Cervical Cancer Screening with Human Papillomavirus Self-Sampling Among Transgender Men in El Salvador

Abstract

Purpose: Sexual and gender minority persons in low-income countries have very limited access to routine health services. This study evaluated the feasibility of using a self-sampled human papillomavirus (HPV) test to increase access to screening for cervical cancer among transgender men in El Salvador.

Methods: We partnered with a local advocacy organization for recruitment. A total of 24 transgender men (men assigned female at birth) ages 19–55 were enrolled and provided consent. Questionnaires assessed sociodemographics, health and sexual histories, and knowledge about HPV and cervical cancer. Screening was performed with a self-sampled HPV test. Participants with a positive test were offered colposcopy and cryotherapy treatment, if appropriate. Those with a negative test were advised to return in 5 years for rescreening.

Results: Out of 24 consenting participants, 23 (95.83%) agreed to conduct HPV self-sampling, and 22/23 (95.65%) expressed willingness to self-sample in the future. Among self-sampled individuals, 3/23 (13%) tested positive and accepted colposcopy and biopsy. Analyses of biopsied tissue revealed one case of cervical intraepithelial neoplasia grade 1.

Conclusion: HPV self-sampling and subsequent procedures were accepted by the majority of participants. This screening method may be a viable alternative to cytology among transgender men in El Salvador.

Introduction

Cervical cancer is a leading cause of death for women in low- and middle-income countries (LMICs).¹ The disease is preventable through the detection of precancerous lesions before progression to invasive disease, but the success of screening programs based on cytology (i.e., Papanicolaou [Pap] test) has been limited to high-income settings.² In addition, sexual and gender minority (SGM) populations, including transgender men (men assigned female at birth), remain difficult to reach.^3,4 Barriers to cervical cancer screening faced by transgender men include systematic discrimination,^5–7 provider misinformation,^8,9 and the psychological discomfort of undergoing an examination usually reserved for women.^10,11 In LMICs, these challenges are compounded by the absence of relevant guidelines and lack of infrastructure required to meet the presentative and therapeutic needs of this population.

As most transgender men do not undergo gender-affirming surgeries, including hysterectomy,^5,12 the risk for cervical cancer remains. The American College of Obstetricians and Gynecologists recommends that transgender men with a cervix follow the same cervical cancer prevention guidelines as cisgender women (women assigned female at birth).¹³ However, in the United States, transgender men are significantly less likely than cisgender women to undergo Pap tests.^14,15 As cervical cancer screening is less common in LMICs,² it is likely that SGM people who live in these settings have even lower screening rates.

The discovery that the human papillomavirus (HPV) is the causal agent of cervical cancer has resulted in the development of alternatives to cytology screening. Assays that detect the presence of high-risk types of HPV have superior sensitivity to the Pap test¹⁶ and offer the possibility of patients collecting the sample themselves. Self-sampled HPV tests have been found to be acceptable to cisgender women in El Salvador and various LMICs,^17–22 circumventing some barriers of conventional screening such as embarrassment at the possibility of being examined by a male doctor and fear of a pelvic examination.^20–22 Similarly, transgender men in the United States have expressed a preference for HPV testing, whether self- or provider-collected, over a Pap test using a speculum.¹¹ In a recent study, 90% of 131 transgender men who completed provider and self-sampled HPV screening preferred the self-sampled test, which correctly identified 15/21 positive results obtained by the provider-sampled tests.²³ Although provider-collected HPV tests have superior sensitivity and specificity than self-sampled tests,²⁴ this must be balanced with the need to reach underscreened and vulnerable populations where self-collection may be more acceptable and cost-effective.²⁵ However, the recommendation in most countries, including El Salvador,²⁶ remains for HPV test samples to be provider-collected.

El Salvador has one of the highest cervical cancer mortality rates of the region (9.4 per 100,000, age-adjusted).²⁷ HPV vaccination is not available in the public sector, but screening via HPV testing was introduced to the country in 2012 and became part of the national cervical cancer control program in 2015.²⁷ Before this, adherence to cytology screening was estimated to be <50%.²⁸ Current guidelines state that (presumably cisgender) women aged between 30 and 59 years should be screened with HPV testing every 5 years, whereas cytology every 2 years is recommended for women aged between 20 and 29 years, women 60 years of age or older, and where HPV tests are unavailable.²⁶ The guidelines include recommendations for vulnerable populations such as sex workers and women living with HIV or AIDS, but not for transgender men or other SGM people.

