Transgender Health

Volume 00, Number 00, 2024 Open camera or QR reader and

ª Mary Ann Liebert, Inc. scan code to access this article
DOI: 10.1089/trgh.2023.0138 and other resources online.

Comparison of Estrone/Estradiol Ratio and Levels

in Transfeminine Individuals on Different Routes
of Estradiol
Navin M. Kariyawasam,1,*. Tehmina Ahmad,1–3 Shohinee Sarma,4 and Raymond Fung1,2,5
Downloaded by University of Toronto via COAHL from www.liebertpub.com at 03/16/24. For personal use only.

Abstract
Purpose: Estradiol for gender-affirming hormone therapy can be taken in different routes: oral, sublingual, trans-
dermal patch or gel, and injectable estradiol. We aimed at comparing the estrone and estradiol ratios and levels
achieved in each of these different routes of estradiol.
Methods: We conducted a retrospective chart review of transfeminine individuals attending an endocrinology
clinic in Toronto, Canada. Study participants were grouped according to the route of estradiol administration:
oral, injectable, transdermal, and sublingual. Our primary outcome was the estrone/estradiol ratio (E1/E2). Our
secondary outcomes were the estradiol and estrone levels in each of these four groups.
Results: We included 286 patients. The oral estradiol group had the highest E1/E2 ratio (9.28), followed by the
sublingual group (6.88). Both the transdermal and injectable groups had substantially lower E1/E2 ratios (2.22 and
0.84, respectively). We observed a large variability of the E1/E2 ratio in the oral and sublingual groups, whereas
the transdermal and the injectable groups’ ratios had much smaller standard deviation. The mean estradiol in the
injectable group (1557 pmol/L, 424.1 pg/mL) was markedly higher than the estradiol levels observed in all other
routes of estradiol.
Conclusion: Our data demonstrate significantly different E1/E2 ratios in the four different routes of estradiol
administration, with oral and sublingual routes having the highest E1/E2 ratios followed by transdermal and
injectable routes.
Keywords: administration routes; estrone; feminizing therapy; gender affirming therapy; hormone therapy;
transgender persons

Introduction In addition, the 17-beta estradiol tablet can be taken

Seventeen-beta estradiol is the recommended type of sublingually, rather than swallowed. There have been
estrogen used in gender-affirming hormone therapy, insufficient data to determine whether one route of
and is generally combined with anti-androgen thera- administration is superior to another with respect to
pies.1–3 17-Beta estradiol is favored over other types inducing feminizing effects. Some transfeminine indi-
of estrogen, such as ethinyl estradiol, because it can viduals request to be on injectable estradiol because
be measured in the serum, and it appears to have a of the belief of superior effects; however, this has not
lower risk of thrombosis.1,3 It is available in different been demonstrated in clinical data.4
forms, including oral 17-beta estradiol, transdermal It is possible that differing routes of estradiol may
estradiol patch or gel, as well as injectable estradiol have different efficacies in inducing desired changes.
(valerate or cypionate). When estradiol is administered orally, 17 beta-
1
Temerty Faculty of Medicine, University of Toronto, Toronto, Canada.
2
Division of Endocrinology, Department of Medicine, Medical Sciences Building, Toronto, Canada.
3
Toronto Western Hospital, University Health Network, Toronto, Canada.
4
Institute of Health Policy, Management, and Evaluation, University of Toronto, Toronto, Canada.
5
Michael Garron Hospital, East York, Canada.

