Retrospective Assessment of Complementary Liquid Biopsy on Tissue Single-Gene Testing for Tumor Genotyping in Advanced NSCLC

Round 1

Reviewer 1 Report

The study of Desmeules et al “Retrospective assessment of complementary liquid biopsy on 2 tissue single gene testing for tumor genotyping in advanced 3 NSCLC” addressed the role of both ctDNA and NGS for molecular analysis in advanced NSCLC. Such an approach is particularly important when tissue testing allows limited biomarker coverage. Overall, this is a well written manuscript.

My only concern regards the success rate of ctDNA testing (47%): it is not clear if this value come from a failure of the experiment or from a negative result of an assessable experiment. In fact the authors discussed this aspect in relation to fusions detections (pag 8 One potential contributing factor in the present study is the inability of the ctDNA assay employed  to capture fusions or isoforms (ALK, ROS1, RET, METex14 et al)….). It is important to better described these results and further clarify the message the authors would like to give: in fact you in the literature you can find data with much higher percentage without fusion testing  (> 80%, example Francaviglia et al 2019 “Identification and monitoring of somatic mutations in circulating cell-free tumor DNA in lung cancer patients”): ctDNA extraction method,  ctDNA quantity recover, NGS method sensitivity .

Author Response

Thank you for your comments, they are much appreciated.  Please see the responses in the attached file.

Please note that we have performed an additional change by reviewing the manuscript:

1-We deleted one line of the Table 3 as the MET D1010 variant is not associated with MET exon 14 skipping alteration; we retrieved this case from the figures but omitted to take off  from Table 3 in the initial submission. 

2-We corrected the truncated legend of Figure 3.

3- In order to integrate the response to the reviewers’ comments, we performed some sentence editing / addition and divided the second paragraph in the discussion.

Author Response File: Author Response.docx

Reviewer 2 Report

Desmeules et al describe their findings on this retrospective study regarding complementary liquid biopsy on tissue single gene testing for tumor genotyping in advanced NSCLC. They aimed to evaluate the value of ctDNA NGS over tissue single gene testing. Ninety-one patients were recruited in this retrospective study, who underwent both ctDNA and tissue single gene testing. ctDNA was informative in 47% (43/91) and identified a targetable biomarker in 21% (19/91) of patients. The likelihood of positive result on ctDNA was higher if patients had extra-thoracic disease (59%) or were off active treatment (59%).  ctDNA provided additional information in 41% but missed a known alteration in 8% of cases.   Therapeutic change for targeted therapy  based on plasma  findings, occurred in 5 patients (5%). In addition, the authors evaluated the turnaround time of ctDNA compared to tissue testing. The median turnaround time of ctDNA was 10 days (range 6-25), shorter (P = 0.002) than the cumulative delays for the tissue testing trajectory until biomarker availability (median 13 d; range 7-1737).

This is a well written manuscript summarizing swiftly the results. Although this study carries several and important limitations, those are well summarized in discussion.

Despite not providing innovative knowledge, this study adds on the current evidence regarding the value of liquid biopsy and to the limitations/benefits of NGS over single gene testing.

As so, I believe that the manuscript should be accepted for publication.

Please consider the following minor comments to the authors for manuscript revision.

1)     Please consider a thorough review for spelling mistakes and syntax, especially in abstract.

2)     A more thorough description of the assay would be valuable, consisting at very least the sensitivity-specificity provided by the vendor and description of which targetable mutations are included.

3)     A more detailed discussion, based on literature evidence, regarding their finding of higher detection rate of mutations on liquid biopsy for patients with extra-thoracic disease would be reasonable. How is this finding explained pathophysiologically? Also, please underline that this finding was not statistically significant.

4)     Please add a remark on discussion regarding the fact that patients off therapy had higher rate of mutations detected. Although not significant, is this finding consistent in literature? How do the authors explain this observation?

5)     Please specify if the 5 patients who switched to TKI therapy based on liquid biopsy’s results were tested negative on tissue genotyping for these mutations or were not tested at all.

6)     I think that there is no reference on HER2 mutant cases in the discussion. Was HER2 tested on your assay? A remark should be added as TDX has been added in NSCLC management.

7)     Was PD-L1 evaluated in your assay? Please specify. If not, I think is confluent to use the term “biomarkers” throughout the text and I would prefer a term reflecting the use of NGS for targetable gene mutations detection (for instance “genotyping”). 

Author Response

Thank you for your comments, they are much appreciated.  Please see the responses in the attached file.

Please note that we have performed an additional change by reviewing the manuscript:

1-We deleted one line of the Table 3 as the MET D1010 variant is not associated with MET exon 14 skipping alteration; we retrieved this case from the figures but omitted to take off  from Table 3 in the initial submission. 

2-We corrected the truncated legend of Figure 3.

3- In order to integrate the response to the reviewers’ comments, we performed some sentence editing / addition and divided the second paragraph in the discussion.

Author Response File: Author Response.docx