Abstract
This study examines prostate-specific antigen values among transgender women in the Veterans Health Administration receiving estrogen.
Among people with prostate cancer, transgender women receiving estrogen are more likely to present with high-grade disease vs cisgender men, suggesting potential delayed diagnosis.1 A possible reason for delayed diagnosis may be the absence of prostate-specific antigen (PSA) reference ranges specific to transgender women receiving gender-affirming hormones. Due to the castrating effects of gender-affirming hormones, PSA, whose production is androgen-regulated,2 would be expected to be lowered. This study determined PSA values among a nationwide cohort of transgender women without known prostate cancer receiving estrogen, the most common gender-affirming hormone.
Methods
After Durham Veterans Affairs institutional review board approval with a waiver of informed consent, Veterans Health Administration (VHA) records between January 2000 and August 2023 were searched for adults with 1 or more diagnostic codes previously determined to be sensitive and specific for transgender identity (Supplement 1).3 Of 14 915 patients identified, detailed chart review of 1000 randomly selected patients aged 40 years or older was performed to establish the cohort. Inclusion criteria were confirmed identity as a transgender woman, no prostate cancer diagnosis, receiving estrogen (regardless of orchiectomy status/other gender-affirming hormone usage), and PSA test performed between ages 40 and 80 years while receiving estrogen for 6 or more consecutive months. Variables abstracted included race, all PSAs between ages 40 and 80 years, receipt of other gender-affirming hormones at the time of PSA, and bilateral orchiectomy status. Statistical analyses were conducted using Stata version 18.0 (StataCorp).
PSA percentiles were calculated using all PSAs. Undetectable PSAs (below the lower limit of assay detection) were assigned a value of 0 ng/mL. Limited non-White patients precluded analyses by race. Sensitivity analyses assessed PSA distributions using only 1 PSA per patient (first PSA regardless of age and taking the median of all PSAs per patient).
Results
In total, 210 patients who underwent 852 PSA tests met inclusion criteria. Most patients (86%) were White, with Black patients comprising only 2% of the group (Table). The mean (SD) age of participants was 60 (8) years. Median estrogen duration at time of PSA assessment was 4.7 years (range, 0.5-29.9).
Table. Baseline Patient Characteristicsa.
| Characteristic | No. (%)b |
|---|---|
| Racec | |
| American Indian or Alaska Native | 7 (3) |
| Asian or Pacific Islander | 3 (1) |
| Black or African American | 4 (2) |
| White/Caucasian | 181 (86) |
| Unknown | 15 (7) |
| Age at PSA measurement, mean (SD), y | 60 (8) |
| Range, y | 40-79 |
| PSA tests by age decile, y | |
| 40-49 | 110 (13) |
| 50-59 | 290 (34) |
| 60-69 | 354 (42) |
| 70-79 | 98 (12) |
| Patients receiving gender-affirming hormone therapy at time of PSA measurement, % | |
| Estrogen | 100 |
| Spironolactone | 89 |
| Progesterone | 39 |
| Finasteride | 30 |
| Goserelin | 3 |
| Leuprolide | 5 |
| Patients who underwent bilateral orchiectomy prior to time of PSA measurement, % | 49 |
| Estrogen duration at time of PSA measurement, median (IQR), y | 5 (2-9) |
Abbreviation: PSA, prostate-specific antigen.
Race reported at the patient level. Other characteristics reported at the PSA level.
Unless otherwise indicated.
Race and ethnicity were self-reported based on prespecified categories by each patient at entry into the Veterans Health Administration. Race and ethnicity were assessed due to their association with prostate cancer screening and differing prostate cancer incidence, morbidity, and mortality.
