Associations of Human Colorectal Adenoma with Serum Biomarkers of Body Iron Stores, Inflammation and Antioxidant Protein Thiols
Round 1
Reviewer 1 Report
The presented review manuscript takes up an essential aspect of searching for the causes of the development of one of the most common and most challenging to treat neoplasms - colorectal cancer.
However, in my opinion, the assumption made in the research and the method of selecting the groups were not completed correctly. Modulation of parameters studied, including inflammation (high-sensitivity C-reactive protein [hs-CRP]) and anti-oxidative capacity (total of thiol groups (-SH) of proteins [SHP]), is not only the results of red / processed meat consumption. It has been known that other nutrients can influence their fluctuations. Additionally, the influence of the parameters in the living environment was completely ignored. Collecting only information on age, sex, smoking status, physical activity, red and processed meat consumption does not solve the assumed problem and does not allow for a reliable analysis. Therefore, I rate the studies as very poor and do not recommend it for publication in Antioxidants.
Author Response
Point 1: The presented review manuscript takes up an essential aspect of searching for the causes of the development of one of the most common and most challenging to treat neoplasms - colorectal cancer.
Response 1: We would like to politely clarify that the aim of our research article with original data was as follows: “The aim of our investigation was to assess whether serum biomarkers of body iron stores (ferritin, transferrin, total serum iron and transferrin saturation), inflammation (hs-CRP) and anti-oxidant capacity (SHP) are associated with overall and advanced CRA in a colonoscopy screening cohort.” (lines 91-94). Thus, the aim of our article was not as broad as stated in the reviewer’s comment.
Point 2: However, in my opinion, the assumption made in the research and the method of selecting the groups were not completed correctly. Modulation of parameters studied, including inflammation (high-sensitivity C-reactive protein [hs-CRP]) and anti-oxidative capacity (total of thiol groups (-SH) of proteins [SHP]), is not only the results of red / processed meat consumption. It has been known that other nutrients can influence their fluctuations.
Response 2: We fully agree that other nutrients than red/processed meat consumption can influence the serum concentrations of biomarkers of inflammation and oxidative stress. However, the rationale to select this nutrient is its known association with the development of colorectal cancer (lines 81-83). We are not aware of enough evidence for other nutrients to consider them as established risk factors for colorectal cancer and we, therefore, focused on red/processed meat consumption. Furthermore, it is likely that other aspects of a healthy diet are correlated with low red/processed meat consumption.
Point 3: Additionally, the influence of the parameters in the living environment was completely ignored. Collecting only information on age, sex, smoking status, physical activity, red and processed meat consumption does not solve the assumed problem and does not allow for a reliable analysis.
Response 3: We assume with the term “living environment” the reviewer referred to exposure to asbestos or radon because we already adjusted for the most important dietary and lifestyle factors, which are also environmental risk factors for colorectal cancer. Asbestos exposure has a weak association with colorectal adenoma (PMID: 31511437) and it is prohibited to use in Germany since 1993. Because radon and its progeny are absorbed mainly by inhaling, and because the radiation they give off travels only a short distance, it is unlikely that radon would affect other tissues than the lung in the body (American Cancer Society; https://www.cancer.org/cancer/cancer-causes/radiation-exposure/radon.html). Thus, as asbestos and radon will not have played a relevant role in the development of colorectal adenoma in our study population, we think that our statistical methods were adequate and the model was adjusted sufficiently for environmental factors.
Point 4: Therefore, I rate the studies as very poor and do not recommend it for publication in Antioxidants.
Response 4: We regret this appraisal of our article and as pointed out in the response to point 1, we think that the reason for the adverse appraisal is a general misconception of the aim of our article.
Reviewer 2 Report
Manuscript ID antioxidants-1271462
Title: Associations of serum biomarkers of body iron stores, inflammation and antioxidant protein thiols with colorectal adenoma
Journal: Antioxidants
Authors: Schöttker et al.
The manuscript describes association between serum biomarkers of body iron stores, inflammation, antioxidant thiols associated with peptides or proteins and colorectal adenoma. The above manuscript is interesting. However, it should be noted that the manuscript contains some errors and deficiencies that need to be corrected prior to publication of the manuscript.
