Integrative Profiling of Tumor and Blood Microenvironments to Uncover Molecular and Immune Determinants of Prognosis and Treatment Efficacy in Metastatic Colorectal Cancer

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This review provides a comprehensive overview of emerging tissue- and blood-based biomarkers in metastatic colorectal cancer (mCRC), with a particular focus on immunological biomarkers and liquid biopsy approaches. The manuscript is generally well organized, timely, and clinically relevant. The figures and summary tables are useful and facilitate comprehension of a rapidly evolving field. Nevertheless, several important aspects of mCRC biology and biomarker stratification remain underdeveloped. In particular, the genomic section overlooks the growing evidence supporting molecular heterogeneity according to primary tumor location and colon segment-specific mutational profiles. Furthermore, some sections present biomarkers in a predominantly descriptive manner without sufficiently discussing their limitations, validation status, and clinical applicability.

Therefore, I recommend Major Revision before the manuscript can be considered for publication.

MAJOR COMMENTS

1. One of the major limitations of the review is the lack of discussion regarding the biological and genomic differences among colorectal cancers arising from different colon segments. Accumulating evidence demonstrates that right-sided and left-sided colorectal cancers exhibit distinct mutational landscapes, molecular evolution patterns, therapeutic sensitivities, and prognostic outcomes. Importantly, heterogeneity exists even within individual colon segments and cannot be fully captured by the traditional right-versus-left classification. The authors should discuss the importance of considering segment-specific biology when evaluating predictive and prognostic biomarkers (e.g., Oliveira DM, Laudanna C, Migliozzi S, Zoppoli P, Santamaria G, Grillone K, et al. Identification of different mutational profiles in cancers arising in specific colon segments by next generation sequencing. Oncotarget. 2018;9:23960–23974). Such information would substantially strengthen Sections 2.1 and 2.2.

2. Although the review discusses SMAD4, POLE/POLD1, TMB, and CIMP, it surprisingly provides limited attention to several clinically relevant molecular determinant, APC and TP53, alterations, consensus patterns of co-mutation (RAS/TP53, RAS/SMAD4, BRAF-related signatures), Copy-number alterations. These alterations may not yet represent routine predictive biomarkers, but they significantly influence prognosis, tumor evolution, metastatic dissemination, and therapeutic response. The authors should either expand the genomic section or explicitly justify why these biomarkers were excluded.

3. The review heavily focuses on immune-related biomarkers, whereas metabolic, stromal, and microbiome-related biomarkers receive comparatively limited discussion.

Given the title ("From Tumor to Blood: Decoding Treatment Response in mCRC"), The authors should better discuss: metabolomic signatures, stromal-derived biomarkers,integrated multi-omics approaches.

4. The review currently presents biomarkers with varying levels of evidence together. The manuscript would benefit from a structured classification according to: validated clinical biomarkers, biomarkers with prospective validation, exploratory biomarkers, preclinical biomarkers. This distinction would greatly improve clinical readability.

Minor Comments

1. The manuscript alternates between MSS/pMMR and  MSI-H/dMMR. The authors should ensure consistent terminology throughout the text.

2. Clarify the role of TMB in MSS mCRC. The discussion on TMB could be interpreted as suggesting a clinically meaningful role of TMB alone in MSS tumors. The authors should more clearly emphasize that TMB is insufficient as a standalone biomarker in MSS mCRC, POLE/POLD1 mutations largely drive the observed responses.

3. The microbiome section is interesting but should discuss lack of methodological standardization, sampling variability, dietary confounders, reproducibility concerns.

5. Several minor grammatical and stylistic issues should be corrected during revision. Examples include: i) occasional inconsistent use of abbreviations, ii) variable punctuation around hazard ratios, iii) minor formatting inconsistencies within Tables 1 and 2. A careful language revision is recommended.

Author Response

MAJOR COMMENTS

We thank the reviewers for their insightful comments and the time spent to improve our manuscript. Below their comments are reported verbatim, followed by our proposed modifications.

1. One of the major limitations of the review is the lack of discussion regarding the biological and genomic differences among colorectal cancers arising from different colon segments. Accumulating evidence demonstrates that right-sided and left-sided colorectal cancers exhibit distinct mutational landscapes, molecular evolution patterns, therapeutic sensitivities, and prognostic outcomes. Importantly, heterogeneity exists even within individual colon segments and cannot be fully captured by the traditional right-versus-left classification. The authors should discuss the importance of considering segment-specific biology when evaluating predictive and prognostic biomarkers (e.g., Oliveira DM, Laudanna C, Migliozzi S, Zoppoli P, Santamaria G, Grillone K, et al. Identification of different mutational profiles in cancers arising in specific colon segments by next generation sequencing. Oncotarget. 2018;9:23960–23974). Such information would substantially strengthen Sections 2.1 and 2.2.

