Synergistic action of folate intake and testosterone associated with breast cancer risk

Journal Pre-proof Synergistic action of folate intake and testosterone associated with breast cancer risk

Luisa María Sánchez-Zamorano, Angélica Angeles-Llerenas, Aaron Salinas-Rodríguez, Eduardo C. Lazcano-Ponce, Isabelle Romieu, Edelmiro Pérez-Rodríguez, Lourdes Flores-Luna, Gabriela Torres-Mejía PII:

S0271-5317(19)30030-2

DOI:

https://doi.org/10.1016/j.nutres.2019.10.002

Reference:

NTR 8052

To appear in:

Nutrition Research

Received date:

18 January 2019

Revised date:

9 August 2019

Accepted date:

4 October 2019

Please cite this article as: L.M. Sánchez-Zamorano, A. Angeles-Llerenas, A. SalinasRodríguez, et al., Synergistic action of folate intake and testosterone associated with breast cancer risk, Nutrition Research(2019), https://doi.org/10.1016/j.nutres.2019.10.002

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof Synergistic action of folate intake and testosterone associated with breast cancer risk

Luisa María Sánchez-Zamorano Population Health Research Center, National Institute of Public Health, Avenida Universidad No. 655, Colonia Santa María Ahuacatitlán, Cuernavaca, Morelos 62100, México, [email protected]

Angélica Angeles-Llerenas Population Health Research Center, National Institute of Public Health, Avenida Universidad No. 655, Colonia Santa María Ahuacatitlán, Cuernavaca, Morelos 62100, México, [email protected]

of

Aaron Salinas-Rodríguez Evaluation and Surveys Research Center, National Institute of Public Health, Avenida

ro

Universidad No. 655, Colonia Santa María Ahuacatitlán, Cuernavaca, Morelos 62100, México, [email protected]

Eduardo C. Lazcano-Ponce Population Health Research Center, National Institute of Public Health, Avenida

-p

Universidad No. 655, Colonia Santa María Ahuacatitlán, Cuernavaca, Morelos 62100, México, [email protected]

Isabelle Romieu International Agency for Research on Cancer (IARC), 150 cours Albert Thomas 69372 Lyon Cedex 08

re

France, [email protected]

lP

Edelmiro Pérez-Rodríguez Hospital Universitario, Dr. José Eleuterio González, Gonzalitos 235, Mitras Centro, 64460 Monterrey, N.L., Mexico, [email protected]

Lourdes Flores-Luna Population Health Research Center, National Institute of Public Health, Avenida Universidad No.

na

655, Colonia Santa María Ahuacatitlán, Cuernavaca, Morelos 62100, México, [email protected]

Gabriela Torres-Mejía, Population Health Research Center, National Institute of Public Health, Avenida Universidad No.

Jo

ur

655, Colonia Santa María Ahuacatitlán, Cuernavaca, Morelos 62100, México, [email protected]

Corresponding author

Gabriela Torres-Mejía, Chronic Diseases Department, Population Health Research Center, National Institute of Public Health, Av. Universidad 655, Col. Sta. Ma. Ahuacatitlán, Cuernavaca, Mor., Mexico, CP 62100, Tel: (52)(777)311 23 43, email: [email protected]

Journal Pre-proof

List of abbreviations

BC; Breast cancer T; Tertile

of

Rpm; Revolutions per minute

ro

FFQ; Food Frequency Questionnaire

-p

IQR; Inter-quartile range

re

ENSANUT; Encuesta Nacional de Salud y Nutrición (in Spanish)

Jo

ur

na

lP

CONACyT; Consejo Nacional de Ciencia y Tecnología (in Spanish)

Journal Pre-proof Abstract

The amount of irreparable DNA damage is a function of the rate of cell division, and the association between sex hormones and the risk of breast cancer has been explained by an increase in cell division. Folate intake insufficiency leads to disturbances in DNA replication and DNA repair. We hypothesized that folate intake insufficiency and high

of

serum concentrations of sex hormones act synergistically on the risk of breast cancer.

ro

The aim of this study was to investigate the interaction between sex hormones

-p

(exposure of interest A) and dietary folate intake (exposure of interest B) on the risk of

re

breast cancer. We included 342 breast cancer primary postmenopausal cases and 294 controls obtained from a large population-based case control study. Multiple conditional

lP

logistic regression models were used for the analysis and interactions were tested. The

na

joint effect of the lowest dietary folate intake (T1 <259.40 mg/d) and the highest serum concentration of testosterone (T3 ≥0.410 on the risk of breast cancer was OR = 9.18

ur

(95% CI 2.56-32.88), when compared to the lowest risk category, namely, the group of

Jo

women with the highest dietary folate intake (T3 >381.29 mg/d) and the lowest serum concentration of testosterone T1 ≤0.25 pg/mL. There were some indications that the estimated join effect was greater than the product of the estimated effects alone (P = 0.001). These findings have important public health implications with respect to reducing the risk of the most frequent cancer in women worldwide.

Keywords Testosterone, Estradiol, Folate, Breast Cancer, Postmenopausal women

Journal Pre-proof 1. Introduction

It is likely that malignant cellular transformation occurs during cell division [1], and the amount of irreparable DNA damage is a function of the rate of cell division [1]. A positive association of the concentration of estrogen and testosterone with the risk of breast

of

cancer has been documented [2, 3], which has been explained by an increase in the

ro

cell division driven by estrogen [4-9] and a stimulating effect on the replication of certain

-p

BC cell lines by testosterone [10-14]. Additionally, it has been reported that both

lP

re

estrogens and androgens favor tumor growth promotion in the mammary gland [15, 16].

na

Some studies have shown that concentrations of androgens, whether in serum [17, 18] or plasma [19], are associated with the development of BC. However, it is not clear if

ur

circulating testosterone, per se, is linked to the growth of the tumor or if it is a surrogate

Jo

for the increased local production of estradiol [13]. The increased risk of BC related to endogenous circulating testosterone has been observed more frequently in longitudinal studies performed in postmenopausal women than in premenopausal women [13, 20].

Folate intake insufficiency leads to disturbances in DNA replication and DNA repair [21, 22]. Some epidemiological studies have shown that dietary folate intake has a protective effect against the risk of breast cancer [23-28].

Journal Pre-proof

Based on the previous statements, we hypothesized that folate intake insufficiency and high serum concentration of sex hormones act synergistically on the risk of breast cancer. The aim of this study was to investigate the interaction between sex hormones (exposure of interest A) and dietary folate intake (exposure of interest B) on the risk of

of

breast cancer using data from a large population-based case control study conducted to

re

-p

ro

assess BC risk factors.

na

lP

2. Methods and materials

Jo

ur

2.1 Design and study population

For the present study, we used information from a subsample of a large multicenter population-based case-control study conducted from 2002 to 2007 in three cities of Mexico: Monterrey (Northern region), Veracruz (East coast) and Mexico City (Center). The motivation, methods, and design of the project entitled “Factors for Breast Cancer in Mexico: Mammographic Patterns, C Peptide, and Growth Factors, a Multi-Center Study (CAMA)”, have been previously described [23, 29-31]. Briefly, the study included pre- and postmenopausal women aged 35 to 69 years, who lived in one of the study areas for the past 5 years.

