Skip to main content
NCBI home page

This is a preprint.

It has not yet been peer reviewed by a journal.

The National Library of Medicine is running a pilot to include preprints that result from research funded by NIH in PMC and PubMed.

medRxiv logoLink to medRxiv
[Preprint]. 2025 Sep 5:2025.09.02.25334965. [Version 1] doi: 10.1101/2025.09.02.25334965

Heterogeneity of effect of Intensive lifestyle intervention on cardiometabolic risk factors by sex hormones in diabetes

Chigolum P Oyeka 1,*, Jianqiao Ma 2,*, Jiahuan Helen He 1,3, Nityasree Srialluri 5, Teresa Gisinger 6, Erin D Michos 3,4, Mark Woodward 7,8, Rita R Kalyani 9, Jeanne M Clark 10, Wendy L Bennett 1,3,†, Dhananjay Vaidya 1,†
PMCID: PMC12424865  PMID: 40950479

Abstract

Background:

Intensive lifestyle intervention(ILI) in type 2 diabetes(T2D) improves cardiometabolic risk factors. Sex differences in these responses may be driven by the sex hormones testosterone(T), estradiol(E2), and sex hormone binding globulin(SHBG). We evaluated whether baseline sex hormone levels modify ILI effects on cardiometabolic risk factors ie, heterogeneity of treatment effect(HTE), and whether these modifications differ by sex.

Methods:

Study included 2,260 Look AHEAD participants(1,093 postmenopausal females; 1,167 males, mean age 60 years) randomized to ILI or diabetes support and education with sex hormone measurements. We used linear mixed-effects models, stratified by sex and adjusted for age, race, study site, medications and baseline weight, to examine the association between baseline T, E2, and SHBG with change in lipids, hemoglobin A1c(HbA1c), systolic and diastolic blood pressure(BP), weight, and waist circumference(WC) over an 8-year follow-up. HTE was assessed using a three way interaction term between baseline hormone levels, time and randomization arm. Permutation tests controlled for multiple comparisons.

Results:

In males, lower baseline E2 and total T significantly augmented ILI-induced triglyceride reductions(p=0.019 and 0.008, respectively) throughout follow up. Higher SHBG enhanced LDL-C lowering in females(p=0.018) and HDL-C increases in males(p=0.043). Higher baseline total T predicted greater long-term weight(p=0.013 females; 0.003 males) and WC loss(p=0.061 females; 0.035 males), while fewer hormone interactions were observed for BP and HbA1c.

Conclusions:

Males with greater endogenous total T achieved larger reductions in triglycerides, weight, and WC, whereas females with higher total T had greater HDL-C improvements and those with higher SHBG experienced more BP lowering. Hormone profiling may guide personalized lifestyle prescriptions to improve long term cardiometabolic benefits.

Introduction

In people living with type 2 diabetes (T2D), intensive behavioral lifestyle interventions that result in weight loss lead to significant improvements in cardiometabolic risk factors, including lipid profiles (triglycerides [TG], low-density-lipoprotein cholesterol [LDL-C], high-density-lipoprotein cholesterol [HDL-C]), glycemic control (HbA1c)), blood pressure (systolic [SBP] and diastolic [DBP]), and waist circumference (WC))1,2. Importantly, sex differences in response to cardiometabolic risk after behavioral weight loss interventions have been previously reported3–8. Females experience greater improvement in fasting glucose, TG, and HDL-C and less improvement in HbA1c and LDL-C4,7,8. Comparatively, males tend to experience greater reductions in visceral adiposity and blood pressure, and less improvements in TG and HDL-C4,5,7. Sex differences in cardiometabolic risk factors may be moderated by biological factors, including hormonal changes and differential fat distributions, highlighting the complexity of sex-specific metabolic adaptations9,10.

Sex hormones (testosterone [T], estrogen [E2]) and sex hormone-binding globulin (SHBG)) are known to modulate lipid profiles, glucose metabolism, and fat distribution, which differentially affect cardiometabolic risk in females and males11,12. Variations in these hormone levels is associated with different female-male metabolic profiles, with low T in males and unusually low or abnormally high E2 levels in females being linked to adverse lipid and glucose outcomes13. Moreover, SHBG is associated with metabolic regulation, where lower concentrations correlate with T2D risk and cardiometabolic events14–16.

The Look AHEAD (Action for Health in Diabetes) randomized controlled trial provided a unique opportunity to investigate the long-term impacts of an intensive lifestyle intervention (ILI) for weight loss on a range of cardiometabolic risk factors in individuals with T2D and overweight or obesity. The trial enrolled over 5,000 adults with overweight and obesity and T2D and compared the effect of an ILI, emphasizing calorie restriction and exercise, vs a diabetes support and education (DSE) control on cardiometabolic morbidity and mortality17. In the Sex Hormone ancillary study to Look AHEAD, we previously measured sex hormones (T, E2, and SHBG) and assessed the role of the intervention on sex hormones and any sex differences among postmenopausal females and older males living with T2D18.

In the current study, we examined whether baseline sex hormone levels modify the response to the ILI targeting cardiometabolic risk factors: TG, LDL-C, HDL-C, HbA1c, SBP, DBP, weight, and WC. We also assessed whether there are differences in these responses by sex. We hypothesized that distinct baseline sex hormone profiles would predict differential trajectories in these cardiometabolic risk factors in response to lifestyle modifications. Elucidating the interactions between the ILI and sex hormones would advance our understanding of sex-specific metabolic adaptations in cardiometabolic risk in people with T2D, as an important step to personalize treatment for cardiometabolic disease prevention.