*Address correspondence to: Navin M. Kariyawasam, BSc, Temerty Faculty of Medicine, University of Toronto, 1 King’s College Circle, M5S 1A8 Toronto, Canada, E-mail:
navin.kariyawasam@mail.utoronto.ca

1
 2 KARIYAWASAM ET AL.

hydroxysteroid dehydrogenase type 2 (17 b-HSD2) cat- Hospital ethics board, and consent was waived. We
alyzes the conversion of estradiol to estrone in the liver, followed study procedures according to the protocol
resulting in higher concentrations of estrone as com- approved by this ethics board. Adult ( > 18 years) trans-
pared with estradiol in the blood.5 Estradiol has been feminine patients who received estradiol therapy and
thought of as the most potent activator of the estrogen had estradiol and estrone values measured within the
receptor, with estrone having only 4% the activity of study timeframe cited earlier were identified through
estradiol.5 chart review and included in the study.
It is therefore biologically conceivable that lower lev- We included every transfeminine patient who
els of estradiol as compared with estrone may result in received estradiol therapy and had routine measure-
less effective feminizing changes. When estradiol is ments of both estradiol and estrone levels measured
administered transdermally or parenterally, there is as part of their follow-up, which generally occurred
less conversion of estradiol to estrone due to circum- every 3 to 6 months. We excluded transfeminine
Downloaded by University of Toronto via COAHL from www.liebertpub.com at 03/16/24. For personal use only.

vention of the first pass effect, resulting in higher estra- patients on other types of estrogen (such as conjugated
diol levels, and much lower estrone levels.5,6 We are equine estrogen) or those not taking any estradiol
unaware of any randomized controlled trials compar- therapy at the time of the blood test.
ing the effectiveness of these different routes of estradi- Study participants were grouped according to the
ol administration.7 route of estradiol administration: oral, injectable, trans-
A prospective cohort study by Tebbens et al. found dermal, and sublingual. All patients in the injectable
no association between estrone concentration or estradiol group were treated with estradiol valerate.
route of administration and changes in fat percent- Study participants were then stratified into low, me-
age or breast development. However, their cohort did dium, and high dosage groups. It was the practice of
not include injectable estradiol and may have been the clinic to advise patients not to obtain their blood
underpowered.8 tested either immediately after taking their estradiol
It is also possible that differing routes of estradiol or just before their next dose is due, but rather some-
have differences in long-term safety. Oral estradiol time between doses.
undergoes first-pass metabolism in the liver, which This is because we otherwise found levels of estradiol
alters production of clotting factors; in several large that did not correlate with their dose, that is, very high
retrospective studies in postmenopausal women, oral levels when the patient was on low dose estradiol, or
estradiol has been associated with increased risk of very low levels when the patient was already on a rela-
venous thromboembolism, whereas transdermal estra- tively high dose. Therefore, for our data collection, we
diol has not.9–11 Some have explained this difference also determined whether the timing of the blood test
through the impact of different routes of estradiol on was done early, mid, or trough time frames with respect
activated protein C resistance,12 whereas others have to when estradiol was taken by the patient before the
proposed that oral estradiol is more thrombogenic blood test draw.
due to estrone production.13 This was done through careful chart review to see
While estrone/estradiol concentrations and ratios whether any notes about the timing were included in
have been studied in some pharmacokinetic studies the medical record. If there was no mention of the tim-
in cis-women,5,6 there is a paucity of clinical data on ing, it was allocated to mid group (See Tables 1 and 2
these levels in transfeminine individuals on feminizing for definition).
hormone therapy. We aimed at studying the estrone to
estradiol concentrations and ratios in a group of trans- Table 1. Low-, Medium-, and High-dose Estradiol definitions
feminine individuals on hormone therapy receiving
Low Medium High
different routes of estrogen and anti-androgen therapy.
Oral (daily) 2 mg or less > 2 to 4 mg > 4 mg
Injectable (per week)a 4 mg or less > 4 to 8 mg > 8 mg
Methods Transdermal gel 1% 2 pumps 3 to 4 pumps > 4 pumps
We conducted a retrospective chart review of transfe- (1.25 g/pump) (daily) or less
Transdermal patch 100 mcg patch 100 to > 200 mcg
minine individuals attending an endocrinology cli- (changed 2 · weekly) or less 200 mcg patch
nic at Michael Garron Hospital in Toronto, Canada Sublingual (daily) 2 mg or less > 2 to 4 mg > 4 mg
between March 1, 2019 and July 1, 2021. The study a
All patients in the injectable group were treated with estradiol
received ethics approval through the Michael Garron valerate.
 ESTRONE/ESTRADIOL ON DIFFERENT ROUTES OF ESTRADIOL 3

