreviated Journal Name Year, Volume, page 365 range.

2. It is necessary to register the protocol before the final publication of the article (for example, in the Open Science Framework).

3. In the Methods section it is stated that two authors conducted the screening and a third verified the final list. Please indicate how disagreements during screening were resolved and whether dual independent data extraction was performed.

Discussion points (optional improvements):

1. It is desirable to provide the NOS scores for each study and the number of works by quality categories.
2. The work would be improved if stratification by clinical context (localized vs. metastatic cancer) were added.

Author Response

Comments 1

The work has undergone significant changes and now looks excellent. For the ideal version, several corrections are required.

Required corrections before final publication:

Response 1: Dear Reviewer,

We sincerely thank you for your constructive guidance and insightful comments throughout the review process. Your recommendations have been instrumental in improving the quality, clarity, and transparency of our manuscript. We are delighted that you consider the revised version to have undergone significant improvement, and we greatly appreciate your positive assessment. We have carefully addressed each of the remaining comments, which we believe have further strengthened the manuscript.

Comments 2:

1. Problem with reference 10.

Das, S.; Dey, M.K.; Devireddy, R.; Gartia, M.R. Biomarkers in Cancer Detection, Diagnosis, and Prognosis. Sensors 2023, 364 24, 37, doi:10.3390/s24010037.Author 1, A.B.; Author 2, C.D. Title of the article. Abbreviated Journal Name Year, Volume, page 365 range.

Response 2: We thank you for drawing our attention to this issue. The formatting error in Reference 10 has been corrected, and the reference has now been updated to comply with the journal's citation style.

Comment 3

2. It is necessary to register the protocol before the final publication of the article (for example, in the Open Science Framework).

Response 3: Thank you for this valuable suggestion. In accordance with the reviewer’s recommendation, the review protocol was retrospectively registered in the Open Science Framework. The Methods section has been revised accordingly to report the registration and provide access to the corresponding OSF record. We would also like to note that the OSF platform was particularly difficult to use during the registration process, with repeated errors occurring while entering and saving the required information, apparently due to a high volume of users. We therefore kindly ask the reviewer and the Editor to take this into consideration should the final approval of the registration take longer than normally expected.

Comments 4

3. In the Methods section it is stated that two authors conducted the screening and a third verified the final list. Please indicate how disagreements during screening were resolved and whether dual independent data extraction was performed.

Response 4: Thank you for this important suggestion. The Methods section has been revised to clarify the screening and data-extraction procedures. Two authors jointly conducted the literature search and screened the retrieved records at both the title/abstract and full-text stages using the predefined eligibility criteria. Any disagreements or uncertainties were resolved through discussion and consensus; when uncertainty remained, a third author independently reassessed the relevant article. Data extraction was performed collaboratively by two authors rather than as independent duplicate extraction, and the completed dataset was subsequently verified by a third author against the source articles. We hope that these clarifications improve the transparency and methodological rigor of the review.

Comments 5

Discussion points (optional improvements):

It is desirable to provide the NOS scores for each study and the number of works by quality categories.

Response 5: We thank you for this helpful suggestion. A new table has been added to present the Newcastle–Ottawa Scale domain scores, total score, and corresponding quality classification for each included study. The number of studies within each quality category has also been summarized in the Results section.

 

Comments 6

The work would be improved if stratification by clinical context (localized vs. metastatic cancer) were added.

Response 6: We thank you for this valuable suggestion. Where such information was available in the source studies, we have added details regarding the clinical context and disease stage. These clarifications have been incorporated into the Results section, for example in Lines 182–184 and 195–196.

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

From both scientific and methodological perspectives, this systematic review remains structurally and evidentially fragile, a limitation that the authors themselves explicitly acknowledge in the Discussion and Conclusions.

For the vast majority of the malignancies evaluated—including prostate cancer, soft tissue sarcoma, renal cell carcinoma, pancreatic cancer, and ovarian cancer—the review is based on a single primary study for each tumor type. Consequently, the available evidence is insufficient to support robust or generalizable conclusions regarding the association under investigation.

Even for the only malignancy with a modest concentration of evidence, breast cancer, the findings remain heterogeneous and inconclusive, reflecting the current state of the literature. A more comprehensive examination of the available evidence readily reveals substantial inconsistency across studies: while some investigations report a positive association, others fail to demonstrate statistical significance, and still others describe discordant or even opposing findings. Such heterogeneity precludes the establishment of a consistent biological or epidemiological relationship and substantially limits the strength of any conclusions derived from the review.

