Macrolide-resistant Mycoplasma genitalium infections in Cuban patients: an underestimated health problem

Mycoplasma genitalium is an emerging sexually transmitted pathogen and the increase of macrolide resistance is considered a health problem globally [1]. In this bacterium, macrolide resistance is mediated mainly by point mutations in the A2058 and A2059 (Escherichia coli numbering) positions in region V of the 23S ribosomal RNA. These mutations are associated with azithromycin treatment failure and high minimal inhibitory concentration (MIC) for macrolides as documented in several M. genitalium strains [2, 3].

The prevalence of strains with macrolide resistance mediating mutations (MRMM) is highly variable and unknown in many regions. In Europe, Australia and Asia, it is generally above 30% and in extreme cases, as in Greenland, 100% of M. genitalium strains carry MRMM [4‐8]. In Cuba, macrolides are used as the first line treatment for urogenital syndromes caused by sexually transmitted infections (STI), including M. genitalium and chlamydia infections. Since 2007, M. genitalium diagnosis has been performed at the Pedro Kourí Tropical Medicine Institute (IPK), and several cases of treatment failure with 1 g single dose and extended azithromycin regimens have been detected in the IPK - STI clinic within recent years. In 2015, a new macrolide resistant M. genitalium strain (B19, A2059G mutation) was isolated from one of these patients using Vero cell co-culture [9], and antimicrobial susceptibility patterns were determined using a cell-assisted procedure [10]. However, no data about the prevalence of MRMM carrying M. genitalium strains in Cuba are available, but clinical experience of macrolide treatment failures and the isolation of at least one MRMM strain suggest the possible circulation of MRMM carrying M. genitalium strains in Cuban patients after 2015, which has direct implications for the effectiveness of syndromic management of STI.

The emergence of MRMM in M. genitalium in several countries is a worrying reality. For several years, this sexually transmitted pathogen has been underestimated as responsible for STI syndromes in men and women. In the Unites States Center for Diseases Control STI treatment guidelines the infection got attention with a separate chapter for the first time in 2015 [19]. In 2016 the European guideline on Mycoplasma genitalium infections was published as the first international guideline addressing diagnosis and treatment [20]. This was the first step to harmonize the international recognition and treatment of M. genitalium as an STI.

Since 2004, M. genitalium infections have been studied at IPK, and diagnosis was established in 2009 as part of an aetiological surveillance of STIs in Cuba. For many years, empiric treatment with macrolides and tetracyclines was used for the control of mycoplasma and chlamydia infections. In 2001, Cuba implemented syndromic management of STI, and the 1 g single dose azithromycin treatment became the recommended treatment for non-gonococcal urogenital infections in men and women. At this time, few studies were available that evaluated the treatment efficacy of macrolides and the emergence of resistance was an unknown problem. The latter, has become recognised as a worldwide problem after 2008 [2].

In the present investigation, we have determined the presence of MRMM in M. genitalium in Cuban clinical specimens archived since 2009, nine years after the initiation of azithromycin syndromic treatment of urogenital syndromes, and until 2016. From our data, it appears that M. genitalium with MRMM was very rare in Cuba before 2014. However, the number of analysed M. genitalium positive samples between 2009 and 2013 was relatively low, so it is difficult to make definite conclusions. Nevertheless, the explosion in the prevalence of M. genitalium with MRMM reaching 90% in 2016 is extremely worrying and threatening the current recommended syndromic management. In Cuba, more than 4000 cases of urogenital syndromes (lower abdominal pain, vaginal discharge and urethral syndrome) are reported to the Ministry of Public Health every year. However, it is not common that M. genitalium tests are ordered, even when macrolide treatment has failed. This finding is very similar in other countries and is partly due to lack of awareness among clinicians, but also due to lack of testing availability in many settings.

The genotyping assay used for MRMM detection was able to characterise more than 70% of the M. genitalium positive specimens, compared with 94.5% reported in the original paper by Kristiansen et al. [14]. However, when we analysed samples from 2016, 15 of the 17 non-amplified specimens were successfully re-amplified by the mgpB qPCR. These samples showed an M. genitalium load < 1000 geq/mL which may explain the failure of amplification in the genotyping assay. Other factors such as destruction of DNA, even when conserved at -80 °C, could be playing a role in the success of the genotyping assay, as has been shown for another M. genitalium molecular test [21]. Unfortunately, it was not possible to provide information about the exact positions of the MRMM as the method applied did not provide this level of discrimination. Most of the samples were used up for culture attempts and additional PCRs for MRMM was not possible. In Cuba, Sanger sequencing is difficult and expensive and consequently, a qPCR for the detection of MRMM directly from the clinical samples was introduced. Furthermore, as the type of mutation is not relevant for the clinical management, and as an increasing amount of surveillance data is produced by PCR based testing, even in more resourced settings, we do not consider this to be a major limitation of the study.

