RNA-Based Next-Generation Sequencing in the Somatic Molecular Testing of Non-Small-Cell Lung Cancer (NSCLC) in a Centralized Model: Real-World Data to Suggest It Is Time to Reconsider Testing Options

Round 1

Reviewer 1 Report

The manuscript entitled "RNA-based next generation sequencing in the somatic molecular testing of non-small cell lung cancer (NSCLC) in a centralized model: Real-world data to suggest it is time to re-consider testing options." examines the real-world context which may account for this failure rate and discusses alternative options for patient care pathways.

- The Authors should provide the expand forms for all acronyms through the text when they first appear.

- Gene acronyms should be written in italics.

- Please consider to adopt "ALK" and "ROS1" acronyms.

- Please indicate company and State after Idylla and Genexus. 

Author Response

Dear Sir/Madam,

Many thanks for your time in reviewing my manuscript. I appreciate your comments and have altered the text accordingly to take into all the changes you suggest. 

Best regards,

Alison Finall

Reviewer 2 Report

Ms. Ref. No.: jmp-1986370 

Title: RNA-based next generation sequencing in the somatic molecular testing of non-small cell lung cancer (NSCLC) in a centralized model: Real-world data to suggest it is time to re-consider testing options. 

This study examine factors to consider for best patient care in the predictive molecular biomarker identification in lung non-squamous NSCLC, adenocarcinoma being the most common type, and discuss whether there is a need for change to current practice and how that might be achieved. This article adds additional interpretation that would advance our understanding of the process of RNA-based next generation sequencing in the somatic molecular testing of NSCLC. Overall, the manuscript is well-written and well-organized.

 Minor comments:

(1).    Figure 1c-1e: Please use a specific software to draw the black circle.

(2).    Figure 2: The suggested algorithm for molecular analysis of somatic NSCLC tissue can be improved by adding more information.

Author Response

Dear Sir/Madam,

Many thanks for your time in reviewing my manuscript.

I appreciate your comments and have made the algorithm much more detailed to more accurtaely reflect the complexity of the decision making process. I have also amended the photomicrographs as suggested.

Best regards,

Alison Finall

Reviewer 3 Report

The author describes the suboptimal and unsatisfactory situation of algorithms and pipelines for predictive biomarker testing in the UK with a focus on RNA-based NGS for gene fusion detection. The main limitations of the current system by NHS are the unacceptably high turn-around times caused by centralization and the high failure rate of RNA-based NGS. Although this paper provides an interesting view on what is happening in the UK with suggestions for improvements such as advocating rapid near patient testing, I have several comments and criticisms for consideration:

1.     Fusion gene testing is not a standalone process but part of broad molecular testing also including DNA. The author does not address the important question of simultaneous vs. sequential testing (i.e. DNA first followed by RNA in case of negative oncogenic driver mutation vs. DNA and RNA simultaneous), although this would clearly impact TAT.

2.     Another means to shorten the TAT would clearly by to cut default empty sections for IHC in all biopsies from patients with suspicion of lung cancer to allow staining for TTF1, p40, PD-L1, ALK and ROS1. Staining for ALK and ROS1identifies almost all patients with ALK or ROS1 fusions with a few days, and ALK IHC positivity alone qualifies to ALK TKI treatment.

3.     The high failure of RNA-based NGS in daily practice in the UK is an important, since similar experience has been made by others. However, the multiplex Idylla approach is probably the magic bullet to solve this problem. When relying on Idylla for DNA and RNA, several, rounds of analyses are needed, and there might not be enough tissue left in for more extensive testing in Idylla-biomarker negative patients. Moreover, Idylla did not perform very well in a recent study on NTRK testing (Sorber L et al, J Mol Diagn 2020). Notably, the performance was even worse when frozen material was used. This paper should be referred to.

4.     One way to maximize the DNA and RNA yield from a patients is triaging all available material including cytology. Cytology is not specifically mentioned in this paper. How is cytology dealt with in the UK? Are all formats for cytology (cell blocks and cellular smears or liquid based preparations) being considered for biomarker analysis, as suggested by current guidelines?

5.     Why does it 14 (!) days to get a FISH analysis done?

Author Response

Dear Sir/Madam,

Many thanks for your time in reviewing my manuscript. I appreciate your well though out and constructive comments. Ihave altered the text accordingly to take into all the changes you suggest. 

1. I have discussed the rationale for synchronous DNA RNA pael testing in our area from line 83 onwards.
2. I have referred to ALK and ROS IHC at the end of the same paragraph.
3. Thank you for highlighting the Sorber study. I have discussed this from line 192. 
4. There is some discussion of the merits of cytology with reference to the Italian study (line 105). I have added in a comment regarding our use of cutology specimens also. 
5. Unfortunately, I cannot explain why it take 2 weeks for FISH reporting at the external lab we use. We are tied in with them for commissioning reasons but am aware it can be done much more quickly. We have challenged this suboptimal performance on a number of occasions with no improvement to date. 

Thank you again for your time and effort in appraising this manuscipt. Your feedback has been most valuable. 

Best regards,

Alison Finall

Reviewer 4 Report

In this work, the authors are trying to examine the existing discrepancies in identifying molecular biomarkers in NSCLC Patients' samples. Further, the authors discuss that RNA-based next-generation sequencing plays a significant role in detecting oncogenic markers and yet the failure rate of this technique is 35% in various stages. This work clearly explains the difficulties in real-world scenarios of processing tissue samples effectively which can be the fundamentals for the follow-up study.

This article as such is well-written and clear to read and I believe, this article is deserved to be published in the journal of molecular pathology.

Author Response

Dear Sir/Madam,

Many thanks for your time in reviewing my manuscript. I appreciate your comments and have reflected upon them.

Best regards,

Alison Finall