Mycoplasma genitalium is frequently associated with urogenital and rectal infections, with the number of cases of macrolide-resistant and quinolone-resistant M. genitalium infection continuing to increase. In this study, we examined the levels of resistance to these two common antibiotic treatments in geographically distinct locations in Queensland, Australia.
KEYWORDS: Australia, Mycoplasma genitalium, Queensland, antimicrobial resistance, azithromycin, macrolide, moxifloxacin, quinolone, sexually transmitted infection
ABSTRACT
Mycoplasma genitalium is frequently associated with urogenital and rectal infections, with the number of cases of macrolide-resistant and quinolone-resistant M. genitalium infection continuing to increase. In this study, we examined the levels of resistance to these two common antibiotic treatments in geographically distinct locations in Queensland, Australia. Samples were screened for macrolide resistance-associated mutations using a commercially available kit (ResistancePlus MG; SpeeDx), and quinolone resistance-associated mutations were identified by PCR and DNA sequencing. Comparisons between antibiotic resistance mutations and location/gender were performed. The levels of M. genitalium macrolide resistance were high across both locations (62%). Quinolone resistance mutations were found in ∼10% of all samples, with a number of samples harboring mutations conferring resistance to both macrolides and quinolones. Quinolone resistance was higher in southeast Queensland than in north Queensland, and this was consistent in both males and females (P = 0.007). The M. genitalium isolates in rectal swab samples from males harbored high levels of macrolide (75.9%) and quinolone (19%) resistance, with 15.5% harboring resistance to both classes of antibiotics. Overall, the lowest observed level of resistance was to quinolones in females from north Queensland (1.6%). These data highlight the high levels of antibiotic resistance in M. genitalium isolates within Queensland and the challenges faced by sexually transmitted infection clinicians in managing these infections. The data do, however, show that the levels of antibiotic resistance may differ between populations within the same state, which has implications for clinical management and treatment guidelines. These findings also support the need for ongoing antibiotic resistance surveillance and tailored treatment.
INTRODUCTION
Mycoplasma genitalium was first identified in the urogenital tract of men in 1981 (1) and has since been confirmed to be a sexually transmissible bacterium that is responsible for a range of sequelae. M. genitalium has been associated with acute and chronic urethritis in men and can cause urethritis, cervicitis, and pelvic inflammatory disease in women (2). Without appropriate treatment, M. genitalium infections are often chronic, with approximately 25% of infections persisting for >12 months, and some infections in women have been shown to persist for up to 2 years (3).
More recently, M. genitalium has gained widespread attention due to its increasing resistance to antibiotic treatments and as a consequence is now listed as a major emerging issue in the sexually transmitted infection (STI) treatment guidelines published by the United States Centers for Disease Control and Prevention (4). Unlike other antibiotic-resistant STI pathogens, like Neisseria gonorrhoeae, in which antibiotic resistance increased somewhat steadily over time, the emergence and spread of antibiotic-resistant M. genitalium appear to have occurred relatively quickly and have prompted recent changes in the Australian and UK treatment guidelines. All these guidelines now recommend a 7-day course of doxycycline as empirical therapy for M. genitalium-associated syndromes and the use of a combined diagnostic-resistance assay, where available, to select an antibiotic to which the M. genitalium isolate is likely to be susceptible (5). This resistance-guided approach involves the use of azithromycin for macrolide-susceptible infections or a quinolone, such as moxifloxacin or sitafloxacin, for macrolide-resistant infections to optimize cure (6, 7). However, there are still worrying trends in the levels of macrolide- and quinolone-resistant strains of M. genitalium that continue to threaten the efficacy of these treatment approaches.
