Circulating Tumour Cell Numbers Correlate with Platelet Count and Circulating Lymphocyte Subsets in Men with Advanced Prostate Cancer: Data from the ExPeCT Clinical Trial (CTRIAL-IE 15-21)
, Lauren Brady 3, Gráinne Sheill 4, Anne-Marie Baird 3, Emer Guinan 5, Bryan Stanfill 6, Jean Dunne 7, Dean Holden 7, Tatjana Vlajnic 8, Orla Casey 9, Verena Murphy 9, John Greene 3, Emma H. Allott 3,10, Juliette Hussey 4, Fidelma Cahill 11, Mieke Van Hemelrijck 11, Nicola Peat 12, Lorelei A. Mucci 13, Moya Cunningham 14, Liam Grogan 15, John J. O’Leary 3,20, Stephen Finn 3,9,20add
Show full author list
Round 1
Reviewer 1 Report
The manuscript focuses on analyzing circulating tumor cells (CTCs), platelets and lymphocytes in the blood of pancreatic cancer patients enrolled in ExPeCT clinical trial. The authors compare the frequency and total numbers of these cell populations in patients subjected to exercise or not and correlate the numbers of CTCs with platelets as well as with T cells, B cells and NK cells. The data is of important value however the way the data is presented is not easy to follow. It is difficult to judge what is the main conclusion of the paper (correlation at all timepoints in all groups? differences between exercise and non-exercise group?) and the conclusion stated in the discussion is not well supported by the data presented.
Major points:
Functional relationship between NK cells, which are mostly mentioned, and CTC/platelets should be shown to draw the suggested conclusion about these cells interacting in the prostate cancer patients.
Figure legends are completely missing making it very difficult to read the figures (which timepoint, which cohort)
How do authors explain increased numbers of lymphocytes/NK cells together with increased numbers of CTCs? Are these NK cells non-functional as suggested by the cited breast cancer study? This should be addressed.
The introduction does not provide sufficient background information while the discussion is filled with background research. This should be rearranged.
The flow cytometry data has to be presented as graphs and /or representative original FACS blots to simplify the interpretation of the data.
Minor points:
In table 1 the ''%'' of total does not sum up to 100% in ‘’cloaked’ platelets.
Numerous typos and spacing mistakes are present throughout the text
Methods section – no subheadings, very difficult to read
Headings in results do not have any statement
Author Response
Responses to Reviewer 1 Comments
We thank Reviewer 1 for their comments, which have been very helpful to us in revising this manuscript.
Reviewer 1 states that “the way the data is presented is not easy to follow. It is difficult to judge what is the main conclusion of the paper”. We have now made substantial changes to the structure of the manuscript in order to emphasise that the main finding is the presence of a correlation between CTC numbers, platelets and NK cells. The differences in lymphocyte concentrations between the exercise and non-exercise group, unrelated to CTC concentrations, is considered a finding of secondary importance and is presented and discussed after the correlation findings. The lack of variation in lymphocyte concentrations over the timepoints of the study is of lesser importance and is relegated to Appendix A.
Reviewer 1 also notes that “the conclusion stated in the discussion is not well supported by the data presented”. The first paragraph of the discussion now notes the relatively stronger correlation with NK cell numbers than with other explanatory variables. The conclusion now states that the findings “provide further evidence of an association” between CTCs and platelets, rather than a close association.
Major points:
Reviewer 1 notes that the “Functional relationship between NK cells, which are mostly mentioned, and CTC/platelets should be shown to draw the suggested conclusion about these cells interacting in the prostate cancer patients”. This information is now presented in a more prominent position in the introduction, and the evidence for a functional relationship is bolstered by an additional reference describing platelet-induced reduced expression of NK-cell activating receptor ligands on CTCs.
Reviewer 1 states that “Figure legends are completely missing making it very difficult to read the figures (which timepoint, which cohort)”. Figure legends have been expanded to include the relevant information.
Reviewer 1 queries “How do authors explain increased numbers of lymphocytes/NK cells together with increased numbers of CTCs? Are these NK cells non-functional as suggested by the cited breast cancer study? This should be addressed.” This is an interesting question and the functional activity of NK cell subsets is now addressed in some detail in the discussion.
Reviewer 1 states that “The introduction does not provide sufficient background information while the discussion is filled with background research. This should be rearranged”. With reference to the first major point noted above, the background information relating CTCs, platelets and their killing by NK cells is now presented in the introduction.
Reviewer 1 states that “The flow cytometry data has to be presented as graphs and /or representative original FACS blots to simplify the interpretation of the data.” While we feel that FACS plots of individual flow cytometry assays are unlikely to greatly benefit the readers’ understanding of the data as a whole, we very much agree that the summary data, particularly the comparisons in lymphocyte populations between the exercise and control groups, would benefit from graphical illustration. Box plots (figures 6 and 7) have been included for this reason. The correlations between flow cytometry-defined lymphocyte populations and CTCs are already presented as scatter plots.
Minor points:
Reviewer 1 states that “In table 1 the ''%'' of total does not sum up to 100% in ‘’cloaked’ platelets”. This has been corrected.
Reviewer 1 states that “Numerous typos and spacing mistakes are present throughout the text”. Typographic errors and spacing inconsistencies have been addressed throughout.
Reviewer 1 notes “Methods section – no subheadings, very difficult to read”. Subheadings have now been included in the methods section to break up the text in a logical fashion.
Reviewer 1 notes that “Headings in results do not have any statement”. The headings in the results section have, where appropriate, been changed to declaratory statements.
Reviewer 2 Report
The manuscript by Hayes et al. describes correlation between circulating tumor cells (CTC) platelet cells and lymphocytes. The authors rely on the previous clinical trial to source their data from and use various statistical analyzes to draw the conclusions. They find positive correlation between the analyzed parameters.