Journal Pre-proof

Cases were recruited from one of twelve participating hospitals that belong to the three major health care systems in Mexico. Cases were excluded from the study if they had received a breast cancer treatment in the past 6 months, were currently using aromatase inhibitors, were pregnant, were HIV positive, or were using antiestrogens. In

of

total, 1,000 women with a new histologically confirmed

ro

diagnosis of breast cancer were enrolled soon after diagnosis (median = 3 days; range

-p

= 0–6 days). The response rate was high (96 % for Mexico City, 94 % for Monterrey,

lP

re

and 97 % for Veracruz).

na

Controls (n = 1074) were selected based on a probabilistic multistage sampling design [29] and were frequency-matched to the cases, according to 5-year age groups, health

ur

care system and place of residence. Personnel, engaged in the study, visited the

Jo

selected households and determined the subjects’ willingness to participate in the study. Once women agreed to participate, signed an informed consent form. An appointment was scheduled for each woman to attend the hospital for anthropometric measurements and a blood sample. The controls’ response rates were high (87 % for Mexico City, 90 % for Monterrey, and 98 % for Veracruz areas). Enrollment of cases and controls occurred at the same time, between 2004 and 2007.

Journal Pre-proof Study subjects provided information about their health, physical activity and diet by means of a face to face interview. Diet information was obtained by asking women about food consumption the year prior to the onset of the symptoms. Diet was assessed using a 104-item semi-quantitative food frequency questionnaire (FFQ) adapted from Willett [32] to the Mexican population and validated in Mexico City [33]. This questionnaire included 104 items and 10 multiple choice frequency categories of

of

consumption. We used the US Department of Agriculture food composition tables and

ro

[34], when necessary, the nutrient database developed by the National Institute of

re

-p

Nutrition in Mexico [35].

This collaborative study was approved by the Institutional

lP

Review Board at the National Institute of Public Health (Number CI-349) and by the

na

equivalent committees at the collaborating hospitals. All participants in the study

Jo

ur

provided written informed consent.

2.2. Sub-sample

For the present study, out of 1000 cases and 1074 controls, we considered 578 postmenopausal cases and 586 controls. Out of them and excluding women who were on hormone replacement therapy (n = 227), we obtained a random subsample in whom

Journal Pre-proof we measured the free serum concentrations of estrogens and testosterone (342 cases and 294 controls) (Fig.1).

of

2.3 Menopause

ro

Menopause was defined either as a woman’s self-report of experiencing natural

-p

menopause (12 months or longer since her last period), a report of induced menopause

re

(with bilateral oophorectomy) or a report of hysterectomy if a woman did not know that her ovaries had been removed but was 48 years or older, given that the mean age of

na

lP

menopause in Mexico for the study period was 48 years [36, 37].

Jo

ur

2.4 Estrogens and testosterone

After each woman had fasted for at least 8 hours, trained nurses obtained blood samples from them by venipuncture. The nurses centrifuged the samples at 2500 rpm for 10 minutes; subsequently, they separated the serum, aliquoted it into cryovials, and stored the samples between -20 and -70°C at the hospital where the samples were obtained. Within a period of no longer than 3 weeks, the samples were transferred from the participating hospitals to the National Institute of Public Health, where they were stored in a freezer at -70°C until analysis. The measurements of serum free fraction

Journal Pre-proof estradiol and testosterone concentrations were performed at the National Institute of Medical Sciences and Nutrition Salvador Zubiran in Mexico City, using liquid chromatography coupled with mass spectrometry (variation coefficient of 9.5%) and radioimmunoassay (variation coefficient of 8.1%), respectively. The laboratory personnel performing the hormonal assays were blinded as to the case-control status of the study subjects. The serum concentrations were classified into tertiles according to

of

their distributions in the control group. The serum free fraction of estradiol was classified

ro

as follows: tertile T1 < 0.181 pg/mL; T2 = 0.181 – 0.32 pg/mL; and T3 > 0.32 pg/mL.

-p

The median free estradiol in controls was 0.26 pg/mL with an IQR (interquartile range)

re

of 0.14-0.45 pg/mL. The serum free fraction of testosterone was classified as follows: T1 < 0.251 pg/mL; T2 = 0.251-0.500 pg/mL; and T3 > 0.500 pg/mL. The median value

Jo

ur

2.5 Dietary folate intake

na

lP

of free testosterone in controls was 0.40 pg/mL with an IQR of 0.26-0.60 pg/ mL.

Folate intake in these women has been previously reported by Beasley et al. (2010)[23]. Briefly, folate intake was computed from their FFQ responses by multiplying the frequency response by the nutrient content of the specified portion sizes. To calculate folate intake, we used the nutrient database that was developed by the National Institute of Nutrition in Mexico [38] and, if necessary, the US Department of Agriculture food composition tables [39]. The participants were asked to report the frequency of

Journal Pre-proof consumption of a typical serving of any item on a previously published list of 104 items during the one year before symptoms were perceived for the first time [30]. The responses were converted to the mean daily consumption. Subsequently, the variables were classified into tertiles according to their distribution in the control group, with dietary folate being classified as follows: T1 < 259.40 g/d; T2 = 259.40 – 381.29 g/d; and T3 > 381.29 g/d. The mean dietary folate intake in controls was 322.7 with an IQR

-p

ro

of

from 248.6 to 436.4.

lP

re

2.6 Statistical analyses

na

We compared sociodemographic characteristics, history of diabetes, benign breast disease, breast cancer family history, and life style variables, Western dietary pattern

ur

and dietary folate intake using the Chi2 test (Table 1). Sex hormones, age, time since

Jo

menopause, anthropometric measurements, breastfeeding, energy and dietary folate intake were compared between cases and controls using Kruskal-Wallis rank sum tests (Table 2). The power to find a statistically significant interaction between the highest tertile of testosterone and the lowest tertile of dietary folate consumption was greater than 80%; for the rest of the combinations, it was less than 50% [40]. To investigate the effect of the interaction between sex hormones (exposure of interest A) and dietary folate intake (exposure of interest B) on the risk of breast cancer (outcome) the following steps were taken [41-43].