Median (IQR) PSA was 0.02 (0-0.2) ng/mL and the 95th percentile value was 0.6 ng/mL (Figure). PSAs were undetectable in 36% of patients (23% and 49% of PSAs in patients without and with orchiectomy, respectively). PSA distributions were similar in sensitivity analyses using first PSA (median [IQR], 0.08 [0-0.3] ng/mL; 95th percentile, 0.7 ng/mL) and median of all PSAs per patient (median [IQR], 0.06 [0.01-0.2] ng/mL; 95th percentile, 0.5 ng/mL) (Figure). The highest PSA in this cohort was 2.21 ng/mL.
Figure. Violin Plots of Prostate-Specific Antigen (PSA) Obtained per Patient.
A, All PSAs obtained (N = 852) in 210 patients meeting inclusion criteria. B, First PSA per patient regardless of age. C, Median of all PSAs per patient. The rectangular shaded area on each plot represents the IQR and the white dot represents the median. Distribution of PSA values in transgender women receiving gender-affirming hormones is based on PSAs after estrogen use for at least 6 consecutive months.
Discussion
This study found the median PSA in transgender women receiving estrogen was 0.02 ng/mL. In studies of similarly aged cisgender men without known prostate cancer, the median (IQR) PSA was higher at 1 (0.6-1.9) ng/mL.4 Moreover, 36% of transgender women receiving estrogen had undetectable PSAs. Thus, PSA values in transgender women should be interpreted cautiously.
The utility of PSA screening in transgender women receiving gender-affirming hormones remains unknown. For those who do undergo screening, the optimal cut point to define “high risk” is unknown. In this study, PSA values were very low among transgender women receiving estrogen, suggesting that the historic cut point of 4 ng/mL, often used as a threshold for further evaluation, is likely far too high a threshold for this population. Indeed, previous work found that median PSA at prostate cancer diagnosis was 7.0 ng/mL among transgender women receiving estrogen, suggesting potentially delayed diagnosis in this population.1
Although the VHA population was composed of approximately 16% Black patients,5 this random sample of transgender women included only 2% Black patients. This aligns with prior studies showing fewer Black transgender women with prostate cancer.1 Additional studies examining the experiences of minority transgender women and inequities in prostate cancer screening are needed.
Limitations include lack of a comparison group, potential biases from not knowing PSA testing indications, and inability to access outside records unless documented in VHA records. Further studies are needed to understand the risks and benefits of PSA screening in this population and to establish ideal prostate cancer screening practices.
Section Editors: Kristin Walter, MD, and Jody W. Zylke, MD, Deputy Editors; Karen Lasser, MD, MPH, Senior Editor.
Supplemental Information
Data Sharing Statement
References
- 1.Nik-Ahd F, De Hoedt A, Butler C, et al. Prostate cancer in transgender women in the Veterans Affairs Health System, 2000-2022. JAMA. 2023;329(21):1877-1879. doi: 10.1001/jama.2023.6028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Yousef GM, Diamandis EP. The new human tissue kallikrein gene family: structure, function, and association to disease. Endocr Rev. 2001;22(2):184-204. doi: 10.1210/edrv.22.2.0424 [DOI] [PubMed] [Google Scholar]
- 3.Nik-Ahd F, Waller J, De Hoedt AM, et al. Seeing the unseen: how can we best identify transgender women within the Veterans Affairs healthcare system’s electronic medical record? J Sex Med. 2023;20(4):559-567. doi: 10.1093/jsxmed/qdac033 [DOI] [PubMed] [Google Scholar]
- 4.Capitanio U, Perrotte P, Zini L, et al. Population-based analysis of normal total PSA and percentage of free/total PSA values: results from screening cohort. Urology. 2009;73(6):1323-1327. doi: 10.1016/j.urology.2008.10.026 [DOI] [PubMed] [Google Scholar]
- 5.Washington DL. National Veteran Health Equity Report 2021. Focus on Veterans Health Administration patient experience and health care quality. Published September 2022. Accessed March 4, 2024. https://www.va.gov/HEALTHEQUITY/docs/NVHER_2021_Report_508_Conformant.pdf
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Information
Data Sharing Statement