Self-taken Penile Meatal Swabs for Chlamydia, Gonorrhoea and Mycoplasma Genitalium Detection

Study Methods The study population will be men and transwomen aged 18 and over presenting to Leeds Sexual Health who wish to be tested for chlamydia, gonorrhoea and M. genitalium (if symptoms of urethritis are present). Those who have taken antibiotics (except metronidazole) in the past 14 days will be excluded. Details of age, ethnicity, past sexually transmitted infection (STI) history, being in contact with a STI and symptoms suggestive of a bacterial STI (urethral discharge, dysuria, testicular pain and swelling) will be collected. Prior to any examination by the clinician the participant will be asked to perform a self-taken penile meatal swab following the written and diagrammatic instructions for self-collected penile meatal swabs using the technique recommended by Hologic. Those with genital symptoms will be examined and the presence of urethral discharge and testicular swelling and tenderness will be recorded. A urethral swab for microscopy and gonococcal culture will be taken by the clinician if there is any evidence of urethral discharge. Those who have no genital symptoms will not be examined and will not have a urethral swab for microscopy or gonococcal culture taken as in routine clinical practice. All men will then be asked to perform a FCU sample. The penile meatal swab and FCU will be tested for M. genitalium if there are symptoms of urethritis. The first 500 participants will be asked to complete a short questionnaire of six questions on how easy/difficult, and comfortable/uncomfortable, they found taking the penile meatal swab was compared to collecting the FCU and their overall sample preference.

Consequently each participant will have two different samples taken:

Meatal swab: Aptima Combo 2 assay; Aptima MGen assay (if symptoms of urethritis)

First-catch urine sample: Aptima Combo 2 assay; Aptima MGen assay (if symptoms of urethritis)

The self-taken penile meatal and the FCU samples will be tested for C. trachomatis and N. gonorrhoeae using the Hologic Aptima Combo 2 assay (AC2), and those from participants with symptoms of urethritis will be tested for M. genitalium using Aptima MGen assay, according to the manufacturer's instructions (Hologic, San Diego, CA, USA). Positive tests for C. trachomatis will be retested using the Aptima CT assay and positive tests for N. gonorrhoeae will be retested using the Aptima GC assay, both of which use a different target to the AC2. A test will only be categorised as clinically positive if the appropriate confirmatory test is also positive. Positive tests for M. genitalium will be tested for M. genitalium macrolide and fluoroquinolone-resistance mutations by in-house PCR using Sanger sequencing to characterise mutants in the 23s gene for macrolide resistance and gyrA and parC genes for fluoroquinolone resistance (as developed by the Sexually Transmitted Bacteria Reference Unit, Public Health England). The resistance test will also act as the confirmatory test if the amplification is successful. It is anticipated that amplification will not be successful in up to 20% of the positive Aptima Mgen assays, because of the better sensitity of the commercial assay, in which case these positive samples for M. genitalium will be frozen, batched and shipped to Hologic, San Diego, CA, for confirmatory testing. The participant will be managed as if M. genitalium positive whilst waiting for these confirmatory results in view of the high specificity of the Aptima MGen assay, and to avoid further morbidity and sexual transmission.

All patients diagnosed as having chlamydia and/or gonorrhoea will be given treatment and managed with partner notification and follow up as per the clinic protocols and national guidelines. All patients tested for M. genitalium will have been diagnosed as having urethritis so will have been treated initially with doxycycline for seven days. Those who are positive for M. genitalium will be tested for M. genitalium macrolide and fluoroquinolone-resistance mutations. The result of this will be used to guide the therapy prescribed to treat the infection as per the United Kingdom (UK) and European guidelines. It is anticipated that about 50% of M. genitalium infections will be resistant to the first-line treatment of an extended regimen of azithromycin. These men will be treated with moxifloxacin. The men with both macrolide and fluoroquinolone resistance will be treated with a guideline recommended third-line treatment. Those where the amplification to perform resistance testing was unsuccessful will have been treated initially with doxycycline for seven days. Following this they will be given an extended course of azithromycin. If there is no clinical resolution, and/or if they are still M. genitalium positive at test of cure, they will be treated with moxifloxacin. A test of cure will be performed at five weeks following the start of treatment using both a self-taken penile meatal swab and a FCU sample. The current sexual partners will be tested for M. genitalium and given the same antibiotic regimen if positive.

Data from the Case Report Forms (CRFs) will be entered using the unique study number onto a secure computer system database which will be password protected. Data entry will be cross checked for errors and the database will contain alerts to identify if data being entered are illogical, inconsistent or incomplete. The master list of clinic unique identification codes and study numbers, and the CRFs will be kept in a secure locked cabinet, within a locked office, within Leeds Sexual Health.

