Intake of the Total, Classes, and Subclasses of (Poly)phenols and Breast Cancer Risk: A Prospective Analysis of the EPIC Study

Round 1

Reviewer 1 Report

Thank you for the opportunity to review the manuscript ID: antioxidants-4157504. This manuscript aimed to examine the association between the intake of total (poly)phenols, and its classes and subclasses, with breast cancer (BC) risk, overall and by subtypes (estrogen, progesterone, and human epidermal growth factor receptor 2 (HER2)), in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.

In order to achieve the goal, the authors conducted a prospective analysis of the EPIC study.
In this paper, the authors in a very high-quality, comprehensive manner in the Introduction section present the gap in the level of knowledge of the link between the intake of total (poly)phenols and BC risk.
Throughout, the authors cite relevant literature. Note: 8/33 references are self-citations.

The authors correctly present the methodology applied in this paper.
The paper presents the results of a cross-sectional study that was conducted in a cohort of participants that were enrolled between 1992 and 2000 in the study.
Comments:
- Describe `Study population`, with a description of eligible persons, and specification of inclusion and exclusion criteria, `Participation rate`, `Response rate`, `Attrition rate`, state the reasons for dropout of respondents from the study.
- Specify `the end of the study period`, that is, until what date did the follow-up period for this manuscript last (that is, specify the date until the cases of BC were identified and included in this study).

Results section:
Clarify whether `n=257,960` or `n=257,961`.

In the Discussion section, the importance of caution should be taken into account when interpreting the described significant associations of the intake of (poly)phenols with the risk of breast cancer in women, given that only the baseline intake of (poly)phenols was analyzed (without a repeated dietary survey during a long follow-up period). 

Strengths and limitations of the study are discussed in detail.
Explain why a repeated survey was not conducted (on all characteristics, and especially on diet), since the follow-up period is too long to be able to draw conclusions in this way about the relationships between the intake of (poly)phenols and the risk of breast cancer in women.   

Thank you for the opportunity to review the manuscript ID: antioxidants-4157504. This manuscript aimed to examine the association between the intake of total (poly)phenols, and its classes and subclasses, with breast cancer (BC) risk, overall and by subtypes (estrogen, progesterone, and human epidermal growth factor receptor 2 (HER2)), in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.

In order to achieve the goal, the authors conducted a prospective analysis of the EPIC study.
In this paper, the authors in a very high-quality, comprehensive manner in the Introduction section present the gap in the level of knowledge of the link between the intake of total (poly)phenols and BC risk.
Throughout, the authors cite relevant literature. Note: 8/33 references are self-citations.

The authors correctly present the methodology applied in this paper.
The paper presents the results of a cross-sectional study that was conducted in a cohort of participants that were enrolled between 1992 and 2000 in the study.
Comments:
- Describe `Study population`, with a description of eligible persons, and specification of inclusion and exclusion criteria, `Participation rate`, `Response rate`, `Attrition rate`, state the reasons for dropout of respondents from the study.
- Specify `the end of the study period`, that is, until what date did the follow-up period for this manuscript last (that is, specify the date until the cases of BC were identified and included in this study).

Results section:
Clarify whether `n=257,960` or `n=257,961`.

In the Discussion section, the importance of caution should be taken into account when interpreting the described significant associations of the intake of (poly)phenols with the risk of breast cancer in women, given that only the baseline intake of (poly)phenols was analyzed (without a repeated dietary survey during a long follow-up period). 

Strengths and limitations of the study are discussed in detail.
Explain why a repeated survey was not conducted (on all characteristics, and especially on diet), since the follow-up period is too long to be able to draw conclusions in this way about the relationships between the intake of (poly)phenols and the risk of breast cancer in women.   

Author Response

Comment 1: Describe `Study population`, with a description of eligible persons, and specification of inclusion and exclusion criteria, `Participation rate`, `Response rate`, `Attrition rate`, state the reasons for dropout of respondents from the study.