Methods

Study Design and Population

The Look AHEAD trial randomized 5,145 overweight or obese adults with T2D to ILI vs DSE and followed them for a median of 9.6 years, including a formal post-trial extension that has provided up to 8 years of additional follow-up for key cardiometabolic outcomes19. The Sex Hormone Ancillary Study was nested within this parent trial8,18. The Johns Hopkins School of Medicine IRB approved the study, and the Look AHEAD steering committee granted ancillary study approval.

Sample Selection

In the Sex Hormone Ancillary Study, we selected a random sample of postmenopausal females (n=1093) and older males (n=1167, total n =2260) from the 5145 females (n=3063) and males (n=2082) in the main Look AHEAD trial to assess sex hormone levels (T, E2, and SHBG) over time18. Sample selection and baseline characteristics have been previously described18,20. Plasma samples were taken at 3 time points (baseline, year 1, and year 4)18. Exclusion criteria were missing stored samples at baseline, year 1 or year 4, females receiving breast cancer treatment or taking exogenous hormones at baseline, females < 55 years of age at baseline and either pre-menopausal or who had hysterectomy without oophorectomy, and males on anti-androgen medications for prostate cancer or androgen replacement at baseline18.

Data Variables

Intervention:

Participants were randomly assigned to ILI vs. DSE. The ILI comprised calorie restriction, physical activity, and frequent individual and group counselling (weekly in year 1) tapering off over the years of follow up21,22. The DSE arm received three group health education sessions annually without specific diet or exercise goals21,23.

Intervention Effect Modifiers:

Sex hormones T, E2, and SHBG were the exposures. Laboratory methods for the sex hormones assessed in this ancillary study has been previously reported in detail18,20. Briefly, total T and E2 were measured using highly sensitive assays (negative electron capture chemical ionization gas chromatography-mass spectrometry [GCMS]) to ensure enhanced detection of the low levels in older adults18,20. SHBG was measured using ELISA (RRID: AB_3255149, Cat# K151G9 K, Mesoscale Discovery, MD)18,20.

Outcomes: Cardiometabolic Risk Factors

All cardiometabolic risk factors were measured at baseline and years 1, 2, 3, 4, 6 and 8.

Lipids (TG, HDL-C, LDL-C).

Fasting blood samples were collected for lipid measurements, including TG, and HDL-C21,24. Lipid concentrations were measured using standardized enzymatic assays in a central laboratory, ensuring consistency across study sites21. LDL-C levels were calculated using Friedewald formula, which is accurate for triglycerides <400 mg/dL24. For participants with triglycerides ≥400 mg/dL, we performed a sensitivity analysis by excluding them.

Blood pressure.

SBP and DBP were assessed with participants seated and after a 5-minute rest period; automated sphygmomanometers were used to obtain multiple readings, which were then averaged to improve measurement accuracy21.

Body weight and WC.

Weight was measured in kg annually with a calibrated digital scale25. WC was measured twice and averaged at each annual visit with a non-stretchable measuring tape between the highest point of the iliac crest and the lowest part of the rib cage21,26.

Glycemic measures.

HbA1c was measured from fasting blood specimens using high-performance liquid chromatography methods, calibrated according to Diabetes Control and Complications Trial (DCCT) standards to ensure reliable assessment of glycemic control21,27.

Other Covariates.

Age, race, and study site were all from baseline measurements. Medication use (statins, antihypertension drugs, and insulins) was self-reported for time varying covariates.

Statistical Analysis

We examined whether baseline sex hormone levels modified the effect of ILI on cardiometabolic outcomes over time using linear mixed-effects models. TG was log-transformed to ensure nearly normal distribution. We logarithmically transformed the sex hormone variables and divided them by their sex-specific standard deviation to allow for comparisons of treatment effect between sex hormones. Models were stratified by sex and adjusted for age, race, study site, and baseline weight. Models with lipids as outcomes were adjusted for statin use, models using weight, WC, and HbA1c as outcomes were adjusted for insulin use. Models using outcomes of SBP and DBP were adjusted by antihypertensive medication use. We evaluated heterogeneity of treatment effects using a three-way interaction term. This assessed the joint effects of baseline log-transformed sex hormones (E2, total T, SHBG), intervention (ILI vs. DSE), and follow-up year (1, 2, 3, 4, 6, 8 as categorical variables).

Sex hormone levels were modeled as continuous predictors in the statistical models. To visualize the estimated treatment effects, we plotted the predicted treatment effect in the cardiometabolic risk factors over time for two representative baseline hormone levels, corresponding to the 25th and 75th percentiles of baseline E2, total T, and SHBG, separately for males and females, with error bars representing ±1 standard error. Additionally, based on linear mixed-effects models above, we estimated the adjusted levels of the cardiometabolic variables for ILI and DSE groups for each follow-up year.

Analyses included multiple null hypothesis tests of effect modification, where the proposed effect modifiers, the sex hormone variables, covaried. Thus, multiple testing corrections based on uncorrelated null hypothesis, such as the Bonferroni correction are overly conservative. Because effect modification is for randomization group, a variable uncorrelated with any regression covariate by design, we generated empirical null distributions of the multiply tested 3-way interactions using 4000 random permutations of the randomized group variable. Diagnostic plots determined that this number of permutations produced stable results. We determined that only 5% of iterations had <=9 null hypothesis tests with p<0.05; thus, for this study if >=9 null hypothesis tests were p<0.05, the overall study had a false positive rate of <= 0.05. The multiple testing procedure has been detailed in supplementary material (Methods supplement).