There is also a dearth of information regarding the experiences of SGM people within the national health system, and transgender men are particularly underrepresented. Existing data focus on transgender women and cisgender men who have sex with men. In those populations, it is estimated that <30% access public health services.²⁹ Reported experiences that discourage attendance at health institutions include harassment from security guards, delay or denial of services, and acts of intimidation and discrimination (e.g., mocking, misgendering).³⁰ If transgender men experience similar marginalization, HPV self-sampling may provide an opportunity to increase cervical cancer screening rates for this population. The purpose of this study was to investigate the feasibility of self-sampled HPV tests as a cervical cancer screening strategy among transgender men in El Salvador.

Methods

Participants

This was a cross-sectional study conducted between August and September 2017. The protocol was developed through a collaboration between Basic Health International (BHI), a nonprofit organization dedicated to cervical cancer prevention, and Asociación Generación Hombres Trans El Salvador (HT El Salvador), an advocacy group for transgender men. Both organizations are members of the Alliance for Cancer Prevention, an initiative of the El Salvador Ministry of Health to increase access to cancer prevention services. Members of HT El Salvador indicated that very few local transgender individuals undergo gender-affirming surgeries that remove the cervix or undergo cervical cancer screenings, placing most of them at risk for cervical cancer.

The goal was to enroll a convenience sample of individuals who met the following inclusion criteria: (1) identified as transgender men who were female sex assigned at birth, (2) age 18 years or older, (3) agreed to participate in the study, and (4) were willing and able to provide informed consent. Exclusion criteria were intended to focus the project on transgender men and included (1) refusal to participate in the study, (2) unwillingness or inability to provide informed consent, (3) not identifying as a transgender male, or (4) uncertainty about gender identity. Preliminary interviews with HT El Salvador members were used to establish eligibility categories that were congruent with the identities and terminologies used in the community, which may not mirror gender and sexual identities that are common elsewhere. For example, there were no mentions of nonbinary identities in interviews or in our literature review of sexual diversity in El Salvador; thus, this category was not included in the eligibility criteria. As transgender men are a hard-to-reach population, group leaders were instrumental in recruiting potential participants from their organization's membership. Recruitment methods also included an educational workshop on health issues considered important by the community, telephone calls, e-mails, and social media. HT El Salvador has a membership roster of 35 individuals and we anticipated that the final sample would be small.

Procedures

All data collection took place at the BHI facilities in San Salvador, El Salvador, which include offices and a clinic staffed by a gynecologist and a nurse. Subject participation involved two or three visits, depending on HPV test results. In the first 90-minute visit, BHI research staff performed informed consent procedures that explained the purpose and methods of the study. As this was a pilot research study focused on the feasibility of self-sampling and included subjects that were not within the target age range for HPV testing established by local guidelines, we did not offer provider-collection. However, local guidelines were explained to subjects during the consent process (transgender men between the ages of 30 and 59 years have the right to undergo provider-collected HPV testing in accordance with local guidelines).²⁶

Individuals who consented were administered a questionnaire (details in the Measures section) and were then offered the opportunity to self-sample using a high-risk HPV test (careHPV; Qiagen, Gaithersburg, VA). Currently, careHPV is the only low-cost HPV test prequalified by the World Health Organization (WHO),³¹ although other tests are in development. The test kit consists of a small cone-shaped brush and a sealable tube filled with collection medium where the brush is stored after sampling. Ideally, the brush is inserted into the vagina until it makes contact with the cervix, but recent research has shown that vaginal swabs are sufficient for the detection of high-risk HPV types.³² After insertion, the brush is rotated several times in place, removed, and placed in the sealable tube. Samples are analyzed in batches using a careHPV machine. As with other HPV tests, performance of the careHPV test differs between self- and provider-collected modalities: a recent meta-analysis provides pooled estimates of 73.6% sensitivity and 88% specificity for cervical intraepithelial neoplasia (CIN) grade 2 or higher with the self-sampled test, compared with 88.1% sensitivity and 83.7% specificity for clinician-collected samples.³³

BHI staff explained the test procedures verbally and with the aid of a visual aid adapted to a transgender male population (Fig. 1).³⁴ Subjects who agreed to self-sample were led to a private room where they could perform the procedure privately. Following HPV testing, participants underwent a routine physical examination including a fasting blood draw (results from these procedures, including levels of exogenous testosterone, will be reported elsewhere).

