Association of High-Density Lipoprotein Cholesterol With Sex Steroid Treatment in Transgender and Gender-Diverse Youth

Mortality and morbidity from cardiovascular disease differ between men and women, and sex differences in lipid profiles may contribute to this difference. Starting at puberty, male individuals have more atherogenic cholesterol profiles, including lower high-density lipoprotein cholesterol (HDL-C) levels, than female individuals.¹ The degree to which sex hormones vs sex chromosomes and other factors contribute to this difference is unknown. The gender-affirming treatment of transgender and gender-diverse youth provides a unique opportunity to study the association of sex steroids with cholesterol in different sex-chromosome contexts.

Adolescents with no prior gonadotropin-releasing hormone analog use were recruited prior to initiating gender-affirming hormones (GAH) at 1 of 4 study sites (Boston Children’s Hospital in Boston, Massachusetts; Children’s Hospital of Los Angeles in Los Angeles, California; Ann & Robert Lurie Children’s Hospital of Chicago in Chicago, Illinois; and Benioff Children’s Hospital in San Francisco, California) between July 2016 and September 2018.² The institutional review board at each site approved the study. Written informed consent was obtained from the participant or their guardian for those younger than 18 years. Laboratory and anthropometric data were collected as part of clinical care at baseline and at 6 and 12 months after beginning GAH. Obesity was defined as a baseline body mass index more than the 95th percentile for designated sex. Comparisons were made via random mixed-effects modeling using Stata version 16 (StataCorp) with a P value less than .05 considered significant. Analysis began May 2020.

Our analysis included 269 transgender and gender-diverse youth. Eighty-three participants (31%) were designated male at birth (DMAB), and 186 (69%) were designated female at birth (DFAB). We previously reported that HDL-C levels are lower at baseline in transgender and gender-diverse youth in our cohort than in a reference adolescent cohort (Figure 1).^2,3 After 6 months of estradiol treatment in participants who were DMAB, mean (SD) HDL-C levels increased by 11.2 (8.8) mg/dL (to convert to millimoles per liter, multiply by 0.0259) (95% CI, 8.6-13.8; P < .001) to fall within the range for female adolescents, and after 6 months of testosterone treatment in participants who were DFAB, decreased by a mean (SD) of 7.2 (10.1) mg/dL (95% CI, 5.3-9.1; P < .001) to fall to a level nearly identical to that of DMAB participants at baseline (Figure 1).

Obesity attenuated the benefit of estradiol treatment on HDL-C levels and exacerbated the association of testosterone treatment with outcomes after 6 months. Participants who were DMAB with obesity had lower HDL-C levels at baseline than participants who were DMAB without obesity (mean [SD], 36.0 [1.0] mg/dL vs 46.1 [8.8 mg/dL]; 95% CI, 35.2-36.8 vs 43.8-48.4; P = .01) and demonstrated no significant change in HDL-C levels with estradiol treatment (mean [SD], 36.0 [1.0] mg/dL vs 41.3 [4.0] mg/dL; 95% CI, 35.2-36.8 vs 37.6-45.0; P = .10). For participants who were DFAB, although baseline HDL-C levels were not significantly different between participants with and without obesity (mean [SD], 48.9 [12.2] mg/dL vs 53.5 [12.4] mg/dL; 95% CI, 44.8-53.0 vs 51.3-55.7; P = .053), the fall in HDL-C levels with testosterone was more pronounced in participants with obesity than without obesity (mean [SD], −12.1 [9.9] mg/dL vs −5.5 [9.7] mg/dL; 95% CI, −16.3 to −7.9 vs −7.7 to −3.4; P = .004) (Figure 2).

Age, race, and tobacco use did not modify GAH-induced changes in HDL-C levels. GAH treatment produced no significant changes in low-density lipoprotein cholesterol or triglyceride levels (data not shown).

The decrease in HDL-C levels with testosterone is consistent with prior reports. An increase in HDL-C levels with estradiol treatment has not been consistently observed in prior studies, possibly because of differing degrees of testosterone suppression, administration in adulthood vs adolescence, and/or smaller sample sizes.^4-6 To our knowledge, this is the first report of the modifying association of obesity with changes in HDL-C levels with GAH treatment, exacerbating the decrease in HDL-C levels with testosterone and blunting the benefit of estradiol treatment. Our results provide direct evidence for the hypothesis that changes in HDL-C levels during puberty and the differences in HDL-C levels between men and women are caused primarily by differences in sex steroids and that sex steroids work synergistically with other risk factors such as obesity.¹

Corresponding Author: Kate Millington, MD, Division of Endocrinology, Boston Children’s Hospital, 300 Longwood Ave, Boston, MA 02115 (kate.millington@childrens.harvard.edu).

Accepted for Publication: October 9, 2020.

Published Online: February 15, 2021. doi:10.1001/jamapediatrics.2020.5620

Author Contributions: Dr Millington had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Millington, Olson-Kennedy, Garofalo, Chan.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Millington.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Millington.

Obtained funding: Olson-Kennedy, Garofalo, Rosenthal.

Administrative, technical, or material support: Finlayson, Olson-Kennedy, Garofalo, Chan.

Supervision: Finlayson, Rosenthal, Chan.

Conflict of Interest Disclosures: Dr Millington reported grants from Doris Duke Charitable Foundation during the conduct of the study. Dr Olson-Kennedy reported grants from the National Institutes of Health during the conduct of the study. Dr Chan reported personal fees from Becker Pharmaceuticals and ENDO Pharmaceuticals outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported by National Institutes of Health (grant R01 HD082554) and Doris Duke Charitable Foundation (grant 2019119).

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We thank Jacob Anderson, BA, BS, of Harvard Medical School for his contributions to the literature review. Mr Anderson did not receive compensation for his contribution.