Macrolide resistance was first reported in M. genitalium in 2008 (8), and since that time the efficacy of these antibiotic treatments has diminished, with some studies reporting treatment success rates as low as 40% (4). The efficacy of moxifloxacin has also begun to decline, with recent meta-analyses showing a decrease in cure rates from 100% in studies conducted prior to 2010 to 89% in studies published after 2010 (9). The mechanisms that underpin M. genitalium resistance to macrolides are well documented, enabling molecular methods to be developed to readily characterize the nucleotide mutations directly within M. genitalium-positive clinical samples. For macrolide resistance, five single nucleotide mutations at positions 2058 and 2059 within the macrolide resistance-determining region (MRDR) of the 23S rRNA gene are each independently associated with the failure of azithromycin (8). While resistance to quinolones has been associated with various mutations in the quinolone resistance-determining regions (QRDR) of the topoisomerase gene, parC, the contribution of specific parC mutations to treatment failure is still being determined. Similarly, the role of nucleotide mutations within the QRDR of the DNA gyrase gene, gyrA, is somewhat unclear with respect to clinical treatment failure (10–14).
In Australia, there have now been several studies examining M. genitalium antimicrobial resistance. In brief, these studies have shown that the current levels of macrolide resistance exceed 50% in the urban centers of Melbourne, Sydney, and Brisbane (15–17) and 40% in backpacker populations in northern Queensland (18), with sexual orientation being an important determinant of resistance, due to higher rates of antibiotic resistance in men who have sex with men (MSM) (19). Quinolone resistance, previously rare in M. genitalium in Australia, was reported to be as high as 13.6% in a study from Melbourne in 2012 and 2013 (14, 20). In the current study, we further investigated and compared the rates of both macrolide and quinolone antibiotic resistance among M. genitalium isolates obtained from two distinct regions within Queensland: southeast Queensland (SEQ; which includes the cities of Brisbane [the capital of Queensland] and the Gold Coast) and north Queensland (NQ; including the city of Townsville and surrounding regional and remote settings). The levels of resistance were further compared between males and females.
MATERIALS AND METHODS
Sample population.
This study was approved by the Children’s Health Queensland Human Research Ethics Committee (HREC/12/QRCH/139). This retrospective study was performed on samples submitted to Pathology Queensland, Australia, for the detection of M. genitalium by PCR between 2013 and 2017. The M. genitalium PCR utilized by Pathology Queensland is an in-house PCR method targeting the M. genitalium MgPa adhesin gene (15). Extracted DNA from M. genitalium-positive samples was sent to The University of Queensland Centre for Clinical Research for screening for the presence of antibiotic resistance markers. The patients primarily comprised those presenting to sexual health clinics with genital symptoms (e.g., urethritis, cervicitis, pelvic inflammatory disease, and proctitis) for whom testing for M. genitalium was indicated. Screening for M. genitalium is not currently recommended in the Australian guidelines. The samples used within this study included urine (n = 280), cervicovaginal swabs (n = 90), urethral swabs (n = 10), anal/rectal swabs (n = 60), throat swabs (n = 1), and samples from unknown sites (n = 6).
Antibiotic resistance screening.
Mutations (A2058G, A2058C, A2058T, A2059G, and A2059C) within the macrolide resistance-determining region (MRDR) of the 23S rRNA gene were detected using a commercially available kit (the ResistancePlus MG kit; SpeeDx, Sydney, Australia) (21–23). The kit is TGA approved and CE-IVD marked; however, it is not currently for sale in the United States.
Mutations within the QRDR of the gyrA and parC genes were detected by PCR, using previously designed primers (14), and Sanger sequencing. Briefly, a Qiagen Quantitect SYBR kit (Qiagen, Australia) was used as the basis for the PCR mix, and PCR products were submitted to the Australian Genome Research Facility (AGRF; Brisbane, Australia) for DNA sequencing. Alterations to the nucleotide sequences within the QRDR were compared to mutations that have been previously reported to be associated with clinical treatment failure or elevated MIC values for quinolones in M. genitalium or other Mycoplasma spp. and Ureaplasma spp. (10–14).
Statistics.
Comparisons of categorical variables (e.g., gender, location) across groups were performed using Pearson’s chi-square test. Statistical significance was accepted as a P value of <0.05.
RESULTS
Sample population.