FIG. 1.

Visual instructions for HPV self-sampling adapted to a population of transgender men (this figure was adapted from our team's previous self-sampling study³⁴ with input from members of Asociación Generación Hombres Trans El Salvador). HPV, human papillomavirus.

Subjects were asked to return after 3 weeks to receive their test results in an individual 45-minute consultation with an internist who provided further explanations about all results, gave recommendations for follow-up procedures, and answered any questions. For those with negative HPV results, the recommendation was to repeat screening in 5 years in accordance with local guidelines.²⁶ Individuals who tested positive for HPV were offered a third visit after 1 week to undergo colposcopy, biopsy, and, if no contraindications were found, cryotherapy treatment as per a screen-and-treat protocol as endorsed by the WHO³⁵ and El Salvador's guidelines.²⁶ The management plan for individuals with high-grade precancer not treatable by cryotherapy or with invasive disease was referral to the National Cancer Institute of El Salvador, but no such cases were found. Colposcopy, biopsy, and treatment were conducted by BHI's gynecologist. Biopsy samples underwent routine histological processing at a local laboratory and were subsequently reviewed by an expert gynecological pathologist. All research procedures and documents were approved by the National Ethics Committee for Clinical Research of El Salvador, a dependency of the Superior Council of Public Health of El Salvador.

Measures

HPV self-samples resulted in either a positive or negative result. The questionnaire consisted of sections administered before and after the HPV test. Portions of the questionnaire were originally developed in collaboration with a sociologist for earlier self-sampling studies with cisgender women in rural areas.^34,36 These questions were retained for comparability purposes. Other questions were added after receiving feedback from informants at HT El Salvador.

The section of the questionnaire to be answered before HPV sample collection consisted of 36 multiple choice and short-answer items divided into seven modules: demographics, tobacco use over the lifetime and in the last month, alcohol consumption in the last year and month, current diet, HPV and cervical cancer history (e.g., date of last screening, site of last screening), sexual and reproductive health (e.g., lifetime sexual partners, contraception use, attendance at routine medical checkups), and knowledge and perceptions of HPV infection and cervical cancer. Participants were then offered HPV self-sampling. Those who completed self-sampling were administered a seven-item multiple choice questionnaire regarding satisfaction with the procedure and two open questions (why the individual would or would not perform the procedure again, and any additional comments regarding HPV self-sampling). Participants who did not agree to self-sample were asked to select applicable reasons for their decision from 12 possible choices (including “Other”) and asked to share additional comments about HPV self-sampling. The questionnaire ended with five closed questions for all participants on individual history of exogenous testosterone use, including duration and procurement.

Analyses

Descriptive statistics were used to analyze questionnaire items. These data were collected and managed using REDCap^37,38 electronic data capture tools hosted at BHI and analyzed using Stata 14 (StataCorp LP, College Station, TX).³⁹ Open responses were translated and analyzed by a trained anthropologist who is a native Spanish speaker (M.S.) to develop a coding scheme. Two other reviewers (K.A. and R.M.) independently coded all responses. The three researchers discussed resulting categories and variations in coding and agreed on major emergent themes.

Results

Out of 35 HT El Salvador members invited to the study, 24 ultimately participated (68.6% enrollment). The remaining 11 did not respond to the study invitation or could not participate due to work or transportation conflicts. Participant mean age was 29 years (standard deviation [SD] = 2.6, range = 19–55). Sociodemographic variables are shown in Table 1. A majority (66.7%) stated that they never attended routine medical checkups and 15 (62.5%) had never had a cervical cancer screening in the past. Other health and socio-sexual history variables are shown in Table 2. Despite the reported nonattendance at health services, most individuals reported basic knowledge of HPV and cervical cancer (Table 3).

Table 1.

Sociodemographic Variables (n = 24)

Age, mean (standard deviation)	29.3 (2.6)
Education level, n (%)
Elementary	6 (25)
High school	7 (29)
University	11 (46)
Marital status, n (%)
Single	15 (63)
Partnered	9 (37)
Have children, n (%)	2 (8)

Table 2.