To assess sex differences in the heterogeneity of the treatment effect (HTE) by baseline hormone levels from the sex-stratified regression models, we calculated the differences of each of the standardized hormone by ILI by categorical time interaction terms between female and male models, as differences in normally distributed variates with the estimated means and standard deviations, standardized by the standard error of the difference. The sum of all these standardized difference terms was tested as a chi-square variate with the appropriate number of degrees of freedom. For this exploratory analysis p<=0.05 was used as the level of statistical significance.

All statistical analyses were performed using Stata 18.0, and results with p<0.05 already corrected for multiple comparisons were highlighted.

Results

Supplement Figure 1 shows the selection process for the analytic sample. Table 1 shows the baseline sample characteristics of 1093 females (ILI: 551, DSE: 542) and 1167 males (ILI: 589, DSE: 578).. The mean age was 60 years. There were more Black females (vs. Black males) in both intervention arms (Black females: ILI, 18.5%, DSE, 22.5%, vs. Black males: ILI: 9.5%, DSE: 9.3%). Across the two arms, females had lower median TG (146.0 vs. 163.0 mg/dL) and higher mean HDL-C (46.7 vs. 38.1 mg/dL), but higher mean LDL-C (118.4 vs. 106.5 mg/dL) compared to males. Females had higher mean BMI (36.3 vs. 34.9 kg/m2) but lower WC (110.2 vs. 117.8 cm) compared to males. HbA1c was similar between sexes (mean 7.2%). SBP was comparable, while females had lower DBP (67.6 vs. 73.1 mmHg).

Table 1.

Baseline Characteristics among Postmenopausal Females and Older Males in the Look AHEAD Sex Hormone Ancillary Study (N=2260)

Female (n=1093) Male (n=1167)
Variable ILI DSE ILI DSE
N 551 542 589 578
Age, mean (SD) 60.4 (5.8) 60.2 (6.0) 60.1 (6.6) 60.3 (6.6)
Race/Ethnicity African American / Black (not Hispanic) 102 (18.5%) 122 (22.5%) 56 (9.5%) 54 (9.3%)
American Indian / Native American / Alaskan Native 5 (0.9%) 6 (1.1%) 1 (0.2%) 0 (0.0%)
Asian/Pacific Islander 6 (1.1%) 6 (1.1%) 6 (1.0%) 4 (0.7%)
White 316 (57.4%) 293 (54.1%) 447 (75.9%) 453 (78.4%)
Hispanic 113 (20.5%) 108 (19.9%) 63 (10.7%) 51 (8.8%)
Other/Mixed 9 (1.6%) 7 (1.3%) 16 (2.7%) 16 (2.8%)
Triglycerides (mg/dl), median (IQR) 148.0 (105.0, 213.0) 146.0 (104.0, 204.0) 161.0 (113.0, 223.0) 166.0 (111.0, 238.0)
HDL cholesterol (mg/dl), mean (SD) 46.4 (11.4) 47.0 (11.9) 38.2 (9.1) 38.1 (9.2)
LDL cholesterol (mg/dl), mean (SD) 118.3 (33.3) 118.5 (32.6) 106.1 (29.4) 106.9 (31.9)
Weight in Kg, mean (SD) 93.6 (16.8) 94.8 (16.6) 108.1 (18.9) 108.7 (18.0)
Waist Circumference, mean (SD) 109.5 (13.1) 111.0 (12.7) 117.8 (13.6) 117.8 (12.9)
Body Mass Index (kg/m2), mean (SD) 34.9 (5.7) 34.9 (5.2) 36.1 (5.9) 36.5 (5.8)
Hemoglobin A1c %, mean (SD) 7.2 (1.1) 7.3 (1.2) 7.2 (1.2) 7.3 (1.2)
Systolic Blood Pressure, mean (SD) 129.7 (18.1) 130.4 (16.6) 128.2 (16.4) 129.1 (16.3)
Diastolic Blood Pressure, mean (SD) 67.2 (9.4) 67.9 (9.0) 72.9 (8.9) 73.4 (8.9)
Testosterone, nmol/L, median (IQR) 0.8 (0.5, 1.2) 0.8 (0.5, 1.1) 15.6 (12.0, 20.0) 15.7 (11.7, 20.0)
Estradiol, pmol/L, median (IQR) 37.5 (23.9, 58.3) 37.6 (23.1, 56.3) 106.9 (78.5, 134.3) 104.9 (77.9, 137.0)
Sex Hormone Binding Globulin, nmol/L, median (IQR) 32.6 (21.7, 51.8) 31.5 (21.4, 51.5) 28.8 (19.7, 45.4) 29.1 (19.8, 43.7)
Open in a new tab

Supplement Figure 2 shows the percent change in the cardiometabolic risk factors over 8 years of follow up by sex in the ILI and DSE arms. Over 8 years, compared with DSE, the ILI arm achieved significant cardiometabolic improvements in both sexes.1

Supplement Table 1 summarizes the estimated associations between baseline hormone levels and cardiometabolic outcomes across the follow-up years, stratified by randomization arm and sex. For each year we presented the effect of each sex hormone on outcomes in the ILI and DSE treatment groups and the differential effect between ILI vs DSE based on the interaction term (treatment effect).

Heterogeneity of Treatment Effects (HTE) of Baseline sex Hormones on Cardiometabolic Risk Factors

Lipids (TG, LDL-C, HDL-C) (Figure 1, Supplement table 2)

Figure 1.

Figure 1.

Open in a new tab

Treatment effects of ILI (vs DSE) on lipids over time

Males with higher baseline E2, and total T experienced TG increases (at year 4, 6 and 8; HTE p = 0.019 for E2, and at year 6 and 8; HTE p = 0.008 for total T) in response to ILI, whereas no similar HTE pattern was observed in females (HTE p =0.800, 0.145, sex difference p = 0.003, 0.001, respectively).