A total of 524 M. genitalium-positive samples were received from Pathology Queensland. Of these, 15 samples resulted in discordant results where we were unable to confirm the presence of M. genitalium, and these were removed from our final data set. We also identified 62 samples that were considered to be either a sample used for test of cure (a sample collected from a patient within 3 months of his or her initial presentation) or a sample from a patient sampled at multiple anatomical sites. These 62 samples were excluded from the primary analysis to determine the proportion of samples with macrolide and quinolone resistance, leaving 447 samples in the primary analysis. Of these additional samples, 50 were included in an additional assessment investigating the temporal changes in antibiotic resistance within individual patients (n = 50), with the remaining 12 samples being excluded from this secondary analysis, as they were collected from multiple anatomical sites of the same patient.
These 447 samples included 14 samples from 2013, 61 samples from 2016, and 372 samples from 2017 and represented samples from 176 females and 269 males; 2 individuals did not disclose their gender. Two hundred nine samples originated from the SEQ region, while 238 M. genitalium samples originated from NQ. Of these, 26 samples did not have complete gyrA sequencing performed due to insufficient DNA; however, these remained within our primary analysis, as mutations in this region are currently of uncertain clinical significance.
MRDR and QRDR mutations identified in samples.
As the ResistancePlus MG assay was used to detect MRDR mutations and we did not perform any DNA sequencing of the 23S rRNA gene, specific MRDR mutations were unable to be reported. However, sequencing of the QRDR of parC and gyrA was undertaken, and specific quinolone resistance-associated amino acid changes were able to be reported. We observed mutations within the topoisomerase gene, parC, and only two mutations of uncertain clinical significance were observed within the DNA gyrase gene, gyrA. In summarizing the most common amino acid mutations observed within the study, we grouped those mutations that were considered likely to be of clinical significance based on previous publications that correlated quinolone resistance mutations with treatment outcomes (20) and found that the parC mutations S83I, D87Y, and D87N were the most frequently observed mutations of clinical significance, and this was consistent across both SEQ and NQ (Table 1). The two gyrA mutations observed within this study were considered to be of uncertain clinical significance, as these were codetected in M. genitalium isolates which also harbored S83I parC mutations, which are known to be associated with treatment failure (24).
TABLE 1.
Amino acid changes in parC and gyrA genes considered to be associated with moxifloxacin failurea
| Location and sex | parC amino acid changes of likely clinical significanceb | parC amino acid changes of uncertain clinical significancec | gyrAd amino acid changes of uncertain clinical significancec |
|---|---|---|---|
| Southeast Queensland | |||
| Male (n = 26) | Ser → Ile 83 (n = 19), Asp → Asn 87 (n = 3), Asp → Tyr 87 (n = 3), Ser → Arg 83 (n = 1) | Gly → Cys 93 (n = 2) | |
| Female (n = 11) | Ser → Ile 83 (n = 6), Asp → Asn 87 (n = 2), Asp → Tyr 87 (n = 1) | Ser → Asn 83 (n = 1), Asp → His 87 (n = 1) | |
| Northern Queensland | |||
| Male (n = 6) | Ser → Ile 83 (n = 4), Asp → Tyr 87 (n = 2) | ||
| Female (n = 2) | Ser → Ile 83 (n = 1), Ser → Arg 83 (n = 1) |
Amino acid position changes are reported according to the amino acid positions within the Mycoplasma genitalium G37 genome.
Clinical evidence from published data suggests that these mutations may be associated with treatment failure and/or elevated MICs of antibiotics.
The clinical significance of these mutations with respect to treatment failure and/or elevated MIC data from published data is uncertain.
No known gyrA mutations that were of known clinical significance were observed.
Proportion of samples with MRDR and QRDR mutations.
Among the 447 patient samples, 277/447 (62.0%) carried strains which harbored MRDR, while a total of 47/447 (10.5%) samples harbored M. genitalium strains with parC or gyrA mutations in their QRDR. A total of 7.8% (35/447) of patient samples harbored both MRDR and QRDR mutations (Table 2), herein referred to as dual-class resistance. There was no evidence that the levels of MRDR or QRDR mutations changed over the study period (data not shown).
TABLE 2.