Table 2. Early, Mid, and Trough Blood Test Timing Definitions

Early Mid Trough

Oral < 2 h of ingestion 2 to 24 h 24 or more hours

Injectable (given weekly) Within 2 days of injection 3 to 6 days of injection 7 or more days post-injection
Transdermal gel (applied daily) < 2 h post-application of gel 2 to 24 h for gel 24 or more hours after gel application
Transdermal patch < 4 h post-application of patch 4 to 72 h for application of patch 72 hours or more after application of patch
(changed 2 · per week)
Sublingual < 2 h of administration 2 to 24 h 24 or more hours

Our primary outcome was to determine the estrone/ excluded these outliers from our final analysis of
estradiol (E1/E2) ratio in each route group. Our sec- estrone and estradiol levels.
ondary outcome was to determine the mean estradiol We used generalized estimating equations (GEE) to
and estrone levels in each of the four route groups, determine the association between the routes of admin-
Downloaded by University of Toronto via COAHL from www.liebertpub.com at 03/16/24. For personal use only.

stratified according to dose group. istration (oral, transdermal, sublingual, and injectable)
over recurrent visits and the E1/E2 ratio as the pri-
Hormone analysis mary outcome. Secondary analyses also used GEE to
Blood samples were collected during clinical care. determine the association between routes of estradiol
Patients went to one of two community laboratories. administration over recurrent visits and the estradiol
For estradiol, laboratory A used the Chemiluminescent level and estrone level, respectively. We assumed a
Microparticle Immunoassay on Architect instrument Gaussian distribution and used the link function for
by Abbott, with coefficient of variation (CV) of 5.9% the GEE analyses for continuous outcomes.
at 330 pmol/L (89.9 pg/mL), and 11% at 1050 pmol/L We used the first-order autoregressive correlation
(286 pg/mL). Laboratory B used electrochemilumines- structure to account for correlations between adja-
cence immunoassay (ECLIA) by Roche, with CV of cent visits for all GEE models. We further adjusted
10%. For estrone, both laboratories used Beckman the models for age and mean body mass index
Coulter radioimmunoassay kit, with CV of 8–12% at (BMI). We reported beta coefficients with associated
385 pmol/L (105 pg/mL), and 6–11% at 1350 pmol/L standard errors and p-values for all GEE models.
(367.7 pg/mL). Of the 434 laboratory results in our All statistical analyses were performed using R Stu-
analysis, 354 (82%) were from laboratory A and 77 dio, version 4.1.2, 2021 (R Foundation for Statistical
(18%) were from laboratory B (three entries were man- Computing, Vienna), and p-values < 0.05 were consid-
ually inputted, with source laboratory unknown). ered statistically significant.

Statistical analysis Results

We determined baseline characteristics at the first Baseline characteristics
recorded visit for each participant and grouped accord- We found 286 patients who fulfilled our inclusion crite-
ing to the routes of administration of estradiol (oral, ria and were included in our study. The mean age of our
transdermal, sublingual, and injectable). We reported study population was 34 ( – 13) years, with a mean BMI
mean and standard deviation for continuous variables of 27 kg/m2 (Table 3). Patients in the transdermal estra-
and percentage (%) for categorical variables. diol group tended to be older than the oral estradiol
We conducted a one-way Analysis of Variance group. The majority of the study population was on spi-
(ANOVA) test to determine differences in mean E1/ ronolactone (40%), with substantial numbers being on
E2 ratio, estrone level, and estradiol level at the last cyproterone (25%) or having had gonadectomy (21%).
visit by route of administration. We then conducted Most of our study population was on medium-dose
two-way ANOVA analyses to account for dose group estradiol, and over 90% of the blood tests were done in
(low, medium, or high), timing of lab draw (early, the mid time range. Table 3 shows the baseline charac-
mid, trough), and type of lab, respectively. teristics of participants by routes of administration of
Only a small minority of the blood tests were drawn estradiol therapy. Given the lower apparent mean LH
in the early or trough groups and given the extreme val- at baseline in the injectable group as compared with
ues that were observed in these groups (very high estra- other routes, a one-way ANOVA was run, which
diol levels even with low doses of estradiol or low levels found no statistically significant difference in LH values
of estradiol with medium- or high-dose estradiol), we across routes at baseline.
 4 KARIYAWASAM ET AL.

