Amy Carpenter
June 2024—Screening in transgender men with cervices and in transgender women with neovaginas was the focus of a CAP23 session that highlighted screening recommendations and morphologic challenges such as detecting high-grade dysplasia in a background of atrophy (trans men) and dysplasia risks (trans women)—and EHR-related improvements for both.
Some transgender men undergo hysterectomy, but many do not, and “that means a large percentage of transgender men still have a cervix and are at risk for cervical cancer and HPV-related dysplasia,” said Margaret Compton, MD, who spoke in the session last fall on cytopathology and LGBTQI health. She is assistant professor, Department of Pathology, Microbiology, and Immunology, Vanderbilt University Medical Center.
American College of Obstetricians and Gynecologists 2021 guidelines, which were reaffirmed in 2024, say any anatomical structure present that warrants screening should be screened, regardless of gender identity. The Centers for Disease Control and Prevention says all persons with a cervix should be screened for cervical cancer, regardless of sexual orientation or gender.
Despite the screening recommendations, LGBT+ patients have a lower rate of cancer screening than heterosexual controls, Dr. Compton said, and the cause may be multifactorial, including prior discriminatory treatment by the health care system, physical and psychological discomfort with the exam, insurance coverage, and lack of awareness about the screening recommendations. “Patients and their physicians may not realize this group of patients needs to be screened,” Dr. Compton said, citing a study of patient and provider perspectives of cervical cancer risk and screening among transmasculine individuals (Agénor M, et al. Cult Health Sex. 2016;18[10]:1192–1206). Coding and billing are problems in some instances, too, with some laboratory information systems unable to apply certain codes for gynecologic cytology if the patient is in the system as male.
While providers may encourage screening by creating a supportive and welcoming environment and employing techniques to make the physical exam more tolerable in the setting of deep atrophy, Dr. Compton said, some patients still may be unable to undergo pelvic examination. In the 2018 U.S. Preventive Services Task Force Recommendation Statement for cervical cancer screening, the authors cite a paucity of data on screening protocols for patients who identify as lesbian or transgender (U.S. Preventive Services Task Force. JAMA. 2018;320[7]:674–686). “This is a very heterogeneous group of patients who have differing sexual practices, so may have different risk profiles for getting HPV,” Dr. Compton said. “So someone who does not, for example, engage in penetrative intercourse may be at low risk for getting HPV, but other transgender men may have a homosexual relationship with cisgender men and may be at higher risk for HPV-related disease.” In these situations, providers and patients can discuss an individualized risk-based approach to screening.
In addition to the challenges in designing an appropriate screening program, Dr. Compton said, “there are morphologic changes and challenges,” which her colleague and co-presenter, Kim Adams Ely, MD, explained.
Dr. Ely, medical director of cytopathology and associate professor, Department of Pathology, Microbiology, and Immunology, Vanderbilt University Medical Center, presented the case of a 25-year-old female for whom there was no last menstrual period date or information about method of birth control.
The patient had a history of two prior abnormal Pap tests: one of atypical squamous cells of undetermined significance (ASC-US) and one of atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion (ASC-H) (Fig. 1). Both Pap tests were followed by colposcopy and negative biopsies. “We pulled those Paps and reviewed them in conjunction with the current Pap and felt that the present changes were like those seen in the prior—ASC-H,” Dr. Ely said.
[All photos courtesy of Margaret Compton, MD, and Kim Adams Ely, MD.]
About 71 percent of transgender men have ever received exogenous hormones, according to a U.S. transgender survey. “So it’s no surprise that the most prevalent finding in their Paps is atrophy involving between 62 and 93 percent of their samples,” Dr. Ely said. “And thus far there is no evidence to suggest that androgens increase a person’s risk of cervical cancer.”
The changes seen in testosterone-induced Pap tests from transgender men are like those seen in Pap tests from cisgender women, Dr. Ely said (Fig. 2): numerous parabasal-type cells dispersed or in clusters with high nuclear-to-cytoplasmic ratios, round nuclear contours, evenly dispersed chromatin, and even hyperchromasia.