Journal Pre-proof

1. First, we estimated the ORs and 95% CIs for the lowest dietary folate intake (T1 <259.40 mg/d) and (T2 = 259.40 – 381.29 mg/d) vs (T3 > 381.29 mg/d) within strata of serum concentrations of sex hormones (tertiles) and for the highest serum concentration of testosterone (T3 ≥0.510 pg/mL) and (T2 = 0.251 - 0.500 pg/mL) vs (T1 ≤0.25 pg/mL) within strata of dietary folate intake (tertiles).

of

Second, we estimated the joint effect of the lowest dietary folate intake and the

ro

highest serum concentration of testosterone (the highest risk category) vs the

-p

lowest risk category (the highest dietary folate intake and the lowest serum

re

concentration of testosterone) (Table 3) by using conditional logistic regression

na

lP

models. Third, we performed the same analyses for estrogens.

2. We calculated the relative excess risk due to interaction (RERI) to estimate the

ur

joint effect on the additive scale [44] and the measure of the ratio of ORs to

Jo

estimate the joint effect on the multiplicative scale [45]. For additive and multiplicative interactions, we calculated the 95% CIs applying the delta method implemented in Stata software [46-48].

3. Logistic regression models were adjusted for design by 5-year age group, health care, institution membership and place of residence. We also adjusted for time since menopause (years), socioeconomic status (SES) in tertiles according to its

Journal Pre-proof distribution in the control group (low, medium and high), breastfeeding (in months), BMI (kg/m2) (normal, overweight and obese, according to WHO), family history of BC (yes/no), history of diabetes (yes/no), total intake of calories (kcal), Western dietary pattern as described previously [30], energy residuals [32], height (cm), benign breast disease (yes/no), physical activity (weekly hours of moderate and vigorous-intensity physical activity), consumption of alcohol (never,

of

less than a gram per day and one or more grams per day) and tobacco

ro

consumption (lifetime history of smoking more than 100 cigarettes) [30]. Energy

-p

residuals were obtained to adjust for the energy intake not explained by the

lP

re

Western dietary pattern.

na

Given that several statistical tests were performed simultaneously, the P-value for each interaction was corrected using Bonferroni method [49] (alpha value = 0.00625). All

Jo

TX, USA).

ur

analyses were performed using STATA v14 software (StataCorp LP, College Station,

3. Results

In Table 1, the distributions of cases and controls according to different characteristics shows that in comparison to controls, cases came from the highest socioeconomic level

Journal Pre-proof (p <0.001), reported less breastfeeding (p<0.001), had a higher history of benign breast disease (p = 0.001), engaged in less physical activity (p <0.001), consumed more alcohol (p = 0.001), had a more Western dietary pattern (p = 0.005) and higher dietary folate intake (p = 0.006). Regarding the distribution of the medians of serum sex hormone concentrations according to cases and controls, the free fraction of

ro

of

testosterone was higher in cases than in controls (p = 0.001) (Table 2).

-p

The effect of the lowest dietary folate intake (T1 <259.40 mg/d vs T3 >381.29 mg/d) on the risk of breast cancer in women with the lowest serum concentration of testosterone

re

T1 ≤0.25 pg/mL was OR = 5.12 (95% CI 1.36- 19.25). The effect of the highest serum

lP

concentration of testosterone (T3 ≥0.410 vs T1 ≤0.25 pg/mL) on the risk of breast

na

cancer in women with the highest dietary folate intake (T3 >381.29 mg/d) was OR = 4.41 (95% CI 1.70-11.44). The joint effect of the lowest dietary folate intake (T1 <259.40

ur

mg/d) and the highest serum concentration of testosterone (T3 ≥0.410) was OR = 9.18

Jo

(95% CI 2.56-32.88), where the highest dietary folate intake (T3 >381.29 mg/d) and the lowest serum concentration of testosterone T1 ≤0.25 pg/mL were the reference category. There was an interaction on the multiplicative scale but not on the additive scale (Table 3 and Fig. 2A). The effect of the lowest dietary folate intake (T1 <259.40 mg/d vs T3 >381.29 mg/d) on the risk of breast cancer in women with the lowest serum concentration of estrogens T1 ≤0.18 pg/mL was OR = 3.02 (95% CI 1.12- 8.13). The effect of the highest serum concentration of estrogens (T3 ≥0. 0.321 vs T1 ≤0.18 pg/mL) on the risk of breast cancer in women with the highest dietary folate intake (T3 >381.29

Journal Pre-proof mg/d) was OR = 1.62 (95% CI 0.75-3.48). We found no interaction between estrogens and dietary folate intake (Table 4 and Fig 2B).

of

4. Discussion

ro

The most remarkable finding of the present study was that the odds of having BC

-p

increased with increasing tertiles of serum testosterone but not significantly with increasing tertiles of estrogens. We also found that the odds of having breast cancer

re

increased with decreasing tertiles of dietary folate intake. Furthermore, we found a joint

lP

effect on the risk of breast cancer of the lowest dietary folate intake (T1 <259.40 mg/d)

na

and the highest serum concentration of testosterone (T3 ≥0.410), when compared to women with the lowest risk according to our hypothesis: highest dietary folate intake (T3

ur

>381.29 mg/d) and the lowest serum concentration of testosterone (T1 ≤0.25 pg/mL).

Jo

We found no interaction between estrogens and dietary folate intake. In addition, 65.8% of the women participating in this study had a lower dietary folate intake than the recommended value (455 μg/d) [50].

We found that the odds of having BC increased with increasing tertiles of serum testosterone but not with increasing tertiles of estrogens. Consistent with our results, a study showed that high circulating levels of total and free testosterone were associated with the risk of developing breast cancer in postmenopausal women, while the

Journal Pre-proof association between circulating estradiol levels and the risk of breast cancer was not statistically significant [18]. Several prospective studies have shown strong associations of testosterone with breast cancer risk in postmenopausal women [51, 52]. Similar conclusions have been obtained from case control studies [52]. Furthermore, a study showed that a high total testosterone level was significantly associated with an increased risk of estrogen receptor-positive cancers, [18] most likely because close to

of

70% of breast tumors are found to express estrogen receptors, and approximately 90%

re

-p

ro

of them are also androgen receptor-positive [11, 53, 54].

Proposed mechanisms that could explain why testosterone increases the risk of BC,

lP

include conversion to estrogen by aromatase enzyme in breast tissue, or direct

na

stimulation of the androgen receptor, which is expressed in the epithelial cells of the normal mammary gland and in BC cell lines [10, 14, 20, 55]. Research performed in BC

ur

cell lines with androgen receptors has shown that testosterone stimulates their

Jo

proliferation independently of the presence of estrogen receptors. Proliferation of MDAMB-231 BC cell line, which is negative for estrogen receptors [56, 57] and the MCF-7 cell line, which is positive for estrogen receptors, was stimulated by testosterone [9]. In addition, dihydrotestosterone (DHT) and testosterone stimulate the proliferation of the MDA-MB-453 cell line, which is positive for androgen receptors and negative for estrogen receptors [14].