Females with higher baseline SHBG experienced more LDL-C reduction (HTE p = 0.018) and excluding those with triglycerides ≥400 mg/dL did not change the results (HTE p-value: 0.022), whereas no effect was seen in males (HTE p =0.914, sex difference p = 0.019).

Males with higher baseline SHBG experienced greater HDL-C increases from year 3 onwards (HTE p = 0.043). Similarly, females with higher baseline SHBG experienced greater HDL-C increases beginning in year 3, but it was not statistically significant. (HTE p = 0.107, sex difference p = 0.960)

Blood Pressure (SBP, DBP) (Figure 2, Supplement table 2)

Figure 2.

Figure 2.

Open in a new tab

Treatment effects of ILI (vs DSE) on blood pressure over time

Females with higher baseline total T experienced greater DBP reduction at year 2 (HTE p=0.030), while a non-statistically significant pattern of treatment effect across varying levels of baseline total T was observed in males (HTE p = 0.181, sex difference p = 0.389). Females with higher baseline SHBG experienced greater SBP reduction, but it was not statistically significant (HTE p=0.054). A different pattern for SBP was seen in males (HTE p = 0.730, sex difference p = 0.001), with no separation of treatment effects on SBP across varying levels of baseline SHBG.

Body Weight and WC (Figure 3, Supplement table 2)

Figure 3.

Figure 3.

Open in a new tab

Treatment effects of ILI (vs DSE) on weight, WC, and HbA1c over time

For weight, both females and males with higher baseline total T experienced greater weight loss, sustained through year 8 in females (HTE p = 0.013) and through year 6 in males (HTE p = 0.003), with interaction by sex (p = 0.049). Males with higher baseline E2 showed greater weight loss through year 6 (HTE p = 0.004). Females showed a similar but non-statistically significant pattern between E2 and weight (HTE p = 0.103, sex difference p = 0.157).

Higher baseline total T was associated with greater reductions in WC in males (HTE p = 0.035), with similar but non-statistically significant pattern observed for WC in females (HTE p = 0.061, sex difference p = 0.869). SHBG showed a sex difference on effect of ILI on WC (sex difference p= 0.006), with higher SHBG associated with greater WC reduction from year 1 onwards in females (HTE p=0.074). In males, there was no treatment effect across different levels of baseline SHBG (HTE p = 0.514).

Hemoglobin A1c (HbA1c) (Figure 3, Supplement Table 2)

Females with lower baseline SHBG exhibited steeper HbA1c declines at year 4 and year 6 (HTE p=0.006), with non-statistically significant pattern observed in males (HTE p = 0.987, sex difference p =0.364). Total T mirrored this trend (HTE p=0.150, sex difference p = 0.485). Males with lower E2 demonstrated more HbA1c reduction due to ILI (HTE p=0.267), with similar and non-statistically significant pattern observed in females (HTE p = 0.526, sex difference p= 0.456)

Discussion

In this secondary analysis of the Look AHEAD trial, we found that the cardiometabolic benefits of an ILI (vs DSE) differed by baseline sex hormone profiles in older adults with T2D, and that these differences were sex specific. Overall, ILI produced significant improvements in weight, and other traditional cardiometabolic risk factors, but the magnitude of changes was modified by baseline sex hormone levels. To our knowledge, this is the first study to assess heterogeneity of treatment effects of an intensive lifestyle intervention by sex hormones and to characterize any sex differences.

Lipid Outcomes (TG, LDL-C, HDL-C).

Baseline total T modulated the ILI treatment effect on TG in males, while SHBG had effects on LDL-C in females, and HDL-C in males.

In females, higher baseline T trended toward steeper ILI-driven TG declines, although not statistically significant (p = 0.145), whereas E2 and SHBG had minimal impact (p>0.7). Conversely, in males, we observed that lower E2 and higher total T at baseline significantly amplified TG reductions during follow-up. These findings are similar to results from cross-sectional studies in cohorts of patients with coronary artery disease. These studies showed that higher circulating E2 correlated with elevated TG in males, and that lower SHBG and lower total T were associated with more atherogenic lipid profiles in males28–31.

For unadjusted changes in LDL-C over time, the DSE group benefited slightly more in LDL-C reduction compared to ILI, which was largely attributed to the DSE group’s increased use of lipid-lowering medications, specifically statins1. However, in our results, higher baseline SHBG in females was associated with greater LDL-C reduction over time, after statins use adjustment. This finding suggests that higher baseline SHBG levels in females may enhance LDL-C reduction following a lifestyle intervention.

In males, baseline SHBG modulated the HDL-C response to ILI, with higher levels associated with greater HDL-C increases. In females, we observed that higher baseline total T and SHBG were associated with greater HDL-C boosting, similar to prior studies, but the difference did not reach statistical significance32,33.

Our finding on the effect of SHBG also supports prior findings of cross-sectional and observational studies that showed SHBG negatively associated with LDL-C and positively associated with HDL-C34–37.

Blood pressure outcomes (SBP, DBP).

Females with higher baseline T and SHBG experienced larger initial reductions in systolic and diastolic blood pressure, but the differences did not reach statistically significance. Benefits in BP reductions were attenuated over the eight years of follow-up. By contrast, male participants’ SBP and DBP trajectories were overlapping across hormone strata, with only a modest sustained SBP reduction linked to higher T (p = 0.088). These sex-specific interactions may reflect the vasodilatory and anti-inflammatory properties of androgens and the role of SHBG as a biomarker of insulin sensitivity and vascular health.