Determination of patient infected status Patients will be classified as infected or not infected based on one positive (confirmed) AC2 test for C. trachomatis; one positive (confirmed) AC2 test for N. gonorrhoeae or a positive culture for N. gonorrhoeae; and one positive Aptima Mgen test.

The Department of Microbiology will perform the nucleic acid amplification assays and the M. genitalium macrolide and fluoroquinolone-resistance mutation assays.

Power calculation and statistical analysis It is anticipated that the sensitivity of the self-taken penile meatal swabs will be similar to the Dize study rather than the Chernesky study as all participants will perform the penile meatal sample first. The investigators predict a sensitivity of 92% for chlamydia and 99% for gonorrhoea. It is anticipated that the sensitivity of the FCU samples will be similar to the Chernesky study rather than the Dize study as a urine collection device will not be used. The investigators predict a sensitivity of 83% for chlamydia and 85% for gonorrhoea. There is only one published study comparing penile meatal swabs and FCU for M. genitalium detection. This reported 100% sensitivity with penile meatal swabs and 83% with FCU.

For the gonorrhoea analysis, 85 positive samples would detect a significant difference between a sensitivity of 85% and 99% with 80% power and 95% probability. For the chlamydia analysis, 260 positive samples would detect a significant difference between a sensitivity of 83% and 92% with 80% power and 95% probability. For the M. genitalium analysis, 80 positive samples would detect a significant difference between a sensitivity of 85% and 99% with 80% power and 95% probability (slightly higher sensitivity of FCU and lower of penile meatal swab used as only one published study).

The urethral gonorrhoea prevalence in the male population to be studied is at least 3.7% including asymptomatic men. Including 2400 symptomatic and asymptomatic participants, will give more than 85 positive individuals with 3.7% prevalence. The urethral chlamydia prevalence in the male population to be studied is at least 11%. 2400 participants will give 260 positive individuals with 11% prevalence. The prevalence of M. genitalium in those with symptoms of urethritis in the population to be studied is not known as a European approved Conformité Européene marked (CE marked) commercial diagnostic test has only just become available. In 2017, 500 cases of male non-gonococcal urethritis (NGU) were seen of which 100 (20%) would be expected to be M. genitalium positive. All with symptomatic gonococcal urethritis will also be tested for M. genitalium, of whom up to 10 men per year may also be positive for M. genitalium. This means the sample size of 80 would be achievable within 15 months with a 50-60% uptake in patients with symptoms of urethritis. It is expected that the uptake will be high, especially in those with symptoms of urethritis, as the study offers testing for M. genitalium which is not yet available in routine clinical practice as there is no funding for this service currently. The investigators have performed and published similar studies in order to identify the optimum sample for detection of chlamydia and gonorrhoea in females where 42 clinicians recruited almost 4000 females in 10 months which was shorter than the planned recruitment period.

All data analysis will be performed using the study number. The overall sensitivity, specificity, positive predictive value and negative predictive value will be determined for self-taken penile meatal swabs and first-catch urine samples by comparing test results with the patient infected status. The comparison between self-taken penile meatal swabs and FCU samples will be by McNemar's test. The analyses for chlamydia and gonorrhoea detection will also be calculated in those with symptoms and those without symptoms.

Utility of Antimicrobial Resistance (AMR)-guided therapy The utility of using AMR-guided therapy will be measured by the numbers that would have been given inappropriate antibiotics if the NGU guideline had been followed by giving doxycycline, then extended azithromycin, then moxifloxacin.

Cost comparison The economic analysis will be a simple cost comparison and will focus on whether using self-taken penile meatal swabs gives a cost saving compared to FCU samples. The analysis will use the correct detection of Chlamydia trachomatis, Neisseria gonorrhoeae and Mycoplasma genitalium as the primary endpoint and adopt the perspective of the UK National Health Service/Prescribed Specialist Services (NHS/PSS) in line with NICE recommendations (NICE 2013). Costs will be calculated from a sub-sample of 240 participants. NHS resource use and costs will be collected prospectively and will include costs of clinic time and resources and laboratory time and resources. Clinic staff will complete a specially designed data collection form for every trial participant during a 5-6 week period of the study. Information on unit costs or prices will be sourced to attach to each resource use item, to enable an overall cost to be calculated (e.g. Curtis \& Burns 2016). The final comparison will be based on cost per correct detection of Chlamydia trachomatis, Neisseria gonorrhoeae and Mycoplasma genitalium using self-taken penile meatal swabs versus FCU samples.