Follow-Up Estradiol Levels Based on Regimen Formulation With Guideline-Concordant Gender-Affirming Hormone Therapy

Abstract

Context

Endocrine Society guidelines for dosing of feminizing gender-affirming hormone therapy (GAHT) have remained essentially unchanged since 2009. The Endocrine Society recommends periodic monitoring of serum estradiol levels, with the goal of maintaining levels in the premenopausal cisgender female range (100-200 pg/mL). However, it is not clear whether guideline-concordant dosing consistently produces guideline-recommended levels across common estradiol formulation types (oral pills, parenteral injections, transdermal patches).

Objective

All transgender and nonbinary patients receiving estradiol-based GAHT between October 2015 and March 2023 were reviewed at a single center, with the goal of determining the frequency with which guideline-concordant dosing with different estradiol formulations led to guideline-recommended estradiol levels.

Methods

Demographics, GAHT regimen, and estradiol levels were obtained via chart review, and data were analyzed descriptively.

Results

The analytic population included n = 35 individuals, including n = 9 prescribed oral estradiol pills, n = 11 prescribed parenteral injections, and n = 15 prescribed transdermal patches. With guideline-concordant doses of oral estradiol (mean 2.8 mg daily), the mean follow-up level was 168 pg/mL; 32% of follow-up levels were subtherapeutic and 14% were supratherapeutic. With guideline-concordant doses of parenteral estradiol (mean 5.8 mg weekly), the mean midpoint follow-up level was 342 pg/mL; 91% of midpoint follow-up levels were supratherapeutic. With guideline-concordant doses of transdermal estradiol (mean 0.09 mg/day), the mean follow-up level was 81.5 pg/mL; 70% of follow-up levels were subtherapeutic.

Conclusion

Supratherapeutic follow-up estradiol levels were common with guideline-concordant parenteral estradiol doses, as were subtherapeutic follow-up levels with guideline-concordant transdermal doses. These findings may suggest the need for revision of guideline-recommended estradiol doses for these formulations.

While prescribing of gender-affirming hormone therapy (GAHT) has increased over the past decade, there remains a need to expand access to evidence-based GAHT. Some transgender and gender-diverse (TGD) individuals may have access to expert providers practicing in dedicated gender medicine clinics, but others may rely on endocrinologists or primary care providers with less experience in gender-affirming care [1]. Such providers are likely to rely on clinical guidelines to inform their prescribing.

Since 2009, Endocrine Society guidelines for feminizing GAHT have remained largely consistent, with only a few minor dose recommendation changes in 2017; dosing recommendations for parenteral estradiol changed from 5 to 20 mg every 2 weeks or 2 to 10 mg every week to 5 to 30 mg every 2 weeks or 2 to 10 mg every week, and recommendations for transdermal estradiol patches changed from 0.1 to 0.4 mg/day twice weekly to 0.025 to 0.2 mg/day twice weekly [2, 3]. Oral estradiol dosing guidelines have remained unchanged (2-6 mg/day). The recommended serum estradiol target range of 100 to 200 pg/mL (obtained at the midpoint between injections for those prescribed parenteral estradiol) is based on premenopausal cisgender female levels, and is intended to balance the benefits of therapy against potential side effects of supraphysiologic estradiol levels, such as venous thromboembolism [4].

Despite these consistent recommendations, we have noted in our practice that TGD individuals prescribed guideline-concordant parenteral estradiol doses often have supratherapeutic serum estradiol levels on follow-up, while those prescribed guideline-concordant transdermal estradiol doses often have subtherapeutic follow-up levels. While these observations align with the few existing studies exploring the effect of estradiol route and dosing on serum estradiol values, prior studies have not explicitly sought to examine estradiol levels with guideline-concordant dosing. Given that many providers depend on clinical guidelines—particularly those practicing outside dedicated gender medicine centers—it is critical to ensure that guideline-concordant estradiol doses reliably produce goal serum estradiol levels. Failure to do so may exacerbate disparities in care quality for rural-dwelling TGD individuals and others with limited access to gender-affirming care.