Levels of MRDR mutations and QRDR mutations associated with antibiotic resistance by region and gender in Queensland, Australia
| Region and gender | Macrolide resistance mutations |
Quinolone resistance mutations |
Dual resistance mutations |
|||
|---|---|---|---|---|---|---|
| No. (%) of patients | 95% CI | No. (%) of patients | 95% CI | No. (%) of patients | 95% CI | |
| Southeast Queensland (n = 209) | 136 (65.1) | 58–72 | 39 (18.7) | 13.6–24.6 | 28 (13.4) | 9.1–18.8 |
| Male (n = 159) | 109 (68.5) | 61–76 | 28 (17.6) | 12–24 | 22 (13.8) | 8.9–20.2 |
| Female (n = 50) | 27 (54.0) | 39–68 | 11 (22) | 12–36 | 6 (12) | 4.5–24.3 |
| Northern Queensland (n = 238) | 141 (59.2) | 53–66 | 8 (3.4) | 2–7 | 7 (3.0) | 1.2–6.0 |
| Male (n = 110) | 68 (61.8) | 52–71 | 6 (5.5) | 2–12 | 5 (4.5) | 1.5–10.3 |
| Female (n = 126) | 71 (56.3) | 47–65 | 2 (1.6) | 1–6 | 2 (1.6) | 2–5.6 |
| Undisclosed (n = 2) | 2 (100.0) | 16–100 | 0 (0.0) | 0–84 | 0 (0.0) | 0–84.2 |
| Total (n = 447) | 277 (62.0) | 57–67 | 47 (10.5) | 7.8–13.7 | 35 (7.8) | 5.5–10.7 |
| Male (n = 269) | 177 (65.8) | 60–72 | 34 (12.6) | 8–16 | 27 (10.0) | 6.1–13.4 |
| Female (n = 176) | 98 (55.7) | 48–63 | 13 (7.4) | 4–12 | 8 (4.5) | 2–8.8 |
| Undisclosed (n = 2) | 2 (100.0) | 16–100 | 0 (0.0) | 0–84 | 0 (0.0) | 0–84.2 |
Regional differences in the levels of antibiotic resistance.
Both the SEQ and the NQ regions of Queensland had similar proportions of samples with MRDR mutations (136/209 [65.1%] and 141/238 [59.2%], respectively; P = 0.21). However, the levels of parC and gyrA mutations were significantly different between the two regions: SEQ M. genitalium samples were significantly more likely to harbor mutations associated with quinolone resistance (39/209, 18.7%) than NQ M. genitalium samples (8/238, 3.4%) (P < 0.001; Table 2). Similarly, the proportion of M. genitalium samples harboring mutations associated with dual-class resistance differed, with SEQ having a significantly higher (28/209, 13.4%) rate of M. genitalium isolates with dual-class resistance than NQ (7/238, 3.0%) (Table 2) (P = 0.0001).
Differences in the levels of antibiotic resistance according to gender.
Of the 176 samples from females within the study, 98 (55.7%) had MRDR mutations and 13 (7.4%) had parC mutations. Of these women, only 8 (4.5%) harbored M. genitalium that had dual-class resistance (Table 2). Macrolide resistance mutations were significantly higher in men than in women (P = 0.03), and this was consistent between SEQ and NQ (Table 2). There was no significant difference in the frequency of QRDR mutations between men (34/269, 12.6%) and women (13/176, 7.4%) (Table 2), and this was again consistent between SEQ and NQ. Men and women from SEQ were more likely to harbor parC or gyrA mutations (men, 28/159 [17.6%]; women, 11/50 [22%]) than men and women from NQ (men, 6/110 [5.5%]; women, 2/126 [1.6%]) (P = 0.007). Women from SEQ were also significantly more likely than women from NQ to harbor QRDR mutations (11/50 [22%] versus 2/126 [1.6%], respectively; P < 0.0001). Overall, the lowest observed level of resistance was for QRDR among females in NQ, at 1.6% (Table 2).