Table 3. Baseline Characteristics at First Visit

Characteristic Overall, N5263a Oral, N5153a Patch/gel, N539a Sublingual, N538a Injectable, N533a

Age 34 (13) 30 (11) 49 (13) 31 (11) 34 (14)

Body mass index (kg/m2) 27 (6) 27 (7) 28 (6) 25 (5) 26 (6)
Estradiol (pmol/L) 555 (936) 362 (671) 342 (472) 773 (835) 1450 (1686)
Estrone (pmol/L) 1737 (1546) 2046 (1519) 472 (401) 2678 (1739) 716 (620)
Luteinizing hormone (IU/L) 5 (8) 5 (8) 6 (10) 5 (11) 2 (3)
Follicle stimulating hormone (IU/L) 5 (12) 5 (11) 8 (13) 8 (19) 2 (4)
Type of anti-androgen therapy
None 23 (8.7%) 12 (7.8%) 1 (2.6%) 3 (7.9%) 7 (21%)
Gonadectomy 54 (21%) 23 (15%) 14 (36%) 9 (24%) 8 (24%)
Cyproterone 66 (25%) 42 (27%) 6 (15%) 12 (32%) 6 (18%)
Spironolactone 106 (40%) 67 (44%) 17 (44%) 13 (34%) 9 (27%)
GnRH agonist 8 (3.0%) 6 (3.9%) 1 (2.6%) 1 (2.6%) 0 (0%)
Bicalutamide 6 (2.3%) 3 (2.0%) 0 (0%) 0 (0%) 3 (9.1%)
Downloaded by University of Toronto via COAHL from www.liebertpub.com at 03/16/24. For personal use only.

Concurrent use of progesterone 27 (10%) 14 (9.2%) 1 (2.6%) 7 (18%) 5 (15%)

Dose group
Low dose 94 (36%) 60 (39%) 16 (41%) 10 (26%) 8 (24%)
Medium dose 137 (52%) 72 (47%) 21 (54%) 22 (58%) 22 (67%)
High dose 32 (12%) 21 (14%) 2 (5.1%) 6 (16%) 3 (9.1%)
Timing of lab draw
Early 12 (4.6%) 6 (3.9%) 1 (2.6%) 3 (7.9%) 2 (6.1%)
Middle 240 (91%) 141 (92%) 38 (97%) 35 (92%) 26 (79%)
Trough 11 (4.2%) 6 (3.9%) 0 (0%) 0 (0%) 5 (15%)
a
Mean (standard deviation), where a % symbol appears: n (%).
GnRH, gonadotropin-releasing hormone.

Estrone to estradiol ratio We observed a large variability of the E1/E2 ratio

The oral estradiol group had the highest E1/E2 ratio in the oral and sublingual groups, whereas the
(9.28) over recurrent visits, followed by the sublingual transdermal and especially the injectable groups’
group (6.88) (Fig. 1). Both the transdermal and inject- ratios were more consistent, and consistently lower.
able groups had substantially lower E1/E2 ratios (2.22 There was no statistically significant difference in
and 0.84, respectively) (Table 4). the mean E1/E2 ratio due to timing of draw or

FIG. 1. Mean E1/E2 ratio across routes of administration. Boxplot of estrone/estradiol ratios according to the
route of administration, with interquartile ranges shown. E1/E2, estrone to estradiol.
 ESTRONE/ESTRADIOL ON DIFFERENT ROUTES OF ESTRADIOL 5

