Skip to main content
NCBI home page
The Journal of Clinical Endocrinology and Metabolism logoLink to The Journal of Clinical Endocrinology and Metabolism
. 2024 Aug 7;110(5):e1488–e1498. doi: 10.1210/clinem/dgae549

Association of Adiposity Phenotypes With 27-Hydroxycholesterol and Sex Hormones: The Multiethnic Cohort Study

Yuqing Li 1,✉, Samantha A Streicher 2, Adrian A Franke 3, Anne N Tome 4, Kami K White 5, Yurii Shvetsov 6, Unhee Lim 7, Veronica W Setiawan 8, Mindy C DeRouen 9, Brenda Y Hernandez 10, Anna H Wu 11, Lynne R Wilkens 12, Loïc Le Marchand 13, Lenora W M Loo 14,2, Iona Cheng 15,2
PMCID: PMC12012796  PMID: 39109840

Abstract

Context

The distribution of body fat has been linked to circulating levels of lipids and sex steroid hormones. The cholesterol metabolite and endogenous selective estrogen receptor modulator, 27-hydroxychlolesterol (27HC), may be influenced by adiposity phenotypes, particularly among females. No study has examined the relationships of 27HC with adiposity phenotypes.

Objective

To investigate the associations of 27HC and steroid hormones with detailed adiposity phenotypes among a multiethnic population of postmenopausal women.

Methods

A cross-sectional study was conducted among 912 postmenopausal women from the Multiethnic Cohort Adiposity Phenotype study. Multivariable linear regression examined the associations of circulating levels of 27HC, steroid hormones, and sex hormone–binding globulin (SHBG) with detailed adiposity phenotypes, adjusting for demographics, lifestyle factors, diabetes status, and use of lipid-lowering drugs. Subgroup analyses were conducted across race and ethnicity.

Results

Total fat mass (P trend = .003), subcutaneous adipose tissue (SAT) (P trend = .006), and superficial subcutaneous adipose tissue (P trend = 4.41 × 10−4) were inversely associated with circulating 27HC levels. In contrast, visceral adipose tissue (VAT) (P trend = .003) and liver fat (P trend = .005) were positively associated with 27HC levels. All adiposity phenotypes were associated with higher levels of free estradiol and testosterone and lower levels of SHBG. Generally, similar patterns of associations were observed across race and ethnicity.

Conclusion

Adiposity phenotypes, such as SAT, VAT, and liver fat, were differentially associated with circulating 27HC, while consistent directions of associations were seen for circulating hormones among postmenopausal women. Future studies are warranted to further understand the biology and relationships of 27HC and adiposity-related diseases.

Keywords: 27-hydroxychlolesterol, steroid hormones, adiposity phenotypes, Multiethnic Cohort Study, Multiethnic Cohort Adiposity Phenotype study

Patterns in the distribution of body fat have been linked to alterations in metabolic profiles such as circulating lipids and sex steroid hormones, which have important health consequences. For instance, greater amounts of visceral adiposity, rather than overall or subcutaneous adiposity, have been associated with high serum total cholesterol, high low-density lipoprotein, and high triglyceride levels among normal and overweight individuals (1-4), and has been strongly associated with obesity-related outcomes such as cancer, cardiovascular disease, and metabolic syndrome (5-8). The cholesterol metabolite, 27-hydroxychlolesterol (27HC), the most-abundant circulating oxysterol and an endogenous selective estrogen receptor modulator (SERM), may be a mediator between obesity and poor health outcomes (9, 10). In addition, excess adiposity and fat tissue among females increases circulating levels of sex steroid hormones, including estrogens and androgens, as well as their carrier protein, sex–hormone binding globulin (SHBG) (11-13).

Greater abdominal adiposity (as measured by waist to hip ratio) has been associated with higher total cholesterol, triglyceride, low-density lipoprotein levels, and lower high-density lipoprotein levels among females across self-identified race and ethnicity and body mass index (BMI) levels (14-19). In a study among African American and White women of the Women's Health Across the Nation (SWAN) cohort, visceral adiposity was positively associated with testosterone levels (20) and liver fat was inversely associated with SHBG levels (21). In a prior study of the Multiethnic Cohort Adiposity Phenotype Study (MEC-APS) among adult males and females, relative amounts of visceral, trunk, and liver fat differed markedly across race and ethnicity with the highest levels observed among Japanese American adults, intermediate levels among Latino, Native Hawaiian, and White adults, and lowest levels among African American adults (22). This pattern of association was also consistent when limited to females alone. The MEC-APS offers a unique opportunity to study detailed adiposity phenotypes (ie, total fat mass, total trunk fat, subcutaneous adipose tissue (SAT), superficial subcutaneous adipose tissue (sSAT), deep subcutaneous adipose tissue (dSAT), visceral adipose tissue (VAT), percent liver fat, and total lean mass) assessed by dual-energy x-ray absorptiometry (DXA) and magnetic resonance imaging (MRI), providing a more precise assessment of adiposity deposition than anthropometric measures. In this study, we examined the associations of adiposity phenotypes with circulating levels of 27HC, steroid hormones, and SHBG among postmenopausal women within the MEC-APS. This is the first study to assess the association between adiposity phenotypes and 27HC levels, offering insight into a potentially important biological link for multiple obesity-related conditions among postmenopausal women.

Materials and Methods

Study Population

The Multiethnic Cohort study (MEC) is an ongoing prospective cohort study designed to examine the association of dietary, lifestyle, and genetic factors in relation to cancer and other chronic conditions (23). The cohort, initiated in 1993-1996, includes 96 810 males and 118 441 females aged 45-75 years of 5 self-reported racial and ethnic groups (African American, Japanese American, Latino, Native Hawaiian, and White) and residents of California (primarily Los Angeles County) and Hawaii. At baseline, participants completed a mailed questionnaire that included information on demographics, anthropometrics, smoking and alcohol use history, diet, medication use, physical activity, and, for females, menstrual and reproductive histories, and hormone use (24).

The MEC-APS is a cross-sectional study within the MEC that aimed to identify predictors of body fat distribution, measured by DXA and MRI scans, and risk factors for obesity-related cancers, as described previously (22). Briefly, a subset of 938 female and 723 male MEC participants (ages 58-74) were selected using stratified sampling by self-reported race and ethnicity and 6 BMI categories (18.5-21.9, 22-24.9, 25-26.9, 27-29.9, 30-34.9, and 35-40 kg/m2 based on weight and height). Exclusion criteria included reported BMI outside the range of 18.5 to 40 kg/m2, smoking in the past 2 years, soft or metal body implants or amputation, insulin or thyroid medications, and serious medical conditions (eg, dialysis, chronic hepatitis, previous cancer diagnosis). Of 938 MEC-APS females, we excluded a total of 26 females due to missing 27HC data (n = 2) and invalid DXA scans because of implants or invalid MRI scans because of motion artifacts or presence of visceral masses (n = 24). For the current study, the final analytic sample included 912 postmenopausal women (Table 1). All MEC-APS participants provided written informed consent and the study was approved by the institutional review boards at the University of Hawai’i Cancer Center (UH) (CHS-#17200), University of Southern California (USC) (#HS-12-00623), and University of California, San Francisco (UCSF) (#17-23399) in agreement with the 1975 Helsinki Declaration.

Table 1.