Journal Pre-proof Another possible mechanism could be that metabolic syndrome is associated with both, increased testosterone levels [58, 59] and breast cancer risk [60]. It has been observed that the postmenopausal ovary is an androgen secreting organ and that levels of testosterone are not directly influenced by the menopausal transition or the occurrence of menopause [61, 62] but that they fall slowly with age [62, 63]. In contrast with our results, several studies suggest that there is no effect or that testosterone reduces the

of

risk of breast cancer by antagonizing the effects of estrogen on mammary tissue [64].

ro

Similarly, another study found no significant associations between breast cancer risk

re

-p

and levels of androgens such as testosterone and free testosterone [65].

lP

Although the etiology of BC is closely linked to high lifetime exposure to estrogens, in our study we did not find a significant increased risk of this cancer in those within the highest stratum of

na

folate consumption. There are discrepancies among studies attempting to establish an

ur

association between high estrogen levels and the risk of BC. Hilakivi-Clarke, et al. (2002) proposed that estrogens can increase, decrease, or have no effect on breast cancer risk,

Jo

depending on the timing of estrogen exposure [66]. Estradiol decreases drastically during

the menopausal transition, while bioavailable testosterone decreases only 28% from 25 to 85 years of age [67]. For example, a Japanese cohort study showed that an increased serum concentration of estrogens during premenopause increased the risk of BC in postmenopausal women [68].

Journal Pre-proof We found that the odds of having BC increased with decreasing tertiles of dietary folate intake. There is evidence that reduced dietary folate intake leads to disturbances in DNA replication and DNA repair [21], and its deficiency has been associated with DNA chain breakage and alterations in DNA repair. All these predispose individuals to carcinogenesis [22]. Our results showed that 65.8% of the women had a lower folate intake than that recommended by the WHO (400 μg/d). This is consistent with what was

of

reported in the National Health and Nutrition Survey (ENSANUT 2012), in which the

re

-p

ro

median folate consumption in Mexican adults was 335.1 μg [69].

We found a joint effect on the risk of breast cancer of the lowest dietary folate intake

lP

and the highest serum concentration of testosterone (highest risk) when compared to

na

women with the highest dietary folate intake and the lowest serum concentration of testosterone (lowest risk). To the best of our knowledge, this interaction has not yet

ur

been documented. Similar to what has been documented for estrogens, it is possible

Jo

that in normal cells, a high rate of cellular proliferation induced by testosterone might lead to the accumulation of DNA adducts and possible mutations. This could be due to cells having less time to repair DNA due to the high rate of cellular proliferation [66], and dietary folate intake deficiency has been associated with DNA chain breakage and alterations in DNA repair, all of which predispose an individual to neoplastic transformation [22]. We found no interaction between estrogens and dietary folate intake probably because serum concentrations of estrogens were measured after menopause, and they do not correspond to the time of exposure necessary to increase

Journal Pre-proof the risk. In our study, the serum concentrations of the sex hormones were lower than those reported by others in postmenopausal women [70-72].

The strengths of our study were that our controls were selected from the same population study base in which the breast cancer cases occurred, such that if they had become cases they would have been selected in our study [29]. We trained and

of

standardized the participating nurses with regard to recruitment, anthropometric

ro

measurements, blood collection, and interviewing women from three different

-p

subpopulations with heterogeneous lifestyles and cancer risks [29]. We measured free

re

testosterone and free estrogens. This is important because less than 2% of testosterone circulates in a free state, 60 to 65% is bound to sex hormone binding protein (SHBG),

lP

35 to 40% is bound to albumin, and less than 5% is bound to corticosteroid-binding

na

globulin [20]. Although both our free estradiol and free testosterone serum concentrations were lower than the concentrations reported by others in

ur

postmenopausal women [70-72], the ratio of testosterone/estrogen was similar to that in

Jo

the review reported by Yasui et al. (2012) [20]. This ratio increases during the menopausal transition and increases significantly in postmenopausal stages [20].

Regarding our limitations, our study is a case control study, and therefore, our results could be due to inverse causation that might be caused by alterations in endogenous hormone levels due to metabolic effects of large tumors, psychological stress and lifestyle changes after cancer diagnosis. To reduce this potential bias, we obtained FFQ information by asking women about their food consumption, for a usual week, the year

Journal Pre-proof prior to the onset of the symptoms [29, 73]. Sex hormone serum concentrations were obtained at the time of the diagnosis of BC. However, it has been shown that the current estradiol and testosterone serum hormone concentrations correlate with measurements five years before (0.80 (95% CI:0.73-0.87) and 0.71 (95% CI:0.62-0.80), respectively) [74]. The postmenopausal ovary is hormonally active, contributing significantly to the circulating pool of testosterone. Furthermore, this contribution appears to persist in

ro

of

women for a long period of time (10 years beyond menopause) [75].

-p

In conclusion, high serum concentration of testosterone may increase the risk of breast

re

cancer. Our results support the hypothesis of an interaction between a high free serum concentration of testosterone and a low dietary intake of folate. The mechanisms of this

lP

interaction need to be further investigated. Our results suggest improving dietary intake

women.

Jo

Acknowledgments

ur

na

recommendations not only for women during pregnancy but also for postmenopausal

We would like to thank the doctors and nurses in the participating hospitals for their support: Germán Castelazo (IMSS, Mexico City, DF), Sinhué Barroso Bravo (IMSS, Mexico City, DF), Joaquín Zarco Méndez (ISSSTE, Mexico City, DF), Edelmiro Pérez Rodríguez (University Hospital, Monterrey, Nuevo León), Jesús Pablo Esparza Cano (IMSS, Monterrey, Nuevo León), Heriberto Fabela (IMSS, Monterrey, Nuevo León), Fausto Hernéndez Morales (ISSSTE, Veracruz, Veracruz), Pedro Coronel Brizio (CECAN SS, Xalapa, Veracruz), Vicente A. Saldaña Quiroz (IMSS, Veracruz,

Journal Pre-proof Veracruz), and Fernando Mainero-Ratchelous (IMSS, Mexico City). Ma. Felix Lazcano López, Silvia Cardoso Muñoz, Libia Zulema Franco Velázquez and Jenny Tejeda Espinoza. Author contributions were as follows. LMSZ: Conceptualization, writing and formal analysis. AALl: Conceptualization, writing, critical review, interpretation of data and acquisition of data. ASR: Critical reading, formal analysis. ECLP: critical reading. IR: critical reading. EPR: acquisition of data, critical reading. LFL: Critical reading. GTM:

of

Conceptualization, acquisition of data, writing, interpretation of data, critical review and

ro

supervision. This work was supported by the Mexican National Council of Science and

-p

Technology (CONACyT: SALUD-2002-C01-7462), which provided financial support to conduct the research (study design and the collection of data). The “Dirección General

re

de Programación Organización y Presupuesto, Ejercicio fiscal 2011, Ramo 12 Salud”

lP

provided financial support for the measurements of the serum concentrations of

Jo

ur

na

hormones. The authors declare no conflicts of interest.