Table 4. Mean Estrone/Estradiol Ratio, Mean Estradiol, The mean estrone levels were highest in the
and Mean Estrone Across Routes of Administration oral (2152 pmol/L, 586.2 pg/mL), and sublingual
at the Last Visit, p-Values from One-Way
Analysis-of-Variance (2340 pmol/L, 637.4 pg/mL) groups (Table 4). As pre-
dicted by the lower E1/E2 ratio, the estrone levels in
Transdermal
Route Oral (patch/gel) Sublingual Injection p the transdermal and injectable groups were much
lower (Fig. 3). Similarly, there was no statistically sig-
Mean E1/E2 9.28 2.22 6.88 0.84 < 0.001*
Ratio nificant difference in the estradiol and estrone levels
Mean 390 456 613 1557 < 0.001* when accounting for type of laboratory assay, although
estradiol
(pmol/L)
estradiol levels were affected by the timing of labora-
Mean 2152 530 2340 772 < 0.001* tory draw (Supplementary Tables S2 and S3).
estrone
(pmol/L)
GEE models for association between routes
Downloaded by University of Toronto via COAHL from www.liebertpub.com at 03/16/24. For personal use only.

*Significant value < 0.001.

Estrone/estradiol ratio.
of administration and outcomes
The results of the GEE models in Supplementary
between the two different laboratory assays (Supple- Tables S4 and S5 show the estimates of the association
mentary Table S1). between routes of administration and the estrone to
estradiol ratio (E1/E2 ratio) over repeat visits. For
Mean estradiol and estrone levels every unit (mg) increase in oral and sublingual estradi-
We analyzed the mean estradiol and estrone levels ol administration, the E1/E2 ratio increased by 7.6 and
at each patient’s last clinic visit during the study pe- 5.0 U over multiple visits, respectively, when compared
riod. The mean estradiol in the injectable group with injectable estradiol (Supplementary Table S4).
(1557 pmol/L, 424.1 pg/mL) was markedly higher than These effects persisted even after adjusting for age
the estradiol levels observed in all other routes of est- and BMI. After adjustment, each unit (mg) increase
radiol (Table 4). Oral, transdermal, and sublingual in oral and sublingual estradiol administration led to
groups achieved similar estradiol levels. The mean est- an increase of the E1/E2 ratio by 7.8 and 5.2 units,
radiol level generally increased with increasing doses of respectively, when compared with injectable estradiol
estradiol administered (Fig. 2). (Supplementary Table S5).

FIG. 2. Mean estradiol level (pmol/L) across dose group and routes of administration. Boxplot of mean
estradiol level according to the route of administration, with interquartile ranges shown.
 6 KARIYAWASAM ET AL.
Downloaded by University of Toronto via COAHL from www.liebertpub.com at 03/16/24. For personal use only.

FIG. 3. Mean estrone level (pmol/L) across dose group and routes of administration. Boxplot of mean estrone
level according to the route of administration, with interquartile ranges shown.

Similarly, for each unit (mg) increase in oral In addition to the low conversion of estradiol to
and sublingual estradiol, the estrone level increa- estrone leading to higher estradiol levels, another rea-
sed by 1478 pmol/L (402.6 pg/mL) and 1769 pmol/L son for the higher levels of estradiol could be that the
(481.9 pg/mL), respectively (Supplementary Table S6). patients were treated with too high of a dose. Most
Estradiol levels did not increase over multiple visits patients in this injectable group were in the medium-
regardless of the route of administration (Supplemen- dose category, receiving between 4 and 8 mg of estradi-
tary Table S7). ol weekly whereas doses of 3.75 mg SC (interquartile
range [IQR], 3–4 mg)to 4 mg IM (IQR 3–5 mg) per
Discussion week were recently reported to achieve therapeutic est-
Over recurrent clinical visits, our data demonstrate radiol levels in a cohort of transfeminine individuals.15
significantly different E1/E2 ratios in the four different Other studies have also reported E1/E2 ratios and lev-
routes of estradiol administration in transfeminine els in transfeminine individuals. Cirrincione et al. looked
individuals on feminizing hormone therapy, with oral at estrone, estradiol, and E1/E2 ratios in a population of
and sublingual routes having the highest E1/E2 ratios transgender women taking sublingual, transdermal, or
followed by transdermal and injectable routes. Even injectable estradiol.16 The E1/E2 ratios found in the sub-
at a cross-sectional snapshot at the last clinic visit, lingual and injectable groups are very similar to our
the mean E1/E2 ratio differed significantly between study, whereas the E1/E2 ratio in their transdermal
routes of administration and was consistently the high- group is slightly lower, closer to 1 versus 2.2 in our study.
est for oral and sublingual followed by transdermal and They also observed that there was a large variation in
the lowest for injectable. the E1/E2 ratio in the sublingual group, as compared with
Our secondary outcomes showed that the injectable the transdermal and injections groups. They did not have
route resulted in the highest mean estradiol levels by an oral comparator group, whereas in our study, we
far, compared with the other three routes. One is there- found a similarly large range in E1/E2 ratios in the oral
fore tempted to speculate whether injectable estradiol group. This suggests that different individuals may have
would result in superior feminizing changes as com- different propensities in their conversion of estradiol to
pared with the other routes. However, whether higher estrone when estradiol is taken orally or sublingually.
levels of estradiol are correlated with more effective It would be clinically relevant to uncover whether
feminization has never been proven.14 there is a difference in effectiveness and safety
 ESTRONE/ESTRADIOL ON DIFFERENT ROUTES OF ESTRADIOL 7

