Circulating Tumor DNA: Less Invasive, More Representative Method to Unveil the Genomic Landscape of Newly Diagnosed Multiple Myeloma Than Bone Marrow Aspirates

Round 1

Reviewer 1 Report

Liu et al. reported in the article "circulating tumor DNA: less invasive, more representative method to unveil the genomic landscape of newly diagnosed multiple myeloma than bone marrow aspirates" about interesting results regarding circulating tumor DNA.  These results are interesting and could improve the understanding of the heterogenous genomic abnormalities in myeloma. 

To better understand the genomic abnormalities in peripheral blood. The authors are comparing bone marrow biopsy with peripheral blood. The results demonstrated that the ctDNA mutations are highly informative. This work is relevant in the field. This is one of the largest comparisons of bone marrow cells versus circulating DNA. The authors prefer the analysis of circulating DNA.  This method can spare bone marrow punctures. The material and methods are well presented. The discussion includes the current knowledge in this field. The references are appropriate.

Author Response

Thanks so much for the comments. 

Reviewer 2 Report

The manuscript entitled “Circulating tumor DNA: less invasive, more representative method to unveil the genomic landscape of newly diagnosed multiple myeloma than bone marrow aspirates” compared the mutational landscapes of ctDNA and BM in patients with newly diagnosed MM and studied the potential of clonal composition analysis of ctDNA to monitor therapeutic response. This manuscript is a very interesting, well-written, and well-designed study that brings novelty to the field. However, some minor issues such be addressed.

 

- Table 1 should be included in the main text. In fact, this Table and all the Figures are missing! The supplemental Table 1 is a gene list, not the patient's demographic and clinical data. Additionally, the Authors need to revise the supplemental Tables numeration.

- Was the BM DNA extraction performed in all BM cells without plasma cell isolation? If plasma cells were not isolated, this fact should be discussed and included as a study limitation. The low-frequency mutations may not have been detected because of the high number of other hematopoietic cells (progenitor and differentiated myeloid and lymphoid cells, and even erythroid precursors).

- Since the follow-up time was short, the Author should also analyze the time-to-next-treatment (TTNT). Did the Authors use ROC analysis to determine the best value to discriminate between live and dead patients? Sometimes this analysis provides a more accurate cut-off value than the mean/median.

Author Response

Thanks so much for the valuable advices and comments.

Table 1 should be included in the main text. In fact, this Table and all the Figures are missing! The supplemental Table 1 is a gene list, not the patient's demographic and clinical data. Additionally, the Authors need to revise the supplemental Tables numeration.

Reply: Thank you for the advice. We reconstructed and modified the manuscript, added Table1 (Clinical characteristics of 82 NDMM patients). the missing Figures had combined in the new PDF. We have revised the supplemental Tables numeration.

 

2. Was the BM DNA extraction performed in all BM cells without plasma cell isolation? If plasma cells were not isolated, this fact should be discussed and included as a study limitation. The low-frequency mutations may not have been detected because of the high number of other hematopoietic cells (progenitor and differentiated myeloid and lymphoid cells, and even erythroid precursors).

Reply: thank you so much for the valuable advice. The BM DNA extraction were performed in all BM cells without plasma cell isolation. We have revised the manuscript and added this limitation in the discussion part (line 474-480) as your suggestion. The revised part is following: Besides, the BM DNA extraction were performed in all BM cells without plasma cell isolation, thus the low-frequency mutations may not have been detected because of the high number of other hematopoietic cells (progenitor and differentiated myeloid and lymphoid cells, and even erythroid precursors). However, from another point of view, ctDNA may also be partially derived from non-plasma cells. Therefore, the analysis of ctDNA, especially some common mutations from the myeloid cells, needs comprehensive analysis.

 

3. Since the follow-up time was short, the Author should also analyze the time-to-next-treatment (TTNT). Did the Authors use ROC analysis to determine the best value to discriminate between live and dead patients? Sometimes this analysis provides a more accurate cut-off value than the mean/median.

Reply: thanks so much for the valuable advices. For ctDNA quantification, we used mTBI indicator, mTBI was analyzed using the mean VAF of clonal mutations. Your suggestions are very valuable. We then used ROC analysis to determine the best value for ctDNA mTBI was 2.765% (p = 0.015, sensitivity 68.9%, specificity 59.5%), whereas 3.25% was the median. After applying 2.765% grouping mTBI as high and low group, still, there were statistical differences in univariate analysis (HR: 2.469, 95%CI: 1.312-4.648, p = 0.005), but not in multivariate analysis. (HR: 1.676, 95%CI: 0.802-3.502, p = 0.170). we added the explanation in the revised manuscript (line 361-364). The cut-off value for other quantitative variables in this study was consistent with values which researchers frequently used, such as 35g/L for Albumin, 100*10^9/L for platelet. 

TTNT is very important. However, on one hand, in this study all patients were treated with BCD/RVD for induction therapy with some dose adjustment depending on the individual patients’ fragility score.  We did not change to next line treatment until PD. On the other hand, we would like to watch and wait cautiously in some low-risk patients although biochemical relapse, thus the TTNT is a little longer than PFS in our center. Taken together, considering that our follow-up time less than 3 years, we did not analyze TTNT.

Author Response File: Author Response.docx