Mycoplasma genitalium treatment strategies in the era of significant macrolide resistance: Findings from a multi-centre audit in Northwest England

Introduction

Mycoplasma genitalium (Mgen) is a sexually transmitted bacterium that predominantly affects the genitourinary tract, causing non-gonococcal urethritis (NGU) or cervicitis, as well as complicated urogenital infections such as epididymo-orchitis in men and pelvic inflammatory disease (PID) in women. It has also been linked to preterm birth and miscarriage. ¹ The prevalence of Mgen in the general UK population is 1–2%, rising to 38% in individuals attending sexual health clinics. ²

The management of Mgen is challenging for several reasons. Firstly, the absence of a cell wall makes penicillins and cephalosporins ineffective treatment against Mgen. Only a few antibiotic classes retain therapeutic activity against mycoplasmas, including tetracyclines, macrolides, fluoroquinolones, and streptogramins (e.g., pristinamycin). Of these, macrolides are the first-line treatment for uncomplicated Mgen in UK and European guidelines.^1,3

Due to limited effective treatment options available, current UK guidelines recommend testing based on symptoms, or for current sexual partners of people infected with Mgen. ¹ Guidelines advise that all Mgen positive specimens should be tested for macrolide resistance associated mutations (MRAMs), but this is logistically challenging as results are often not received in a timely manner to allow for prompt and definitive treatment decision making. In this audit, we describe the epidemiology of Mgen, MRAMs and fluoroquinolone resistance associated mutations (QFRAMs) in our region and review our current management against UK guidelines to aid service improvement.

Methods

Thirteen sexual health clinics across Cheshire, Manchester, and Merseyside participated in this audit. We undertook a retrospective case note review of all attendees with a positive Mgen NAAT from September 2022 to December 2023. Patients were identified using the Genitourinary Medicine Clinical Activity Dataset (GUMCAD), a mandatory STI surveillance system in the UK, with Mgen coded as C16. Data abstracted from electronic patient records included demographics, reason for testing, site(s) of infection, antimicrobial resistance reports, treatment regimen received, and test of cure results.

For all clinics, Mgen NAAT was performed on the Roche Cobas 8800 and samples for antibiotic resistance testing were sent to the United Kingdom Health Security Agency (UKHSA) sexually transmitted infections reference laboratory where inhouse real time polymerase chain reaction (RT-PCR) was undertaken (targets Mgen adhesin P1 (MgPa) and glyceraldehyde-3-phosphate dehydrogrenase (gap)). ² Sanger sequencing was then performed for region V of the 23S rRNA gene (MRAM) and in some samples the QRDR of the parC gene or gyrA gene (FRAMs).

All data were anonymized, and identifiable individual-level data were removed. We performed descriptive analysis of the collated data using Microsoft Excel. We also used the Statistical Package for the Social Sciences (SPSS version 29) utilising chi square testing to assess association between antimicrobial regimen received and TOC rates.

Results

We reviewed 1067 samples from 1036 patients. A total of 528 persons were seen in Liverpool and Cheshire, and 508 were seen in Manchester.

Macrolide and fluoroquinolone resistance

As summarised in Figure 1: Flow chart of clinical outcomes of MRAM-determined samples, 753 samples (72.9%) were successfully tested of which 418 (55.5%) had MRAMs detected. It was not possible to test 27.1% (n = 280) of samples due to factors such as low DNA titres. Routine fluoroquinolone resistance testing was available for a brief period at the start of the audit but was subsequently discontinued by UKHSA. Results were available for 3.3% (n = 34) of samples, as summarised in Figure 2: Flow chart of clinical outcomes of FRAM-determined samples. All 7 patients with FRAMs also had MRAMs. FRAMs included S83I (5 samples), D87 N (1 sample) D87Y (1 sample). One sample with a S83I mutation also had a G93 C mutation (of unknown significance).OPEN IN VIEWER

Figure 1.

Flow chart of clinical outcomes of MRAM-determined samples.

OPEN IN VIEWER

Figure 2.

Flow chart of clinical outcomes of FRAM-determined samples.