Table 3 provides a summary of the sample types and associated levels of resistance. While sexual orientation data for men and women were not available for this study and we were unable to determine the true proportion of male samples that were from MSM, rectal swab samples from males are likely from MSM since this sample type is rarely, if ever, collected from heterosexual men. Urine samples within this study, however, likely represent a mixture of samples from both MSM and heterosexual individuals. When comparing genitourinary samples from females to rectal swab samples from males (a proxy for samples from MSM), there were differences in the proportion of MRDR mutations from male rectal swab samples (44/58, 75.9%) and female genitourinary samples (93/167, 55.7%) (P = 0.007), and we also observed significant differences in the levels of dual-class resistance between male rectal swab samples (9/58, 15.5%) and female genitourinary samples (7/167, 4.2%) (P = 0.03). Sample types with less than 3 specimens were not included in these comparisons.
TABLE 3.
Antibiotic resistance according to sample site and gender
|
Sample site and gender |
No. of samples with the indicated mutation/total no. of specimens (%) |
||
|---|---|---|---|
| Macrolide resistance mutations | Quinolone resistance mutations | Dual resistance mutations | |
| Male urine/urethral swab samplesa | 131/205 (63.9) | 23/205 (11.2) | 18/205 (8.8) |
| Female urine/cervicovaginal swab samplesa | 93/167 (55.7) | 12/167 (7.2) | 7/167 (4.2) |
| Anal/rectal swab samples | |||
| Male (n = 58) | 44/58 (75.9) | 11/58 (19.0) | 9/58 (15.5) |
| Female (n = 2) | 1/2 (50.0) | 1/2 (50.0) | 1/2 (50.0) |
| Total | 45/60 (75.0) | 12/60 (20.0) | 10/60 (16.7) |
| Throat swab samples | |||
| Male (n = 1) | 0/1 (0.0) | 0/1 (0.0) | 0/1 (0.0) |
| Total | 0/1 (0.0) | 0/1 (0.0) | 0/1 (0.0) |
| Unknown sampling site | |||
| Male (n = 4) | 2/4 (50.0) | 0/4 (0.0) | 0/4 (0.0) |
| Female (n = 8) | 4/8 (50.0) | 0/8 (0.0) | 0/8 (0.0) |
| Total | 6/12 (50.0) | 0/12 (0.0) | 0/12 (0.0) |
Two urine samples within this study were from patients who chose not to disclose their gender. Data for these were not included within the table, but both urine samples harbored macrolide resistance mutations but no quinolone resistance mutations.
Urine/urethral swab samples from males and urine/cervicovaginal swab samples from women harbored similar levels of macrolide resistance (males, 177/269 [65.8%]; females, 98/176 [55.7%]; P = 0.1) and quinolone resistance (males, 34/269 [12.6%]; females, 13/176 [7.4%]; P = 0.1); however, there was a significant difference in dual-class resistance according to gender (males, 27/269 [10%]; females, 8/176 [4.5%]; P = 0.04). Specimen numbers for the remaining sample types were too low to make meaningful comparisons in relation to gender (Table 3).
Patients sampled over time had changes in their M. genitalium antibiotic resistance profile.
The results for the 50 patients sampled on multiple occasions are summarized in Table 4. For 21/50 (42%) patients, the presence of antibiotic-resistant M. genitalium appeared to persist following suspected treatment of the infection. The resistance profiles of isolates from male and female urine samples did not significantly differ from one another; however, male rectal swab samples were more likely to have a persistent antibiotic resistance profile (P = 0.017), in keeping with our previous finding that male rectal swab samples harbor high levels of antibiotic resistance. For 12/50 (24%) patients, the emergence of antibiotic-resistant strains appeared following antibiotic treatment, which may be consistent with de novo antibiotic resistance, where the strains in the baseline samples were susceptible to the antibiotic (Table 4, posttreatment), but may also be associated with a new infection with M. genitalium strains with different susceptibility patterns. For example, patient 2 had no evidence of MRDR mutations prior to antibiotic treatment, while MRDR mutations were observed 22 days after treatment. Likewise, patient 10 had MRDR mutations at both the pre- and posttreatment sample collection times but no QRDR mutations; however, at the posttreatment sample collection time, an S83I quinolone antibiotic resistance mutation was evident. In 28% (14/50) of patients, we observed a potential loss of antibiotic resistance within the patient samples between the pre- and posttreatment sample collection points; however, for 2 of these (patients 9 and 35), there was variation in resistance over time. For both of these patients, there was a lack of MRDR mutations observed in sample 2 (posttreatment); however, at the time that a third sample was collected, MRDR mutations were again evident. It is important to note that due to the lack of clinical data, this change in resistance profile may also indicate reinfection of the patient with a susceptible strain of M. genitalium. In 3/50 (6%) patients, no change in the antibiotic resistance profiles was observed (Table 4).