outcomes between those who have higher E1/E2 ratios Conclusion

(e.g., > 10) as compared with those who have relatively In summary, this study yields valuable insights into
lower E1/E2 ratios (e.g., 5) in the oral estradiol group. the variations in estrone to estradiol concentrations
Another explanation for the large variability is that and ratios among transfeminine individuals undergo-
some in the sublingual group could have swallowed ing feminizing hormone therapy. Our study found
their estradiol on occasion whereas others in the oral that the route of administration of estradiol affects
group could have taken their estradiol sublingually these concentrations and ratios, with injectable estra-
some of the time; we were unable to capture these diol resulting in the lowest E1/E2 ratio, highest levels
subtleties in our data. of estradiol, and lower levels of estrone compared
Tebbens et al. examined estradiol and estrone con- with all other routes. Our study, therefore, highlights
centrations in a group of trans women on oral estradiol the importance of future research into whether inject-
compared with transdermal estradiol. Similar to our able estradiol exhibits any differences in efficacy for
Downloaded by University of Toronto via COAHL from www.liebertpub.com at 03/16/24. For personal use only.

study, they observed much higher estrone levels, as clinical feminization as compared with other routes
well as a higher E1/E2 ratio in the oral estradiol group. of estradiol.
In a subset of 23 trans women, they were able to com- Finally, our study contributes to an emerging pool
pare the breast growth between the oral and transdermal of data for this population with future studies directed
groups, and did not find a significant difference. toward understanding the effectiveness of feminiza-
They also compared fat percentage change in a sub- tion, and determination of the optimal route and dos-
set of 53 patients and found that it was not correlated to age of estrogen administration that minimizes the risk
estrone levels. They concluded that measuring estrone of adverse health outcomes, while maximizing the ben-
levels was, therefore, not useful, but the study was lim- efits of hormone therapy in transfeminine individuals.
ited by low sample sizes, and did not include injectable
estradiol. Our study found a large difference in the E1/ Strengths and limitations
E2 ratio between oral and injectable estradiol groups Our study has several important strengths and uni-
and a much higher estradiol level achieved in the inject- que contributions to advancing our understanding of
able group, and future study on clinical implications of estradiol therapy in transfeminine individuals. First,
these patterns is necessary. we included individuals on all four different routes of
Moreover, the results of this study are novel and estradiol, whereas previous studies were limited to
important because they provide insight into the associa- two or three routes. Second, we used a range of doses
tion between routes of estrogen administration and the in the different route groups, within guideline recom-
estrone to estradiol ratio (E1/E2 ratio) in transfeminine mended parameters, providing a realistic estimation
individuals over multiple visits and recurrent laboratory of what clinicians can expect when they measure estra-
values. Our study found that oral and sublingual admin- diol – estrone levels in their own patients.
istration of estradiol led to a significantly higher E1/E2 This retrospective study used real-world data with
ratio compared with injectable estradiol, and this effect currently used anti-androgen agents, and biochemical
persisted even after adjusting for age and BMI. measurements were done at two different laboratories,
Our modeling showed that each unit increase in oral therefore extending its generalizability. Third, we have
and sublingual estradiol administration led to a signif- tried to eliminate spurious results from our data by cat-
icant increase in the estrone levels. These findings have egorizing the timing of the blood test with respect to
significant implications for the clinical management when the estradiol hormone was taken.
of transfeminine individuals, as they suggest that the However, our study has its limitations, including the
route of estrogen administration impacts the E1/E2 absence of data on the direct effects of E1/E2 ratios
ratio and estrone levels over time. and estradiol levels on feminizing changes. Data on
This may be particularly relevant for the risk of testosterone levels were not available due to limited
adverse health outcomes associated with high estrone data collection, but would have been valuable in further
levels, such as venous thromboembolism and breast elucidating any differences in route of administration.
cancer.13,17 Clinicians may need to consider the poten- Further, the use of immunoassays, while common in
tial risks and benefits of different routes of estrogen clinical settings, may introduce variability and are
administration when prescribing hormone therapy to more prone to error compared with more precise liq-
transfeminine patients. uid chromatography mass spectrometry methods.
 8 KARIYAWASAM ET AL.