Descriptive characteristics of postmenopausal female participants in the Multiethnic Cohort Adiposity Phenotype Study (n = 912)a

Characteristic n (%)
Age at blood draw, years; mean (SD) 69.1 (2.71)
Race and ethnicity, n (%)
 African American 176 (19.30)
 Japanese American 202 (22.15)
 Latino 186 (20.39)
 Native Hawaiian 155 (17.00)
 White 193 (21.16)
Lipid-lowering drug use, n (%)b
 No 459 (50.33)
 Yes 401 (43.86)
Hormone replacement usec
 Current use estrogen or progesterone 53 (5.92)
 Never or previously use estrogen or progesterone 843 (94.08)
Diabetes, n (%)
 No 784 (85.96)
 Yes 128 (14.04)
METs of physical activity per day 1.62 (0.26)
Open in a new tab

Abbreviations: MET, metabolic equivalent.

a Age, race, and ethnicity, lipid-lowering drug use, diabetes status, and physical activity collected at clinic visit.

b Fifty-two participants have no information of lipid-lowering drug use.

c Sixteen participants have no information of hormone replacement use.

Measurement of 27HC, Sex Hormones, and SHBG

The measurements of 27HC, sex hormones, and SHBG were conducted at the University of Hawai’i Cancer Center Analytical Biochemistry Shared Resource. Overnight fasting blood samples were collected at the time of body composition measurement, processed into components, and stored at −80 °C (25). For 27HC (ng/mL), plasma or serum levels were determined by stabilizing plasma/serum with butylated hydroxytoluene followed by hydrolysis with potassium hydroxide, extraction with hexane, picolinyl derivatization, re-extraction into hexane, and liquid chromatography-orbitrap mass spectrometry analysis using ultrahigh-performance liquid chromatography and isotope dilution high-resolution high-accuracy orbitrap-based mass spectrometry (26). For unconjugated estradiol (pg/mL) (n = 910), unconjugated estrone (pg/mL) (n = 910), and unconjugated testosterone (pg/mL) (n = 907), plasma/serum levels were measured according to our previous reports (26, 27). This method for steroid hormone measurement is 2- to 50-fold more sensitive than traditional liquid chromatography-orbitrap mass spectrometry methods and was validated by comparisons to the National Institute for Standards and Technology controls. Plasma SHBG (nmol/L) (n = 912) levels were assayed by double-antibody enzyme-linked immunosorbent assay (R and D Systems Cat# DSHBG0B, RRID:AB_3101841) according to the manufacturer's specifications to assess SHBG. Briefly, the concentration of free estradiol, free estrone, and free testosterone were calculated based on the concentration of unconjugated hormone and SHBG (28). Details on this calculation and quality control procedures can be found in the Supplementary methods (29). Our analysis focused on free estradiol, free estrone, and free testosterone as they represent bioavailable steroid hormones that are able to cross cell membranes and bind to steroid receptors (30)

Replicate samples of plasma/serum for 27HC and each sex steroid hormone, and SHBG were included in each of the 3 batches for quality control measures (Table S1 (29)). The coefficient of variation (CV) percent for 27HC was 6.8 within batch. The CV% for unconjugated estradiol, unconjugated estrone, unconjugated testosterone, and SHBG within batches were 15.9%, 23.5%, 9.2% and 8.4%, respectively. The larger CVs likely reflect several samples close to the lower limit of detection (Table S1 (29)). Hormone analytes below the lower limit of detection were assigned a value half of the lower limit of detection.

Adiposity Phenotypes

Trained technicians obtained measurements of height, weight, circumferences of the waist and hip, and chest depth (22). Abdominal scans using 3-T MRI scanners (Siemens TIM Trio at UH and General Electric HDx at USC) were acquired to quantify VAT, SAT, and liver fat areas (square centimeters) at 4 intervertebral segments of the intra-abdominal cavity (L1-L2, L2-L3, L3-L4, L4-L5) using an axial gradient-echo sequence with breath holds (22). The average VAT and SAT across segments L1 to L5 was used in analysis. Segmentation of the superficial fascia between dSAT and sSAT was performed manually. In order to quantify the amount of dSAT at L1 to L5, first dSAT was measured at L3 to L4, then the fraction of dSAT to SAT at L3 to L4 was calculated, and finally the mean SAT across segments L1 to L5 was multiplied by the fraction of dSAT to SAT at L3 to L4. The amount of sSAT at L1 to L5 was calculated using the same steps as the dSAT calculation. Whole-body composition, including total fat mass, total lean mass, and trunk fat mass were determined by a DXA scan (Hologic Discovery A fan-beam densitometer at UH and USC, Bedford, MA) (22). Extensive details regarding the imaging protocol, as well as quality control calibration and estimation of VAT and SAT area were previously published (22). Figure 1 displays the distribution of adiposity phenotypes (total fat mass, total trunk mass, SAT, sSAT, dSAT, VAT, percent liver fat, and total lean mass) and anthropometric measures (BMI, waist to hip ratio, and waist circumstance) overall and by race and ethnicity.

Figure 1.

Figure 1.

Open in a new tab

Distribution of adiposity phenotypes among 912 MEC-APS females among all and by race and ethnicity. The central line within each box represents the median value (50th percentile), while the box contains the 25th to 75th percentiles of this study population. The whiskers extend to the 5th and 95th percentiles, and values beyond these upper and lower bounds are considered outliers and marked with dots.

Abbreviations: A, All; B, African American; J, Japanese American; L, Latino; H, Native Hawaiian; W, White.

Statistical Analysis

Descriptive characteristics were examined in the overall study population using means and frequencies. Geometric means of levels of 27HC, sex steroid hormones, and SHBG were calculated for the overall study population and by race and ethnicity. As 27HC, sex steroid hormones, SHBG, and adiposity phenotypes were not normally distributed, we applied a log2-transformation to achieve normal distributions. The Pearson correlation coefficient (r) was used to examine correlations between log2-transformed 27HC, free estrogen, free estrone, free testosterone, and SHBG (Table S2 (29)). In addition, correlations between log2 transformed adiposity phenotypes were assessed (Table S3 (29)).

Adiposity phenotypes were categorized into tertiles based on the overall study population. Multivariable linear regression was conducted to examine the association of circulating 27HC, free estrogen, free estrone, free testosterone, and SHBG with the adiposity phenotypes. Models were adjusted for age at blood draw (continuous), physical activity (continuous as metabolic equivalents), diabetes (Yes, No), and use of lipid-lowering drugs (Yes, No). Additional adjustment for total fat mass was conducted for VAT area and percent liver fat as the correlations with total fat mass was r < 0.7 (Table 2; Table S3 (29)). Similar associations were observed with adjustment for current use of hormone replacement therapy (Table S4 (29)). Covariate-adjusted geometric means and 95% CI of each biomarker by adiposity tertiles were estimated based on the log2-transformed values of the least square means. P trends for the adiposity–biomarker associations were estimated using a continuous adiposity term in our linear regression models. Subgroup analyses was conducted by race and ethnicity, and by use of lipid-lowering drugs. Heterogeneity in associations by subgroups was tested using the Type III Sum of Squares global test of the cross-product terms for the adiposity phenotype and subgroup. All analyses were performed using SAS version 9.4 (SAS Institute, Inc, Cary, NC, USA). P values were 2-sided with statistical significance defined as P < .0083 based on a Bonferroni correction of 6 independent tests in our overall analysis.