TABLE 4.
Investigation of antibiotic resistance over time in 50 patients with multiple samples submitted to Pathology Queensland
| Retested patient no. | Sample region | Specimen site (gendera ) | Sample no. | No. of days between sample collections | Macrolide resistance | Quinolone resistanceb | Dual resistance | Resistance profilec |
|---|---|---|---|---|---|---|---|---|
| 1 | SEQ | Urine (M) | 1 | 7 | No | Yes (D87N) | No | Persistence |
| 2 | No | Yes (D87N) | No | |||||
| 2 | SEQ | Urine (M) | 1 | 22 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 3 | SEQ | Urine (M) | 1 | 31 | Yes | Yes (S83I) | Yes | Persistence |
| 2 | Yes | Yes (S83I) | Yes | |||||
| 4 | SEQ | Urine (M) | 1 | 127 | Yes | Yes (D87N) | Yes | Loss |
| 2 | Yes | No | No | |||||
| 5 | SEQ | Urine (M) | 1 | 58 | Yes | Yes (S83I) | Yes | Loss |
| 2 | Yes | No | No | |||||
| 6 | SEQ | Urine (M) | 1 | 35 | Yes | No | No | Posttreatment |
| 2 | Yes | Yes (D87N) | Yes | |||||
| 7 | SEQ | Urine (M) | 1 | 28 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 8 | SEQ | Rectal (M) | 1 | 50 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 9 | SEQ | Urine (M) | 1 | 55 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 3 | 16 | Yes | No | No | ||||
| 10 | SEQ | Rectal (M) | 1 | 176 | Yes | No | No | Posttreatment |
| 2 | Yes | Yes (S83I) | Yes | |||||
| 11 | SEQ | Rectal (M) | 1 | 16 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 12 | SEQ | Urine (M) | 1 | 37 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 13 | SEQ | Urine (M) | 1 | 18 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 14 | SEQ | Rectal (M) | 1 | 149 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 15 | SEQ | Urine (M) | 1 | 35 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 16 | SEQ | Rectal (M) | 1 | 29 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 17 | SEQ | Urine (M) | 1 | 48 | Yes | No | No | Posttreatment dual resistance |
| 2 | Yes | Yes (S83I) | Yes | |||||
| 18 | SEQ | Urine (M) | 1 | 25 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 19 | SEQ | Urine (M) | 1 | 36 | Yes | Yes (S83I) | Yes | Loss |
| 2 | Yes | No | No | |||||
| 20 | SEQ | Urine (M) | 1 | 29 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 21 | SEQ | Urine (M) | 1 | 19 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 22 | SEQ | Urine (M) | 1 | 25 | No | No | No | No change |
| 2 | No | No | No | |||||
| 23 | SEQ | Urine (M) | 1 | 37 | No | No | No | No change |
| 2 | No | No | No | |||||
| 24 | SEQ | Urine (M) | 1 | 113 | Yes | Yes (S83I) | Yes | Loss |
| 2 | Yes | No | No | |||||
| 25 | SEQ | Cervix (F) | 1 | 31 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 26 | SEQ | Urine (M) | 1 | 169 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 27 | SEQ | Rectal (M) | 1 | 36 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 28 | SEQ | Unspecified site (M) | 1 | 43 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 29 | SEQ | Vaginal (F) | 1 | 40 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 30 | NQ | Vaginal (F) | 1 | 127 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 31 | NQ | Urine (M) | 1 | 80 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 32 | NQ | Rectal (M) | 1 | 98 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 33 | NQ | Urine (F) | 1 | 124 | No | No | No | No change |
| 2 | No | No | No | |||||
| 34 | NQ | Urine (F) | 1 | 34 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 3 | 50 | Yes | No | No | ||||
| 35 | NQ | Urine (M) | 1 | 54 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 3 | 43 | Yes | No | No | ||||
| 36 | NQ | Vaginal (F) | 1 | 103 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 37 | NQ | Rectal (M) | 1 | 82 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 38 | NQ | Urine (M) | 1 | 113 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 39 | NQ | Urine (M) | 1 | 26 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 40 | NQ | Urine (unkd ) | 1 | 37 | Yes | No | No | Loss |