Transitional cell metaplasia is another feature seen in transgender men and cisgender women, though it’s less common in the latter, Dr. Ely said (Fig. 3). “Here, the exfoliated cells are arranged in flat sheets and take on a kind of streaming look. The nuclei are oval to spindled with somewhat tapered ends, the chromatin is powdery, and the nucleoli are very small.” Longitudinal grooves extending along the diameter of the nuclei give the cell “a coffee-bean-like appearance,” she said.
The presence of small cells representing severely atrophic parabasal cells is another common finding (Fig. 4). “These cells are arranged in groups, have high N:C ratios, nuclear hyperchromasia, and indistinct nucleoli, but the nuclear envelopes are round and smooth,” Dr. Ely said, noting this has been seen in patients on tamoxifen therapy.
The importance of these small cells in the transgender population lies in not misinterpreting them as endometrial cells because that could trigger an unnecessary biopsy and colposcopy, Dr. Ely said, especially if the patient is over age 45.
Testosterone-induced atrophy contributes to statistically higher numbers of unsatisfactory samples, with rates as high as 23.4 percent, she said, and the rate increases with length of time on androgen therapy. This is in contrast to a rate of three percent for cisgender patients with atrophy (Plummer RM, et al. J Am Soc Cytopathol. 2021;10[3]:255–260).
“It has been suggested that the cellularity threshold for transgender men on testosterone be lowered from 5,000 to 2,000 just as it has been done for cisgender women who have undergone chemotherapy or radiation, have a surgically absent cervix, or are postmenopausal,” she said. This would be especially impactful for transgender men because an unsatisfactory diagnosis can result in their becoming discouraged by the screening process. “It can take them five times longer to return for retesting compared with cisgender women.”
The Bethesda System for Reporting Cervical Cytology doesn’t address the effects of testosterone on cellularity, Dr. Ely said, but the explanatory notes in the chapter on specimen adequacy say strict objective criteria may not be applicable to every case and in some clinical circumstances a lower cell number may be considered adequate, requiring professional judgment. The Vanderbilt laboratory has thus composed a comment: “Please note that testosterone therapy often results in lower-cellularity samples. If a reduced threshold for specimen adequacy (>2,000 cells) is applied for patients in this population, this sample would/would not meet the lower cellularity threshold.” Said Dr. Ely, “We’re letting the clinician know we see the patient is on testosterone and have taken that into account when determining cellularity.”
The Vanderbilt laboratory performs cotesting on all samples from transgender men regardless of age, Dr. Ely said, to provide the physician and patient with more data to develop follow-up plans. “If the HPV is negative, this will allow for an extended screening interval. If it is positive and the Pap is unsatisfactory, this can help guide future management decisions.”
Dr. Ely presented the case of a 25-year-old transgender man who had been on testosterone for two months to illustrate the challenge of detecting high-grade squamous intraepithelial lesion in a background of atrophy (Fig. 5). “This goes for cisgender women as well,” she said. Morphologic features (Fig. 6) supportive of HSIL include cells with a higher N:C ratio, more nuclear contour irregularity, and hyperchromatic coarse chromatin.
There is limited data on the rates of epithelial abnormalities and high-risk HPV in transgender men, she said. Epithelial abnormality rates range broadly from 5.6 percent to 29.2 percent. Small sample study sizes and differing sexual practices within the cohorts may explain some of the variation, she said. “And we all have difficulties in interpreting Paps with atrophy, especially if there’s background inflammation, so we may feel more comfortable calling them ASC-H or an ASC-US. Interestingly, the high-risk HPV positivity rates in those same studies mirrored the epithelial abnormality rates, with a range of eight percent to 33 percent.”