As recommended, we have expanded the description of the Study Population section in the revised manuscript (page 3, lines 113-120). The text now reads: “Overall, 367,903 female participants were recruited from the general population from defined areas in each country with the exception of women who were members of a health insurance programme for state school employees (France), women attending breast cancer screening (Utrecht, the Netherlands and Florence, Italy), blood donors (some centres in Italy and Spain) and vegetarians (the ‘health conscious’ cohort in Oxford, UK) of whom 33,053 were excluded because of prevalent cancer at baseline, missing information on dietary or lifestyle variables or due to an extreme ratio between energy intake and energy requirement.”

Regarding eligibility criteria, EPIC is a large multicentre cohort, and specific eligibility criteria vary slightly by centre/country. To maintain clarity and readability, we prefer to refer readers to the original cohort description, where this information is comprehensively detailed (see table below, ref: Riboli E, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection Public Health Nutr. 2002;5(6B):1113-24.

Concerning participation rate, response rate, attrition rate, and reasons for dropout, these data are not available at the level of the entire EPIC cohort but only for specific centres. For example, in Germany, the participation rate was 22.7% in Potsdam and 38.3% in Heidelberg, with a considerable variation by municipality and gender [Boeing H, et al. Recruitment Procedures of EPIC-Germany. Ann. Nutr. Metab. 1999; 43:205–215]. In the Netherlands, response rates were the highest in Doetinchem (68%), intermediate in Maastricht (45%) and lowest in Amsterdam (34%) [Beulens JWJ, et al. Cohort profile: the EPIC-NL study. Int J Epidemiol. 2010;39(5):1170-8.] and Utrecht (34.5%) [Boker LK, et al. Prospect-EPIC Utrecht: Study Design and Characteristics of the Cohort Population. Eur. J. Epidemiol. 2001;17:1047–1053]. Given the centre-specific nature of these data and the extensive body of previously published EPIC manuscripts describing recruitment procedures, we believe that including all these details in the present manuscript would substantially reduce its conciseness. However, if the Editor considers this information essential, we will incorporate it in the revised version.

Comment 2: Specify `the end of the study period`, that is, until what date did the follow-up period for this manuscript last (that is, specify the date until the cases of BC were identified and included in this study).

Thank you for pointing this out. Each centre has a different data of end of follow-up. We have now added this information in the revised manuscript (page 3, lines 137-138): “Complete follow-up censoring dates varied among centers, ranging between June 2008 and December 2013.”

In the statistical section, it was already stated the beginning and the ending of the study period (page 4, lines 166-168): “Entry time was age at recruitment and exit time was age at diagnosis, death, or censoring date (lost or end of follow-up), whichever occurred first.”

Comment 3: Results section: Clarify whether `n=257,960` or `n=257,961`.

Thank you for pointing this mistake out. We have now corrected to 257,960 participants everywhere.

Comment 4: In the Discussion section, the importance of caution should be taken into account when interpreting the described significant associations of the intake of (poly)phenols with the risk of breast cancer in women, given that only the baseline intake of (poly)phenols was analysed (without a repeated dietary survey during a long follow-up period)  and Comment 5:  Strengths and limitations of the study are discussed in detail.
Explain why a repeated survey was not conducted (on all characteristics, and especially on diet), since the follow-up period is too long to be able to draw conclusions in this way about the relationships between the intake of (poly)phenols and the risk of breast cancer in women.   

We fully agree with the reviewer that administering a second dietary questionnaire during follow-up would have been highly valuable to capture potential changes in dietary habits over time. Unfortunately, the absence of repeated dietary assessment is due solely to financial constraints. Although funding for this purpose was requested on several occasions, these applications were not successful. We, therefore, can only acknowledge this as a limitation of the study and explicitly stated it as such in the manuscript (see page 12, lines 374-375): “Another limitation of this study is that we only have baseline data, which does not allow us to assess changes over the follow-up period.”