To understand the relationship between guideline-concordant feminizing GAHT and follow-up estradiol levels, we analyzed data from TGD individuals at our center.

Materials and Methods

Overview

We conducted a retrospective cohort study based on chart review of electronic health record data from our center. Regulatory oversight was provided by the institutional review board of the Durham Veterans Affairs Health Care System (protocol 2130).

Inclusion/Exclusion Criteria

Our cohort was identified through review of all clinical referrals for gender-affirming care between October 1, 2015, and March 1, 2023. Included patients were prescribed gender-affirming estradiol therapy by an endocrinologist or a women's health provider, with at least one in-person or virtual follow-up assessment during this period. Included patients had to be either started on GAHT for the first time or have their existing regimen changed prior to measurement of a follow-up estradiol level. For inclusion, this follow-up level had to be obtained on a guideline-concordant, verifiable dose and route of estradiol. Patients were excluded if they did not have follow-up estradiol levels obtained, did not have their GAHT regimen adjusted at any point (to mitigate prevalent-user biases), could not confirm the dose associated with their estradiol level(s), or (for those prescribed parenteral estradiol) if they did not have any values measured midway between injections (3-4 days after weekly parenteral injections; none of our patients used every-other-week dosing).

Measures

Baseline demographic data were obtained from the first available clinic visit (index visit) within the analytic time frame, including sex assigned at birth, gender identity, new or prevalent user, age at time of first eligible estradiol level, race, ethnicity, zip code, and whether the patient had undergone orchiectomy at the time of the index visit. Estradiol formulations and doses, laboratory values, laboratory draw timing for patients on parenteral estradiol, and doses of adjunctive medications like spironolactone, finasteride, and progesterone were also recorded.

Given the complexity of comparing baseline and follow-up estradiol levels across incident and prevalent users, our primary data set included the first follow-up estradiol value obtained after a patient's index visit. Specifically, we chose the earliest follow-up value for which the following criteria were met: the estradiol value was obtained between 3 weeks and 18 months after the prescription of estradiol; in instances in which patients had multiple estradiol values after their index visit, the first value meeting the aforementioned criteria was used.

To explore the effect of treatment changes after the index visit, we created an expanded data set that included additional estradiol levels obtained following subsequent changes to estradiol route or dose, spironolactone dose, progesterone dose, finasteride dose, and/or orchiectomy after the index visit. Subsequent estradiol values were included when the following criteria were met: the estradiol value was obtained between 3 weeks and 18 months after a treatment change; estradiol dose and route were guideline-concordant and could be verified via chart review; there was no evidence of treatment nonadherence; and, in the case of injection therapy, it could be reasonably concluded from chart review that the estradiol value was obtained at the midpoint between injections as described earlier. If no changes were made to an individual's regimen, subsequent estradiol levels were not included.

Analyses

We used descriptive statistics to examine demographic data. We calculated mean estradiol level at the index visit and in follow-up, both overall and by estradiol formulation (oral, parenteral, transdermal). Finally, for each formulation, we calculated the percentage of values that fell into the subtherapeutic (<100 pg/mL), therapeutic (100-200 pg/mL), and supratherapeutic (>200 pg/mL) ranges based on Endocrine Society guidelines.

Results

We included 35 individuals seeking feminizing GAHT (Table 1). The mean age of this cohort was 45.5 years; 97% primarily identified as transgender women, 9% identified as gender nonbinary, 3% identified as intersex, and 3% identified as gender fluid (some individuals identified with more than one category). Eighty-five percent of individuals identified as White, 9% identified as Black, and 6% as other. Three percent identified as Hispanic. Fifty-one percent were new hormone users and 49% had already been prescribed GAHT (either at an earlier Veterans Administration visit or by an outside provider) at the time of their index visit. Seventeen percent of individuals had undergone orchiectomy at time of first appointment. In terms of estradiol route, 31% were prescribed parenteral estradiol, 26% were prescribed oral estradiol, and 43% were prescribed transdermal estradiol patches.