In a study published in 2022, Lin, et al., evaluated the accuracy of high-risk HPV testing for the detection of high-grade squamous intraepithelial lesions in transmasculine persons receiving testosterone (Lin LH, et al. Diagn Cytopathol. 2022;50[11]:518–524). “Because of the low numbers of patients in their institution who had cotesting, they combined their cases with others in the literature for a total of around 72,” Dr. Ely said. The authors reported HPV testing has a sensitivity of 100 percent, a specificity of 87 percent, and positive and negative predictive values of 36 and 100 percent, respectively, for the detection of HSIL, she said. They write, “High-risk HPV testing can aid in the detection of precursor lesions due to its high sensitivity for HSIL detection.”
The same authors noted the use of self-collected vaginal swabs as an alternative for patients who decline screening due to discomfort with pelvic examination. Dr. Ely cited a study by Reisner, et al., of 131 transmasculine people among whom self-collection and provider collection for HPV testing were evaluated (Reisner SL, et al. PLoS One. 2018;13[3]:e0190172). Compared with the provider-collected cervical high-risk HPV DNA sample, the self-collected vaginal hrHPV DNA test demonstrated a sensitivity of 71.4 percent, specificity of 98.2 percent, and a negative predictive value approaching 95 percent.
“Over 90 percent of the participants in that study preferred the self-collected vaginal swab to the provider-collected cervical swab,” Dr. Ely said.
In mid-May, Roche and BD announced Food and Drug Administration approval for the use of self-collected vaginal specimens for HPV testing.
Appropriate cancer screening and dysplasia risks are less clear for people who have neovaginas, Dr. Compton said, which can be seen in transgender women and patients who undergo vaginal reconstruction after exenteration for gynecologic cancer or surgery for variations in sexual development.
A woman with Mayer-Rokitansky-Küster-Hauser syndrome—a form of idiopathic Müllerian agenesis—who had undergone vaginoplasty surgery with creation of a sigmoid neovagina (Fig. 7) reported engaging in penetrative intercourse with a cisgender male partner and asked her physician about appropriate screening. The patient’s physician then asked Dr. Compton, who was Dr. Ely’s fellow at the time, if the patient should be screened for HPV-related cancer. Organizational guidelines differ. The American College of Obstetricians and Gynecologists committees on gynecologic practice and health care for underserved women said in 2021 that a neovagina does not require routine cytologic screening. The Endocrine Society advised monitoring clinically for reproductive organ cancer risk (Hembree WC, et al. J Clin Endocrinol Metab. 2017;102[11]:3869–3903). The Canadian Rainbow Health Coalition recommended in 2006 that transgender women who have undergone penile inversion vaginoplasty and have risk factors, such as a history of genital warts or HIV infection, be screened every year, and women without risk factors every two years.
“So there’s a bit of a debate with pros and cons,” Dr. Compton said. One screening con: lack of a cervical transformation zone because there is no cervix. “And even in those neovaginas that are created from squamous epithelium, the epithelial composition, which could be from a split-thickness skin graft or from penile epithelium, is different from that of the cervix,” she said. Cisgender women who have had adequate screening and no history of dysplasia can exit cervical cancer screening programs after hysterectomy.
In the pro screening category: “We’ve seen that penile epithelium can develop HPV-related cancers,” Dr. Compton said. “And cohort studies have shown that patients with neovaginas created due to either vaginal aplasia or in transgender women have shown persistent low- and high-risk HPV infection.”
If the patient is screened, several questions arise, one of which is how to report it. “Does this fall under the Bethesda reporting for cervical cytopathology? Should we process it as a nongynecologic sample? Should we put it on an imaging system? Or does it need to undergo full manual review, as imaging systems are not FDA approved for this sample type? Does it need to be examined by a pathologist even if negative, or can it be signed out by a cytotechnologist?” In addition: What is the surgical technique? “The cytomorphology can vary greatly based on the surgical technique,” Dr. Compton said.
Fig. 8 is a Pap smear from a 32-year-old transgender woman who underwent feminizing genitoplasty via penile skin inversion. “Uncharted waters” is how Dr. Ely described it. Laboratories that follow the 5,000-nucleated-squamous-cell requirement of the Bethesda criteria would consider it unsatisfactory due to hyperkeratosis, she said.