| 2 | No | No | No | |||||
| 41 | NQ | Urine (M) | 1 | 36 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 42 | NQ | Urine (M) | 1 | 9 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 43 | NQ | Rectal (M) | 1 | 28 | Yes | Yes (D87N) | Yes | Persistence |
| 2 | Yes | Yes (D87N) | Yes | |||||
| 44 | NQ | Urine (M) | 1 | 41 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 45 | NQ | Rectal (M) | 1 | 45 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 46 | NQ | Urine (F) | 1 | 42 | No | No | No | Posttreatment |
| 2 | Yes | No | No | |||||
| 47 | NQ | Rectal (M) | 1 | 54 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 3 | 43 | Yes | No | No | ||||
| 48 | NQ | Urine (unkd ) | 1 | 39 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 49 | NQ | Urine (M) | 1 | 15 | Yes | No | No | Persistence |
| 2 | Yes | No | No | |||||
| 50 | NQ | Rectal (M) | 1 | 29 | Yes | No | No | Persistence |
| 2 | Yes | No | No |
F, female; M, male.
Quinolone resistance-associated mutations within the parC topoisomerase gene, per M. genitalium G37 amino acid numbering.
Changes in resistance profiles from patient samples were defined as follows: persistence, the resistance mutation was persistently detected but in the absence of accompanying clinical data may reflect treatment failure, a lack of treatment, or reinfection; posttreatment, the appearance of posttreatment (de novo) resistance mutations when the baseline sample was wild type (no resistance mutation); loss, a loss of detectable resistance, in which the resistance was detected in the baseline sample and the follow-up sample was wild type; in the absence of clinical data, this may reflect reinfection with a susceptible strain or an inability of the assay to detect a resistance mutation in a low-load MRDR infection; no change, no changes in antibiotic resistance mutations were observed.
unk, unknown (these patients chose not to disclose their gender).
DISCUSSION
Antibiotic resistance in M. genitalium has become a significant problem, impacting the successful treatment of these infections, which is epitomized by the fact that treatment failures following recommended therapies are now commonplace (4). This is compounded due to the paucity of available and effective therapies and a lack of alternative antibiotic treatment choices for M. genitalium. Here, we present further evidence of the high rates of macrolide and quinolone resistance and the emergence of dual-class resistance in M. genitalium isolates obtained from Queensland, Australia, and show some regional and gender differences in resistance that likely reflect differences in ethnicity and sexual orientation, which have significant implications for clinical care.
Consistent with the findings of our previous pilot study conducted in Queensland (17), the rate of macrolide resistance was high, with 62% of all samples harboring mutations associated with macrolide resistance. This level of macrolide-resistant M. genitalium is consistent with other estimates in urban centers elsewhere, including Melbourne (16) and Sydney (25) and urban centers in New Zealand (26), Japan (24, 27, 28), and the United States (29), and these MRDR mutations have previously been shown to be associated with elevated MIC values for macrolide antibiotics (8, 30) and the failure of clinical treatment with macrolides (31, 32). The level of quinolone resistance was approximately 10% across both SEQ and NQ and is similar to the rates presented in recent reports from the Asia-Pacific region; for example, 10.7% of cases harbored QRDR mutations in Japan (28) and 13.5% of cases harbored QRDR mutations in Melbourne (20). Of note, we observed that 7.8% of M. genitalium isolates harbored both MRDR and QRDR mutations, consistent with resistance to both classes of antibiotics, which is also in line with the findings of a recent study conducted in Melbourne that reported that dual-class resistance was observed in 8.6% of specimens (20).