Journal Pre-proof References

Jo

ur

na

lP

re

-p

ro

of

[1] Preston-Martin S, Pike MC, Ross RK, Jones PA, Henderson BE. Increased cell division as a cause of human cancer. Cancer Res 1990;50:7415-21. [2] Grattarola R. Androgens in breast cancer. I. Atypical endometrial hyperplasia and breast cancer in married premenopausal women. Am J Obstet Gynecol 1973;116:423-8. [3] Grattarola R, Secreto G, Recchione C, Castellini W. Androgens in breast cancer. II. Endometrial adenocarcinoma and breast cancer in married postmenopausal women. Am J Obstet Gynecol 1974;118:173-8. [4] Pike MC, Spicer DV, Dahmoush L, Press MF. Estrogens, progestogens, normal breast cell proliferation, and breast cancer risk. Epidemiol Rev 1993;15:17-35. [5] Russo J, Russo IH. Genotoxicity of steroidal estrogens. Trends Endocrinol Metab 2004;15:211-4. [6] Travis RC, Key TJ. Oestrogen exposure and breast cancer risk. Breast Cancer Res 2003;5:239-47. [7] Deroo BJ, Korach KS. Estrogen receptors and human disease. J Clin Invest 2006;116:561-70. [8] Endogenous H, Breast Cancer Collaborative G, Key TJ, Appleby PN, Reeves GK, Roddam AW, et al. Circulating sex hormones and breast cancer risk factors in postmenopausal women: reanalysis of 13 studies. Br J Cancer 2011;105:709-22. [9] Lippman M, Bolan G, Huff K. The effects of estrogens and antiestrogens on hormone-responsive human breast cancer in long-term tissue culture. Cancer Res 1976;36:4595-601. [10] Garay JP, Karakas B, Abukhdeir AM, Cosgrove DP, Gustin JP, Higgins MJ, et al. The growth response to androgen receptor signaling in ERalpha-negative human breast cells is dependent on p21 and mediated by MAPK activation. Breast Cancer Res 2012;14:R27. [11] Gonzalez LO, Corte MD, Vazquez J, Junquera S, Sanchez R, Alvarez AC, et al. Androgen receptor expresion in breast cancer: relationship with clinicopathological characteristics of the tumors, prognosis, and expression of metalloproteases and their inhibitors. BMC Cancer 2008;8:149. [12] Fourkala EO, Blyuss O, Field H, Gunu R, Ryan A, Barth J, et al. Sex hormone measurements using mass spectrometry and sensitive extraction radioimmunoassay and risk of estrogen receptor negative and positive breast cancer: Case control study in UK Collaborative Cancer Trial of Ovarian Cancer Screening (UKCTOCS). Steroids 2016;110:62-9. [13] McNamara KM, Moore NL, Hickey TE, Sasano H, Tilley WD. Complexities of androgen receptor signalling in breast cancer. Endocr Relat Cancer 2014;21:T161-81. [14] Birrell SN, Bentel JM, Hickey TE, Ricciardelli C, Weger MA, Horsfall DJ, et al. Androgens induce divergent proliferative responses in human breast cancer cell lines. J Steroid Biochem Mol Biol 1995;52:459-67. [15] Dorgan JF, Longcope C, Stephenson HE, Jr., Falk RT, Miller R, Franz C, et al. Relation of prediagnostic serum estrogen and androgen levels to breast cancer risk. Cancer Epidemiol Biomarkers Prev 1996;5:533-9. [16] Kim YI. Does a high folate intake increase the risk of breast cancer? Nutr Rev 2006;64:468-75. [17] Secreto G, Venturelli E, Meneghini E, Greco M, Ferraris C, Gion M, et al. Testosterone and biological characteristics of breast cancers in postmenopausal women. Cancer Epidemiol Biomarkers Prev 2009;18:2942-8. [18] Sieri S, Krogh V, Bolelli G, Abagnato CA, Grioni S, Pala V, et al. Sex hormone levels, breast cancer risk, and cancer receptor status in postmenopausal women: the ORDET cohort. Cancer Epidemiol Biomarkers Prev 2009;18:169-76.

Journal Pre-proof

Jo

ur

na

lP

re

-p

ro

of

[19] Hankinson SE, Willett WC, Manson JE, Colditz GA, Hunter DJ, Spiegelman D, et al. Plasma sex steroid hormone levels and risk of breast cancer in postmenopausal women. J Natl Cancer Inst 1998;90:1292-9. [20] Yasui T, Matsui S, Tani A, Kunimi K, Yamamoto S, Irahara M. Androgen in postmenopausal women. J Med Invest 2012;59:12-27. [21] Kawakita D, Lee YA, Gren LH, Buys SS, La Vecchia C, Hashibe M. The impact of folate intake on the risk of head and neck cancer in the prostate, lung, colorectal, and ovarian cancer screening trial (PLCO) cohort. Br J Cancer 2018;118:299-306. [22] Kim YI. Folate and carcinogenesis: evidence, mechanisms, and implications. J Nutr Biochem 1999;10:66-88. [23] Beasley JM, Coronado GD, Livaudais J, Angeles-Llerenas A, Ortega-Olvera C, Romieu I, et al. Alcohol and risk of breast cancer in Mexican women. Cancer Causes Control 2010;21:863-70. [24] Fagherazzi G, Vilier A, Boutron-Ruault MC, Mesrine S, Clavel-Chapelon F. Alcohol consumption and breast cancer risk subtypes in the E3N-EPIC cohort. Eur J Cancer Prev 2014. [25] Castillo LC, Tur JA, Uauy R. [Folate and breast cancer risk: a systematic review]. Rev Med Chil 2012;140:251-60. [26] Lajous M, Romieu I, Sabia S, Boutron-Ruault MC, Clavel-Chapelon F. Folate, vitamin B12 and postmenopausal breast cancer in a prospective study of French women. Cancer Causes Control 2006;17:1209-13. [27] Ericson U, Sonestedt E, Gullberg B, Olsson H, Wirfalt E. High folate intake is associated with lower breast cancer incidence in postmenopausal women in the Malmo Diet and Cancer cohort. Am J Clin Nutr 2007;86:434-43. [28] Chen P, Li C, Li X, Li J, Chu R, Wang H. Higher dietary folate intake reduces the breast cancer risk: a systematic review and meta-analysis. Br J Cancer 2014;110:2327-38. [29] Angeles-Llerenas A, Ortega-Olvera C, Perez-Rodriguez E, Esparza-Cano JP, Lazcano-Ponce E, Romieu I, et al. Moderate physical activity and breast cancer risk: the effect of menopausal status. Cancer Causes Control 2010;21:577-86. [30] Sanchez-Zamorano LM, Flores-Luna L, Angeles-Llerenas A, Romieu I, Lazcano-Ponce E, MirandaHernandez H, et al. Healthy lifestyle on the risk of breast cancer. Cancer Epidemiol Biomarkers Prev 2011;20:912-22. [31] Fejerman L, Romieu I, John EM, Lazcano-Ponce E, Huntsman S, Beckman KB, et al. European ancestry is positively associated with breast cancer risk in Mexican women. Cancer Epidemiol Biomarkers Prev 2010;19:1074-82. [32] Willett W. Nutritional Epidemiology. Third edition ed: Oxford 2012. [33] Hernandez-Avila M, Romieu I, Parra S, Hernandez-Avila J, Madrigal H, Willett W. Validity and reproducibility of a food frequency questionnaire to assess dietary intake of women living in Mexico City. Salud Publica Mex 1998;40:133-40. [34] Service AR. USDA National Nutrient Database for Standard Reference In: Agriculture USDo, editor. Nutrient Data Laboratory Home Page2009. [35] Morales de León J BV, Bourges H, Camacho PM. Tables of composition of Mexican foods [Published in CD-ROM Interactive Multimedia]. INCMNSZ, Mexico; . Mexico: Published in CD-ROM Interactive Multimedia. INCMNSZ; 2000. [36] Salazar-Martinez E, Lazcano-Ponce EC, Gonzalez Lira-Lira G, Escudero-De los Rios P, SalmeronCastro J, Hernandez-Avila M. Reproductive factors of ovarian and endometrial cancer risk in a high fertility population in Mexico. Cancer Res 1999;59:3658-62. [37] Mayagoitia S. La edad de la menopausia en México Rev Endocrinol Nutr 2006;14:4.

