Identification of Key Molecular Pathways and Associated Genes as Targets to Overcome Radiotherapy Resistance Using a Combination of Radiotherapy and Immunotherapy in Glioma Patients

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Article " Identification of key molecular pathways and associated genes as targets to overcome radiotherapy resistance using a combination of radiotherapy plus immunotherapy in glioma patients " Authors: Tianqi Zhang and at. all.  This is study that addresses a clinically relevant issue in problem in cancer therapy: overcoming resistance to the treatment of glioma using combination therapy methods - radiation and immunotherapy. The authors of the work pay special attention to studying the mechanisms of changes in the microenvironment after radiation therapy. The work is based on studying the molecular pathways of inhibition of the colony-stimulating factor-1 receptor (CSF-1R) in combination with radiation therapy. The article corresponds to the subject of the journal "Int. J. Mol. Sci.". The article is well structured, written in a clear and understandable language, the conclusions are logical, the literature corresponds to the stated topic. Interesting article. There are no critical comments about the work. I ask the authors to correct the typo in the title of the “Discussion” and “Conclusion” sections. I recommend publishing your work

Author Response

Response to Reviewer #1:

Comment 1. I ask the authors to correct the typo in the title of the “Discussion” and “Conclusion” sections.

Response to Comment 1: Thank you for the comment. We have thoroughly revised the manuscript and corrected all typos including those in the “Discussion” and “Conclusion”.

Reviewer 2 Report

Comments and Suggestions for Authors

Strengths:

- The article raises the extremely important problem of resistance to radiotherapy in patients with glioma and looks for solutions in the form of combined treatment with radiotherapy and immunotherapy. This is certainly a key challenge in the treatment of gliomas.

- The authors thoughtfully use data from mouse preclinical studies and translationally transfer the identified molecular pathways and genes to human patient cohorts. This is an interesting research strategy combining research on animal models and humans.

- The statistical and bioinformatic methods used are correct and well selected. In particular, meaningful Cox models were constructed, and their accuracy was assessed using appropriate measures.

 

 Weaknesses:

- The main disadvantage is the small sample size, both in studies on mice and in human cohorts, which limits the reliability of the results obtained.

- The authors focus on macrophages, ignoring their communication with cancer cells, which plays an important role in the progression of gliomas.

- This study focused only on low-grade gliomas, while the mechanisms of radiotherapy resistance may differ between high-grade gliomas (GBMs).

 

Conclusions:

Despite some limitations, this article describes an interesting attempt to translate results from animal models to patients to identify the mechanisms of resistance to radiotherapy and the synergistic effects of radiotherapy with immunotherapy. The presented research strategy using data from preclinical studies in animal models and patient cohorts may constitute a promising development path for future translational oncology. The identified signalling pathways and genes represent potential targets for immunotherapy in combination with radiotherapy. The study results provide grounds for planning future clinical trials of combination therapies for radiotherapy-resistant gliomas. Despite its limitations, this study presents a promising direction for further work to overcome radiotherapy resistance in gliomas through the synergistic effects of radiotherapy and immunotherapy. Its methodology and results can be used in the design of future advanced therapies in oncology.

Author Response

Response to Reviewer #2:

Comment 1. The main disadvantage is the small sample size, both in studies on mice and in human cohorts, which limits the reliability of the results obtained.

Response to Comment 1: As we acknowledged in the Discussion section of the manuscript, small sample size is indeed a limitation of this study. However, the animal study dataset that we used is from a well-designed randomized mouse study by Akkari L, et al. (2020, Science Translational Medicine). In the animal study, they measured dynamic changes of the proportion of different macrophages as well as RNA-seq expression levels in murine glioma macrophages. The study team has successfully used similar study designs in their earlier murine glioma studies including Pyontek et al (2013, Nature Medicine) and Quail et al (2016, Science). Despite the limitation on sample size, the mouse sample size is adequate to reliably identify key molecular pathways.  Also, in the translational part of our study, although the dimension of all candidate DEGs is high, we essentially first conducted dimension reduction by focusing on DEGs in the three selected pathways where the number of DEGs is limited. We further used LASSO-based dimension reduction in Cox regression analysis when needed. After dimension reduction for the candidate DEGs, the sample size of the training TCGA data set (n=295 RT-treated LGG patients) provided adequate statistical power to robustly identify genetic signatures and evaluate 

them using an extra independent cohort of RT-treated LGG patients from CGGA. It is true that the evaluation of the predictive performance of the identified signature genes in the non-irradiated cohorts is exploratory because the sample size seems to be limited. Nevertheless, the results for the non-irradiated cohorts are as expected given our knowledge from the existing literature.

Comment 2. The authors focus on macrophages, ignoring their communication with cancer cells, which plays an important role in the progression of gliomas.

Response to Comment 2: Yes, we have previously established that the crosstalk between tumor cells (TCs) and tumor-associated macrophages (TAMs) can be used to further improve predictive power of the genetic signature when interactions are considered in the prognostic models [He X, Sun X, Shao Y. 2023. IEEE Journal of Biomedical and Health Informatics.] In the current study, we only have RNA-seq gene expression data of the mouse macrophages, no data from mouse tumors available to us from the Akkari et al (2020) study. We have acknowledged this as a limitation in the Discussion section. It is indeed more reasonable to consider the interaction between macrophages and tumor cells when data are available.

Comment 3. This study focused only on low-grade gliomas, while the mechanisms of radiotherapy resistance may differ between high-grade gliomas (GBMs).

Response to Comment 3: More than half of the LGG patients can evolve and progress to GBMs. Thus, developing effective therapies for LGG patients is crucial for improving their survival and preventing progression from LGGs to GBMs. Advancements in understanding the biology of these LGGs and identifying targeted therapies can significantly enhance treatment outcomes and quality of life for LGG patients. Moreover, our translational approach can be easily applied to GBM patients, as pointed out by the reviewer, we also expect that the mechanisms of radiotherapy resistance may differ between LGG and GBMs based on findings of a newly added reference [Lu Y and Shao Y 2022, Cancer Drug Resistance]. GBM patients are very heterogeneous with four subtypes. Different subtypes of GBM patients can have very different responses to treatments. Due to the small overall sample size for GBM patients in TCGA, the number of GBM patients in each subtype is inadequate for robust statistical modeling. Moreover, very few GBM subjects with radiation therapy data, therefore we did not build a predictive model for the GBMs. Future studies might focus on GBMs when larger sample sizes are available. We have mentioned these points in the Discussion section.

There are extra improvements made in this revision beyond the requests of the reviewers. (a) We went through the whole paper carefully and improved quality of all figures and their legends and corrected typos and texts lack of clarity. (b) We newly added an additional reference (Reference #1) on the functions of CSF-1R which was accidentally omitted in the submitted version. (c) We also newly added Figure S1 on the Kaplan-Meier plots of all the key genes identified in the paper for easy references of general readers. We believe that these revisions have further strengthened the quality and clarity of our manuscript.