List of deficiencies and errors:
- Section 1. Background (page 2, lines 63-64). The authors wrote that “In blood, the most frequent proteins with thiol groups include albumin, glutathione and members of the thioredoxin family”. Glutathione is not a protein but a tripeptide and for this reason the authors should replace “proteins” with “peptides and proteins”.
- Title of the manuscript. According to comment # 1, the authors should reconsider changing “antioxidant protein thiols” to “antioxidant peptide and protein thiols”.
- Abstract, line 20 and other parts of the manuscript. According to comment 1 and 2, the authors should consider changing “thiol groups (-S-H) of proteins [SHP]” to “thiol groups (-S-H) of peptides and proteins (SHPP)”.
- Abstract, line 18, 20 and other parts of the manuscript. The authors did not measure total serum iron or “total of thiol groups”. They measured concentration of total serum iron or concentration of thiol groups. This should be corrected.
- Section 2.3. Laboratory measurements (page 3, lines 116-119. The authors wrote that “Blood samples were collected from an antecubital vein, and serum samples were stored at -80°C until analysis. For measurements of selected biomarkers, serum samples were aliquoted and were shipped on dry ice to the Laboratory for Health Protection Research (Bilthoven, the Netherlands)”. This statement is unclear and sequence of events is unknown. Serum should be divided into separate samples immediately after collection, at least one for each parameter and then frozen at -80°C. Previously frozen samples should be shipped on dry ice in a thermal pack.
- Section 2.3. Laboratory measurements. The authors should present the idea and more details of each method used to measure the parameters tested.
- Section 2.3. Laboratory measurements, line 125. The authors present the formula to determine the transferrin saturation and support this formula by reference 1. However, the reference 1 does not provide the formula to determine transferrin saturation. The authors should provide an appropriate reference.
- Discussion. One of the most important risk factor for colorectal adenoma and colorectal cancer is obesity. The authors should consider writing some words on ghrelin, a gastrointestinal hormone produced mainly in the stomach. Previous studies showed that ghrelin stimulates food intake and serum level of ghrelin is negatively correlated with body mass index, food intake and serum level of glucose. On the other hand, fasting, low level of glucose and anorexia nervosa increase serum level of this hormone (PMID: 22300084; PMID: 25716961). Animal studies showed that ghrelin exhibits protective and therapeutic effects in experimental colitis and these effects are related to its ant-inflammatory and anti-oxidative activities (PMID: 26713317; PMID: 27598133). In addition, animals studies showed that administration of ghrelin suppress inflammation-induced colorectal carcinogenesis in mice (PMID: 26094822). These findings are additionally supported by clinical studies showing that low serum level is significantly associated with increased risk of colorectal cancer development (PMID: 28814486).
Author Response
General appraisal of the reviewer: The manuscript describes association between serum biomarkers of body iron stores, inflammation, antioxidant thiols associated with peptides or proteins and colorectal adenoma. The above manuscript is interesting. However, it should be noted that the manuscript contains some errors and deficiencies that need to be corrected prior to publication of the manuscript.
Point 1: Section 1. Background (page 2, lines 63-64). The authors wrote that “In blood, the most frequent proteins with thiol groups include albumin, glutathione and members of the thioredoxin family”. Glutathione is not a protein but a tripeptide and for this reason the authors should replace “proteins” with “peptides and proteins”.
Response 1: Changed as suggested (line 76)
Point 2: Title of the manuscript. According to comment # 1, the authors should reconsider changing “antioxidant protein thiols” to “antioxidant peptide and protein thiols”.
Response 2: Changed as suggested (line 3).
Point 3: Abstract, line 20 and other parts of the manuscript. According to comment 1 and 2, the authors should consider changing “thiol groups (-S-H) of proteins [SHP]” to “thiol groups (-S-H) of peptides and proteins (SHPP)”
Response 3: Changed as suggested (lines 19, 73).
Point 4: Abstract, line 18, 20 and other parts of the manuscript. The authors did not measure total serum iron or “total of thiol groups”. They measured concentration of total serum iron or concentration of thiol groups. This should be corrected.
Response 4: Changed as suggested (lines 29, 31, 73, 139, 143, 173, ....).