Weight and WC.

Higher baseline T predicted more rapid and durable weight loss due to ILI in both female (p = 0.013) and male participants (p = 0.003). Female participants maintained greater relative reductions in weight through year 8 (sex difference p = 0.049). This supports prior findings in males where higher T was associated with larger weight and metabolic improvements compared to other males with low T38,39. Our study demonstrates that postmenopausal females likewise experience superior, longer-lasting weight loss when relatively androgen rich. E2 influenced male weight trajectories. Male participants with higher baseline E2 lost and sustained more weight loss through year 6, reflecting the known positive correlation between adiposity and peripheral E2 production via aromatization in males40–42. However, higher E2 may be a marker for greater baseline obesity, capacity for loss rather than directly driving weight change40. Higher baseline E2 may support maintenance of weight loss through effects on energy homeostasis and appetite regulation40,43.

Both female and male participants with higher baseline total T achieved the largest reductions in central adiposity, consistent with evidence that androgens promote visceral fat mobilization during calorie restriction44,45. Other sex hormones had modest WC effects (all HTE p>0.07). The lack of sex difference in total T’s impact on WC suggests that endogenous androgenicity facilitated loss of harmful visceral fat in both males and females. Our results support other studies that showed endogenous testosterone concentrations inversely correlated with central adiposity46,47. Modest, non-significant WC effects of E2 and SHBG in our cohort further highlight testosterone as the principal hormonal determinant of ILI-driven central fat reduction.

These results suggest that androgen status at baseline, not sex alone, modulated responsiveness of weight and WC to diet and exercise interventions in T2D.

Hemoglobin A1C.

ILI induced robust HbA1c reductions (~10% at year 1) that partially rebounded by year 8. Baseline hormone levels did not significantly modify these glycemic benefits in either sex (HTE p > 0.20). This largely uniform response in A1c suggests that weight loss is the driver in glucose control, with sex hormones having a minor role. Our finding that glycemic benefits did not differ by baseline hormone levels corroborates results from smaller trials using testosterone-therapy which similarly showed that glycemic improvements correlate closely with changes in adiposity and muscle mass, rather than initial hormone concentrations48–53. From a clinical standpoint, this emphasizes the universal benefit of lifestyle modification for glycemic improvements regardless of sex hormone profiles. It reinforces current guidelines that promote ILI as first-line therapy for glycemic improvements in both males and females with T2D54,55.

Limitations.

First, all female participants were postmenopausal; premenopausal or perimenopausal dynamics could differ. Second, other unmeasured hormones like DHEA or progesterone may have related effects which we have not assessed. Third, the ILI was a comprehensive program (diet, exercise, behavioral support); we cannot disentangle which components drove each effect. Finally, medication changes during follow up, though adjusted for, may confound the effects of lipids, A1c and BP.

Our data suggests that pre-intervention sex hormone profiling could help guide the overall approach to recommendations for weight loss in people with T2D. For example, males with lower E2 or higher T may anticipate greater TG-lowering, whereas females with higher SHBG may derive more BP benefits. Incorporating endocrine markers into risk stratification may guide adjunctive lipid lowering treatment, such as targeting TG-lowering in males with higher E2 or antihypertensives in females with lower SHBG. Future research should validate these hormone-intervention interactions in diverse populations, explore underlying mechanisms in lipoprotein and vascular biology, and test combined lifestyle–hormone modulation strategies (e.g., selective androgen receptor modulators, targeted estrogen modulation) to enhance cardiometabolic outcomes.

Conclusion

Our analysis of Look AHEAD data shows that baseline sex hormones influence ILI benefits in T2D in a sex-specific manner. Males with greater endogenous androgenicity achieved larger reductions in TG, weight, and waist circumference, whereas females with higher total testosterone had greater HDL-C improvements and those with higher SHBG experienced more pronounced blood pressure lowering. These findings extend prior knowledge by identifying hormone-dependent heterogeneity in lifestyle response, and they support a precision medicine approach: assessing a patient’s endocrine status may help personalize lifestyle prescriptions for metabolic health.

Clinical Perspective.

  • What Is New? Baseline sex hormone levels especially estradiol in males and testosterone/SHBG in females, sex-specifically modify the cardiometabolic benefits of intensive lifestyle intervention in type 2 diabetes.

  • What Are the Clinical Implications? Endogenous hormone profiles may help clinicians personalize diet and exercise programs and guide adjunctive therapies to maximize lipid, blood pressure, and adiposity outcomes in females and males with T2D.

Acknowledgement

The authors would like to thank Dr. Allen D Everett and the members of his laboratory. We would also thank Dr. David Graham and former members of the Molecular Determinants Core. We also thank all the participants of the Look AHEAD study.

Funding Sources

This work was funded by NIH-NIDDK grants: R01DK127222 and U01DK57149

The Look AHEAD (Action for Health in Diabetes) trial’s ClinicalTrials.gov number is NCT00017953.

Disclosures:

Disclosure Summary: The authors had full access to all the data in this study and take complete responsibility for the integrity of the data and the accuracy of the data analysis. JMC reports serving as a Scientific Advisor to Boehringer Ingelheim and receiving writing support from Novo Nordisk in the last 3 years. Unrelated to this work, Dr Michos has served as a consultant for Amgen, Arrowhead, AstraZeneca, Bayer, Boehringer Ingelheim, Edwards Life Science, Esperion, Ionis, Eli Lilly, Medtronic, Merck, New Amsterdam, Novartis, Novo Nordisk, and Zoll.