The mean time between the index visit and first qualifying follow-up visit was 5.8 months (oral: 3.1 months, parenteral: 8.9 months, transdermal: 4.3 months). All prescribed estradiol doses were concordant with Endocrine Society guideline recommendations. Mean estradiol doses by route were 2.8 mg daily (oral; range, 2-5 mg daily), 5.8 mg weekly (parenteral; range, 2-10 mg per week), and 0.09 mg/24 hours (transdermal; range, 0.025 mg–0.2 mg/24 hours). The mean estradiol level at first qualifying follow-up was 182 pg/mL (oral: 168 pg/mL, parenteral: 342 pg/mL, transdermal: 81.5 pg/mL). Among oral estradiol users, 67% had subtherapeutic estradiol levels, 11% had therapeutic levels, and 22% had supratherapeutic levels. Among parenteral users, none had subtherapeutic estradiol levels, 9% had therapeutic levels, and 91% had supratherapeutic estradiol levels. Among 15 transdermal estradiol users, 80% had subtherapeutic estradiol levels, 7% had therapeutic levels, and 13% had supratherapeutic levels (Table 2).

Using our expanded data set, we assessed additional subsequent estradiol levels from included individuals. Among qualifying estradiol levels obtained from oral estradiol users (n = 22), 41% were subtherapeutic, 45% were therapeutic, and 14% were supratherapeutic. Among levels obtained from parenteral estradiol users (n = 31), 6% were subtherapeutic, 23% were therapeutic, and 71% were supratherapeutic. Among levels obtained from transdermal estradiol users (n = 40), 70% were subtherapeutic, 20% were therapeutic, and 10% were supratherapeutic (Fig. 1A-1C).

Discussion

Our findings indicate that prescribing guideline-concordant estradiol doses as part of GAHT often results in follow-up estradiol levels that fall outside the therapeutic range. In particular, supratherapeutic levels were common with guideline-concordant parenteral estradiol doses, and subtherapeutic levels were common with guideline-concordant transdermal doses. Our findings support the clinical observations that led us to conduct these analyses, and suggest that revision of guideline-recommended estradiol doses—particularly for parenteral and transdermal formulations—may support more effective GAHT, as well as more equitable outcomes for those receiving GAHT outside major centers.

Overall, we found that levels outside the recommended therapeutic range were common with guideline-concordant prescribing across all estradiol formulations. Given the high number of prevalent users at first qualifying visit, the nontherapeutic levels were not solely attributable to new GAHT use. Furthermore, analyses of our expanded data set showed that levels outside the therapeutic range often persisted despite estradiol dose adjustments within the guideline-recommended ranges and adjustments to other ancillary medications. It is worth noting that despite many of the estradiol levels associated with guideline-based dosing falling outside the recommended range, levels slightly outside the recommended range may not warrant immediate adjustment. However, the information gathered from these individuals indicate that guideline-based starting doses may need to be reconsidered—even if eventual adjustments did or would bring patients into the recommended range. Moreover, some individuals, based on their gender-affirming goals, may be comfortable with a regimen that yields lower estradiol levels than the recommended range provided by Endocrine Society guidelines.

In looking at estradiol levels by route and dose (see Fig. 1A-1C), our findings suggest that to achieve a goal estradiol level of 100 to 200 pg/mL, starting doses of 2-mg oral estradiol may be too low. A starting dose of between 4 and 6 mg may be more likely to achieve therapeutic levels without a significant risk of being supratherapeutic. For injectable estradiol, a starting dose of between 2 and 4 mg weekly may be reasonable, and few should need doses greater than 6 mg to achieve a therapeutic estradiol level. For transdermal dosing, doses below 0.1 mg are often unable to achieve therapeutic estradiol levels—though they are often used for specific purposes, such as for patients who are high risk or who do not wish to achieve full feminization. For all others, 0.2 mg/day twice weekly may be more likely to achieve therapeutic estradiol levels, and some may need higher doses.