Furthermore, there is a lack of a baseline of strong, clean evidence. The structural issues deeply affect the validity of the data; studies are retrospective observational cohorts. Retrospective studies are highly susceptible to selection bias, misclassification, and incomplete data capture, placing them lower on the hierarchy of scientific evidence, which are not irreparable if the authors may eventually change the route of the study, maybe to be more descriptive or epidemiological? The review fails to demonstrate that the primary studies effectively isolated tumor grade from these massive systemic confounders. The paper tries to bridge a massive crosstalk between a localized histopathological trait (tumor differentiation) and a highly volatile systemic marker (peripheral blood counts)... Again, NLR is a volatile, non-specific marker of general systemic stress and catalogues how that stress presents across different cancer types and populations.

Author Response

Comments 1:

From both scientific and methodological perspectives, this systematic review remains structurally and evidentially fragile, a limitation that the authors themselves explicitly acknowledge in the Discussion and Conclusions.

 

For the vast majority of the malignancies evaluated—including prostate cancer, soft tissue sarcoma, renal cell carcinoma, pancreatic cancer, and ovarian cancer—the review is based on a single primary study for each tumor type. Consequently, the available evidence is insufficient to support robust or generalizable conclusions regarding the association under investigation.

 

Even for the only malignancy with a modest concentration of evidence, breast cancer, the findings remain heterogeneous and inconclusive, reflecting the current state of the literature. A more comprehensive examination of the available evidence readily reveals substantial inconsistency across studies: while some investigations report a positive association, others fail to demonstrate statistical significance, and still others describe discordant or even opposing findings. Such heterogeneity precludes the establishment of a consistent biological or epidemiological relationship and substantially limits the strength of any conclusions derived from the review.

 

Furthermore, there is a lack of a baseline of strong, clean evidence. The structural issues deeply affect the validity of the data; studies are retrospective observational cohorts. Retrospective studies are highly susceptible to selection bias, misclassification, and incomplete data capture, placing them lower on the hierarchy of scientific evidence, which are not irreparable if the authors may eventually change the route of the study, maybe to be more descriptive or epidemiological? The review fails to demonstrate that the primary studies effectively isolated tumor grade from these massive systemic confounders. The paper tries to bridge a massive crosstalk between a localized histopathological trait (tumor differentiation) and a highly volatile systemic marker (peripheral blood counts)... Again, NLR is a volatile, non-specific marker of general systemic stress and catalogues how that stress presents across different cancer types and populations.

Response 1:

We sincerely thank the reviewer once again for the time devoted to evaluating our manuscript and for sharing these thoughtful comments. We fully acknowledge that the currently available evidence addressing the association between NLR and histological tumor grade is limited, heterogeneous, and predominantly based on retrospective observational studies. These limitations are explicitly recognized throughout our manuscript and are discussed in detail in both the Discussion and Conclusions.

 

We have intentionally presented the available evidence in a transparent manner. For tumor types represented by only a single eligible study, this has been clearly stated, and no attempt has been made to draw generalized conclusions. Likewise, studies with relatively small cohorts were specifically identified to allow readers to appropriately interpret the strength of the available evidence. We also explicitly highlighted the heterogeneity and, in some instances, contradictory findings, particularly for breast and colorectal cancer, where the available studies yielded inconsistent results.

 

We further agree that NLR is a non-specific and dynamic systemic inflammatory biomarker that may be influenced by numerous clinical and biological confounders. This important limitation is discussed extensively in the revised manuscript, together with the need to interpret NLR alongside established clinicopathological parameters rather than in isolation.

 

Nevertheless, we believe that the heterogeneity of the currently available evidence does not diminish the relevance of this systematic review. Rather, it underscores the existing knowledge gap and highlights the need for further well-designed prospective studies. Our intention was not to establish definitive conclusions, but to provide a structured and critical synthesis of the available literature, identify areas of agreement and disagreement, and generate hypotheses for future research. Given the inherent biological and clinical heterogeneity of oncology, we believe that systematically documenting these inconsistencies represents an important step toward a better understanding of the potential relationship between systemic inflammatory biomarkers and histological tumor differentiation.