Health and Socio-Sexual History Variables

	n (%)
Tobacco use in the last 30 days	15 (62.5)
Alcohol consumption in the last 30 days	13 (54.16)
Daily water consumption <1 L	10 (41.7)
Past sexual partners—cisgender males	10 (41.7)
If yes, how many (range)	2 (1–8)
Past sexual partners—cisgender females	20 (83.3)
If yes, how many (range)	5.5 (1–50)
Contraception use
Condoms	6 (25.0)
Oral pills	2 (8.3)
Injection	0
Intrauterine device	0
Other	4 (16.7)
Never	14 (58.3)

Table 3.

Knowledge and Perceptions of Human Papillomavirus and Cervical Cancer

	Yes, n (%)	No, n (%)	Not sure, n (%)
Ever heard about HPV	21 (87.5)	3 (12.5)	0
Perception of likely having had HPV infection in the past	3 (12.5)	16 (66.7)	5 (20.8)
Perception of likely having HPV infection in the future	13 (54.2)	4 (16.7)	7 (29.1)
Perception of lifetime risk of cervical cancer	14 (58.3)	6 (25)	4 (16.7)
Perception of HPV infection as serious/extremely serious	20 (83.3)	0	4 (16.7)
Perception of cervical cancer as serious/extremely serious	19 (79.2)	2 (8.3)	3 (12.5)

HPV, human papillomavirus.

Out of 24 participants, 23 (95.8%) agreed to perform HPV self-sampling and the individual who refused the test stated as his reason that he had never participated in penile/vaginal penetrative sexual intercourse. All 24 participants returned for the second visit to receive results from the HPV test, the physical examination, and the blood draw (as mentioned earlier, results from the latter two procedures will be reported elsewhere). Of these, 3/23 (13%) who performed the self-sampling tested HPV positive. These three individuals returned for the third visit to undergo colposcopy and biopsy. All were eligible for ablation and received cryotherapy treatment. Subsequent histological analyses of biopsied tissue revealed one case of CIN grade 1, while the other two samples showed no CIN.

Just over half of the subjects (n = 13) reported using some form of testosterone hormone therapy in the past 30 days. Of these, four self-prescribed, six used a prescription from a medical professional, and the remaining three did not respond. Mean duration of therapy was 326.07 days (SD = 292.59, range = 14–730), with three individuals taking multiple types of testosterone for varying amounts of time (the longest usage duration period reported by each person was used to calculate the sample mean). The sample size was too small to calculate the association between HPV positivity and testosterone use, but 2/3 of the HPV-positive subjects reported current use of testosterone therapy.

Perceptions regarding the self-sampled HPV test by those who accepted self-sampling were mostly positive (Table 4). All but one individual stated that they would perform the self-sampling again. Reasons given for willingness to perform the self-sampling in the future are presented in Table 5.

Table 4.

Perceptions About Self-Sampling (1 = Least, 5 = Most)

	Mean	Mode	Median (range)
Overall satisfaction	4.3	5	4 (3–5)
Instruments felt comfortable	3.6	4	4 (1–5)
Satisfaction with oral explanation	4.8	5	5 (3–5)
Satisfaction with other explanation materials	4.8	5	5 (4–5)
Pain	1.3	1	1 (0–4)
Discomfort during sampling	1.5	2	2 (0–4)
Confidence that self-sampling was done correctly	2.8	3	3 (1–4)

Data are from 23 participants (95.8%) who accepted self-sampling.

Table 5.

Selected quotes from participants

	Would perform again	Would not perform again
	Much more comfortable, it is less painful, it is private	Because it is not necessary due to my hysterectomy
	Because I do not need a doctor and I can do it myself
	Less painful and more private
	To prevent
	Because it is for my own good
Question: Why would you or would you not perform self-sampling again?	The check-up is important, it is very easy and I felt very comfortable
	It is more private and I can do it more delicately
	It was easy, quick, and with no pain
	Comfortable and more private than cytology
	It is more intimate
	For my health, to be in charge of the screening
	I like the initiative, it is new and quick
	The test is important for the trans population, (to) improve ease and boost campaigns for trans men
	Like it and helps a lot
	It could be done with a longer instrument, the sampling would be easier to do lying down
	If someone as not used a tampon, it will be harder for them
Question: Additional comments about the test	I think it is good, I hope they can implement this in regular appointments, at least as an option
	I could recommend it to other people
	It is uncomfortable because we are not used to this test, but it is more comfortable than going to the gynecologist
	I liked the test and would recommend it to other people
	They could include a video of how to do the test. It would also be good to have more information of the difference between cytology and the test
	The sampling is very easy
	It is useful, it does not have to be comfortable