Journal Pre-proof

Jo

ur

na

lP

re

-p

ro

of

[38] Bourges Rodríguez H MdLJ, Camacho Parra ME, Escobedo Olea G. Tablas de composición de alimentos. Edición de Aniversario 50th ed. México D. F.: INNSZ; 1996. [39] Agriculture UDo. USDA National Nutritient Database for Standard Reference. In: USDA, editor.2009. [40] Dupont WD. Power calculations for matched case-control studies. Biometrics 1988;44:1157-68. [41] Knol MJ, VanderWeele TJ. Recommendations for presenting analyses of effect modification and interaction. Int J Epidemiol 2012;41:514-20. [42] Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC, Mulrow CD, Pocock SJ, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Epidemiology 2007;18:805-35. [43] VanderWeele TJaK, M.J. A tutorial on Interaction. Epidemiol Methods 2014;3:33-72. [44] KJ. R. Modern Epidemiology. Boston, MA: Little, Brown and Company; 1986. [45] MJ VTaK. A tutorial on interaction. Epidemiol Methods 2014;3:33-72. [46] Hosmer DW, Lemeshow S. Confidence interval estimation of interaction. Epidemiology 1992;3:4526. [47] Feiveson AH. What is the delta method and how is it used to estimate the standard error of a transformed parameter? FAQs: STATA; 1999. [48] Oehlert GW. A note on the delta method. . The American Statistician 1992;46. [49] Bland JM, Altman DG. Multiple significance tests: the Bonferroni method. BMJ 1995;310:170. [50] U.S. Department of Agriculture ARS. What we eat in America, 2013-2014. 2017. [51] Key T, Appleby P, Barnes I, Reeves G, Endogenous H, Breast Cancer Collaborative G. Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies. J Natl Cancer Inst 2002;94:606-16. [52] Missmer SA, Eliassen AH, Barbieri RL, Hankinson SE. Endogenous estrogen, androgen, and progesterone concentrations and breast cancer risk among postmenopausal women. J Natl Cancer Inst 2004;96:1856-65. [53] Collins LC, Cole KS, Marotti JD, Hu R, Schnitt SJ, Tamimi RM. Androgen receptor expression in breast cancer in relation to molecular phenotype: results from the Nurses' Health Study. Mod Pathol 2011;24:924-31. [54] Ring A, Dowsett M. Mechanisms of tamoxifen resistance. Endocr Relat Cancer 2004;11:643-58. [55] Dimitrakakis C, Bondy C. Androgens and the breast. Breast Cancer Res 2009;11:212. [56] Lin HY, Sun M, Lin C, Tang HY, London D, Shih A, et al. Androgen-induced human breast cancer cell proliferation is mediated by discrete mechanisms in estrogen receptor-alpha-positive and -negative breast cancer cells. J Steroid Biochem Mol Biol 2009;113:182-8. [57] Fioretti FM, Sita-Lumsden A, Bevan CL, Brooke GN. Revising the role of the androgen receptor in breast cancer. J Mol Endocrinol 2014;52:R257-65. [58] Pasanisi P, Berrino F, De Petris M, Venturelli E, Mastroianni A, Panico S. Metabolic syndrome as a prognostic factor for breast cancer recurrences. Int J Cancer 2006;119:236-8. [59] Weinberg ME, Manson JE, Buring JE, Cook NR, Seely EW, Ridker PM, et al. Low sex hormone-binding globulin is associated with the metabolic syndrome in postmenopausal women. Metabolism 2006;55:1473-80. [60] Xue F, Michels KB. Diabetes, metabolic syndrome, and breast cancer: a review of the current evidence. Am J Clin Nutr 2007;86:s823-35. [61] Burger HG. Androgen production in women. Fertil Steril 2002;77 Suppl 4:S3-5. [62] Judd HL, Judd GE, Lucas WE, Yen SS. Endocrine function of the postmenopausal ovary: concentration of androgens and estrogens in ovarian and peripheral vein blood. J Clin Endocrinol Metab 1974;39:1020-4.