Point 5: Section 2.3. Laboratory measurements (page 3, lines 116-119. The authors wrote that “Blood samples were collected from an antecubital vein, and serum samples were stored at -80°C until analysis. For measurements of selected biomarkers, serum samples were aliquoted and were shipped on dry ice to the Laboratory for Health Protection Research (Bilthoven, the Netherlands)”. This statement is unclear and sequence of events is unknown. Serum should be divided into separate samples immediately after collection, at least one for each parameter and then frozen at -80°C. Previously frozen samples should be shipped on dry ice in a thermal pack.
Response 5: The sentences were modified so that the sequence of events gets clear (line 131-133). It is true that only one serum aliquot was needed for the measurements because the different parameters were done on the same autoanalyzer.
Point 6: Section 2.3. Laboratory measurements. The authors should present the idea and more details of each method used to measure the parameters tested.
Response 6: The idea why these biomarkers were chosen has been outlined in detail in the introduction. In the methods section about the laboratory measurements, we list assay names, suppliers, name of lab and autoanalyzer and the coefficient of variations, which should be sufficient for reproducibility of the results. Thus, we did not add further details about the laboratory measurements.
Point 7: Section 2.3. Laboratory measurements, line 125. The authors present the formula to determine the transferrin saturation and support this formula by reference 1. However, the reference 1 does not provide the formula to determine transferrin saturation. The authors should provide an appropriate reference.
Response 7: Thank you for finding this important issue. The (1) was actually not meant as a reference for the equation but as the number of the equation. However, as no 2nd equation follows, we now took out this number. We now added the missing reference for the equation (line 141). However, the cited equation used transferrin in a different unit (mg/dl instead of g/L) and therefore we now also report transferrin valued in the unit mg/dl. As we had a little rounding error in the conversion of units before, some transferrin saturation values in the manuscript changed by 1 point in the first digit after the comma. However, the rounding error did not have any influence on the results that used the transferrin saturation in tertiles or in correlation analyses (Tables 4 and 5).
Point 8: Discussion. One of the most important risk factor for colorectal adenoma and colorectal cancer is obesity. The authors should consider writing some words on ghrelin, a gastrointestinal hormone produced mainly in the stomach. Previous studies showed that ghrelin stimulates food intake and serum level of ghrelin is negatively correlated with body mass index, food intake and serum level of glucose. On the other hand, fasting, low level of glucose and anorexia nervosa increase serum level of this hormone (PMID: 22300084; PMID: 25716961). Animal studies showed that ghrelin exhibits protective and therapeutic effects in experimental colitis and these effects are related to its anti-inflammatory and anti-oxidative activities (PMID: 26713317; PMID: 27598133). In addition, animals studies showed that administration of ghrelin suppress inflammation-induced colorectal carcinogenesis in mice (PMID: 26094822). These findings are additionally supported by clinical studies showing that low serum level is significantly associated with increased risk of colorectal cancer development (PMID: 28814486).
Response 8: Thank you for this suggestion that we gladly included in the discussion (lines 476-483).
Round 2
Reviewer 1 Report
Unfortunately, I hold my previous opinion about the rejection of the presented manuscript. The authors in their explanation tried to base on the theory supposedly that so far, it is not enough evidence for other nutrients than red meat to consider them as established risk factors for colorectal cancer. If authors would like to based on that hypothesis, then they should confirm that in the preliminary studies, the results of the presented research would be reliable. I regret to say that in its current form, the evidence of the basis for the research assumptions is too weak to consider the results valid.
Author Response
Point 1: Unfortunately, I hold my previous opinion about the rejection of the presented manuscript. The authors in their explanation tried to base on the theory supposedly that so far, it is not enough evidence for other nutrients than red meat to consider them as established risk factors for colorectal cancer. If authors would like to based on that hypothesis, then they should confirm that in the preliminary studies, the results of the presented research would be reliable. I regret to say that in its current form, the evidence of the basis for the research assumptions is too weak to consider the results valid.