Table 2.

Geometric mean levels of circulating 27-hydroxycholesterol (ng/mL) by adiposity phenotypes among postmenopausal women overall and by race and ethnicity, Multiethnic Cohort Adiposity Phenotype Study (N = 912)a

  Overall African American Japanese American Latino Native Hawaiian White  
Adiposity phenotype n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI) P het
Total fat mass (kg)
 ≤23.02 303 68.00 (64.68, 71.49) 20 62.09 (51.94, 74.23) 116 68.48 (63.00, 74.44) 28 62.64 (52.77, 74.35) 61 67.39 (60.12, 75.54) 78 65.86 (57.35, 75.64)
 23.02-30.41 305 65.26 (62.19, 68.48) 53 65.08 (58.00, 73.02) 67 64.46 (58.19, 71.40) 66 65.10 (58.00, 73.07) 52 71.35 (64.27, 79.21) 67 63.23 (55.09, 72.56)
 >30.41 304 63.48 (60.49, 66.61) 103 61.79 (56.63, 67.41) 19 64.78 (54.07, 77.63) 92 57.22 (51.96, 63.02) 42 75.80 (67.30, 85.37) 48 72.77 (64.09, 82.63)
 P trend .003 .927 .817 .048 .486 .776 .520
Total trunk fat (kg)
 ≤11.89 300 67.31 (63.90, 70.91) 40 63.61 (55.42, 73.00) 97 67.52 (61.38, 74.27) 31 63.47 (54.10, 74.45) 55 67.45 (59.84, 76.03) 77 66.69 (57.85, 76.87)
 11.89-15.56 282 64.04 (60.93, 67.32) 48 60.12 (53.62, 67.42) 67 65.21 (59.04, 72.03) 64 63.50 (56.34, 71.57) 46 68.33 (60.69, 76.92) 57 65.80 (56.77, 76.26)
 >15.56 302 64.88 (61.83, 68.09) 80 64.10 (58.30, 70.47) 29 67.39 (57.49, 79.00) 86 57.74 (52.29, 63.75) 53 74.85 (67.41, 83.10) 54 70.16 (61.92, 79.50)
 P trend .17 .360 .794 .090 .351 .577 .280
Subcutaneous adipose tissue area (cm2)
 ≤209.92 298 68.49 (65.18, 71.96) 27 62.21 (52.74, 73.39) 114 68.29 (63.02, 74.00) 33 69.46 (59.25, 81.42) 45 67.17 (59.56, 75.76) 79 68.88 (60.01, 79.07)
 209.92-292.05 310 65.00 (61.91, 68.24) 54 66.16 (59.23, 73.89) 69 64.86 (58.50, 71.91) 65 61.03 (54.34, 68.53) 58 71.32 (64.15, 79.30) 64 65.00 (56.76, 74.44)
 >292.05 298 63.40 (60.43, 66.52) 92 61.11 (55.91, 66.78) 18 66.69 (55.75, 79.77) 87 57.59 (52.24, 63.49) 51 74.88 (67.12, 83.53) 50 70.91 (62.29, 80.72)
 P trend .006 .897 .680 .017 .410 .752 .280
Superficial subcutaneous adipose tissue area (cm2)
 ≤130.3 304 68.84 (65.49, 72.36) 28 63.62 (54.24, 74.62) 101 68.71 (63.05, 74.86) 41 69.32 (60.07, 80.00) 50 69.02 (61.34, 77.67) 84 68.76 (59.80, 79.06)
 130.3-182.9 296 65.62 (62.44, 68.97) 51 67.34 (59.91, 75.70) 72 64.37 (58.25, 71.14) 61 61.37 (54.40, 69.24) 57 70.28 (63.14, 78.23) 55 66.04 (57.22, 76.22)
 >182.9 302 62.91 (60.01, 65.94) 94 60.74 (55.66, 66.28) 27 67.96 (58.61, 78.81) 82 57.01 (51.66, 62.90) 47 74.65 (66.64, 83.62) 52 69.59 (61.22, 79.09)
 P trend 4.41 × 10−4 .296 .823 .024 .689 .297 .540
Deep subcutaneous adipose tissue area (cm2)
 ≤711.2 298 67.01 (63.84, 70.34) 30 56.55 (48.66, 65.72) 117 69.03 (63.87, 74.60) 39 64.28 (55.46, 74.49) 42 68.10 (60.35, 76.85) 70 66.50 (57.92,76.35)
 711.2-109.2 308 65.89 (62.78, 69.16) 56 63.36 (56.79, 70.69) 63 64.46 (57.89, 71.78) 67 61.45 (54.91, 68.77) 63 71.50 (64.71, 79.01) 59 71.24 (61.89,82.02)
 >109.2 296 63.56 (60.49, 66.79) 87 64.07 (58.53, 70.13) 20 59.61 (50.03, 71.03) 78 58.04 (52.17, 64.57) 49 74.58 (66.27, 83.92) 62 68.88 (60.58,78.32)
 P trend .136 .266 .784 .065 .226 .456 .210
Visceral adipose tissue area (cm2)a
 ≤101.05 301 61.25 (57.93, 64.77) 75 62.57 (56.05, 69.84) 66 64.31 (56.83, 72.78) 36 47.98 (40.49, 56.86) 47 66.31 (56.90, 77.28) 77 65.49 (56.65,75.70)
 101.05-157.84 306 66.61 (63.41, 69.97) 61 57.54 (51.66, 64.08) 63 69.77 (62.76, 77.56) 65 62.86 (56.14, 70.40) 51 74.40 (66.58, 83.14) 66 69.27 (60.08,79.86)
 >157.84 299 68.64 (65.33, 72.12) 37 69.92 (61.20, 79.87) 72 67.43 (60.73, 74.86) 84 64.99 (58.34, 72.40) 56 71.34 (63.17, 80.57) 50 70.80 (61.42,81.62)
 Trend .003 .057 .150 .820 .477 .956 .500
Percent liver fata
 ≤2.61 299 61.85 (58.69, 65.17) 102 62.18 (56.42, 68.53) 42 63.48 (55.52, 72.59) 45 53.70 (46.25, 62.36) 39 61.81 (53.51, 71.39) 71 66.54 (57.85,76.53)
 2.61-5.1 306 65.85 (62.69, 69.16) 53 67.67 (60.34, 75.90) 58 65.99 (58.87, 73.97) 78 62.74 (56.67, 69.46) 46 70.47 (62.69, 79.21) 71 68.24 (59.08,78.81)
 >5.1 296 68.31 (65.17, 71.60) 20 58.10 (49.35, 68.40) 100 68.47 (62.67, 74.80) 61 62.25 (55.33, 70.04) 65 77.05 (69.20, 85.80) 50 70.17 (61.16,80.51)
 P trend .005 .357 .397 .107 .059 .597 .440
Total lean mass (kg)
 ≤38.5 298 66.69 (63.42, 70.14) 28 67.07 (57.33, 78.45) 103 68.96 (63.15, 75.29) 80 63.98 (57.44, 71.26) 37 63.37 (55.57, 72.25) 50 65.48 (56.39,76.02)
 38.5-44.46 307 66.24 (63.11, 69.53) 46 65.84 (58.62, 73.95) 70 66.00 (59.76, 72.89) 65 60.47 (54.07, 67.62) 54 73.97 (66.18, 82.68) 72 66.15 (57.52,76.07)
 >44.46 307 63.73 (60.78, 66.81) 102 60.58 (55.61, 66.01) 29 62.35 (54.09, 71.86) 41 55.02 (47.96, 63.12) 64 73.11 (66.41, 80.48) 71 70.22 (62.17,79.30)
 P trend .23 .756 .852 .014 .179 .606 .180
Open in a new tab

a Percent liver fat and visceral adipose tissue area adjusted for age, physical activity, diabetes status, lipid-lowering drug use, and total fat mass; all other adiposity phenotypes adjusted only for age, physical activity, diabetes status, and lipid-lowering drug use.