Our study was retrospective and dependent on chart 4. Deutsch MB. Information on estrogen hormone therapy. UCSF Trans-
gender Care: San Francisco, California, USA; 2020. Available from: https://
review, and there may have been misclassification in how transcare.ucsf.edu/article/information-estrogen-hormone-therapy [Last
the patient was taking their estradiol (swallowing, orally accessed: December 29, 2022].
5. Kuhl H. Pharmacology of estrogens and progestogens: Influence of dif-
vs. putting underneath their tongue, sublingually), as ferent routes of administration. Climacteric 2005;8 Suppl 1:3–63; doi: 10
well as timing of the blood test with respect to estradiol .1080/13697130500148875
6. Setnikar I, Rovati LC, Vens-Cappell B, et al. Pharmacokinetics of estradiol
ingestion/administration based on the clinical note and of estrone during repeated transdermal or oral administration of
reviewed. We did assume a timing of ‘‘mid’’ when no tim- estradiol. Arzneimittelforschung 1996;46(8):766–773.
ing was otherwise specified. In addition, the study’s focus 7. Iwamoto SJ, Defreyne J, Rothman MS, et al. Health considerations for
transgender women and remaining unknowns: A narrative review. Ther
on one clinic in a specific urban center (Toronto, Canada) Adv Endocrinol Metab 2019; doi: 10:2042018819871166.
may limit its generalizability to the broader population of 8. Tebbens M, Heijboer AC, T’Sjoen G, et al. The role of estrone in feminizing
hormone treatment. J Clin Endocrinol Metab 2022;107(2):e458–e466; doi:
transfeminine individuals undergoing hormone therapy. 10.1210/clinem/dgab741
Lastly, this study did not investigate the long-term 9. Vinogradova Y, Coupland C, Hippisley-Cox J. Use of hormone replace-
Downloaded by University of Toronto via COAHL from www.liebertpub.com at 03/16/24. For personal use only.