Discussion

Our resistance data show a macrolide resistance rate of 55.5% across our region. Although this is lower than the national average of 62.2% reported by the Mgen Antimicrobial Resistance Surveillance, it remains high. ² High rates of macrolide-resistant Mgen infection have been reported globally. For instance, rates of up to 66% have been reported in the Netherlands, and all 31 Mgen-positive samples in a study in Greenland demonstrated macrolide resistance.^5,6

In our setting, 28.7% people diagnosed with Mgen required second-line treatment due to macrolide resistance, antimicrobial intolerance, or treatment failure. The implications of this are significant, potentially leading to increased costs due to extra prescribing and multiple patient reattendances, greater strain on sexual health services, risk of loss to follow-up, and higher patient dissatisfaction rates. We propose local resistance testing as a solution to this challenge for two reasons. Firstly, local testing ensures results are available to the clinician more quickly compared to the current arrangement of testing at a central reference laboratory with a turnover of 2–3 weeks. If results are available sooner, then patients can be counselled to wait for resistance results before commencing treatment. Unnecessary exposure to macrolides may contribute to further antimicrobial resistance and lead to adverse effects, such as gastrointestinal discomfort, which could be avoided if treatment were delayed until resistance reports were available. Available evidence shows that resistance-guided management can be cost-effective, reduces rates of treatment failure, and prevents overuse of fluoroquinolones.^7,8

Equally worryingly, the rate of fluoroquinolone resistance in our cohort is 20.6%. Fluoroquinolone resistance testing is not routinely available, and therefore our data should be interpreted with caution due to our small sample size (n = 37). Fluoroquinolone resistance rates of 12.1% were reported in the latest MARS report, where all samples from reporting clinics were tested over a 2-month period. ² Fluoroquinolone-resistant Mgen already presents a management dilemma for clinicians, and some opt for a trial of fluoroquinolone regardless of resistance status. More research is needed to determine the optimal management strategy for this patient cohort.

The 2018 guideline, current at the time of this project, recommends that all patients have a test of cure performed 5 weeks post-treatment to confirm microbiological clearance of infection and identify treatment failures, with a duty to report confirmed treatment failures to the UKHSA for surveillance purposes. ¹ Our data show that only 56% of patients had a test of cure performed. The reasons for this high rate of nonattendance are unclear. However, it is possible that a home test kit, if available, may be a better testing option, as it does not require clinic attendance.

The 2025 UK guideline (currently in draft) is no longer recommending routine TOC, only advising TOC in individuals who remain symptomatic and where there is suspicion of persistent infection. ⁹ Given that microbiological cure can be challenging, especially in the presence of resistance associated mutations, the more pragmatic approach taken by the new UK guidelines seems sensible, despite the potential risk of the presence of ongoing antibiotic-resistant infection in some patients.

UK guidelines recommend moxifloxacin monotherapy where a fluoroquinolone is required, and this recommendation is supported by our findings of comparable microbiological cure rates among those who received moxifloxacin alone and those who received doxcycline and moxifloxacin. USA guidelines however continue to recommend doxycycline pre-treatment to preserve moxifloxacin effectiveness.^1,4

At least 1 in 10 individuals in our cohort were inappropriately screened for Mgen. This is comparable with a UK national audit in 2022 where 8.6% of individuals did not have a guideline recommended or suggested reason for Mgen testing. ¹⁰ While Mgen screening is increasingly available, it is paramount that screening is reserved only for those meeting BASHH criteria. ¹ We recommend strict screening criteria that discourages asymptomatic screening. This will help avoid unnecessary exposure to antibiotics, with the concomitant risk of driving further antimicrobial resistance.

Limitations

Our audit was geographically restricted to the Northwest of England and may not be generalizable to the rest of the UK or internationally. Resistance reports were not available for a small proportion of patients, due to the transfer of patient records from paper to electronic.

Conclusions

The high and rising rates of macrolide-resistant Mgen are an issue of public health concern, and clinicians are increasingly faced with challenges on how to address this. Reported fluoroquinolone resistance is currently low; however, we would not recommend fluoroquinolones as first-line treatment due to their side effect profile and the risk of disabling musculoskeletal and neuropsychiatric toxicity. In January 2024, the UK Medicines Regulator (MHRA) further limited fluoroquinolone use only to situations where other classes of antibiotics are inappropriate. ¹¹ There are also wider issues to consider regarding the lack of large-scale data on baseline fluoroquinolone resistance, as well as expected increases in fluoroquinolone resistance with more prevalent use. We propose a multifaceted approach that includes strict adherence to testing criteria, local resistance testing, resistance-guided management, as well as innovative approaches to improve test-of-cure rates, such as home testing kits.