Results

The study population comprised 912 females with an average age of 69.1 years at blood collection. The self-reported race and ethnicity composition was as follows: African American (19%), Japanese American (22%), Latino (20%), Native Hawaiian (17%), or White (21%) (Table 1). Forty-four percent of females reported taking lipid-lowering drugs and 14% reported having diabetes (Table 1). The overall unadjusted geometric mean levels were 27HC (68.6 ng/mL), free estradiol (0.03 pg/mL), free estrone (0.3 pg/mL), free testosterone (1.31 pg/mL), and SHBG (66.94 nmol/L) (Table S5 (29)). Native Hawaiian females had the highest level of circulating 27HC (74.82 ng/mL) compared to the other racial ethnic groups. African American females had the highest levels of free estradiol (0.04 pg/mL), free estrone (0.34 pg/mL), and free testosterone (1.42 pg/mL). White females had the highest level of SHBG (80.58 nmol/L).

Table 2 presents the associations of adiposity phenotypes with circulating levels of 27HC overall and by race and ethnicity. Overall, we observed a statistically significant inverse association between total fat mass (P trend = .003), SAT area (P trend = .006), and sSAT area (P trend = 4.4 × 10−4) with circulating 27HC. In contrast, we observed a statistically significant positive association of VAT area (P trend = .003) and percent liver fat (P trend = .005) with 27HC, after adjusting for total fat mass. A similar positive association between the ratio of VAT/SAT and 27HC was observed (data not shown). Inverse associations were observed for total trunk fat, dSAT area, and total lean mass with 27HC, although these associations were not statistically significant. Generally, similar patterns of associations were seen across racial and ethnic groups (P heterogeneity > .18). Notably, among Latino females only, inverse associations for total fat mass (P trend = .048), SAT area (P trend = .017), sSAT area (P trend = .024), and total lean mass (P trend = .014) with 27HC were observed.

Table 3 presents associations of adiposity phenotypes with levels of circulating 27HC by use of lipid-lowering drugs. Among females with no use of lipid-lowering drugs, statistically significant inverse associations were observed for total fat mass (P trend = .002), SAT area (P trend = .002), and sSAT area (P trend = 2.91 × 10−4) with levels of 27HC. In addition, positive associations were observed for VAT area (P trend = .027) and percent of liver fat (P trend = .003) with 27HC. Among females who used lipid-lowering drugs, a positive association was also seen between VAT and 27HC (P trend = .028). There was no evidence of heterogeneity in associations by the use of lipid-lowering drug across adiposity phenotypes (P heterogeneity > .07).

Table 3.

Geometric mean levels of circulating 27-hydroxycholesterol (ng/mL) by adiposity phenotypes among postmenopausal women by lipid-lowering drug use, Multiethnic Cohort-Adiposity Phenotype Study (n = 860)a

  No lipid-lowering drug use Lipid-lowering drug use  
Adiposity Phenotype n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI) P het
Total fat mass (kg)
 ≤23.02 167 69.42 (63.93, 75.39) 124 64.81 (60.43, 69.51)
 23.02-30.41 154 64.59 (59.77, 69.79) 135 64.81 (60.63, 69.27)
 >30.41 138 62.33 (57.38, 67.72) 142 63.02 (59.16, 67.13)
 P trend .002 .285 .184
Total trunk fat (kg)
 ≤11.89 171 67.74 (62.32, 73.64) 114 64.86 (60.14, 69.94)
 11.89-15.56 139 64.85 (59.70, 70.44) 129 61.56 (57.56, 65.85)
 >15.56 133 62.37 (57.47, 67.69) 147 65.47 (61.49, 69.72)
 P trend .050 .900 .073
Subcutaneous adipose tissue area (cm2)
 ≤209.92 183 68.24 (63.10, 73.80) 135 64.99 (60.84, 69.42)
 209.92-292.05 132 66.53 (61.27, 72.25) 133 62.59 (58.39, 67.08)
 >292.05 141 61.29 (56.58, 66.40) 131 64.48 (60.40, 68.84)
 P trend .002 .406 .190
Superficial subcutaneous adipose tissue area (cm2)
 ≤130.3 167 68.77 (63.47, 74.51) 121 66.79 (62.21, 71.71)
 130.3-182.9 148 68.23 (63.04, 73.85) 135 61.35 (57.30, 65.68)
 >182.9 140 59.77 (55.22, 64.70) 140 64.39 (60.44, 68.59)
 P trend 2.91 × 10−4 .223 .168
Deep subcutaneous adipose tissue area (cm2)
 ≤711.2 163 67.87 (62.58, 73.62) 121 64.47 (60.21, 69.03)
 711.2-109.2 151 65.77 (60.72, 71.23) 142 64.62 (60.50, 69.01)
 >109.2 141 62.58 (57.75, 67.82) 133 63.23 (59.12, 67.62)
 P trend .062 .894 .354
Visceral adipose tissue area (cm2)a
 ≤101.05 175 62.10 (56.97, 67.70) 106 58.57 (53.96, 63.56)
 101.05-157.84 161 67.16 (61.94, 72.81) 131 64.51 (60.29, 69.03)
 >157.84 120 68.09 (62.53, 74.15) 162 67.19 (63.15, 71.48)
 Trend .027 .028 .421
Percent liver fata
 ≤2.61 170 61.90 (57.05, 67.16) 105 60.00 (55.50, 64.86)
 2.61-5.1 153 64.95 (59.94, 70.37) 139 65.44 (61.22, 69.95)
 >5.1 130 69.99 (64.54, 75.91) 153 65.49 (61.60, 69.62)
 P trend .003 .237 .667
Total lean mass (kg)
 ≤38.5 162 66.57 (61.25, 72.34) 123 65.29 (60.86, 70.05)
 38.5-44.46 159 65.82 (60.75, 71.32) 130 65.22 (61.02, 69.71)
 >44.46 138 63.62 (58.74, 68.91) 148 62.37 (58.60, 66.38)
 P trend .287 .494 .628
Open in a new tab

a Percent liver fat and visceral adipose tissue area adjusted for age, physical activity, diabetes status, lipid-lowering drug use, and total fat mass; all other adiposity phenotypes adjusted only for age, physical activity, diabetes status, and lipid-lowering drug use.

Table 4 presents associations of adiposity phenotypes with circulating sex steroid hormones and SHBG. For free estradiol, free estrone, and free testosterone, we observed positive associations across all adiposity phenotypes, including several highly statistically significant associations for free estradiol. For SHBG, we observed statistically significant inverse associations across all adiposity phenotypes. SHBG was inversely correlated with all circulating hormones (Table S2 (29)). Similar patterns of associations were observed across racial and ethnic groups (Table S6-9 (29)).