Journal Pre-proof

Jo

ur

na

lP

re

-p

ro

of

[63] Burger HG, Dudley EC, Cui J, Dennerstein L, Hopper JL. A prospective longitudinal study of serum testosterone, dehydroepiandrosterone sulfate, and sex hormone-binding globulin levels through the menopause transition. J Clin Endocrinol Metab 2000;85:2832-8. [64] Braunstein GD. Safety of testosterone treatment in postmenopausal women. Fertil Steril 2007;88:117. [65] Danforth KN, Eliassen AH, Tworoger SS, Missmer SA, Barbieri RL, Rosner BA, et al. The association of plasma androgen levels with breast, ovarian and endometrial cancer risk factors among postmenopausal women. Int J Cancer 2010;126:199-207. [66] Hilakivi-Clarke L, Cabanes A, Olivo S, Kerr L, Bouker KB, Clarke R. Do estrogens always increase breast cancer risk? J Steroid Biochem Mol Biol 2002;80:163-74. [67] Khosla S, Melton LJ, 3rd, Atkinson EJ, O'Fallon WM, Klee GG, Riggs BL. Relationship of serum sex steroid levels and bone turnover markers with bone mineral density in men and women: a key role for bioavailable estrogen. J Clin Endocrinol Metab 1998;83:2266-74. [68] Kabuto M, Akiba S, Stevens RG, Neriishi K, Land CE. A prospective study of estradiol and breast cancer in Japanese women. Cancer Epidemiol Biomarkers Prev 2000;9:575-9. [69] Denova-Gutierrez E, Ramirez-Silva I, Rodriguez-Ramirez S, Jimenez-Aguilar A, Shamah-Levy T, Rivera-Dommarco JA. Validity of a food frequency questionnaire to assess food intake in Mexican adolescent and adult population. Salud Publica Mex 2016;58:617-28. [70] Tamimi RM, Byrne C, Colditz GA, Hankinson SE. Endogenous hormone levels, mammographic density, and subsequent risk of breast cancer in postmenopausal women. J Natl Cancer Inst 2007;99:1178-87. [71] Phillips GB, Jing TY, Laragh JH. Serum sex hormone levels in postmenopausal women with hypertension. J Hum Hypertens 1997;11:523-6. [72] Berrino F, Muti P, Micheli A, Bolelli G, Krogh V, Sciajno R, et al. Serum sex hormone levels after menopause and subsequent breast cancer. J Natl Cancer Inst 1996;88:291-6. [73] Deschamps V, de Lauzon-Guillain B, Lafay L, Borys JM, Charles MA, Romon M. Reproducibility and relative validity of a food-frequency questionnaire among French adults and adolescents. Eur J Clin Nutr 2009;63:282-91. [74] Jones ME, Schoemaker MJ, Rae M, Folkerd EJ, Dowsett M, Ashworth A, et al. Reproducibility of estradiol and testosterone levels in postmenopausal women over 5 years: results from the breakthrough generations study. Am J Epidemiol 2014;179:1128-33. [75] Fogle RH, Stanczyk FZ, Zhang X, Paulson RJ. Ovarian androgen production in postmenopausal women. J Clin Endocrinol Metab 2007;92:3040-3.

Journal Pre-proof Table 1 - Characteristics of postmenopausal women stratified by breast cancer status in Mexico from 2004-2007 Cases n=342

Controls n=294

Pbc

57 (52.6-63.2) %

56.5 (51.8-61.1) %

0.07

31.9 35.6 32.5

0.001

ro

<0.001

-p

34.3 33 32.7

of

32.8 33.1 34.1

Jo

ur

na

Socioeconomic status Low 32.3 Medium 24.7 High 43.0 Reproductive Breastfeeding (months)b 0 to 12 52.9 13 to 44 24.5 >44 22.6 Time since menopause (years) <6.61 31.9 6.61 to 14.1 31.7 >14.1 36.4 Morbidity History of diabetes No 73.8 Yes 26.2 History of a benign breast disease No 87.1 Yes 12.9 Breast cancer family history No 96.5 Yes 3.5 Lifestyles Moderate-vigorous physical activity (hrs) 0 to 7 hrs 54.8 7.1 to 20 hrs 32.6 >20 hrs 12.6 Alcohol consumption per day (gr) Never 32.6 <1gr 45.6 ≥1gr 21.8 Cigarettes smoked over lifetime <100 77.8 ≥100 22.2 Western dietary pattern T1 (lowest) 20.8 T2 32.2 T3 (highest) 47 Dietary folate intake (mg/d ) T1 < 259.40 23.4 T2 = 259.40 – 381.29 32.9 T3 > 381.29 43.7

Characteristics by cases and controls a Medians and interquartile range b Kruskal-Wallis rank sum test for continuous variables c 2 Chi test for categorical variables

0.613

77.8 22.1

0.195

94.9 5.1

<0.001

98.1 1.9

0.157

37.8 34.5 27.7

<0.001

43.8 45.7 10.5

0.001

80.2 19.8

0.5

27.4 39.3 33.3

0.005

32.1 34.9 33.0

0.006

re

Sociodemographic a Age

lP

Characteristics

Journal Pre-proof

Table 2 - Anthropometry and serum concentrations of sex hormones in postmenopausal women by breast cancer status in Mexico from 2004-2007

Variables Height (cm) 2

Body mass index (kg/m ) Energy consumption (Kcal) Estradiol, free fraction (pg/ml)

Cases

Controls

n=342

n=294

Medians

Q25

Q75

151.8

147.9

155.8

29.3

26.4

33.3

2022.5

1620.6

2534.8

0.24

0.15

0.43

0.5

0.3

Testosterone/estradiol ratio

1.81

0.91

Dietary folate intake (mg/day) 362.83 Characteristics of cases and controls a Kruskal-Wallis rank sum test

271.33

l a

Jo

n r u

0.8

f o Q25

Q75

147.5

155

0.058

27.7

35.4

0.001

1744.7

1439.4

2159.5

<0.001

0.26

0.14

0.45

0.841

o r p

e

r P

Testosterone, free fraction (pg/ml)

Medians

Pa

150.9 30.7

0.4

0.26

0.6

0.001

4

1.48

0.88

3.33

0.122

477.39

322.67

248.62

436.36

0.002

Journal Pre-proof Table 3 - Joint effect of testosterone and dietary folate intake on the risk of breast cancer in postmenopausal women in Mexico from 2004-2007

ORs (95% CI) for dietary folate intake (T2,T1 vs. T3) within strata of testosterone serum concentration

Dietary folate intake (mg/d) Sex hormones

T2 = 259.40 – 381.29

T3 > 381.29 a

Case

Control

OR

95% CI

≤0.25

25

27

1.00

0.251 - 0.500

50

43

2.82

1.13-7.05

≥0.510

63

37

4.41

1.70-11.44

P

Case

Control

OR

19

25

0.026

37

0.002

51

a

95% CI

P

Case

Control

4.03

1.16-13.95

0.028

16

30

46

4.01

1.43-11.19

0.008

45

43

42

6.05

2.18-16.73

0.001

28

29

Testosterone (pg/mL)

l a

r P

OR

f o

a

ro

p e

p for interaction ORs (95% CI) for testosterone serum concentration (T2,T3 vs. T1) within strata of dietary folate intake

T2 = 259.40 – 381.29

T1 < 259.40

5.12 6.04 9.18

95% CI

P

OR

1.36-19.25

0.016

1.82-20.05

0.003

2.56-32.88

0.001

1.13-7.05

0.026

0.99

0.35-2.81

0.993

1.18

0.43-3.23

0.748

T3 ≥ 0.510

4.41

1.70-11.44

0.002

1.50

0.53-4.24

0.444

1.79

0.57-5.57

0.313

-0.60

-3.2, 1.90

0.636

-0.30

-4.48, 3.86

0.886

-0.17

-3.93, 3.59

0.929

1.94

-8.03, 11.93

0.702

0.35

0.08, 0.84

0.009

0.41

0.10, 0.98

0.045

0.34

0.13, 0.81

0.006

0.40

0.09, 1.00

0.051

Measure of interaction on multiplicative scale: Ratio of ORs (95%CI)