A variety of tissues and techniques have been used for vaginal reconstruction (Hembree WC, et al. J Clin Endocrinol Metab. 2009;94[9]:3132–3154). “The gold standard for transgender women is penile skin inversion with or without scrotal flaps,” Dr. Ely said. For those with congenital or traumatic vaginal absence, intestinal transplant is the preferred method. Other sources that have been used for neovaginal creation are dura, peritoneum, myocutaneous flaps, amniotic membrane, and split-thickness skin grafts from nongenital sites. “Regardless of the materials or methods used,” Dr. Ely said, “those grafted tissues are subject to stressors, which can predispose them to the development of precancerous lesions and invasive carcinoma.”
The neovaginal malignancies that develop appear to depend upon the tissues from which they originated, with adenocarcinomas arising from intestinal transplants and squamous cell carcinomas stemming from skin flaps or grafts, Dr. Ely said (Schober JM. J Pediatr Urol. 2007;3[3]:167–170). “To date there have been eight cases of adenocarcinoma following bowel vaginoplasty, none of which were in transgender individuals,” she said, adding there does not appear to be an association with HPV. “Transplanted bowel retains its basic properties, like those genetic factors that can predispose one to polyposis syndrome, inflammatory bowel disease, and cancer.”
There have been more cases of neovaginal squamous cell carcinoma than there have been of neovaginal adenocarcinoma, Dr. Ely said. Five cases of squamous cell carcinoma have been reported in transfeminine patients with neovaginas constructed by penile inversion with scrotal skin flaps, she said, “and this number will likely increase as gender-affirming surgery becomes more accessible and frequently performed.” Four of the five reported SCC cases were tested for HPV—three were positive for high-risk HPV and one had a history of low-risk HPV infection. The four SCC cases found to be HPV positive “would support the role for HPV in cervical neoplasia in transgender women with neovaginas created by penile skin inversion,” Dr. Ely said. In the fifth case of SCC, lung metastasis was the initial presentation (Wang G, et al. Case Rep Oncol. 2020;13[1]:17–22).
Microinjury from intercourse and from recurrent dilatation of the vagina to prevent it from shortening, contracting, and eventually closing and from exposure to chemicals in semen may also be at play, Dr. Ely said. “But we don’t know—there’s not sufficient data.”
Determining the prevalence of neovaginal HPV and cytologic abnormalities in transgender women post-vaginoplasty is difficult, she said, because the studies are few and have limitations. Some included cisgender (with congenital or traumatic absence of the vagina) and transgender women in their cohorts or combined HPV prevalences for both transgender males and females. Others did not indicate the types of HPV tested or the location from which the HPV sample was collected (including samples from the anus, oral cavity, and neovagina). Still others did not record the vaginoplasty procedure type.
In a systematic review of the literature, Nandwana, et al., identified 15 studies of neovaginal HPV prevalence among transfeminine individuals (Nandwana D, et al. Obstet Gynecol. 2023;142[2]:296–306). “A few of them looked at the high-risk HPV prevalence, which ranged from 8.3 percent to 20 percent,” Dr. Ely said.
None of the studies looked at the pre-vaginoplasty HPV status of the donor tissue from the penile or scrotal skin, she said, “and this would be important to know because there is a high prevalence of HPV in male genital sites”—28 percent in the foreskin, 24 percent in the penile shaft, and 16 percent in the glans.
Of 212 patients in the 15 studies, the authors reported the aforementioned five cases of squamous cell carcinoma. “In addition, there were two high-grade squamous intraepithelial lesions, three LSIL, two ASC-US, and three condylomas. These numbers are probably an underestimate of the true incidence of dysplasia and carcinoma in this population,” she said.