Table 4.

Geometric mean levels of circulating hormones (pg/mL) and sex hormone–binding globulin (nmol/L) among postmenopausal females, Multiethnic Cohort Adiposity Phenotype Studya

  Free estradiol (pg/mL) (N = 910) Free estrone (pg/mL) (N = 908) Free testosterone (pg/mL) (N = 910) SHBGb (nmol/L) (N = 912)
Adiposity phenotype n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI) n Adjusted geometric mean (95% CI)
Total fat mass(kg)
 ≤23.02 301 0.011 (0.009, 0.013) 301 0.21 (0.17, 0.25) 301 1.04 (0.93, 1.17) 303 68.29 (63.32, 73.66)
 23.02-30.41 305 0.028 (0.023, 0.034) 305 0.31 (0.26, 0.38) 305 1.38 (1.23, 1.54) 305 55.01 (51.15, 59.17)
 >30.41 304 0.044 (0.036, 0.053) 304 0.39 (0.32, 0.47) 302 1.57 (1.4, 1.75) 304 52.23 (48.56, 56.18)
 P trend 3.42 × 10−35 2.07 × 10−6 4.02 × 10−9 9.5 × 10−13
Total trunk fat (kg)
 ≤11.89 299 0.011 (0.009, 0.014) 299 0.21 (0.17, 0.26) 299 1.06 (0.95, 1.19) 300 72.37 (67.04, 78.12)
 11.89-15.56 282 0.026 (0.022, 0.032) 282 0.31 (0.25, 0.38) 281 1.34 (1.2, 1.5) 282 54.08 (50.25, 58.20)
 >15.56 302 0.042 (0.035, 0.051) 302 0.36 (0.30, 0.44) 301 1.56 (1.4, 1.73) 302 51.63 (48.10, 55.43)
 P trend 1.21 × 10−33 4.27 × 10−6 3.95 × 10−10 7.96 × 10−20
Subcutaneous adipose tissue area (cm2)
 ≤209.92 332 0.013 (0.011, 0.016) 332 0.24 (0.20, 0.29) 332 1.11 (0.99, 1.23) 333 66.49 (61.91, 71.41)
 209.92-292.05 278 0.026 (0.021, 0.033) 278 0.28 (0.23, 0.35) 278 1.47 (1.31, 1.66) 279 54.91 (50.81, 59.34)
 >292.05 294 0.045 (0.037, 0.055) 294 0.38 (0.31, 0.47) 292 1.49 (1.33, 1.66) 294 51.70 (48.06, 55.62)
 P trend 1.65 × 10−27 2.89 × 10−4 8.92 × 10−6 4.51 × 10−11
Superficial subcutaneous adipose tissue area (cm2)
 ≤130.3 303 0.011 (0.009, 0.014) 303 0.20 (0.17, 0.25) 303 1.1 (0.97, 1.23) 304 67.73 (62.80, 73.05)
 130.3-182.9 295 0.026 (0.021, 0.031) 295 0.30 (0.25, 0.37) 295 1.4 (1.25, 1.58) 296 55.99 (51.92, 60.38)
 >182.9 302 0.044 (0.037, 0.053) 302 0.39 (0.32, 0.48) 300 1.48 (1.33, 1.65) 302 52.03 (48.44, 55.88)
 P trend 3.41 × 10−26 9.63 × 10−5 2.74 × 10−5 1.13 × 10−10
Deep subcutaneous adipose tissue area (cm2)
 ≤711.2 297 0.015 (0.012, 0.018) 297 0.26 (0.21, 0.31) 297 1.15 (1.02, 1.29) 298 65.11 (60.48, 70.09)
 711.2-109.2 307 0.026 (0.021, 0.031) 307 0.28 (0.23, 0.34) 307 1.43 (1.27, 1.6) 308 57.04 (53.00, 61.40)
 >109.2 296 0.04 (0.032, 0.049) 296 0.37 (0.30, 0.45) 294 1.44 (1.28, 1.62) 296 51.37 (47.64, 55.39)
 P trend 3.42 × 10−16 0.018 2.84 × 10−4 1.46 × 10−7
Visceral adipose tissue area (cm2)a
 ≤101.05 300 0.016 (0.013, 0.02) 300 0.22 (0.18, 0.28) 300 1.08 (0.95, 1.23) 301 76.52 (70.57, 82.96)
 101.05-157.84 305 0.026 (0.022, 0.032) 305 0.33 (0.27, 0.4) 304 1.34 (1.2, 1.51) 306 56.98 (53.06, 61.19)
 >157.84 299 0.026 (0.021, 0.031) 299 0.31 (0.26, 0.38) 298 1.45 (1.29, 1.63) 299 49.59 (46.16, 53.27)
 Trend 1.35 × 10−3 0.015 6.70 × 10−5 1.12 × 10−22
Percent liver fata
 ≤2.61 298 0.018 (0.015, 0.022) 298 0.26 (0.21, 0.32) 297 1.11 (0.98, 1.25) 299 76.75 (71.29, 82.63)
 2.61-5.1 305 0.022 (0.018, 0.026) 305 0.28 (0.23, 0.35) 305 1.24 (1.11, 1.39) 306 62.11 (57.96, 66.57)
 >5.1 296 0.027 (0.023, 0.033) 296 0.31 (0.26, 0.38) 295 1.5 (1.34, 1.67) 296 46.78 (43.78, 49.98)
 P trend 8.41 × 10−4 0.107 7.00 × 10−5 2.15 × 10−27
Total lean mass (kg)
 ≤38.5 297 0.014 (0.011, 0.017) 297 0.23 (0.19, 0.29) 297 1.14 (1.01, 1.28) 298 68.68 (63.66, 74.08)
 38.5-44.46 306 0.02 (0.017, 0.025) 306 0.26 (0.21, 0.32) 305 1.33 (1.19, 1.49) 307 56.92 (52.91, 61.22)
 >44.46 307 0.046 (0.038, 0.056) 307 0.40 (0.33, 0.49) 306 1.49 (1.33, 1.66) 307 50.81 (47.32, 54.56)
 P trend 3.32 × 10−22 2.80 × 10−5 6.24 × 10−6 3.14 × 10−13
Open in a new tab

a Percent liver fat and visceral adipose tissue area adjusted for age, physical activity, diabetes status, lipid-lowering drug use, and total fat mass; all other adiposity phenotypes adjusted only for age, physical activity, lipid-lowering drug use, and diabetes status.

The associations of BMI, waist to hip ratio, and waist circumference with 27HC, sex steroid hormones, and SHBG were also examined (Table S10 (29)). For 27HC, an inverse association was observed with BMI (P trend = .021) and a positive association was seen for waist to hip ratio (P trend = .03). For free estradiol, positive associations were observed with BMI, waist to hip ratio, and waist circumference (P trend < .0001). For free estrone, positive associations were seen with BMI and waist circumference (P trend < .0001). For free testosterone, positive associations were also seen with BMI and waist circumference (P trend < .0001). For SHBG, inverse associations were observed for BMI, waist to hip ratio, and waist circumference (P trend < .0001).