 

This review also represents a continuation of our previous work published in Diagnostics, in which we systematically examined the relationship between C-reactive protein and tumor grading across malignant tumors. Taken together, these reviews form part of a broader research direction focused on the potential associations between systemic inflammatory biomarkers and histopathological tumor characteristics. We consider such evidence syntheses particularly important because they help define current knowledge gaps, identify methodological limitations, and provide a rationale for the development of prospective studies. In particular, this review will serve as a foundation for future prospective studies planned by our research group in colorectal cancer, aimed at further investigating the relationship between systemic inflammatory biomarkers and histopathological tumor characteristics.

 

Șiancu, P.; Oprinca, G.-C.; Vulcu, A.-C.; Pătran, M.; Croitoru, A.E.; Tănăsescu, D.; Bratu, D.; Boicean, A.; Tănăsescu, C. The Significance of C-Reactive Protein Value and Tumor Grading for Malignant Tumors: A Systematic Review. Diagnostics 2024, 14, 2073. https://doi.org/10.3390/diagnostics14182073

Round 2

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

The paper can be accepted after minor adjustments to references.

ref 28 is an ovarian malignancy paper, not breast cancer. 

ref 32 is Jadoon et al., an invasive breast cancer study, not bladder cancer.

Everything else appears internally consistent and well supported by the cited literature. These two corrections should be made before submission.

Comments on the Quality of English Language

ok

Author Response

Comments 1:

The paper can be accepted after minor adjustments to references.

ref 28 is an ovarian malignancy paper, not breast cancer. 

ref 32 is Jadoon et al., an invasive breast cancer study, not bladder cancer.

Everything else appears internally consistent and well supported by the cited literature. These two corrections should be made before submission.

 

Response 1:

We sincerely thank the reviewer for carefully identifying these two reference inconsistencies and bringing them to our attention. The indicated corrections have been made accordingly, and these revisions are highlighted in red in the manuscript.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

“The Association Between Neutrophil-to-Lymphocyte Ratio and Histological Tumor Differentiation in Solid Malignancies: A Narrative Review” is a systematic review analyzing the association between NLR and tumor grade in 14 articles from the literature. The research aim is good and worth investigating; however, the study requires major revision due to serious issues in the methodological part and in the presentation and discussion of the results. These issues should be resolved before considering this article for acceptance. I recommend major revision.

The authors should change the type of the article to a systematic review. The current manuscript is not a literature review as described. Additionally, replace all instances of “narrative review” with “systematic review” throughout the text, including in the title and line 25.

Introduction comments:

In the introduction, the authors accurately describe the dual roles of neutrophils in cancer (Lines 68-74). However, the discussion of T lymphocytes (Lines 75-80) does not address the dual roles of these cells in cancer progression. Specifically, while CD8+ T cell infiltration into the tumor microenvironment is associated with a favorable prognosis, regulatory T cells (Tregs) can suppress anti-tumor immunity and correlate with worse clinical outcomes. While the current statement is not incorrect, incorporating this specificity would meaningfully strengthen the introduction.

In the introduction (Line 96-100), the authors should specify what they mean by “a histological characterization of the tumor” since in this sentence the histological characterization is included in the diagnostic process. I understand that the authors intend to refer to the prognostic value of all the histological parameters described in the following paragraph (lines 101–109). However, in its current form, the sentence is misleading. It would be preferable to rephrase it to emphasize that histological characterization not only confirms the diagnosis but also provides insightful prognostic information based on the tumor architecture and all other morphological features.

It is not logical to first discuss the NLR, then the histological grade, and then return to the NLR again. Therefore, lines 110–113 should be integrated into the paragraph on NLR (lines 89–94). And then the authors are encouraged to explain why it is of interest to study the relationship between the NLR and histology.

Materials and Methods:

In the methods section the authors should write out the precise search line exactly as entered in the database, including all “AND / OR / NOT” and applied filters. The current description allows for multiple possible interpretations of the actual search string.

The search could be like this:

(neutrophil-to-lymphocyte ratio) AND ((tumor grading in malignant cance) OR (histopathological grade in malignant tumor) OR (tumor differentiation in malignant cancer))

or like this:

(“neutrophil-to-lymphocyte ratio”[MeSH] OR “neutrophil to lymphocyte ratio”[Title/Abstract]) AND (“tumor grading”[MeSH] OR “histopathological grade”[Title/Abstract] OR “tumor differentiation”[Title/Abstract])

The authors should be precise. Providing the exact syntax will ensure that the search is fully reproducible, which is essential for a systematic review.