Weyers, et al., reported that only four percent of transgender women ever had a gynecologic exam after their confirmation surgeries were complete (Weyers S, et al. Cytopathology. 2010;21[2]:111–115). “This raises concern as to whether patients with neovaginas are aware of their cancer risk and their need to participate in cancer screening programs,” Dr. Ely said. There’s a paucity of data to determine if HPV cotesting or primary HPV screening detects preneoplastic lesions and prevents invasive cancer in this population, she said.
For those who pursue cytologic screening, she said, pathologists and cytotechnologists encounter problems with adequacy requirements and the cellular composition of the grafts when strictly following the Bethesda criteria for their evaluation (Mohr S, et al. Eur J Obstet Gynecol Reprod Biol. 2021;260:177–182).
A small number of anucleated cells can be encountered in Pap tests from cisgender women, Dr. Ely said (Weyers S, et al. Cytopathology. 2010;21[2]:111–115). “But abundant anucleated cells is a constant finding in samples from [penile] skin-lined neovaginas,” she said, “so if you’re adhering to that 5,000-nucleated-squamous-cell requirement, many [as in the prior case] would be considered unsatisfactory due to hyperkeratosis, with nondiagnostic rates ranging from 3.8 percent to 28 percent. It would seem reasonable to use a lower cellularity threshold closer to the 2,000-cell requirement in place for anal cytology, another specimen known to yield abundant hyperkeratotic material.”
Intestinal-type grafts (Fig. 9), seen more commonly in cisgender women, are not expected to contain a significant squamous component, she said. “So reporting them as unsatisfactory is a problem because no matter how many times you go back, you’re not going to increase your yield of squamous cells. Nor is it appropriate to use ‘negative for squamous intraepithelial lesion or malignancy,’ as smears from intestinal grafts do not typically contain adequate squamous cellularity.”
The term “benign glandular cells consistent with neovagina” may be more appropriate in this setting, Dr. Ely said. “And this also raises the question of whether the traditional kits we’re using for cisgender women are best suited for transgender patients.”
To avoid misdiagnoses and prevent missed screenings for transgender people, Dr. Ely recommends increasing dialogue with clinicians, perhaps in the multidisciplinary colposcopy conferences, where they can be educated on the effects of testosterone use on cytological samples and the need for thorough documentation.
An inclusive and complete health record that addresses the unique needs of the transgender community is important but comes with IT and other hurdles.
Easier to implement in the IT system might be “expanding our Pap specimen sources to go beyond cervix/endocervix and vagina to include neovagina,” she said, with boxes to check off regarding how the neovagina was made, be it skin graft or intestinal transplant. Likewise, options for hormone therapy could capture testosterone, as well as estrogen and progesterone. “Some laboratories may want to consider specifying the length of testosterone use,” Dr. Ely said, because the unsatisfactory rate for transgender men on testosterone for less than six months is under five percent, and for those receiving exogenous hormones for greater than six months is 20 percent (Peitzmeier SM, et al. J Gen Intern Med. 2014;29[5]:778–784).
“Our colleagues in the clinical pathology world would also like to know about testosterone use because it can impact other laboratory indices like hemoglobin, cardiac troponin, and creatinine,” she said (Cheung AS, et al. J Clin Endocrinol Metab. 2021;106[3]:893–901).
In a retrospective slide review by Moatamed, et al., the authors found that clinical information regarding whether a subject is transgender and/or receiving testosterone therapy is crucial to avoiding Pap test overcalls (Moatamed NA, et al. Cytopathology. 2023;34[2]:120–129). When testosterone use history was not readily available, Dr. Ely said, 58.2 percent of 122 cases were diagnosed initially as negative for intraepithelial lesion without atrophy, and 26.2 percent were diagnosed as NILM with atrophy. Seven tests, or 5.7 percent, were diagnosed initially as abnormal. When that clinical history was made known, she said, “the number of cases in which atrophy was picked up more than doubled, at 62.3 percent, whereas the abnormals nearly halved, to 3.3 percent.” The two ASC-H diagnoses were downgraded.
“These findings reinforce the importance of appropriate history in the diagnoses of samples from transgender patients.”
Amy Carpenter is CAP TODAY senior editor.