Response 1: We fully agree that other nutrients than red/processed meat consumption can influence the serum concentrations of biomarkers of inflammation and oxidative stress. However, the rationale to select this nutrient is its known association with the development of colorectal cancer (see meta-analyses referenced as reference no. 4 and 22). We are not aware of as good evidence for other nutrients to consider them as established risk factors for colorectal cancer and we, therefore, focused on red/processed meat consumption. In addition, the focus on red/processed meat consumption was chosen because this biomarker oriented paper assessed the iron metabolism biomarkers, which are influenced by dietary haem intake and meat is the main source of dietary haem. Other nutrients do not contain relevant amounts of dietary haem.
Reviewer 2 Report
Manuscript ID antioxidants-1271462
Title: Associations of serum biomarkers of body iron stores, inflammation and antioxidant protein thiols with colorectal adenoma
Journal: Antioxidants
Authors: Schöttker et al.
The manuscript describes the relationship between serum biomarkers of body iron stores, inflammation, antioxidant thiols associated with peptides or proteins and colorectal adenoma. The above manuscript is interesting and its current version is much better than the previous one. However, there are still some errors that need to be corrected:
- Previous comment # 6. Laboratory measurement. The authors should provide an idea and more details about each method used to measure the parameters tested. As the authors rightly noted, the introduction provides information on why the authors assessed particular parameters. The reviewer’s comment, however, concerns the procedures used to evaluate parameters tested. For example, in the case of immunoassay, the authors should specify whether they used ELISA or turbidimetric method. What was the sensitivity, specificity, level of detection and repeatability of results? The authors should easily find all these data in chemistry information sheets provided by the manufacturer. These data allow the readers to check the quality and validity of the results presented..
- Previous comment # 8. Discussion. One of the most important risk factor for colorectal adenoma and colorectal cancer is obesity. The authors should consider writing some words on ghrelin, a gastrointestinal hormone produced mainly in the stomach. Previous studies showed that ghrelin stimulates food intake and serum level of ghrelin is negatively correlated with body mass index, food intake and serum level of glucose. On the other hand, fasting, low level of glucose and anorexia nervosa increase serum level of this hormone (PMID: 22300084; PMID: 25716961). Animal studies showed that ghrelin exhibits protective and therapeutic effects in experimental colitis and these effects are related to its ant-inflammatory and anti-oxidative activities (PMID: 26713317; PMID: 27598133). In addition, animals studies showed that administration of ghrelin suppress inflammation-induced colorectal carcinogenesis in mice (PMID: 26094822). These findings are additionally supported by clinical studies showing that low serum level is significantly associated with increased risk of colorectal cancer development (PMID: 28814486). The authors have presented some suggested articles. However, suggested articles form a logical cause-and-effect sequence. It is not possible to remove some elements without disturbing the meaning of the statement. For this reason, the authors should used all these article.
Author Response
Point 1
Previous comment # 6. Laboratory measurement. The authors should provide an idea and more details about each method used to measure the parameters tested. As the authors rightly noted, the introduction provides information on why the authors assessed particular parameters. The reviewer’s comment, however, concerns the procedures used to evaluate parameters tested. For example, in the case of immunoassay, the authors should specify whether they used ELISA or turbidimetric method. What was the sensitivity, specificity, level of detection and repeatability of results? The authors should easily find all these data in chemistry information sheets provided by the manufacturer. These data allow the readers to check the quality and validity of the results presented.
Response 1:
The paragraph “The kit to measure SHP was purchased from Rel Assay Di-119 agnostics (Gaziantep, Turkey) and the kits for hs-CRP, ferritin, transferrin and total serum iron were purchased from Beckman Coulter Diagnostics (Woerden, the Netherlands). All above mentioned assays were conducted by following manufacturers’ manuals and were adapted to an automatic analyzer (LX20-Pro, Beckman-Coulter Diagnostics, Woerden, the Netherlands)."
Was replaced by:
The following assays were used by following manufacturers’ manuals and were adapted to an automatic analyzer (LX20-Pro, Beckman-Coulter Diagnostics, Woerden, the Netherlands).