We agree that colorectal cancers arising from different colon segments exhibit distinct biological and genomic characteristics that may influence treatment response. We have now incorporated this aspect into the manuscript (lines 90-93).

2. Although the review discusses SMAD4, POLE/POLD1, TMB, and CIMP, it surprisingly provides limited attention to several clinically relevant molecular determinant, APC and TP53, alterations, consensus patterns of co-mutation (RAS/TP53, RAS/SMAD4, BRAF-related signatures), Copy-number alterations. These alterations may not yet represent routine predictive biomarkers, but they significantly influence prognosis, tumor evolution, metastatic dissemination, and therapeutic response. The authors should either expand the genomic section or explicitly justify why these biomarkers were excluded.

We have expanded the Introduction to briefly describe the major molecular events underlying CRC development. However, as the primary aim of this review is to summarize biomarkers with established or emerging clinical relevance for prognosis and treatment response in metastatic colorectal cancer, we chose to focus the genomic section on biomarkers that are currently evaluated or implemented in clinical studies and therapeutic decision-making. Since alterations such as APC and TP53 are not routinely used as predictive or prognostic biomarkers in clinical practice, they were not discussed in detail (lines 58-64). We also justified our selection of papers by adding a small methodology section (lines 122-127).

3. The review heavily focuses on immune-related biomarkers, whereas metabolic, stromal, and microbiome-related biomarkers receive comparatively limited discussion.

Given the title ("From Tumor to Blood: Decoding Treatment Response in mCRC"), The authors should better discuss: metabolomic signatures, stromal-derived biomarkers, integrated multi-omics approaches.

As suggested also by the two other reviewers, we have revised the title to better reflect the primary focus of the review.

The emphasis on immune-related biomarkers was chosen because this field currently has a broader body of evidence and a stronger translational framework, with several biomarkers being evaluated in clinical studies and demonstrating potential relevance for patient stratification. Nevertheless, we considered it important to introduce other biological dimensions, including metabolic, stromal, and microbiome-related alterations, to highlight the multifactorial complexity of tumor biology and the need for integrated approaches in understanding treatment response. We have clarified this rationale in the revised manuscript (lines 118-127).

We have also added a section on liver-specific immune tolerance, which helps explain the strong emphasis in the literature on immune-related biomarkers (lines 69-74 in the Introduction and lines 606-609 in the Conclusion).  

4. The review currently presents biomarkers with varying levels of evidence together. The manuscript would benefit from a structured classification according to: validated clinical biomarkers, biomarkers with prospective validation, exploratory biomarkers, preclinical biomarkers. This distinction would greatly improve clinical readability.

In the revised manuscript, biomarkers have already been structured according to their translational status, with clinically validated biomarkers discussed separately from more exploratory biomarkers. In addition, the prognostic and predictive value of each biomarker is systematically summarized in Tables 1 and 2. We chose to provide this information in a tabulated format to improve clarity, avoid unnecessary repetition throughout the text, and allow readers to more easily compare the available evidence (pages 14 and 17).

Minor Comments

1. The manuscript alternates between MSS/pMMR and MSI-H/dMMR. The authors should ensure consistent terminology throughout the text. We corrected this.
2. Clarify the role of TMB in MSS mCRC. The discussion on TMB could be interpreted as suggesting a clinically meaningful role of TMB alone in MSS tumors. The authors should more clearly emphasize that TMB is insufficient as a standalone biomarker in MSS mCRC, POLE/POLD1 mutations largely drive the observed responses.

We have revised this section to emphasize that TMB alone is insufficient as a standalone biomarker for predicting response to immune checkpoint inhibitors in MSS mCRC, and that the clinical benefit observed in this setting is largely driven by specific molecular contexts, particularly POLE/POLD1 mutations (lines 159-163 and 169-171).

3. The microbiome section is interesting but should discuss lack of methodological standardization, sampling variability, dietary confounders, reproducibility concerns.

We have expanded the microbiome section to include the impact of methodological variability, including differences in sampling procedures, technical approaches, dietary and clinical confounders, as well as concerns regarding reproducibility and cross-cohort validation (lines 408-416).

