Method for the Intraoperative Detection of IDH Mutation in Gliomas with Differential Mobility Spectrometry

Round 1

Reviewer 1 Report

Dear Editor, thank you so much for inviting me to revise this manuscript.

This study addresses a current topic.

The manuscript is quite well written and organized. English could be improved.

Figures and tables are comprehensive and clear.

The introduction explains in a clear and coherent manner the background of this study.

We suggest the following modifications:

• Introduction section: although the authors correctly included important papers in this setting, we believe some studies regarding IDH inhibitors and IDH in other solid tumors should be cited within the introduction ( PMID: 32487595 ; PMID: 33799004; PMID: 33592561 ), only for a matter of consistency. We think it might be useful to introduce the topic of this interesting study.
• Methods and Statistical Analysis: nothing to add.
• Discussion section: Very interesting and timely discussion. Of note, the authors should expand the Discussion section, including a more personal perspective to reflect on. For example, they could answer the following questions – in order to facilitate the understanding of this complex topic to readers: what potential does this study hold? What are the knowledge gaps and how do researchers tackle them? How do you see this area unfolding in the next 5 years? We think it would be extremely interesting for the readers.

However, we think the authors should be acknowledged for their work. In fact, they correctly addressed an important topic, the methods sound good and their discussion is well balanced.

One additional little flaw: the authors could better explain the limitations of their work, in the last part of the Discussion.

We believe this article is suitable for publication in the journal although some revisions are needed. The main strengths of this paper are that it addresses an interesting and very timely question and provides a clear answer, with some limitations.

We suggest a linguistic revision and the addition of some references for a matter of consistency. Moreover, the authors should better clarify some points.

Author Response

Please see the attachment.

This study addresses a current topic.

The manuscript is quite well written and organized. English could be improved.

Figures and tables are comprehensive and clear.

The introduction explains in a clear and coherent manner the background of this study. 

Thank you for the positive feedback. Please find our response to the detailed comments separately under each comment.

We suggest the following modifications:

• Introduction section: although the authors correctly included important papers in this setting, we believe some studies regarding IDH inhibitors and IDH in other solid tumors should be cited within the introduction ( PMID: 32487595 ; PMID: 33799004; PMID: 33592561 ), only for a matter of consistency. We think it might be useful to introduce the topic of this interesting study.

Authors’ answer: This is an important addition. Thank you for proposing the additional references. We have looked into these and expanded the Introduction with these and some other relevant new references regarding the extent of resection and its effect on survival.

• Methods and Statistical Analysis: nothing to add.

• Discussion section: Very interesting and timely discussion. Of note, the authors should expand the Discussion section, including a more personal perspective to reflect on. For example, they could answer the following questions – in order to facilitate the understanding of this complex topic to readers: what potential does this study hold? What are the knowledge gaps and how do researchers tackle them? How do you see this area unfolding in the next 5 years? We think it would be extremely interesting for the readers. However, we think the authors should be acknowledged for their work. In fact, they correctly addressed an important topic, the methods sound good and their discussion is well balanced.

Authors’ answer: Thank you for the comment. We believe that in the near future AI-based intraoperative tissue identification systems, including the DMS, will become a widely used tool for tumor surgery. Already in the latest WHO brain tumor classification, molecular features of the tumor have become more prominent in the classification than classical histological appearance. Thus, the ability to recognise such features in real time will become a necessity in tumor surgery, which requires novel tissue identification methods. Only few such methods have been introduced, including mass spectrometry and Raman spectroscopy -based solutions, but their clinical use has been hindered due to high cost, impracticality and maintenance issues. We propose a simple and straightforward new solution for the problem. We have expanded the Discussion section to better address future prospects.

 

One additional little flaw: the authors could better explain the limitations of their work, in the last part of the Discussion.

Authors’ answer: This is a good point. We have clarified and expanded the limitations of the study regarding e.g. limited number of samples, the usage of frozen samples instead of fresh tumor and potential consequences of simple sample preparation to the DMS signal strength and our ideas on how to tackle that in the future.

We believe this article is suitable for publication in the journal although some revisions are needed. The main strengths of this paper are that it addresses an interesting and very timely question and provides a clear answer, with some limitations.

We suggest a linguistic revision and the addition of some references for a matter of consistency. Moreover, the authors should better clarify some points.

Authors’ answer: Thank you for the comment. We have carefully double-checked the language.  

Author Response File: Author Response.pdf

Reviewer 2 Report

Please describe how control IDH testing was done. IDH R132 H immunhistochemistry? Pyrosequencing covering also rare IDH mutations?

To make it more palpable for the reader: I do only see a tiny difference between the two groups in Fig. 1G - is there a better figure where the difference is more visible?

How could you improve sensitivity and specificity? More samples? Augmented training set? 

Please discuss more in detail and/ or speculate:

How does fresh tissue directly from the OR work in this analysis?

Does tissue have to be snap frozen first?

How about "contamination" of the samples with blood, bone, dura, cerebrospinal fluid, 5-ALA, fluorescin, etc.? Does this affect results?

Have you tried suspending tumor tissue without having to freeze and cut it?

How much time do you estimate from removal to tissue diagnosis? How much time does tissue preparation need?

What is the cost of hardware and per sample?

Can you distinguish between metastasis, glioma, ependymoma and lymphoma?