However, if the authors used selected search terms with various modifications and combinations, this could be stated in the Methods section as: "Search terms included variations and combinations of …". Nevertheless, providing the exact search syntax remains preferable to ensure full reproducibility. If this is the case, this limitation should be acknowledged in the Limitations section.

It is recommended to include the research question before presenting the PICOS framework. Additionally, the inclusion and exclusion criteria for the analyzed studies should be described more clearly and comprehensively. It is also recommended to specify whether study selection was conducted independently by multiple researchers to minimize selection bias. If applicable, the authors should describe how disagreements between researchers were resolved (e.g., through discussion to reach consensus or consultation with a third researcher). Please see similar systematic reviews published in the same field for reference (https://doi.org/10.3390/jcm14196822 , doi: 10.1002/lio2.1161 ,  https://doi.org/10.1038/s41598-024-84890-3 , https://doi.org/10.1016/j.clineuro.2025.108996 , https://doi.org/10.3390/diseases11010051).

In the PICOS section, the description of C (“Comparison between NLR values and tumor differentiation grades”) is not appropriate. The C element should refer to groups or conditions being compared, not to two different variables. NLR and tumor differentiation grades are not “groups”; they are two variables whose association is being studied. The authors should first clearly define and state the study objective and then map it accurately to the PICOS framework. For example, C could be reformulated as: C: Comparison of NLR levels across different tumor differentiation grades, or, more specifically, C: Comparison between patients with low NLR and patients with high NLR across different tumor differentiation grades.

In Table 1, please add the corresponding reference number after each first author to facilitate assessment of the cited studies. It is recommended to provide the full description of cancer type for each study included (e.g., for De La Cruz-Ku et al., specify "metastatic triple-negative breast cancer" rather than just "breast cancer"). Please add Newcastle–Ottawa Scale result for every study and gender of patients (Male/Female n/n). Additionally, please clarify the timing of NLR measurement for each study: was it pre-treatment NLR, post-treatment, or N/A…..?

Major comment: Systematic reviews typically include only primary/original studies. The authors have included a systematic review (Tang et al., reference 23), which may represent an error in study selection. The authors should either provide a clear justification for including this secondary study or exclude this study from the analysis. Please clarify the rationale for this inclusion or remove it to maintain methodological consistency.

 

Results:

Visualizing the results would strengthen this section. Consider adding an effect direction plot, a harvest plot, or another summarizing figure as a visual synthesis. A plot would complement the results nicely.

Lines 176 and 231–232: Please clarify what “marginal” means statistically (specify the exact p-value or define the p-value range used). It is recommended to provide precise statistical information. This comment is also about Table 2.

Line 188: Please quantify “weak” by reporting the correlation coefficient (r) or effect size alongside the p-value. Descriptive terms like "weak" should be supported by specific statistical values.

Discussion:

This section needs major revision. There is a loss of focus. I suggest rewriting the discussion starting with the key findings of the systematic review, then comparing with the literature and other systematic reviews linking NLR with tumor grade or general prognostic factors in cancer. Following that, add suggested mechanisms and biological explanations for why NLR might be associated with tumor grade. Then add clinical implications and finally discuss the limitations in detail. In the limitations section, it would be a good idea to add the point that the number of studies analyzed was relatively small.

Line 239: The phrase "non-significant association" is ambiguous. I suggest combining this study with the two other studies that reported no association.

It is not appropriate to include histological figures in the discussion section of a systematic review. However, where are these figures from? If they are taken from other articles or books, the authors must obtain permission and add a reference under each figure. If the authors prepared these figures themselves, they should describe how they were created in the Methods section. Nevertheless, I recommend removing them altogether.

Minor comments:

• Terminology suggestion: instead of “degree of tumor differentiation” or “Histological Tumor Differentiation” consider using “tumor grade”.
• Line 227: “The results of two studies indicate a possible association between the report and tumor grading.” Please correct “the report” to “the NLR”, as this appears to be a typographical error.

 

Good luck!

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript addresses a relevant and clinically significant topic: the relationship between the available inflammatory biomarker NLR and the histological grade of tumor malignancy. The authors collected data from 14 studies involving 8,611 patients.

However, in its current form, the work suffers from methodological inconsistencies, incomplete data presentation, and conceptual overload.