ment therapy and risk of venous thromboembolism: Nested case-control

safety of the different routes of estradiol therapy. studies using the QResearch and CPRD databases. BMJ 2019;364:k4810;
Therefore, it is imperative to conduct further inves- doi: 10.1136/bmj.k4810
10. Canonico M, Oger E, Plu-Bureau G, et al. Hormone therapy and venous
tigations to confirm these findings and thoroughly thromboembolism among postmenopausal women: Impact of the route
examine potential differences in long-term safety of estrogen administration and progestogens: The ESTHER Study. Circu-
among the various routes of estradiol administration. lation 2007;115(7):840–845; doi: 10.1016/j.maturitas.2015.02.044
11. Sweetland S, Beral V, Balkwill A, et al. Venous thromboembolism risk in
relation to use of different types of postmenopausal hormone therapy in
Authors’ Contributions a large prospective study. J Thromb Haemost 2012;10(11):2277–2286;
doi: 10.1111/j.1538-7836.2012.04919.x
N.M.K.: Investigation, data curation, writing—original 12. Toorians AWFT, Thomassen MCLGD, Zweegman S, et al. Venous throm-
draft, review, and editing. T.A.: Conceptualization, bosis and changes of hemostatic variables during cross-sex hormone
treatment in transsexual people. J Clin Endocrinol Metab 2003;88(12):
methodology, writing—original draft, review, and 5723–5729; doi: 10.1210/jc.2003-030520
editing, and project administration. S.S.: Methodology, 13. Bagot CN, Marsh MS, Whitehead M, et al. The effect of estrone on
writing, editing, software, and statistical analysis. thrombin generation may explain the different thrombotic risk between
oral and transdermal hormone replacement therapy. J Thromb Haemost
R.F.: Conceptualization, investigation, methodology, 2010;8(8):1736–1744; doi: 10.1111/j.1538-7836.2010.03953.x
writing—original draft, review, and editing, resources, 14. Nolan BJ, Cheung AS. Relationship between serum estradiol concentra-
tions and clinical outcomes in transgender individuals undergoing fem-
and supervision. inizing hormone therapy: A narrative review. Transgend Health 2021;6(3):
125–131; doi: 10.1089/trgh.2020.0077
15. Herndon JS, Maheshwari AK, Nippoldt TB, et al. Comparison of the sub-
Author Disclosure Statement cutaneous and intramuscular estradiol regimens as part of gender-
No competing financial interests exist. affirming hormone therapy. Endocr Pract 2023;29(5):356–361; doi: 10
.1016/j.eprac.2023.02.006
16. Cirrincione LR, Winston McPherson G, Rongitsch J, et al. Sublingual
Funding Information estradiol is associated with higher estrone concentrations than transder-
No funding was received for this research. mal or injectable preparations in transgender women and gender non-
binary adults. LGBT Health 2021;8(2):125–132; doi: 10.1089/lgbt.2020.0249
17. Qureshi R, Picon-Ruiz M, Sho M, et al. Estrone, the major postmenopausal
Supplementary Material estrogen, binds ERa to induce SNAI2, epithelial-to-mesenchymal transi-
tion, and ER + breast cancer metastasis. Cell Rep 2022;41(7):111672; doi:
Supplementary Table S1 10.1016/j.celrep.2022.111672
Supplementary Table S2
Supplementary Table S3
Supplementary Table S4
Cite this article as: Kariyawasam NM, Ahmad T, Sarma S, Fung R
Supplementary Table S5 (2024) Comparison of estrone/estradiol ratio and levels in
Supplementary Table S6 transfeminine individuals on different routes of estradiol, Transgender
Health X:X, 1–8, DOI: 10.1089/trgh.2023.0138.
Supplementary Table S7

References
1. Hembree WC, Cohen-Kettenis PT, Gooren L, et al. Endocrine treatment of
gender-dysphoric/gender-incongruent persons: An Endocrine Society
Abbreviations Used
clinical practice guideline. J Clin Endocrinol Metab 2017;102(11):3869– ANOVA ¼ Analysis of Variance
3903; doi: 10.1210/jc.2017-01658. BMI ¼ body mass index
2. Bourns A. Guidelines for Gender-Affirming Primary Care with Trans and CV ¼ coefficient of variation
Non-Binary Patients. Rainbow Health Ontario, Sherbourne Health Centre: E1/E2 ¼ estrone/estradiol ratio
Toronto, Ontario, Canada; 2019. ECLIA ¼ electrochemiluminescence immunoassay
3. Safer JD, Tangpricha V. Care of transgender persons. N Engl J Med 2019; GEE ¼ generalized estimating equations
381(25):2451–2460; doi: 10.1056/NEJMcp1903650