The manuscript is well written and organized. It contributes additional data to the research community involved in the prostate cancer field, although the correlation found is not particularly strong. However, I fear that the data package is not strong enough to warrant publication in Cancers. Besides, there are few suggestions and questions I would like to ask:
- Line 128: When were the samples examined after being taken at each timepoint. How were they stored in between?
- The authors should explain why there were only 162 blood draws instead of 183 (3 timepoints, 61 participants) or any other deviation from the expected number of samples.
- Table 1 has 0 and 161 values in CTCs absent/present row which does not add up to 162.
- The manuscript can be benefit from rounding to the significant numbers (especially Table 3).
- Figures with correlations might be better described with r on the graph.
Author Response
Responses to Reviewer 2 Comments
We thank Reviewer 2 for their comments, which have been very useful to us in revising this manuscript.
Reviewer 2 notes that “the data package is not strong enough to warrant publication in Cancers”. It is true that the correlation with platelet concentrations is not strong, but the correlation with NK cell numbers is of moderate strength. It is likely to be of interest to readers of this special issue of Cancers concerned with “The Role of Inflammation in Prostate Cancer”, being derived from a clinical trial directly concerned with the interface between exercise, obesity, systemic inflammation, platelet cloaking and the immune response in the metastatic cascade.
Reviewer 2 notes: “Line 128: When were the samples examined after being taken at each timepoint. How were they stored in between?” The flow cytometry samples were transported directly to the laboratory and processed immediately, within a few hours of specimen acquisition from the participant. This information has now been clarified in the text. The blood samples for CTC filtration were processed within four hours of acquisition from the participant and the ScreenCell filters were stored at 4°C before and after staining with May-Grunwald-Giemsa, as per the manufacturer’s instructions. They were transported on ice packs and examined microscopically several months later. This storage period had no impact on the processed filters.
Reviewer 2 indicates that “The authors should explain why there were only 162 blood draws instead of 183 (3 timepoints, 61 participants) or any other deviation from the expected number of samples.” Several participants withdrew over the course of the study, and so no blood draws were taken from them at later timepoints. Rarely was an individual participant unable to provide a blood sample at a particular timepoint. This information is now explained in section 3.1 of the results.
Reviewer 2 notes that “Table 1 has 0 and 161 values in CTCs absent/present row which does not add up to 162”. This is a typographic error and has been corrected.
Reviewer 2 notes that “The manuscript can be benefit from rounding to the significant numbers (especially Table 3).” Table 3 (which is now renamed Table 5) has been rounded to 1 significant figure, and a similar approach has been taken to the percentages presented in Table 1 and elsewhere in the text. Most of the remaining tables present figures derived from statistical tests – the results are stated as precisely as possible.
Reviewer 2 notes that “Figures with correlations might be better described with r on the graph”. We agree and the correlation graphs now include r, r squared and the relevant p-value. Some additional graphs (figures 6, 7, A1, A2) have been included in response to the comments of another reviewer and these are also annotated with p-values.
Reviewer 3 Report
In this article the authors tend to describe the correlation of CTC and platelet count and circulating lymphocyte in men with advanced prostate cancer. This is an exploratory analysis of data from previously publish ExPeCT clinical trial.
The topic is in general terms very interesting and well described, however I have several concerns:
Major revision:
The main results of these articles are describing correlation between CTC amount and other variables. However, the description of the methodology in respect is very scant and missing any definition what is consider according to the authors as a positive correlation and how was it calculated
Similarly, in the presented figure is absent R2 value that may indicate whether the presented data correlate or not.
Just after the proper definition of a correlation analysis and correct data description authors may draw conclusion is respect.
Minor revisions:
Even thou the results from ExPeCT have been already published elsewhere, it would be informative to summarize them briefly in the introduction.
Revise the spaces between words over document
Author Response
Responses to Reviewer 3 Comments
We thank Reviewer 3 for their comments, which have been very helpful to us in revising the manuscript.
Reviewer 3 notes that “The main results of these articles are describing correlation between CTC amount and other variables. However, the description of the methodology in respect is very scant and missing any definition what is consider according to the authors as a positive correlation and how was it calculated. Similarly, in the presented figure is absent R2 value that may indicate whether the presented data correlate or not. Just after the proper definition of a correlation analysis and correct data description authors may draw conclusion is respect.”
The methods section relating to the correlation and regression has been greatly expanded with these important points in mind. R squared values are presented more prominently, including on the relevant figures (Figures 1-5), and the relative strength of the correlations are discussed – the best fit being for the relationship between CTC numbers and circulating NK cells. The relative weakness of the correlations with platelets is acknowledged.
Reviewer 3 notes that “Even thou the results from ExPeCT have been already published elsewhere, it would be informative to summarize them briefly in the introduction.” The main findings of ExPeCT have now been included in the introduction, as suggested.
Reviewer 3 advises us to “Revise the spaces between words over document”. Typographic errors and spacing inconsistencies throughout the document have been addressed.
Round 2
Reviewer 1 Report
Thw authors significantly improved the manuscrpit and defined their goals and conclusions now clearly. The introduction provides now a good background knowledge.
Minor points:
Chapter 3.2. still has a title which is rather misleading ''Flow cytometry-derived''? PLease state the main conlcusion of the chapter.
The correlation figures always have the same figure legend. However it would be much easier to follow if the legened claims in which group the correaltion was performed.
Reviewer 2 Report
The authors addressed the raised issues.
Reviewer 3 Report
Dear Authors. Thank you for the revision of your manuscript. I now recommend it for publication