Major comments:

1. I see a critical discrepancy between the stated article type and the methodology. The manuscript is labeled as a narrative review, yet the authors use the PICOS strategy and a PRISMA-like flow chart. This is methodologically incorrect. The manuscript should either be redesigned as a systematic review or the PICOS, PRISMA flow chart, and description of the systematic article selection should be removed. In the former case, it is also recommended to elaborate the description of the article selection criteria, in line with the guidelines, and perform a meta-analysis.
2. Much of the "Results" section essentially replicates Table 1, without any original thought. This section needs to be rewritten and expanded with more analysis.
3. The Discussion section is confused by large sections of text describing breast cancer immunohistochemistry. This is redundant and unnecessary. However, Figures 1-4 are particularly perplexing. Where did they come from? Did the authors create them themselves? Then where is their description in the Materials and Methods? If this is taken from other articles, why are there no references? The authors should provide a detailed explanation for the presence of this text and figures in the article.
4. According to the journal profile, information about the number of articles, patients, etc. should be added to the abstract. Please check.
5. In the Methods (at least as a footnote in the Supplementary), a detailed description of the Newcastle-Ottawa Scale (NOS) quality assessment should be added.
6. The Limitations chapter is sorely lacking. It could have been expanded to include why only English-language articles were included, all the questions about the differences in data processing between articles, and the issues with NLR thresholds.

Minor comments

1. There is a funny discrepancy in the dates - the headers and footers show 2025, although the database queries show 2026.
2. The authors should explain the need for Figure 5.
3. Authors should provide an explanation of the term Marginal in Tables 1 and 2. Preferably in the Materials and Methods, Statistics section.

Comments on the Quality of English Language

As far as I can tell, the quality of the English language is very good.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript addresses a potentially relevant topic: the association between neutrophil-to-lymphocyte ratio (NLR) and histological tumor differentiation across solid malignancies. The topic is clinically interesting because NLR is inexpensive, widely available, and has been extensively investigated as a systemic inflammatory biomarker in oncology. However, in my opinion, the manuscript has substantial methodological and conceptual limitations that preclude publication in its current form.

 

Although the authors describe the article as a narrative review, the manuscript uses elements of a systematic review methodology, including database searches, PICOS framing, PRISMA-inspired study selection, and Newcastle–Ottawa Scale assessment. However, the methodological reporting does not meet the standards expected for a systematic review, while the interpretive structure is also not sufficiently rigorous for a high-quality narrative review. Therefore, the manuscript remains methodologically unclear and insufficiently robust.

 

Major comments

 

1. The review type and methodology are not clearly defined

 

The authors describe the manuscript as a narrative review, but the Methods section includes a structured database search, PICOS strategy, PRISMA-inspired selection flow, predefined inclusion/exclusion criteria, and NOS quality assessment. This creates ambiguity regarding whether the article is intended to be a narrative review or a systematic review. The Methods section states that PubMed and Web of Science were searched between February 4, 2026, and March 28, 2026, and that 14 retrospective studies were included after screening 788 records.

 

If the manuscript is intended to be a systematic review, the authors should provide a complete reproducible search strategy, detailed Boolean strings, date limits, screening procedures, reasons for exclusion at full-text review, and ideally protocol registration. If it is intended to be a narrative review, the use of PRISMA/PICOS/NOS language should be reduced or clearly justified. In its current form, the manuscript falls between these two formats.

 

2. Study eligibility appears inconsistent

 

The inclusion criteria state that clinical studies evaluating NLR and tumor grading were included, while book chapters, editorials, letters, commentaries, and conference abstracts were excluded. However, at least one included study, Tang et al. 2017 in biliary tract cancer, appears to be a systematic review and meta-analysis rather than a primary clinical cohort. This is problematic because it may duplicate primary data and is not methodologically comparable with the other included retrospective studies.

 

The authors should verify whether each included reference directly reports an association between NLR and histological differentiation/grade in primary patient-level data. Studies primarily evaluating prognosis, survival, recurrence, or treatment outcomes should not be included unless grade-specific NLR analyses are explicitly available.

 

3. The manuscript does not sufficiently distinguish tumor grade from prognosis

 

A central conceptual issue is that NLR has been extensively studied as a prognostic marker, but this review specifically claims to evaluate its relationship with histological differentiation. Several included studies appear to focus mainly on prognosis or clinical outcomes rather than grade as the primary endpoint. This distinction is critical. A biomarker associated with survival or recurrence is not necessarily associated with histological differentiation.

 

For example, some cited breast cancer and colorectal cancer studies appear to focus on disease outcomes, mortality, or prognosis, while grade-related analyses may be secondary or incomplete. This weakens the validity of the overall synthesis.

