OP1-1 - Hormone therapy and risk of venous thromboembolism in transgender women

Background:

Some transgender people receive gender-affirming hormone therapy (GAHT) to achieve a closer match between their physical appearance and their gender identity. For transfeminine individuals (TF), GAHT typically includes estrogen, alone or in combination with testosterone-lowering medications. Previous research has shown an elevated risk of venous thromboembolism (VTE) among TF taking estrogen compared to cisgender (i.e., non-transgender) men and women.

Objectives:

To compare risk of VTE across subgroups of TF receiving different types of feminizing GAHT to the risk observed in cisgender individuals of either sex.

Methods:

The Study of Transition, Outcomes and Gender (STRONG) is a cohort of transgender people enrolled in Kaiser Permanente health systems.  The present analysis included 3,325 TF STRONG cohort members who initiated GAHT. Each TF participant was matched to 10 male and 10 female cisgender enrollees on year of birth, race/ethnicity, site, and health plan enrollment year. Incident cases of VTE were ascertained from January 1, 2006 through December 31, 2019 with follow up for each participant starting at the time of the first filled estrogen prescription.  VTE rates were compared in TF cohort members and their matched cisgender referents using Cox regression models with results expressed as adjusted hazard ratios (HR) and the corresponding 95% confidence intervals (CI). In subgroup analyses, the associations of GAHT with VTE risk were further explored according to route of estrogen administration and drug combinations. Dose effects were investigated by categorizing TF taking oral estradiol (the most common active ingredient) into 2 groups based on initial daily dose (≤2mg vs. >2mg per day).

Results:

The HR estimates for VTE among TF individuals who started and remained on oral estrogen were 1.9 (95% CI: 1.0, 3.5) and 2.5 (95% CI: 1.4, 4.5) compared with cisgender males and cisgender females, respectively.  The corresponding HR estimates for VTE for cohort members who started and remained on parenteral estrogen were 1.0 (95% CI: 0.4, 2.4) and 1.1 (95% CI: 0.4, 2.6). For TF persons receiving estrogen in combination with testosterone-lowering medications, subgroup analyses found the risk of VTE might be less elevated among those using spironolactone (HR: 2.3; 95% CI: 1.5, 3.5) than among those using other antiandrogen medications such as finasteride (HR: 3.4; 95% CI: 1.2, 9.9). The HR for VTE among persons who remained on oral estradiol was 1.6 (95% CI: 0.7, 3.6) for those starting on ≤2mg/day and 3.3 (95% CI: 1.0, 10.8) for those starting on >2mg/day, compared with cisgender males.

Conclusions:

This study suggests that TF participants taking oral estrogen have a higher risk of VTE than those who receive estradiol via parenteral routes.  It is also possible that this increase in VTE risk may be dose dependent.  Future analyses should account for changes in GAHT type, route and dose over time and focus on the associations between levels of sex hormones and VTE risk.