Discussion

In this cross-sectional study of a multiethnic population of postmenopausal women, we found inverse associations of 27HC levels with total fat mass, SAT area, and sSAT area, and positive associations with VAT and percent liver fat, which was independent of total fat mass. In addition, higher levels of circulating free estradiol, free estrone, free testosterone, along with lower levels of SHBG were associated with increased adiposity across all measured phenotypes. This is the first study to assess the association of detailed adiposity phenotypes with 27HC, an abundant oxysterol metabolite of cholesterol and an endogenous SERM. It highlights 27HC as a potentially important biological link for multiple adiposity-related outcomes among postmenopausal women. In addition, this is the first report to examine the relationships of adiposity phenotypes with sex hormones within the MEC-APS.

The literature is limited to a few studies evaluating 27HC and anthropometric measurements (31, 32). The inverse associations we observed for 27HC with BMI confirm similar findings among 1163 postmenopausal women in a prior report of the Women's Health Initiative (31). These inverse associations between 27HC and BMI align with the inverse association we observed between 27HC and total fat mass. In contrast, a study of 1036 premenopausal and postmenopausal females from the European Prospective Investigation into Cancer and Nutrition (EPIC)-Heidelberg reported no association between 27HC and BMI (32). Possible differences in the results may be due to the inclusion of premenopausal and primarily White European females in the EPIC-Heidelberg study population.

27HC is enzymatically generated by a P450 enzyme encoded by the CYP27A1 gene. The CYP27A1 gene has been shown to be expressed differentially in adipose tissue types. Specifically, CYP27A1 has been found to be expressed at low levels in SAT and high levels in VAT (33, 34). High and low expression levels of CYP27A1 in VAT and SAT, respectively, may contribute to the positive association seen for 27HC and VAT area, and the inverse association of 27HC with SAT and sSAT areas. In addition, findings from prior studies of mouse models (35) provide support for the positive association of 27HC and VAT area. Higher circulating levels of 27HC in mice, achieved through 27HC administration, was found to promote an increase in visceral white adipose tissue (VWAT) mass. This increase in VWAT mass was found to be mediated by estrogen receptor-alpha (ER-α) activity (35, 36). Specifically, as 27HC can bind to and activate both ER-α and the liver X receptor (LXR), Asghari et al delineated that the increase in VWAT mediated by 27HC required the ER-α receptor, and not LXR. The 27HC-mediated increase in VWAT was not observed in mice lacking the ER-α receptor (ER-α null mice) but was observed in mice lacking the LXR (LXR null mice) (35). The authors also showed that the increase in VWAT mass was the result of adipose tissue hyperplasia and not hypertrophy. These results suggest that higher CYP27A1 expression in VAT as well as 27HC promoting VAT mass by ER-α signaling and adipose tissue proliferation may be potential mechanisms contributing to the positive association between 27HC levels and VAT area. A positive association was also observed between 27HC and percent liver fat in our study. Oxysterols, including 27HC, play an important role in regulating cholesterol homeostasis in the liver by binding to the LXR. The LXR is a key component in regulating fat deposition in hepatic tissues by regulating the expression of genes involved in lipid metabolism (37-39). The LXR also promotes lipogenesis, the synthesis of fatty acids and triglycerides, while inhibiting cholesterol absorption and transport (40). In a small case–control study of nonalcoholic fatty liver disease conducted in Japan, higher levels of 27HC were observed in nonalcoholic fatty liver disease cases than in controls (41), aligning with the positive association we observed between 27HC and percent liver fat.

Among female participants who reported not taking lipid-lowering drugs, we observed statistically significantly inverse associations for 27HC with total fat mass, SAT area, and sSAT area with positive associations for 27HC with VAT area and percent liver fat. In contrast, among those taking lipid-lowering drugs, we observed no statistically significant associations across adiposity phenotypes, with the exception of 27HC and VAT area. Lipid-lowering drugs can have varying effects on different types of adipose tissues (42, 43). Laboratory studies have demonstrated varying biological effects of lipid-lowering drugs on adipokine secretion and lipogenesis in adipocytes originating from visceral vs subcutaneous adipose tissue (42, 43). Our findings among females not taking lipid-lowering drugs provide supportive evidence of associations between 27HC and adiposity phenotypes that are unbiased to the effects of lipid-lowering drugs.

In our previous publications (25, 44), we highlighted the potential utility of the VAT score in further exploring the association of visceral fat with the risk of postmenopausal breast cancer and type 2 diabetes. Given the positive association observed between VAT and 27HC, it is conceivable that circulating levels of 27HC could be serve as a cost-effective indicator for VAT and consequently for cancer and diabetes risk (25, 44), considering that VAT typically requires measurement via MRI.

We observed higher levels of circulating sex hormones (free estradiol, free estrone, and free testosterone) associated with greater total fat, total lean mass, and other adiposity phenotypes. The most notable associations were observed for total fat mass and total trunk fat, particularly for estradiol where strong associations were seen for SAT and sSAT areas relative to VAT area and percent liver fat. These findings are consistent with previous reports of positive associations for increasing adiposity and anthropometric measures with circulating estradiol in postmenopausal women (12, 13, 45-47). Moreover, when evaluating the relationship between the levels of estradiol with adipose deposition, estradiol is produced in both the VAT and SAT areas, but it is produced more efficiently through 17β-HSD enzyme activity in SAT than in VAT (48), supporting our observation of a stronger positive association for SAT area and circulating estradiol (P trend < .0001) than VAT area and circulating estradiol (P trend = .001).

To our knowledge, this is the first study to evaluate associations of adiposity phenotypes with circulating 27HC. The strengths of the study include the inclusion of five racial and ethnic groups with detailed adiposity assessments in a single study population. In addition, we were able to account for key covariates, including demographic factors, physical activity, diabetes status, and use of lipid-lowering drugs. However, there are some limitations. Our study is cross-sectional and we are unable to assess temporal relationships of adiposity phenotypes with 27HC and hormone levels. Furthermore, the number of tests conducted raises the possibility of false positive associations. Additionally, our study population consists of postmenopausal women, we are unable to generalize our findings to males and younger females.

In conclusion, our study reveals novel associations for 27HC, a cholesterol metabolite and endogenous SERM, with specific adiposity phenotypes in a multiethnic population of postmenopausal women. These findings warrant further investigation in larger study populations to extend our understanding of the biology and relationships of 27HC with adiposity and adiposity-related diseases.

Acknowledgments

We thank the Multiethnic Cohort Study participants who generously donated their time and effort for the Multiethnic Cohort Adiposity Phenotype Study.

Abbreviations

27HC

27-hydroxychlolesterol

BMI

body mass index

CV

coefficient of variation

DXA

dual-energy x-ray absorptiometry

ER-α

estrogen receptor-alpha

LXR

liver X receptor

MEC-APS

Multiethnic Cohort Adiposity Phenotype Study

MRI

magnetic resonance imaging

SERM

selective estrogen receptor modulator

SHBG

sex hormone–binding globulin

SAT

subcutaneous adipose tissue

sSAT

superficial subcutaneous adipose tissue

SWAN

Women's Health Across the Nation

VAT

visceral adipose tissue

VWAT

visceral white adipose tissue

Contributor Information

Yuqing Li, Department of Epidemiology and Biostatistics, School of Medicine, University of California San Francisco, 550 16th Street, San Francisco, CA 94538, USA.

Samantha A Streicher, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Adrian A Franke, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Anne N Tome, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Kami K White, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Yurii Shvetsov, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Unhee Lim, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Veronica W Setiawan, Center for Genetic Epidemiology, Department of Population and Public Health Sciences, Keck School of Medicine, University of Southern California, 1845 N Soto St, Los Angeles, CA 90032, USA.