a

J

u o

P

OR

4.03

1.16-13.95

0.028

1.42

0.62-3.25

0.406

1.37

0.58-3.23

0.470

a

95% CI

P

5.12

1.36-19.25

0.016

2.13

0.80-5.71

0.129

2.08

0.67-6.41

0.202

0.082

2.82

rn

95% CI

b

T2 = 0.251 - 0.500

Measure of interaction on additive scale: RERI (95%CI)

a

T1 < 259.40

Models were adjusted for design by: 5-year age group, health care institution membership and place of residence. We also adjusted for: time since menopause (years), SES (socioeconomic status) in tertiles according to its distribution in the control group, breastfeeding (in months), BMI (kg/m2) (normal, overweight and obese according to the WHO), family history of BC (yes/no), history of diabetes (yes/no), total calories intake (kcal), Western dietary pattern, energy residuals, height (cm), benign breast disease (yes/no), physical activity (weekly hours of moderate and vigorous-intensity physical activity), consumption of alcohol (never, less than a gram per day and one or more grams per day) and tobacco consumption (lifetime history of smoking more than 100 cigarettes). Energy residuals were obtained to adjust for the energy intake not explained by the Western dietary pattern. b P for interaction for the probability of at least one contrast was statistically significant: p=0.082, the significant p value calculated according to Bland and Altman, 1995 [43]. By using Bonferroni correction, the significant alpha value for any contrast (category) was 0.00625.

Journal Pre-proof

Table 4 - Joint effect of estrogen and dietary folate intake on the risk of breast cancer in postmenopausal women in Mexico from 2004-2007

f o

Dietary folate intake (mg/d) Sex hormones

T2 = 259.40 – 381.29

T3 > 381.29 a

Case

Control

OR

95% CI

≤0.18

42

49

1.00

0.181 - 0.32

55

43

1.24

0.58-2.66

≥0.321

49

40

1.62

0.75-3.48

P

Case

Control

OR

42

44

0.572

42

0.211

36

a

95% CI

P

3.38

1.48-7.73

0.004

45

2.12

0.92-4.88

0.076

47

2.03

0.88-4.70

0.095

p for interaction

T2 = 0.181 - 0.32

1.24

0.58-2.66

0.572

T3 ≥ 0.321

1.62

0.75-3.48

0.211

o J

ur

Measure of interaction on additive scale: RERI (95%CI)

Measure of interaction on multiplicative scale: Ratio of ORs (95%CI)

l a n

T2 = 259.40 – 381.29

T1 < 259.40

o r p

Case

Estradiol (pg/mL)

ORs (95% CI) for estradiol serum concentration (T2,T3 vs. T1) within strata of dietary folate intake

ORs (95% CI) for dietary folate intake (T2,T1 vs. T3) within strata of estradiol serum concentration

r P

e 52 60 40

Control

OR

34

a

95% CI

P

OR

3.02

1.12-8.13

0.029

37

3.28

1.30-8.29

29

2.56

0.86-7.65

a

T1 < 259.40

95% CI

P

OR

3.38

1.48-7.73

0.004

0.012

1.70

0.71-4.07

0.090

1.25

0.53-2.93

a

95% CI

P

3.02

1.12-8.13

0.029

0.229

2.63

1.01-6.89

0.048

0.600

1.58

0.52-4.75

0.415

b

0.754

0.62

0.29-1.35

0.233

1.08

0.47-2.47

0.842

0.60

0.27-1.29

0.195

0.85

0.31-2.27

0.746

-1.52

-3.67, 0.62

0.163

0.32

-2.66, 3.31

0.833

-1.57

-3.71, 0.57

0.151

-0.76

-3.12, 1.58

0.523

0.50

0.03, 1.04

0.072

0.87

0.10, 1.84

0.798

0.37

0.02, 0.76

0.002

0.52

0.12, 1.17

0.149

a

Models were adjusted for design by: 5-year age group, health care institution membership and place of residence. We also adjusted for: time since menopause (years), SES (socioeconomic status) in tertiles according to its distribution in the control group, breastfeeding (in months), BMI (kg/m2) (normal, overweight and obese according to the WHO), family history of BC (yes/no), history of diabetes (yes/no), total calories intake (kcal), Western dietary pattern, energy residuals, height (cm), benign breast disease (yes/no), physical activity (weekly hours of moderate and vigorous-intensity physical activity), consumption of alcohol (never, less than a gram per day and one or more grams per day) and tobacco consumption (lifetime history of smoking more than 100 cigarettes). Energy residuals were obtained to adjust for the energy intake not explained by the Western dietary pattern. b P for interaction for the probability of at least one contrast was not statistically significant: p=0.754, the significant p value calculated according to Bland and Altman, 1995 [43]. By using Bonferroni correction, the significant alpha value for any contrast (category) was 0.00625.

28

Journal Pre-proof

Fig. 1 - Study population flow-chart.

Study population n=2074

Original study population

Cases n=1000

Original study classification

l a n

Postmenopausal women

Women without antecedent of Hormone Replacement Therapy

Women with Testosterone, Estradiol and Folate information.

J

r u o

o r p

f o

Controls n=1074

e

r P

n=578

n=586

N=439

n=498

n=342

n=294

29

Journal Pre-proof

10 8 6 4 2 0

of

259.40-381.29

>381.29

Testosterone (pg/mL) Testosterone (pg/mL) Testosterone (pg/mL) <259.40

Adjusted ORa

A) Testosterone

Dietary folate intake (mg/d)

re na

lP

10 9 8 7 6 5 4 3 2 1 0

259.40-381.29

Estradiol (pg/mL) Estradiol (pg/mL) Estradiol (pg/mL) >381.29

Jo

<259.40

ur

Adjusted ORa

-p

ro

B) Estradiol

Dietary folate intake (mg/d)

Fig. 2 - Adjusted ORs of the Interaction between dietary folate intake and free serum fraction of testosterone (A) and estrogens (B) on the risk of breast cancer in postmenopausal women in Mexico a

Models were adjusted for design by: 5-year age group, health care institution membership and place of residence. We adjusted also for: time since menopause (years), SES (socioeconomic status) in tertiles according to its distribution in the control group, breastfeeding (in 2

months), BMI (kg/m ) (normal, overweight and obese according to WHO), family history of BC (yes/no), history of diabetes (yes/no), total calories intake (kcal), Western dietary pattern, energy residuals, height (cm), benign breast disease (yes/no), physical ac tivity (weekly hours of moderate and vigorous-intensity physical activity), consumption of alcohol (never, less than a gram per day and one or more grams per day) and tobacco consumption (lifetime history of smoking more than 100 cigarettes). Energy residu als were obtained to adjust for the energy intake not explained by the Western dietary pattern.

30

Journal Pre-proof Highlights The odds of having BC increased with increasing tertiles of serum testosterone. The odds of having breast cancer increased with decreasing tertiles of dietary folate intake.

Jo

ur

na

lP

re

-p

ro

of

There was a joint effect between testosterone and dietary folate intake on the risk of breast cancer

31