Most of the responses to open questions were only a few words or sentences. Among the 22 participants who agreed that they would be willing to self-sample again in the future, themes that emerged as factors in the decision included the perception of the procedure as physically comfortable, private, and good for one's health. Echoing previous research on the negative psychological impact of conventional screening examinations,^10,11 one person expressed that cytology is “much more invasive and traumatic. That's why I would do self-sampling again,” while another stated, “In my case as a trans man, it is much more comfortable, more private. Also, I can do it myself and avoid the shame of standing in line with women who are in line to see the gynecologist.” The one person who would not repeat the procedure stated that he had a previous hysterectomy, and thus did not believe it was necessary (Table 5).

Additional comments regarding self-sampling fell into two categories: positive aspects of the test (e.g., fast, easy, and comfortable) and suggestions for improving the experience (e.g., a longer brush to facilitate taking the sample, availability of the test at public health centers, and an instructional video). Approximately one-third of participants (n = 7) did not offer additional comments.

Discussion

To our knowledge, this is the first study to evaluate the feasibility of HPV self-sampling as a screening strategy for transgender men in an LMIC. The method was perceived favorably by most participants. With the exception of the procurement of hormone therapy, participants in this sample had infrequent contact with health services. A majority reported that they had never attended routine physical checkups and more than half had never received a cervical cancer screening. Despite this, all participants returned to receive their HPV test results and the three individuals who tested positive accepted colposcopy and treatment.

Although lack of access to health care is not unusual in LMICs, it is usually associated with poverty and low education levels. In contrast, these individuals were more educated than the national average (the middle school enrollment rate in the country is 33.2%).⁴⁰ Moreover, they demonstrated knowledge of both HPV and cervical cancer, and were cognizant of the risks posed by both.

Avoidance of medical services in this population may reflect the systematic bias and discrimination that SGM people encounter in health care systems, as has been observed in other settings.⁵ HPV prevalence in this sample (13%) was similar to estimates for cisgender women (12%) in the country.⁴¹ In addition to barriers to accessing adequate health services, the transgender men in this study exhibited cervical cancer risk behaviors, including tobacco and alcohol use and multiple past sexual partners. Health risk behaviors have been reported among other SGM subpopulations in El Salvador and linked to stigma-related stress.⁴²

Just over half of the transgender men in this study reported exogenous testosterone use. In transgender men, testosterone therapy has been linked to high rates of inconclusive Pap test results, possibly as a result of testosterone-induced cervical atrophy.⁴³ Although no effect of testosterone therapy has been found on cervical or vaginal HPV sampling,²³ this is a possible avenue for further research.

Limitations

A significant limitation of this study was the small sample size, which restricts the generalizability of these findings. Most participants were members of HT El Salvador, which is located in a metropolitan area and may not be representative of the broader community of transgender men. Some items on the questionnaire were retained for comparability with earlier studies, but future research would benefit from more precisely tailoring questions to a population of transgender men.

Conclusion

In LMICs, pervasive bias and lack of adequate resources mean that transgender men face significant barriers to accessing preventive health services.⁴⁴ Self-sampled HPV testing is an alternative screening method that circumvents pelvic examinations and offers a degree of privacy that may be particularly valued by SGM populations. Although current guidelines in El Salvador do not include self-sampling, there are ongoing efforts to make this strategy available for use with reluctant or hard-to-reach groups. If this occurs, possible venues to reach transgender men may include the screening centers of the current national program, HIV clinics that serve large numbers of SGM individuals, or nonprofit organizations with relevant missions.

Disclaimer

The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Diabetes and Digestive and Kidney Diseases. A preliminary version of the study abstract was presented at the 6th Annual Symposium on Global Cancer Research held in New York, NY, USA, on March 15, 2018, and published in the Journal of Global Oncology as part of the meeting proceedings.

Author Disclosure Statement

M.C. is president and founder of BHI, on the speaker bureau for Merck & Co. (Kenilworth, NJ) and CooperSurgical, Inc. (Trumbull, CT), and on the advisory board of Mylan Pharmaceuticals (Canonsburg, PA). M.S. is a former employee of BHI. No other competing financial interests exist.

Funding Information

This study was funded by the Einhorn Family Charitable Trust. J.O.M. was partially supported by grant K12DK111028 from the National Institute of Diabetes and Digestive and Kidney Diseases.