5. Several minor grammatical and stylistic issues should be corrected during revision. Examples include: i) occasional inconsistent use of abbreviations, ii) variable punctuation around hazard ratios, iii) minor formatting inconsistencies within Tables 1 and 2. A careful language revision is recommended. We corrected this.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have presented a review of the present knowledge regarding various biomarkers that can predict response to therapy in colorectal cancer. The article is well written and provides a synopsis of a wide body of literature on the topic.

Some minor concerns:

The title of the manuscript " From Tumor to Blood: Decoding Treatment Response in mCRC?" or the abstract  "This review summarizes the most promising biomarkers currently being investigated in metastatic colorectal cancer.....", appear to indicate that biomarkers for all therapies for CRC will be discussed. However, the manuscript predominantly discusses biomarkers involved in response to immunotherapy only.  It is suggested that the title and abstract be modified to reflect the content of the manuscript.

Switching paragraph (lines 609-617) and paragraph (618-623) may improve  flow of discussion.

Including a statement/discussion on how all these multimodal analysis (transcriptomics, proteomics, metabolomics etc) can realistically be implemented to study biomarkers in treatment response (e.g., how cost associated with these assays will be met, time for analyses of large volumes of data etc.) will likely help link expectations with reality.

Font sizes for Table 1 and Table 2 need to be revised as it is very hard to read the content of these tables.

Author Response

We thank the reviewers for their insightful comments and the time spent to improve our manuscript. Below their comments are reported verbatim, followed by our proposed modifications.

Some minor concerns:

1. The title of the manuscript " From Tumor to Blood: Decoding Treatment Response in mCRC?" or the abstract "This review summarizes the most promising biomarkers currently being investigated in metastatic colorectal cancer...", appear to indicate that biomarkers for all therapies for CRC will be discussed. However, the manuscript predominantly discusses biomarkers involved in response to immunotherapy only.  It is suggested that the title and abstract be modified to reflect the content of the manuscript.

As suggested also by the two other reviewers, we have modified the title of this review.

We agree that the previous title and abstract could have suggested a broader discussion of biomarkers across all therapeutic strategies in mCRC, whereas the focus of this review is on immune-related biomarkers and their potential role in predicting prognosis and treatment response, particularly in the context of immunotherapy. Accordingly, we have revised the title and removed the lipidomics and metabolomics from the abstract to better reflect the scope and focus of the manuscript.

2. Switching paragraph (lines 609-617) and paragraph (618-623) may improve flow of discussion. We corrected this.
3. Including a statement/discussion on how all these multimodal analysis (transcriptomics, proteomics, metabolomics etc) can realistically be implemented to study biomarkers in treatment response (e.g., how cost associated with these assays will be met, time for analyses of large volumes of data etc.) will likely help link expectations with reality.

We have expanded this section to address key considerations, including the cost and scalability of omics-based assays, the complexity and volume of generated datasets, and the need for advanced bioinformatic pipelines and computational expertise for data integration and interpretation. These additions highlight the importance of multidisciplinary approaches combining experimental technologies with robust computational analyses to enable the identification and validation of clinically relevant biomarkers (in the Limits, challenges and perspectives section: lines 639-664; 675-678; 687-707 and in the Conclusion: lines 717-738).

4. Font sizes for Table 1 and Table 2 need to be revised as it is very hard to read the content of these tables. We corrected this.

Reviewer 3 Report

Comments and Suggestions for Authors

In this manuscript, Benidovskaya et al. provide a narrative review of emerging predictive and prognostic biomarkers in metastatic colorectal cancer. The review is comprehensive and is timely, the tissue-to-circulating structure is good, and the summary tables are scientifically useful.
However, there are several areas that need improvement. Below are my observations:
1. There is no description of how the literature was screened for this analysis. Even for a narrative review, it will help if the time scope and rationale will be added to prevent any bias. Please add a short description or  paragraph describing how candidate biomarkers and supporting studies were identified.
2. The review lacks clear synthesis. Given the "from tumor to blood" framing, the manuscript would be substantially stronger if it is connecting tissue and circulating markers such as concordance, or complementarity, and prioritized the clinical implementation. The authors can improve this in their revision.
4. The artificial-intelligence and multiomics perspective is not specifically discussed in depth.  Please concentrate more on CRC-specific work.
5. The authors can refine the title and abstract to more closely reflect the actual content of the manuscript for example: title centers on "treatment response," whereas a large portion of the content concerns prognostic markers

Author Response

We thank the reviewers for their insightful comments and the time spent to improve our manuscript. Below their comments are reported verbatim, followed by our proposed modifications.