     

      

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This topic is interesting and very current, but some points need to be revised:

• Lines 24-25: "We prospectively obtained tumor samples from 22 patients that included 11 IDH mutated and 11 IDH wild type tumors" - Are these patients consecutive? 
• 11 IDH mutated/11 IDH wild type. Is this a coincidence or is it a choice of the authors? Please explain in the text.
• Lines 50-54: "IDH mutation status is an important factor for surgical decision-making.. resection to avoid neurological deficit is favored" At this point, authors should discuss the impact of extent of resection at first and second surgery on survival. Look at these refs: -- Surgical outcome and molecular pattern characterization of recurrent glioblastoma multiforme: A single-center retrospective series. Clin Neurol Neurosurg. 2021 Aug;207:106735. doi: 10.1016/j.clineuro.2021.106735. --  Overcoming the blood-brain tumor barrier for effective glioblastoma treatment. Drug Resist Updat. 2015 Mar;19:1-12. doi: 10.1016/j.drup.2015.02
• Lines 99-107: "We evaluated the accuracy of several machine learning algorithms for the detection... Forest (RF), Decision Tree (DT), Support Vector Machines (SVM) and XGBoost (XGB)" This part seems Methods. Move it above in the correct section.
• Lines 127-129: "Results should be better reported. The most difficult tumor type for the classifier was gr. IV GBM. " What did authors mean?
• Lines 155-156: "This was expected since GBMs harbor a lot of intratumoral heterogeneity and proved to be difficult to classify in our previous DMS study already". Discuss more about the intratumoral heterogeneity of GBM.  doi: 10.3389/fimmu.2020.01402. -- doi: 10.1038/onc.2016.85.  Discuss also more about the role of the "GBM microenvironment for new groundbreaking strategies" in Pubmed.
• Lines 168-170: "Our results show that DMS is able to differentiate IDH mutated and IDH wild type tumors with good accuracy..." - What benefits can DMS bring to patients with glioblastoma? Add here.

Author Response

Please see the attachment.

• Lines 24-25: "We prospectively obtained tumor samples from 22 patients that included 11 IDH mutated and 11 IDH wild type tumors" - Are these patients consecutive? 

Authors’ answer: All the adult patients with intraparenchymal brain tumors operated in Tampere or Helsinki University Hospitals were requested to give tumor samples for the study, so our material was prospectively obtained and non-selected. Naturally, there were a lot more IDH wild type tumors, so we continued with the patient recruitment until we had a sufficient amount of IDH mutated tumors for balanced classes. Then we randomly selected IDH wild type tumors from the material to make balanced groups. Thank you for this relevant notion, we have clarified the selection process in the revised manuscript.

• 11 IDH mutated/11 IDH wild type. Is this a coincidence or is it a choice of the authors? Please explain in the text.

Authors’ answer: It was our choice. For the training of machine learning classifiers it is often preferred to have balanced classes, otherwise the classifier may begin to emphasize the larger class. The limiting factor for the group sizes was the availability of IDH mutated tumors. This is an important notion and should have been better addressed in the text, which we have now done.

• Lines 50-54: "IDH mutation status is an important factor for surgical decision-making.. resection to avoid neurological deficit is favored" At this point, authors should discuss the impact of extent of resection at first and second surgery on survival. Look at these refs: -- Surgical outcome and molecular pattern characterization of recurrent glioblastoma multiforme: A single-center retrospective series. Clin Neurol Neurosurg. 2021 Aug;207:106735. doi: 10.1016/j.clineuro.2021.106735. --  Overcoming the blood-brain tumor barrier for effective glioblastoma treatment. Drug Resist Updat. 2015 Mar;19:1-12. doi: 10.1016/j.drup.2015.02

Authors’ answer: Thank you for the suggestion. We have looked into these references and expanded the Introduction to discuss the impact of extent of resection further.

• Lines 99-107: "We evaluated the accuracy of several machine learning algorithms for the detection... Forest (RF), Decision Tree (DT), Support Vector Machines (SVM) and XGBoost (XGB)" This part seems Methods. Move it above in the correct section.

Authors’ answer: You are correct, thank you for the notion. We have moved the chapter into the Methods section.

• Lines 127-129: "Results should be better reported. The most difficult tumor type for the classifier was gr. IV GBM. " What did authors mean?

Authors’ answer: Thank you for your sharp-sighted notion. This was actually an erroneous statement written by a mistake. It was originally meant that GBMs would produce more erroneous classifications. However, out of the 22 original samples, 13 were GBMs. In total, 14 samples had one or more of their individual incisions incorrectly classified. Out of these 14, 9 were GBMs. Thus, the proportion of GBMs in wrongly classified samples is roughly the same as in all the samples. We have removed the erroneous sentence from the text.

Lines 155-156: "This was expected since GBMs harbor a lot of intratumoral heterogeneity and proved to be difficult to classify in our previous DMS study already". Discuss more about the intratumoral heterogeneity of GBM.  doi: 10.3389/fimmu.2020.01402. -- doi: 10.1038/onc.2016.85.  Discuss also more about the role of the "GBM microenvironment for new groundbreaking strategies" in Pubmed.

Authors’ answer: Intratumoral heterogeneity in GBMs is truly a fascinating topic, and also a potential source of error in DMS analysis. This should be addressed in more detail in our paper. We have looked into your references and expanded the text in the Discussion section.

• Lines 168-170: "Our results show that DMS is able to differentiate IDH mutated and IDH wild type tumors with good accuracy..." - What benefits can DMS bring to patients with glioblastoma? Add here.

Authors’ answer: We performed this study according to contemporary WHO tumor classification, which had IDH mutated GBM as an entity. These are tumors that histologically and radiologically resemble GBMs but still harbor IDH mutation. Also, these tumors have generally better prognosis than IDH wild type GBMs and the patients get additional survival benefit from a gross total resection. With classical frozen section analysis, it is impossible to differentiate IDH mutated and IDH wild type GBMs from another. Therefore, we believe that GBM patients are an excellent example of patients that would benefit from intraoperative identification of the tumor with DMS. We have discussed this with more detail in the text.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors addressed all our queries.

We recommend Acceptance.

Reviewer 3 Report

Authors solved all my criticisms.