The three new Cuban M. genitalium strains isolated from samples with MRMM have an antimicrobial resistance profile with susceptibility to tetracyclines and moxifloxacin according to CLSI guidelines for M. pneumoniae [22]. The patient from whom the B25 strain was isolated is a Cuban man living in Switzerland, where he was infected, diagnosed and treated for M. genitalium. His Cuban male partner never lived outside of Cuba and was infected after unprotected sex during a visit of patient B25. Thus, introduction of the MRMM M. genitalium strain from Europe is the most plausible explanation, and confirmation of the same mgpB type (Cuba3) in clinical samples and MRMM strains, support this. Fortunately, both the B25 and B26 strains were susceptible to doxycycline and the patients were cured after combined treatment with doxycycline and levofloxacin for 14 days. The B30 female patient was a woman having had sex with foreign men and she was the sexual contact reported by the B19 male patient described in our previous study [10] at the time of M. genitalium diagnosis. The same mgpB type (type 4) was found in the clinical specimens and isolated strains B19 and B30, reaffirming the sexual transmission between the partners. However, it is impossible to determine if the MRMM strain was imported or not. Both patients were cured with levofloxacin plus doxycycline.

In this investigation, the overall macrolide resistance rate was 32% but with all of the resistant strains detected within the last three years of the study. The A2058G/A2059G was the most frequently detected MRMM genotype detected in M. genitalium positives specimens as in most other studies. Comparable rates of macrolide resistance have been reported recently from Denmark [4], United Kingdom [6], and Australia [8], where the prevalence of macrolide resistance–associated mutations was 40, 41 and 43%, respectively. These studies also found similar distributions of the most common A2058G and A2059G mutations but did not see the same dramatic increase in the prevalence of MRMM strains. The A2058C/T was detected in 19% of Cuban MRMM specimens. Unfortunately, neither of the isolated strains had this mutation pattern but no influence on the MICs values for macrolides have been recorded in other investigations for the different mutation-types [2, 3].

Based on the findings reported here, a new molecular diagnostic strategy for M. genitalium at IPK has been established since January 2017. In this strategy, we use the mgpB qPCR as screening test and positive specimens are immediately confirmed by the 5’nuclease genotyping assay and by the mgpA qPCR for mgpB positive samples with less than 1000 geq/mL. This approach is in accordance with the European guideline on M. genitalium infections [20] where determination of MRMM is recommended for all M. genitalium positive samples.

The prevalence of MRMM carrying M. genitalium strains has been high after 2014 in Cuba. Since 2015, all M. genitalium positive patients with MRMM seen at the STD clinic at the IPK Hospital have demonstrated treatment failure after azithromycin. In these 46 cases, MRMM M. genitalium was successfully eradicated with a 14-days second-line combined therapy of levofloxacin plus doxycycline (Orestes Blanco, unpublished). This regimen has been used due to the limited availability of moxifloxacin in Cuba and may be applicable elsewhere. The choice of this combination regimen was based on the synergistic in vitro activity of doxycycline and moxifloxacin in fluoroquinolone susceptible M. genitalium strains [23]. Experimental work documenting the synergy between levofloxacin and doxycycline is currently underway. The high efficacy was somewhat unexpected as doxycycline monotherapy has a microbiological cure rate of approximately 30% and that of levofloxacin monotherapy is rarely exceeding 50% [24].

No data about MRMM prevalence in other Latin-American countries are available, but it will probably be found at a similar rate in countries with similar macrolide treatment strategy for STIs as Cuba.

The results obtained in this investigation showed for the first time the rapid emergence and high prevalence of MRMM carrying M. genitalium strains in Cuban patients. No data about macrolide failure in syndromic STI management outside IPK are available, but macrolide resistance in M. genitalium-infected Cuban patients could be an underestimated health problem. We suggest that Cuban guidelines for STI treatment are modified to include testing for M. genitalium and detection of macrolide resistance–mediating mutations at least in patients failing syndromic treatment to ensure that in these cases the treatment will be guided by the aetiological diagnosis.