1. There is no description of how the literature was screened for this analysis. Even for a narrative review, it will help if the time scope and rationale will be added to prevent any bias. Please add a short description or paragraph describing how candidate biomarkers and supporting studies were identified.

We added a paragraph in the Introduction describing the database used (PubMed), the time frame, the general search strategy, and the criteria used to identify candidate biomarkers and relevant supporting studies. (lines 122-127)

2. The review lacks clear synthesis. Given the "from tumor to blood" framing, the manuscript would be substantially stronger if it is connecting tissue and circulating markers such as concordance, or complementarity, and prioritized the clinical implementation. The authors can improve this in their revision.

To the best of our knowledge, relatively few clinical studies have evaluated both tissue and circulating immune biomarkers within the same patient cohorts. Most published studies focus on either the tumor immune microenvironment or circulating immune markers independently, making direct assessment of concordance, complementarity, or their combined clinical utility challenging. We have revised the manuscript to better emphasize this limitation of the current literature and to highlight the need for future studies integrating both compartments to facilitate clinical implementation (lines 669-672).

3. /
4. The artificial-intelligence and multi-omics perspective is not specifically discussed in depth.  Please concentrate more on CRC-specific work.

We have expanded this section by including additional CRC-specific examples and providing further discussion of their potential role in biomarker discovery and clinical translation (lines 687-707).

5. The authors can refine the title and abstract to more closely reflect the actual content of the manuscript for example: title centers on "treatment response," whereas a large portion of the content concerns prognostic markers. As suggested also by the two other reviewers, the title has been refined.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have carefully revised the manuscript according to my suggestion. However, my first suggestion: 

"One of the major limitations of the review is the lack of discussion regarding the biological and genomic differences among colorectal cancers arising from different colon segments. Accumulating evidence demonstrates that right-sided and left-sided colorectal cancers exhibit distinct mutational landscapes, molecular evolution patterns, therapeutic sensitivities, and prognostic outcomes. Importantly, heterogeneity exists even within individual colon segments and cannot be fully captured by the traditional right-versus-left classification. The authors should discuss the importance of considering segment-specific biology when evaluating predictive and prognostic biomarkers (e.g., Oliveira DM, Laudanna C, Migliozzi S, Zoppoli P, Santamaria G, Grillone K, et al. Identification of different mutational profiles in cancers arising in specific colon segments by next generation sequencing. Oncotarget. 2018;9:23960–23974). Such information would substantially strengthen Sections 2.1 and 2.2."

was insufficiently considered by authors who write three lines that do not discuss the point:

"Primary tumor sidedness is a major prognostic and predictive factor guiding treatment decisions. RAS-WT patients with left-sided tumors tend to have a better prognosis and derive greater benefit from anti-EGFR therapy, whereas right-sided tumors are associated with worse outcomes and reduced anti-EGFR efficacy"

Author Response

Reviewer's comment [One of the major limitations of the review is the lack of discussion regarding the biological and genomic differences among colorectal cancers arising from different colon segments. Accumulating evidence demonstrates that right-sided and left-sided colorectal cancers exhibit distinct mutational landscapes, molecular evolution patterns, therapeutic sensitivities, and prognostic outcomes. Importantly, heterogeneity exists even within individual colon segments and cannot be fully captured by the traditional right-versus-left classification. The authors should discuss the importance of considering segment-specific biology when evaluating predictive and prognostic biomarkers (e.g., Oliveira DM, Laudanna C, Migliozzi S, Zoppoli P, Santamaria G, Grillone K, et al. Identification of different mutational profiles in cancers arising in specific colon segments by next generation sequencing. Oncotarget. 2018;9:23960–23974). Such information would substantially strengthen Sections 2.1 and 2.2. This point was insufficiently considered in the previous revision, in which the authors added only three lines that did not discuss it.]

Response: We agree that our initial revision did not adequately address this point, and we thank the reviewer for their patience in re-raising it. We have now added the dedicated subsection described above (Section 2.1.5), which goes beyond a brief mention of tumor sidedness to explicitly discuss segment-level molecular heterogeneity, citing Oliveira et al. as recommended.

Change made: As above (new Section 2.1.5).