Nonstandard Abbreviations and Acronyms

T2D

Type 2 Diabetes

ILI

Intensive Lifestyle Intervention

DSE

Diabetes Support and Education

TG

Triglycerides

LDL-C

Low Density Lipoprotein Cholesterol

HDL-C

High Density Lipoprotein Cholesterol

HbA1c

Glycated Hemoglobin

SBP

Systolic Blood Pressure

DBP

Diastolic Blood Pressure

WC

Waist Circumference

Total T

Total Testosterone

E2

Estradiol

SHBG

Sex Hormone–Binding Globulin

GCMS

Gas Chromatography–Mass Spectrometry

ELISA

Enzyme Linked Immunosorbent Assay

HTE

Heterogeneity of Treatment Effect

Funding Statement

This work was funded by NIH-NIDDK grants: R01DK127222 and U01DK57149

The Look AHEAD (Action for Health in Diabetes) trial’s ClinicalTrials.gov number is NCT00017953.

Data availability.

Restrictions apply to the availability of some, or all data generated or analyzed during this study to preserve patient confidentiality or because they were used under license. The corresponding author will on request detail the restrictions and any conditions under which access to some data may be provided.

References

  • 1.Wing RR. Long-term effects of a lifestyle intervention on weight and cardiovascular risk factors in individuals with type 2 diabetes mellitus: four-year results of the Look AHEAD trial. Arch Intern Med. 2010;170(17):1566–1575. doi: 10.1001/archinternmed.2010.334 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Pi-Sunyer X. The Look AHEAD Trial: A Review and Discussion Of Its Outcomes. Curr Nutr Rep. 2014;3(4):387–391. doi: 10.1007/s13668-014-0099-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Miller CK, Nagaraja HN, Cheavens JS, Fujita K, Lazarus SA, Brunette DS. Sex Differences in Early Weight Loss Success during a Diabetes Prevention Intervention. Am J Health Behav. 2023;47(2). doi: 10.5993/AJHB.47.2.13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Domaszewski P, Konieczny M, Dybek T, et al. Comparison of the effects of six-week time-restricted eating on weight loss, body composition, and visceral fat in overweight older men and women. Exp Gerontol. 2023;174. doi: 10.1016/j.exger.2023.112116 [DOI] [PubMed] [Google Scholar]
  • 5.Kuk JL, Ross R. Influence of sex on total and regional fat loss in overweight and obese men and women. Int J Obes. 2009;33(6). doi: 10.1038/ijo.2009.48 [DOI] [PubMed] [Google Scholar]
  • 6.Susanto A, Burk J, Hocking S, Markovic T, Gill T. Differences in weight loss outcomes for males and females on a low-carbohydrate diet: A systematic review. Obes Res Clin Pract. 2022;16(6). doi: 10.1016/j.orcp.2022.09.006 [DOI] [PubMed] [Google Scholar]
  • 7.Zhu R, Craciun I, Bernhards-Werge J, et al. Age- and sex-specific effects of a long-term lifestyle intervention on body weight and cardiometabolic health markers in adults with prediabetes: results from the diabetes prevention study PREVIEW. Diabetologia. 2022;65(8). doi: 10.1007/s00125-022-05716-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Oyeka CP, He JH, Ma J, et al. Role of Sex Hormones in Mediating Adiposity Changes from Weight Loss in People with Type 2 Diabetes: Look AHEAD Sex Hormone Study. J Clin Endocrinol Metab. Published online May 20, 2025:dgaf287. doi: 10.1210/clinem/dgaf287 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Tchernof A, Després JP. Pathophysiology of human visceral obesity: An update. Physiol Rev. 2013;93(1). doi: 10.1152/physrev.00033.2011 [DOI] [PubMed] [Google Scholar]
  • 10.Mauvais-Jarvis F. Sex differences in energy metabolism: natural selection, mechanisms and consequences. Nat Rev Nephrol. 2024;20(1). doi: 10.1038/s41581-023-00781-2 [DOI] [PubMed] [Google Scholar]
  • 11.Testosterone Grossmann M. and glucose metabolism in men: Current concepts and controversies. Journal of Endocrinology. 2014;220(3). doi: 10.1530/JOE-13-0393 [DOI] [PubMed] [Google Scholar]
  • 12.Krishnan KC, Mehrabian M, Lusis AJ. Sex differences in metabolism and cardiometabolic disorders. Curr Opin Lipidol. 2018;29(5). doi: 10.1097/MOL.0000000000000536 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.De Paoli M, Zakharia A, Werstuck GH. The Role of Estrogen in Insulin Resistance: A Review of Clinical and Preclinical Data. American Journal of Pathology. 2021;191(9). doi: 10.1016/j.ajpath.2021.05.011 [DOI] [PubMed] [Google Scholar]
  • 14.Hedderson MM, Capra A, Lee C, et al. Longitudinal Changes in Sex Hormone Binding Globulin (SHBG) and Risk of Incident Diabetes: The Study of Women’s Health Across the Nation (SWAN). Diabetes Care. 2024;47(4). doi: 10.2337/dc23-1630 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Li J, Zheng L, Chan KHK, et al. Sex Hormone-Binding Globulin and Risk of Coronary Heart Disease in Men and Women. Clin Chem. 2023;69(4). doi: 10.1093/clinchem/hvac209 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Ding EL, Song Y, Manson JE, et al. Sex Hormone–Binding Globulin and Risk of Type 2 Diabetes in Women and Men. New England Journal of Medicine. 2009;361(12). doi: 10.1056/nejmoa0804381 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Wing RR, Bolin P, Brancati FL, et al. Cardiovascular effects of intensive lifestyle intervention in type 2 diabetes. N Engl J Med. 2013;369(2):145–154. doi: 10.1056/NEJMoa1212914 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Bennett WL, He JH, Michos ED, et al. Weight loss differentially impacts sex hormones in women and men with type 2 diabetes: Look AHEAD sex hormone study. J Clin Endocrinol Metab. Published online 2024. doi: 10.1210/clinem/dgae584 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Wing RR. Effects of Intensive Lifestyle Intervention on All-Cause Mortality in Older Adults With Type 2 Diabetes and Overweight/ Obesity: Results From the Look AHEAD Study. Diabetes Care. 2022;45(5). doi: 10.2337/dc21-1805 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Bennett W, He Jiahuan, Michos E, et al. LookAHEAD Aim 1 Supplement-Revised 8-6-24. September 12, 2024. Accessed November 25, 2024. https://osf.io/8uzh3?view_only=4879e53e2d1b4ef58c09fe2b17a78bb5