Mindy C DeRouen, Department of Epidemiology and Biostatistics, School of Medicine, University of California San Francisco, 550 16th Street, San Francisco, CA 94538, USA.

Brenda Y Hernandez, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Anna H Wu, Center for Genetic Epidemiology, Department of Population and Public Health Sciences, Keck School of Medicine, University of Southern California, 1845 N Soto St, Los Angeles, CA 90032, USA.

Lynne R Wilkens, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Loïc Le Marchand, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Lenora W M Loo, University of Hawaii Cancer Center, University of Hawaii at Mānoa, 701 Ilalo St, Honolulu, HI 96813, USA.

Iona Cheng, Department of Epidemiology and Biostatistics, School of Medicine, University of California San Francisco, 550 16th Street, San Francisco, CA 94538, USA.

Funding

This work was supported by the National Cancer Institute (R01CA229815 to L.W.M.L. and I.C. and P01 CA168530 to L.L.M.). The Multiethnic Cohort Study is supported by U01 CA164973 to L.L.M., L.R.W. Additional funding included NCI (P30 CA071789 to University of Hawaii Cancer Center Shared Resources for Biostatistics, Analytical Biochemistry, Genomics, and Nutrition Support services) and the National Center for Advancing Translational Science, NIH, for recruitment activities at USC (UL1TR000130 to Southern California Clinical and Translational Science Institute). S.A.S. was funded by a NCI training grant (T32 CA229100).

Disclosures

The authors have nothing to disclose.

Data Availability

The Multiethnic Cohort investigators and institutions affirm their intention to share the research data consistent with all relevant NIH resource/data sharing policies. Data requests should be submitted through Multiethnic Cohort online data request system at https://www.uhcancercenter.org/for-researchers/mec-data-sharing.