Reviewer 2 Report

Strengths

This: is an interesting manuscript detailing the outcomes from a well-controlled EPIC study conducted in 7 European countries including France, UK, and Italy. Their study design, participants recruitment and monitoring, data collection, data analysis, data presentation, and interpretation were all highly acceptable. Their study conclusions were based on their observations and statistical analysis. Overall, this highly interesting and important study and manuscript, which will contribute significant scientific knowledge to the field as well as provide results indicating that there no associations between intake of total polyphenols and BC incidence and protections.

There are several strengths in this manuscript, which include its novelty, the large sample size, excellent experimental designs, inclusion of various breast cancer subtypes, different classes of polyphenols, and the lengthy period for following.

the study participants. In addition, the manuscript is well-written and has adequate citations as well as lists some limitations, The study conclusions are highly relevant. Certainly, the manuscript will attract large array of readers and contribute significant scientific information to the field. Therefore, this is a highly enthusiastic manuscript. However, there are a few questions that the authors must address in revised version.

Recommendation: Minor Revision because the strengths far outweigh the concerns.

Minor Questions/Concerns

1. The study did not ask the number of packs of cigarettes the participants who are smokers actually smoked per week. This may or may not be relevant to the study outcomes, Please comment on it.
2. Self-reporting of anthropometric data in UK and France as compared to the measurement at centers in the other countries. Please, briefly discuss how important this was.
3. Some of the participating European countries especially France and UK have significant percent of residents and citizens of African descent. France has about 8 % to 10 % citizen and residents of African descents. In England it may be between 2.5 and 4 %. What percent of the study participants of 257961 are Blacks and/or of African descent? If there was none or if the number is too small, what strategies were used to improve their participation especially since overall, more Blacks die from BC when compared to people of European descent.
4. Were there any genetic information for breast cancer susceptibility genes (BRCA1 and 2) collected about the participants considering that the percent of TNBC was high and compounded with the observation that France has significant percent of citizens and residents of African descent and who may have inherited BRCA1 and 2 and may not necessarily benefit from protection by polyphenols.
5. Under the discussion, the authors of this manuscript referred to the outcomes of the Southern Cohort Community Studies (SCCS) conducted in the United States, which the authors reported some potential protective benefit from polyphenols against BC. This reference is nice. However, it should be noted that the SCCS participants include 66 % to 70 % African Americans. This was particularly important especially in the US where African Americans die disproportionally from BC than Caucasian Americans. But if the EPIC study has truly little or no participation by people of African descent, it may be difficult to make direct comparison with the outcomes of the SCCS. Please, comment on it.

1. Some of the participating European countries especially France and UK have significant percent of residents and citizens of African descent. France has about 8 % to 10 % citizen and residents of African descents. In England it may be between 2.5 and 4 %. What percent of the study participants of 257961 are Blacks and/or of African descent? If there was none or if the number is too small, what strategies were used to improve their participation especially since overall, more Blacks die from BC when compared to people of European descent
2. Were there any genetic information for breast cancer susceptibility genes (BRCA1 and 2) collected about the participants considering that the percent of TNBC was high and compounded with the observation that France has significant percent of citizens and residents of African descent and who may have inherited BRCA1 and 2 and may not necessarily benefit from protection by polyphenols.

Author Response

Comment 1: The study did not ask the number of packs of cigarettes the participants who are smokers actually smoked per week. This may or may not be relevant to the study outcomes, please comment on it.

Thank you for this insightful comment. The EPIC study includes comprehensive information on smoking habits, covering both qualitative variables (e.g., smoking status) and quantitative measures (e.g., number of cigarettes per day). To avoid unnecessary overcomplication and overadjustment of the statistical models, we chose to adjust for smoking status only, as we have done in our previous analyses of diet–cancer associations within EPIC. An exception is made for cancer sites where tobacco is a major risk factor (e.g., lung cancer), where more detailed adjustment is mandatory. In the present analysis, we additionally ran the main models with and without the variable number of cigarettes per day and the results remained the same. Therefore, to maintain model parsimony, we prefer to retain the simpler model adjusting for smoking status only.