  • 21.Ryan DH, Espeland MA, Foster GD, et al. Look AHEAD (Action for Health in Diabetes): design and methods for a clinical trial of weight loss for the prevention of cardiovascular disease in type 2 diabetes. Control Clin Trials. 2003;24(5):610–628. doi: 10.1016/s0197-2456(03)00064-3 [DOI] [PubMed] [Google Scholar]
  • 22.Wadden TA, West DS, Delahanty L, et al. The Look AHEAD study: a description of the lifestyle intervention and the evidence supporting it. Obesity (Silver Spring). 2006;14(5):737–752. doi: 10.1038/oby.2006.84 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Wesche-Thobaben JA. The development and description of the comparison group in the Look AHEAD trial. Clin Trials. 2011;8(3):320–329. doi: 10.1177/1740774511405858 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Kaze AD, Santhanam P, Musani SK, Ahima R, Echouffo-Tcheugui JB. Metabolic dyslipidemia and cardiovascular outcomes in type 2 diabetes mellitus: Findings from the look ahead study. J Am Heart Assoc. 2021;10(7). doi: 10.1161/JAHA.120.016947 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Stewart TM, Bachand AR, Han H, Ryan DH, Bray GA, Williamson DA. Body image changes associated with participation in an intensive lifestyle weight loss intervention. Obesity (Silver Spring). 2011;19(6):1290–1295. doi: 10.1038/oby.2010.276 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Pownall HJ, Schwartz A V, Bray GA, et al. Changes in regional body composition over 8 years in a randomized lifestyle trial: The look AHEAD study. Obesity (Silver Spring). 2016;24(9):1899–1905. doi: 10.1002/oby.21577 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Disclosures: Standards of Medical Care in Diabetes-2019. Diabetes Care. 2019;42. doi: 10.2337/dc19-Sdis01 [DOI] [PubMed] [Google Scholar]
  • 28.Tsygankova O V., Timoshchenko O V., Latyntseva LD, Veretyuk V V. Lipid profile parameters in men with coronary heart disease in different age categories in connection with sex hormone level. Ateroscleroz. 2023;19(4). doi: 10.52727/2078-256x-2023-19-4-404-414 [DOI] [Google Scholar]
  • 29.Wranicz JK, Cygankiewicz I, Rosiak M, Kula P, Kula K, Zareba W. The relationship between sex hormones and lipid profile in men with coronary artery disease. Int J Cardiol. 2005;101(1). doi: 10.1016/j.ijcard.2004.07.010 [DOI] [PubMed] [Google Scholar]
  • 30.Tomaszewski M, Charchar FJ, Maric C, et al. Association between lipid profile and circulating concentrations of estrogens in young men. Atherosclerosis. 2009;203(1). doi: 10.1016/j.atherosclerosis.2008.06.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Wu S, Wu Y, Fang L, Zhao J, Cai Y, Xia W. A negative association between triglyceride glucose-body mass index and testosterone in adult males: a cross-sectional study. Front Endocrinol (Lausanne). 2023;14. doi: 10.3389/fendo.2023.1187212 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Haffner SM, Dunn JF, Katz MS. Relationship of sex hormone-binding globulin to lipid, lipoprotein, glucose, and insulin concentrations in postmenopausal women. Metabolism. 1992;41(3). doi: 10.1016/0026-0495(92)90271-B [DOI] [PubMed] [Google Scholar]
  • 33.Davis SR, Azene ZN, Tonkin AM, Woods RL, McNeil JJ, Islam RM. Higher testosterone is associated with higher HDL-cholesterol and lower triglyceride concentrations in older women: an observational study. Climacteric. 2024;27(3). doi: 10.1080/13697137.2024.2310530 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Bataille V, Perret B, Evans A, et al. Sex hormone-binding globulin is a major determinant of the lipid profile: The PRIME study. Atherosclerosis. 2005;179(2). doi: 10.1016/j.atherosclerosis.2004.10.029 [DOI] [PubMed] [Google Scholar]
  • 35.Ali Mohammadrezaei, Abnoos Mokhtari Ardekani, Mahdieh Abbasalizad-Farhangi, Mehran Mesgari-Abbasi, Reihaneh Mousavi. Association Between Sex Hormone-Binding Globulin, Atherogenic Indices of Plasma Among Young Sedentary Males. Nutr Metab Insights. 2023;16:11786388231155006. doi: 10.1177/11786388231155006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Yang J, Zhou J, Liu H, et al. Blood lipid levels mediating the effects of sex hormone-binding globulin on coronary heart disease: Mendelian randomization and mediation analysis. Sci Rep. 2024;14(1):11993. doi: 10.1038/s41598-024-62695-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Vaidya D, Dobs A, Gapstur SM, et al. The association of endogenous sex hormones with lipoprotein subfraction profile in the Multi-Ethnic Study of Atherosclerosis. Metabolism. 2008;57(6):782–790. doi: 10.1016/j.metabol.2008.01.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Francomano D, Lenzi A, Aversa A. Effects of five-year treatment with testosterone undecanoate on metabolic and hormonal parameters in ageing men with metabolic syndrome. Int J Endocrinol. 2014;2014. doi: 10.1155/2014/527470 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Saad F, Haider A, Doros G, Traish A. Long-term treatment of hypogonadal men with testosterone produces substantial and sustained weight loss. Obesity. 2013;21(10). doi: 10.1002/oby.20407 [DOI] [PubMed] [Google Scholar]