 

4. Heterogeneity is acknowledged but not analytically handled

 

The manuscript includes highly heterogeneous tumor types, grading systems, disease stages, NLR cutoffs, and clinical settings. The included studies use different grade definitions, including Gleason score, Nottingham/SBR grade, low- versus high-grade classifications, and well/moderate/poor differentiation categories. The authors summarize these across cancer types without a clear harmonization framework.

 

The manuscript states that 14 retrospective studies including 8611 patients were included, but there is no structured synthesis separating adjusted from unadjusted findings, tumor-specific evidence strength, or cutoff derivation methods. A simple “significant / not significant / inconsistent” summary is insufficient for such heterogeneous evidence.

 

The authors should add a detailed evidence table including:

 

tumor type and stage distribution

timing of NLR measurement

NLR analyzed as continuous or categorical variable

cutoff derivation method

histological grading system

univariate versus multivariable findings

covariates included in adjusted models

exclusion of infection, steroid use, hematologic disease, or treatment-related confounders

 

Without this information, the conclusion that NLR may reflect tumor differentiation remains speculative.

 

5. Risk-of-bias assessment is insufficiently reported

 

The authors state that all included studies had high methodological quality with NOS scores of at least 7. However, no itemized NOS table is provided. The reader cannot verify which domains were judged as low or high risk of bias. This is particularly important because all included studies were retrospective, and biomarker studies of NLR are highly vulnerable to selection bias, confounding, and measurement heterogeneity.

 

The authors should provide a complete per-study NOS table or consider a more appropriate tool such as ROBINS-I for nonrandomized biomarker-exposure studies.

 

6. Confounding is a major unresolved limitation

 

NLR is a non-specific systemic inflammatory marker. The manuscript itself acknowledges that NLR can be affected by age, obesity, infection, diabetes, emotional stress, steroid exposure, hematologic disorders, cytotoxic therapy, and G-CSF use. However, these confounders are not systematically evaluated across included studies.

 

This is a major limitation. Associations between high NLR and poor differentiation may be confounded by advanced stage, tumor burden, infection, treatment exposure, comorbidity, or systemic inflammation unrelated to tumor grade. The authors should clarify whether included studies excluded acute infection or steroid/G-CSF exposure and whether multivariable models adjusted for key clinical factors.

 

7. The figures are not well aligned with the main research question

 

Figures 2–4 show representative immunohistochemical profiles of breast cancer subtypes. These figures may be educational, but they are not directly relevant to the central question of whether peripheral NLR is associated with histological tumor differentiation. They distract from the main focus of the review. I recommend removing these figures or moving them to supplementary material, unless the authors can clearly explain how they contribute to the NLR–grade relationship.

 

8. The conclusions are somewhat overstated

 

The conclusion states that NLR has the potential to reflect tumor aggressiveness and may contribute to clinical decision-making. Given the retrospective nature of all included studies, heterogeneous tumor types, inconsistent findings, variable cutoffs, and unresolved confounding, this statement should be substantially softened.

 

A more appropriate conclusion would be that the available evidence is preliminary, inconsistent, and insufficient to support the use of NLR as a reliable marker of histological differentiation. NLR may warrant further prospective tumor-specific evaluation, but it should not currently be interpreted as a surrogate for tumor grade.

 

Minor comments

The search strategy should be rewritten using exact reproducible Boolean terms. The current wording is too general and does not allow replication.

The PRISMA flow diagram should include reasons for exclusion at full-text review.

Table 1 should be revised for consistency, including uniform terminology for NLR cutoffs, grading systems, p-values, and direction of association.

The term “marginal” statistical significance should be avoided unless a precise p-value and prespecified interpretation are provided.

The manuscript would benefit from language editing to improve clarity and reduce repetitive statements.

The authors should avoid presenting prognostic NLR literature as if it directly supports an association with histological differentiation.

The limitations section is too brief and should explicitly discuss retrospective design, confounding, lack of standardized cutoffs, tumor-type heterogeneity, and publication bias.

Recommendation

 

Despite the clinical relevance of the topic, the current manuscript has substantial methodological limitations, inconsistent eligibility criteria, insufficient risk-of-bias reporting, inadequate handling of heterogeneity, and conclusions that exceed the strength of the evidence. Therefore, I do not recommend publication in its current form.

 

A substantially redesigned review with a clearly defined systematic methodology, stricter eligibility criteria, transparent quality assessment, and tumor-specific structured synthesis would be required before the topic could be reconsidered.