SHP
The kit to measure SHP was purchased from Rel Assay Diagnostics (Gaziantep, Turkey). The SHP test quantifies the presence of sulfhydryl (–SH) groups in the biological sample. When serum/plasma is used as a sample, the -SH groups are present mainly in proteins. The method is based on the capacity that the –SH groups react with 5,5’-dithiobis-2-nitrobenzoic acid, followed by a development of a colored complex that can be measured with an autoanalyzer at 405 nm absorbance. L-cysteine has been used as standard. The range of detection is 0.1-1.8 mmol SH-groups in proteins. The reference range for SHP is 450-680 μmol/L. At a level of level 660 μmol/L, the interassay coefficient of variation is 2.5% (N=9).
hs-CRP
The kits for hs-CRP were purchased from Beckman Coulter Diagnostics (Woerden, the Netherlands). The high-sensitive CRP assay is based on a highly sensitive particle immunoassay rate methodology. An anti-CRP antibody-coated particle binds to CRP in the patient sample resulting in the formation of insoluble aggregates causing turbidity. The autoanalyzer monitors the change in absorbance at 940 nm. This change in absorbance is proportional to the concentration of C-reactive protein in the sample. The dynamic range of the assay is 0.2-80 mg/L, with a limit of detection of 0.2 mg/L. At a level of 1.75 mg/L, the interassay coefficient of variation is 2,9% (N=9).
Ferritin
The ferritin assay was obtained from Dialab, Vienna, Austria. ( nr.A01551). The assay of ferritin is based on turbidimetric measurement. Turbidity is caused by the formation of an insoluble immuno complex of ferritin with a highly-specific antibody. The dynamic range of the assay is 10-300 ng/mL, with a limit of detection of 10 ng/mL. At a level of 240 ng/mL, the interassay coefficient of variation is 2.4% (N=9). Ferritin could not be measured in samples with a high lipemic index (n=62, 31%).
Transferrin
The kits for transferrin were purchased from Beckman Coulter Diagnostics (Woerden, the Netherlands). The transferrin assay is based on a turbidimetric method. In the reaction, transferrin combines with highly specific antibody to form an insoluble antigen-antibody complex. The sensitivity for transferrin assay is 0.7 g/L, defined as the lowest measurable concentration which can be distinguished from zero with 95% confidence. At a level of 3.2 g/L, the interassay coefficient of variation is 1.2% (N=9).
Total serum iron
The kits for total serum iron were purchased from Beckman Coulter Diagnostics (Woerden, the Netherlands). The total iron assayis a colorimetris assay. The iron reagents are used to measure the iron concentration by a timed-endpoint method. In the reaction, iron is released from transferrin by acetic acid and is reduced to the ferrous state by hydroxylamine and thioglycolate. The ferrous ion is immediately complexed with the FerroZine Iron Reagent (3-2-pyridyl-5,6-diphenyl-1,2,4-triazine-p,p'-disulfonic acid monosodium salt hydrate). The sensitivity for the iron assy is 0.9 μmol/L, defined as the lowest measurable concentration, which can be distinguished from zero with 95% confidence. At a level of 28 μmol/L, the interassay coefficient of variation is 2.5% (N=9).
Point 2
Previous comment # 8. Discussion. One of the most important risk factor for colorectal adenoma and colorectal cancer is obesity. The authors should consider writing some words on ghrelin, a gastrointestinal hormone produced mainly in the stomach. Previous studies showed that ghrelin stimulates food intake and serum level of ghrelin is negatively correlated with body mass index, food intake and serum level of glucose. On the other hand, fasting, low level of glucose and anorexia nervosa increase serum level of this hormone (PMID: 22300084; PMID: 25716961). Animal studies showed that ghrelin exhibits protective and therapeutic effects in experimental colitis and these effects are related to its ant-inflammatory and anti-oxidative activities (PMID: 26713317; PMID: 27598133). In addition, animals studies showed that administration of ghrelin suppress inflammation-induced colorectal carcinogenesis in mice (PMID: 26094822). These findings are additionally supported by clinical studies showing that low serum level is significantly associated with increased risk of colorectal cancer development (PMID: 28814486). The authors have presented some suggested articles. However, suggested articles form a logical cause-and-effect sequence. It is not possible to remove some elements without disturbing the meaning of the statement. For this reason, the authors should used all these article.
Response 2: We now cite all suggested 6 articles (see references 39-45, page 14).
Round 3
Reviewer 2 Report
Manuscript is almost ready for publication. Page 4, line 160. There are two typing errors that need to be corrected.
Author Response
Thank you very much for pointing this out. The typing errors are now corrected.