  • 40.Rubinow KB. Estrogens and Body Weight Regulation in Men. In: Mauvais-Jarvis F, ed. Sex and Gender Factors Affecting Metabolic Homeostasis, Diabetes and Obesity. Springer International Publishing; 2017:285–313. doi: 10.1007/978-3-319-70178-3_14 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Ahmed F, Hetty S, Laterveer R, et al. Altered Expression of Aromatase and Estrogen Receptors in Adipose Tissue From Men With Obesity or Type 2 Diabetes. J Clin Endocrinol Metab. Published online January 21, 2025:dgaf038. doi: 10.1210/clinem/dgaf038 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Mair KM, Gaw R, MacLean MR. Obesity, estrogens and adipose tissue dysfunction – implications for pulmonary arterial hypertension. Pulm Circ. 2020;10(3). doi: 10.1177/2045894020952023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Vigil P, Meléndez J, Petkovic G, Del Río JP. The importance of estradiol for body weight regulation in women. Front Endocrinol (Lausanne). 2022;13. doi: 10.3389/fendo.2022.951186 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Lovejoy JC, Bray GA, Bourgeois MO, et al. Exogenous androgens influence body composition and regional body fat distribution in obese postmenopausal women - A clinical research center study. Journal of Clinical Endocrinology and Metabolism. 1996;81(6). doi: 10.1210/jc.81.6.2198 [DOI] [PubMed] [Google Scholar]
  • 45.Blouin K, Boivin A, Tchernof A. Androgens and body fat distribution. J Steroid Biochem Mol Biol. 2008;108(3):272–280. doi: 10.1016/j.jsbmb.2007.09.001 [DOI] [PubMed] [Google Scholar]
  • 46.Wang P, Li Q, Wu L, et al. Association between the weight-adjusted-waist index and testosterone deficiency in adult males: a cross-sectional study. Sci Rep. 2024;14(1):25574. doi: 10.1038/s41598-024-76574-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Jang S, Ryder JR, Kelly AS, Bomberg EM. Association between endogenous sex hormones and adiposity in youth across a weight status spectrum. Pediatr Res. Published online 2024. doi: 10.1038/s41390-024-03578-6 [DOI] [PubMed] [Google Scholar]
  • 48.Group DPPR. Reduction in the Incidence of Type 2 Diabetes with Lifestyle Intervention or Metformin - NEJMoa012512. New England Journal of Medicine. 2002;346(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Kumar S, Khatri M, Memon RA, et al. Effects of testosterone therapy in adult males with hypogonadism and T2DM: A meta-analysis and systematic review. Diabetes and Metabolic Syndrome: Clinical Research and Reviews. 2022;16(8). doi: 10.1016/j.dsx.2022.102588 [DOI] [PubMed] [Google Scholar]
  • 50.Srikanthan P, Karlamangla AS. Relative muscle mass is inversely associated with insulin resistance and prediabetes. Findings from the Third National Health and Nutrition Examination Survey. Journal of Clinical Endocrinology and Metabolism. 2011;96(9). doi: 10.1210/jc.2011-0435 [DOI] [PubMed] [Google Scholar]
  • 51.Lovejoy JC, Champagne CM, De Jonge L, Xie H, Smith SR. Increased visceral fat and decreased energy expenditure during the menopausal transition. Int J Obes. 2008;32(6). doi: 10.1038/ijo.2008.25 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Salpeter SR, Walsh JME, Ormiston TM, Greyber E, Buckley NS, Salpeter EE. Meta-analysis: Effect of hormone-replacement therapy on components of the metabolic syndrome in postmenopausal women. Diabetes Obes Metab. 2006;8(5). doi: 10.1111/j.1463-1326.2005.00545.x [DOI] [PubMed] [Google Scholar]
  • 53.Ding EL, Song Y, Manson JE, et al. Sex Hormone–Binding Globulin and Risk of Type 2 Diabetes in Women and Men. New England Journal of Medicine. 2009;361(12). doi: 10.1056/nejmoa0804381 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.LeBlanc ES, Patnode CD, Webber EM, Redmond N, Rushkin M, O’Connor EA. Behavioral and pharmacotherapy weight loss interventions to prevent obesity-related morbidity and mortality in adults updated evidence report and systematic review for the US preventive services task force. JAMA - Journal of the American Medical Association. 2018;320(11). doi: 10.1001/jama.2018.7777 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Elsayed NA, Aleppo G, Aroda VR, et al. 1. Improving Care and Promoting Health in Populations: Standards of Care in Diabetes—2023. Diabetes Care. 2023;46(supp). doi: 10.2337/dc23-S001 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Data Availability Statement

Restrictions apply to the availability of some, or all data generated or analyzed during this study to preserve patient confidentiality or because they were used under license. The corresponding author will on request detail the restrictions and any conditions under which access to some data may be provided.

Articles from medRxiv are provided here courtesy of Cold Spring Harbor Laboratory Preprints

RESOURCES