Comment 2: Self-reporting of anthropometric data in UK and France as compared to the measurement at centres in the other countries. Please, briefly discuss how important this was.

Thank you for highlighting this important point. Although anthropometric data was self-reported in France and UK, these measurements were validated for identifying relationships in epidemiological studies. We have now clarified this in the main text (page 4, lines 154-155), as follows: Additionally, anthropometric data were mostly measured in all centers at recruitment, except in Oxford (UK) and France where they were self-reported; however, these self-reported measures were previously validated (Spencer EA, et al. Validity of self-reported height and weight in 4808 EPIC-Oxford participants. Public Health Nutr. 2002;5(4):561–5.)

Comment 3: Some of the participating European countries especially France and UK have significant percent of residents and citizens of African descent. France has about 8 % to 10 % citizen and residents of African descents. In England it may be between 2.5 and 4 %. What percent of the study participants of 257961 are Blacks and/or of African descent? If there was none or if the number is too small, what strategies were used to improve their participation especially since overall, more Blacks die from BC when compared to people of European descent.

Thank you for raising this important point. In the EPIC cohort, the proportion of participants from races other than Caucasian is negligible. At the time the study was initiated in the nineties, the proportion of individuals of African descent or other racial/ethnic minorities in the participating countries was considerably lower than it is today, and participation rates among these groups were also limited. Moreover, no specific recruitment strategies were implemented to increase participation from underrepresented racial/ethnic groups. We agree that this limits the generalizability of our findings. We have now explicitly added this as a limitation in the revised manuscript (page 12, lines 386-387): “The findings of this study cannot be generalized to other racial/ethnic groups, as the study population consisted almost exclusively of Caucasian participants.”

Comment 4: Were there any genetic information for breast cancer susceptibility genes (BRCA1 and 2) collected about the participants considering that the percent of TNBC was high and compounded with the observation that France has significant percent of citizens and residents of African descent and who may have inherited BRCA1 and 2 and may not necessarily benefit from protection by polyphenols.

We thank the reviewer for this insightful comment. data on breast cancer susceptibility genes are available for only approximately 3,000 women with invasive breast cancer (around 27% of total cases), which makes it unfeasible to adequately evaluate genetic differences in triple-negative breast cancer within our dataset. Moreover, EPIC, including the E3N cohort (EPIC-France), predominantly recruited Caucasian participants. Although the reviewer’s hypothesis is scientifically plausible, it is not possible to assess it with our data. For this reason, we prefer not to include this hypothesis in the manuscript, as we cannot appropriately evaluate it in our study population.

Comment 5: Under the discussion, the authors of this manuscript referred to the outcomes of the Southern Cohort Community Studies (SCCS) conducted in the United States, which the authors reported some potential protective benefit from polyphenols against BC. This reference is nice. However, it should be noted that the SCCS participants include 66 % to 70 % African Americans. This was particularly important especially in the US where African Americans die disproportionally from BC than Caucasian Americans. But if the EPIC study has truly little or no participation by people of African descent, it may be difficult to make direct comparison with the outcomes of the SCCS. Please, comment on it.

We agreed with reviewer comment. Our study population consists almost exclusively of Caucasian participants, therefore, ethnicity may partly explain some of the observed differences. We have now incorporated this point in the text (page 12, lines 353-): “Furthermore, some protective associations appear to vary by menopausal status and tumor receptor subtypes, as observed in a similar study based in the Southern Community in the United States, a population composed with a 66.5% of African American, for total (poly)phenol intake they observed a reduced risk of BC incidence in ER/PR positive subtype and between phenolic acids and overall BC in postmenopausal women [31]. These findings add complexity to the overall understanding and suggest potential differences by ethnicity [31], particularly considering that in EPIC predominantly Caucasian participants were recruited.” and in the limitations (page 12, lines 386-387): “The findings of this study cannot be generalized to other racial/ethnic groups, as the study population consisted almost exclusively of Caucasian participants.”.