References

  • 1. DeNino WF, Tchernof A, Dionne IJ, et al. Contribution of abdominal adiposity to age-related differences in insulin sensitivity and plasma lipids in healthy nonobese women. Diabetes Care. 2001;24(5):925‐932. [DOI] [PubMed] [Google Scholar]
  • 2. Matsuzawa Y, Shimomura I, Nakamura T, Keno Y, Kotani K, Tokunaga K. Pathophysiology and pathogenesis of visceral fat obesity. Obes Res. 1995;3(Suppl 2):187S‐194S. [DOI] [PubMed] [Google Scholar]
  • 3. Nakamura T, Tokunaga K, Shimomura I, et al. Contribution of visceral fat accumulation to the development of coronary artery disease in non-obese men. Atherosclerosis. 1994;107(2):239‐246. [DOI] [PubMed] [Google Scholar]
  • 4. Fox CS, Massaro JM, Hoffmann U, et al. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the framingham heart study. Circulation. 2007;116(1):39‐48. [DOI] [PubMed] [Google Scholar]
  • 5. Keum N, Lee DH, Kim R, Greenwood DC, Giovannucci EL. Visceral adiposity and colorectal adenomas: dose-response meta-analysis of observational studies. Ann Oncol. 2015;26(6):1101‐1109. [DOI] [PubMed] [Google Scholar]
  • 6. Patel P, Abate N. Role of subcutaneous adipose tissue in the pathogenesis of insulin resistance. J Obes. 2013;2013:489187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Pou KM, Massaro JM, Hoffmann U, et al. Visceral and subcutaneous adipose tissue volumes are cross-sectionally related to markers of inflammation and oxidative stress: the framingham heart study. Circulation. 2007;116(11):1234‐1241. [DOI] [PubMed] [Google Scholar]
  • 8. Ritchie SA, Connell JM. The link between abdominal obesity, metabolic syndrome and cardiovascular disease. Nutr Metab Cardiovasc Dis. 2007;17(4):319‐326. [DOI] [PubMed] [Google Scholar]
  • 9. Nelson ER, Wardell SE, Jasper JS, et al. 27-Hydroxycholesterol links hypercholesterolemia and breast cancer pathophysiology. Science. 2013;342(6162):1094‐1098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Wu Q, Ishikawa T, Sirianni R, et al. 27-Hydroxycholesterol promotes cell-autonomous, ER-positive breast cancer growth. Cell Rep. 2013;5(3):637‐645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Pasquali R. Obesity and androgens: facts and perspectives. Fertil Steril. 2006;85(5):1319‐1340. [DOI] [PubMed] [Google Scholar]
  • 12. Bezemer ID, Rinaldi S, Dossus L, et al. C-peptide, IGF-I, sex-steroid hormones and adiposity: a cross-sectional study in healthy women within the European prospective investigation into cancer and nutrition (EPIC). Cancer Causes Control. 2005;16(5):561‐572. [DOI] [PubMed] [Google Scholar]
  • 13. Marchand GB, Carreau AM, Weisnagel SJ, et al. Increased body fat mass explains the positive association between circulating estradiol and insulin resistance in postmenopausal women. Am J Physiol Endocrinol Metab. 2018;314(5):E448‐EE56. [DOI] [PubMed] [Google Scholar]
  • 14. Anderson AJ, Sobocinski KA, Freedman DS, Barboriak JJ, Rimm AA, Gruchow HW. Body fat distribution, plasma lipids, and lipoproteins. Arterioscler Thromb Vasc Biol. 1988;8(1):88‐94. [DOI] [PubMed] [Google Scholar]
  • 15. Seidell JC, Cigolini M, Deurenberg P, Oosterlee A, Doornbos G. Fat distribution, androgens, and metabolism in nonobese women. Am J Clin Nutr. 1989;50(2):269‐273. [DOI] [PubMed] [Google Scholar]
  • 16. Freedman DS, Jacobsen SJ, Barboriak JJ, et al. Body fat distribution and male/female differences in lipids and lipoproteins. Circulation. 1990;81(5):1498‐1506. [DOI] [PubMed] [Google Scholar]
  • 17. Pascot A, Lemieux S, Lemieux I, et al. Age-related increase in visceral adipose tissue and body fat and the metabolic risk profile of premenopausal women. Diabetes Care. 1999;22(9):1471‐1478. [DOI] [PubMed] [Google Scholar]
  • 18. Barzi F, Woodward M, Czernichow S, et al. The discrimination of dyslipidaemia using anthropometric measures in ethnically diverse populations of the Asia-pacific region: the obesity in Asia collaboration. Obes Rev. 2010;11(2):127‐136. [DOI] [PubMed] [Google Scholar]
  • 19. Azizi F, Esmaillzadeh A, Mirmiran P, Ainy E. Is there an independent association between waist-to-hip ratio and cardiovascular risk factors in overweight and obese women? Int J Cardiol. 2005;101(1):39‐46. [DOI] [PubMed] [Google Scholar]
  • 20. Janssen I, Powell LH, Kazlauskaite R, Dugan SA. Testosterone and visceral fat in midlife women: the study of Women's health across the nation (SWAN) fat patterning study. Obesity (Silver Spring). 2010;18(3):604‐610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Kavanagh K, Espeland MA, Sutton-Tyrrell K, Barinas-Mitchell E, El Khoudary SR, Wildman RP. Liver fat and SHBG affect insulin resistance in midlife women: the study of Women's health across the nation (SWAN). Obesity (Silver Spring). 2013;21(5):1031‐1038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Lim U, Monroe KR, Buchthal S, et al. Propensity for intra-abdominal and hepatic adiposity varies among ethnic groups. Gastroenterology. 2019;156(4):966‐975. e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Kolonel LN, Henderson BE, Hankin JH, et al. A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol. 2000;151(4):346‐357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Epplein M, Franke AA, Cooney RV, et al. Association of plasma micronutrient levels and urinary isoprostane with risk of lung cancer: the multiethnic cohort study. Cancer Epidemiol Biomarkers Prev. 2009;18(7):1962‐1970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Le Marchand L, Wilkens LR, Castelfranco AM, et al. Circulating biomarker score for visceral fat and risks of incident colorectal and postmenopausal breast cancer: the multiethnic cohort adiposity phenotype study. Cancer Epidemiol Biomarkers Prev. 2020;29(5):966‐973. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Honda A, Yamashita K, Ikegami T, et al. Highly sensitive quantification of serum malonate, a possible marker for de novo lipogenesis, by LC-ESI-MS/MS. J Lipid Res. 2009;50(10):2124‐2130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Li X, Franke AA. Improved profiling of estrogen metabolites by orbitrap LC/MS. Steroids. 2015;99(Pt A):84‐90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in Serum. The Journal of Clinical Endocrinology & Metabolism. 1999;18:3666‐3672. [DOI] [PubMed] [Google Scholar]
  • 29. Li Y, Streicher SA, Franke AA, et al. Supplementary material and tables for the manuscript “Association of Adiposity Phenotypes with 27-Hydroxycholesterol and Sex Hormones: The Multiethnic Cohort Study”. Zenodo Repository. 2024. Deposited 15 June 2024. 10.5281/zenodo.11682335 [DOI] [PMC free article] [PubMed]
  • 30. Bernstein L, Ross RK. Endogenous hormones and breast cancer risk. Epidemiol Rev. 1993;15(1):48‐65. [DOI] [PubMed] [Google Scholar]
  • 31. Rossouw JE, Prentice RL, Manson JE, et al. Relationships of coronary heart disease with 27-hydroxycholesterol, low-density lipoprotein cholesterol, and menopausal hormone therapy. Circulation. 2012;126(13):1577‐1586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Le Cornet C, Johnson TS, Lu DL, Kaaks R, Fortner RT. Association between lifestyle, dietary, reproductive, and anthropometric factors and circulating 27-hydroxycholesterol in EPIC-Heidelberg. Cancer Causes Control. 2020;31(2):181‐192. [DOI] [PubMed] [Google Scholar]
  • 33. Li J, Daly E, Campioli E, Wabitsch M, Papadopoulos V. De novo synthesis of steroids and oxysterols in adipocytes. J Biol Chem. 2014;289(2):747‐764. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Wamberg L, Christiansen T, Paulsen SK, et al. Expression of vitamin D-metabolizing enzymes in human adipose tissue – the effect of obesity and diet-induced weight loss. Int J Obes (Lond). 2013;37(5):651‐657. [DOI] [PubMed] [Google Scholar]
  • 35. Asghari A, Ishikawa T, Hiramitsu S, et al. 27-Hydroxycholesterol promotes adiposity and mimics adipogenic diet-induced inflammatory signaling. Endocrinology. 2019;160(10):2485‐2494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Asghari A, Umetani M. Obesity and cancer: 27-hydroxycholesterol, the missing link. Int J Mol Sci. 2020;21(14):4822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Gill S, Chow R, Brown AJ. Sterol regulators of cholesterol homeostasis and beyond: the oxysterol hypothesis revisited and revised. Prog Lipid Res. 2008;47(6):391‐404. [DOI] [PubMed] [Google Scholar]
  • 38. Janowski BA, Willy PJ, Devi TR, Falck JR, Mangelsdorf DJ. An oxysterol signalling pathway mediated by the nuclear receptor LXR alpha. Nature. 1996;383(6602):728‐731. [DOI] [PubMed] [Google Scholar]
  • 39. Nakamuta M, Fujino T, Yada R, et al. Impact of cholesterol metabolism and the LXRalpha-SREBP-1c pathway on nonalcoholic fatty liver disease. Int J Mol Med. 2009;23(5):603‐608. [DOI] [PubMed] [Google Scholar]
  • 40. Schulman IG. Liver X receptors link lipid metabolism and inflammation. FEBS Lett. 2017;591(19):2978‐2991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Ikegami T, Hyogo H, Honda A, et al. Increased serum liver X receptor ligand oxysterols in patients with non-alcoholic fatty liver disease. J Gastroenterol. 2012;47(11):1257‐1266. [DOI] [PubMed] [Google Scholar]
  • 42. Labuzek K, Buldak L, Dulawa-Buldak A, et al. Atorvastatin and fenofibric acid differentially affect the release of adipokines in the visceral and subcutaneous cultures of adipocytes that were obtained from patients with and without mixed dyslipidemia. Pharmacol Rep. 2011;63(5):1124‐1136. [DOI] [PubMed] [Google Scholar]
  • 43. Aguirre L, Hijona E, Macarulla MT, et al. Several statins increase body and liver fat accumulation in a model of metabolic syndrome. J Physiol Pharmacol. 2013;64(3):281‐288. [PubMed] [Google Scholar]
  • 44. Maskarinec G, Raquinio PA, Setiawan VW, et al. Biomarker-based visceral adiposity score and incident type 2 diabetes in the multiethnic cohort. Ann Epidemiol. 2021;63:29‐34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Dashti SG, Simpson JA, Karahalios A, et al. Adiposity and estrogen receptor-positive, postmenopausal breast cancer risk: quantification of the mediating effects of fasting insulin and free estradiol. Int J Cancer. 2020;146(6):1541‐1552. [DOI] [PubMed] [Google Scholar]
  • 46. Baglietto L, English DR, Hopper JL, et al. Circulating steroid hormone concentrations in postmenopausal women in relation to body size and composition. Breast Cancer Res Treat. 2009;115(1):171‐179. [DOI] [PubMed] [Google Scholar]
  • 47. Chen YW, Hang D, Kvaerner AS, Giovannucci E, Song M. Associations between body shape across the life course and adulthood concentrations of sex hormones in men and pre- and postmenopausal women: a multicohort study. Br J Nutr. 2022;127(7):1000‐1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Hetemaki N, Savolainen-Peltonen H, Tikkanen MJ, et al. Estrogen metabolism in abdominal subcutaneous and visceral adipose tissue in postmenopausal women. J Clin Endocrinol Metab. 2017;102(12):4588‐4595. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Citations

  1. Li Y, Streicher SA, Franke AA, et al. Supplementary material and tables for the manuscript “Association of Adiposity Phenotypes with 27-Hydroxycholesterol and Sex Hormones: The Multiethnic Cohort Study”. Zenodo Repository. 2024. Deposited 15 June 2024. 10.5281/zenodo.11682335 [DOI] [PMC free article] [PubMed]

Data Availability Statement

The Multiethnic Cohort investigators and institutions affirm their intention to share the research data consistent with all relevant NIH resource/data sharing policies. Data requests should be submitted through Multiethnic Cohort online data request system at https://www.uhcancercenter.org/for-researchers/mec-data-sharing.

Articles from The Journal of Clinical Endocrinology and Metabolism are provided here courtesy of The Endocrine Society

RESOURCES