Our findings speak to the challenges of achieving guideline-recommended estradiol targets in patients receiving GAHT. Factors that contribute to this difficulty may include treatment nonadherence (potentially including taking both less than and more than the recommended estradiol dose), interindividual and interroute differences in estradiol absorption, as well as pharmacokinetic differences among routes that may lead to variable estradiol levels depending on the timing of laboratory draws after estradiol administration [5, 6].

While our observations are consistent with limited studies exploring estradiol dosing and measurement, our study provides unique observations about guideline-concordant GAHT. Three such studies were conducted in Europe, where the choice of antiandrogen, estrogen formulation, and rate of orchiectomy make comparing European and US TGD populations challenging [7-9]. Of studies conducted in the United States, one prospectively analyzed a limited number of TGD patients over a 6-month period [10]. The largest and most relevant study to ours analyzed 134 TGD individuals receiving feminizing therapy [11]. While the authors arrived at similar conclusions regarding the potential for supratherapeutic levels in parenteral users, a significant number of individuals in their study's cohort used estradiol regimens either below or above guideline-recommended dosing. A majority of the study's patch users, for example, were on doses above the upper limit recommended by the Endocrine Society. Trough estradiol levels for individuals on parenteral estradiol were preferred, and per the authors, the timing of individuals’ laboratory draws related to their injection could not be confirmed. Lastly, it is unclear how many estradiol values in this study were obtained on a steady (unchanged) GAHT regimen. As our explicit goal with this study was to examine outcomes in the setting of guideline-concordant dosing, our findings build on and add to existing knowledge.

Limitations of our study include the small sample size and the single-center cohort; future research should examine larger populations across multiple centers. Due to the small sample size, we were unable to examine the moderating effect of demographics and social drivers of health on formulation choice and attainment of therapeutic estradiol levels. The retrospective nature of the study additionally limited our ability to standardize estradiol comparisons in terms of follow-up timing. We are not able to confirm timing of estradiol laboratory draws relative to oral estradiol administration. Our approach to selecting eligible estradiol levels for our expanded data set excluded measures obtained while patients were on unchanged GAHT regimens, so may have underestimated attainment of estradiol levels in the therapeutic range; however, we intentionally pursued this selection strategy to evaluate the effects of dose changes within the guideline-recommended range. Another limitation of our study was not having enough testosterone levels paired with estradiol levels to include this information in our analyses. It would be useful in future studies to include paired testosterone levels to determine if there was any difference in testosterone suppression achieved by route or dose. Furthermore, monitoring luteinizing hormone and follicle-stimulating hormone levels in patients with intact testes not on gonadotropin-releasing hormone agonist therapy was not performed in our cohort, but this information would be useful in future studies looking at dose requirements for achieving therapeutic estradiol levels and gender-affirming goals.

Ultimately, our goal in performing this study was to confirm or refute our clinical impressions that guideline-based parenteral estradiol doses often yield supratherapeutic estradiol levels and transdermal doses often yield subtherapeutic levels. Our findings support our clinical observations and echo similar recent calls for revisiting parenteral dosing recommendations [12, 13]. While larger, prospective studies are needed to further assess and compare the adequacy, efficacy, and safety of the 3 main routes of gender-affirming estradiol therapy, our study suggests that it may be prudent to revisit Endocrine Society GAHT guidelines, especially with regard to recommended parenteral and transdermal estradiol doses.

Funding

Drs Carlson and Jeevananthan acknowledge support from the National Institutes of Health (NIH) (T32DK007012). Dr Crowley acknowledges funding from the NIH (1R01NR019594), the Veterans Affairs Quality Enhancement Research Initiative (VA QUE 20-012), and the Veterans Affairs Office of Rural Health.

Disclosures

The authors have no conflicts of interest to report.

Data Availability

Some or all data sets generated during and/or analyzed during this study are not publicly available but are available from the corresponding author on reasonable request.

References

Abbreviations

 
• GAHT

  gender-affirming hormone therapy

 
• TGD

  transgender and gender-diverse