These data highlight the very real challenges faced by our STI clinicians in managing M. genitalium infections and help explain why they continue to experience treatment failures. These data also reinforce the need for using combined diagnostic-resistance assays to directly inform patient treatment rather than relying on empirical treatment. In a recent study conducted in Melbourne, Read et al. (7) were able to achieve cure rates of >92% (compared to macrolide treatment failure rates of 39% in a previous study [19]) when patients were tested for the presence of macrolide-resistant M. genitalium at the time of presentation. The patients were treated with a 7-day course of doxycycline, followed by either azithromycin or pristinamycin, depending on the results of the mutation screening. The implementation of resistance testing-guided therapy has recently been shown to have considerable potential to improve the first-line cure of M. genitalium infections, and based on the results of our and other studies, a method that can rapidly detect QRDR mutations of known clinical significance may also be of considerable benefit to further enhance this approach.
Notwithstanding the information presented above, our data also suggest that empirical treatment may still be viable in certain populations, so long as adequate resistance surveillance data are available to directly inform local treatment practices and guidelines. While there was no significant difference in the rates of macrolide resistance between SEQ (65.1%) and NQ (59.2%), there were much lower quinolone resistance levels in NQ (3.4%) than in SEQ (18.7%). Notably, the rate of quinolone resistance was only 1.6% in females in NQ, suggesting that quinolones may be highly efficacious among heterosexuals in NQ. Further investigations are required to explore why quinolone resistance was significantly lower among females in NQ than among females in SEQ (22%), but it is likely that the differences are attributable to differences in the populations, with the NQ region also including many indigenous populations. Differences in the antibiotic susceptibility of gonococci between indigenous and nonindigenous Australians are well documented, with gonococci in many indigenous communities remaining susceptible to antibiotics, such as penicillin and ciprofloxacin, that are no longer recommended for use elsewhere due to widespread resistance (33). Unfortunately, we did not have access to information on patient indigenous status to allow such comparisons to be made for M. genitalium.
A further limitation of this study was the fact that we were unable to accurately correlate sexuality with our antibiotic resistance data. Studies elsewhere have shown an association between high rates of macrolide-resistant M. genitalium and MSM (29, 34), with high rates of dual resistance to both macrolides and quinolones being seen among MSM populations (29). It is, however, likely that many of the samples obtained from males within our study came from MSM, and this is supported by the fact that we observed higher levels of macrolide resistance and dual-class resistance in rectal swab samples from males (a sample type which represents MSM) than in genitourinary samples from females (more likely to represent heterosexual individuals), and this difference was statistically significant.
Changes in resistance among patient samples were also observed over time. In the absence of clinical data, these samples may have been collected from individuals prior to antibiotic treatment and then following antibiotic treatment as a test of cure, but they may also have come from individuals reinfected with a new strain of M. genitalium. We observed that a persistent antimicrobial resistance genotype was more commonly found in male rectal swab samples than in samples from other anatomical sites, which also supports our previous finding that male rectal swab samples harbor isolates with high levels of antimicrobial resistance mutations. Future studies of interest would include pairing of molecular antimicrobial resistance typing alongside clinical data, as well as the use of molecular strain typing, in order to investigate the likelihood of reinfection and the generation of resistance during antimicrobial treatments.
In summary, we have identified high levels of resistance to both macrolides and quinolones, as well as dual-class resistance, among cases of M. genitalium infection in Queensland. The data highlight the need for ongoing M. genitalium resistance surveillance as well as the importance of using molecular assays to tailor the treatment of patients infected with M. genitalium.
ACKNOWLEDGMENTS
We thank Cameron Buckley and Amanda Bordin for their assistance with this work.
D.M.W. reports research funding from SpeeDx Pty Ltd. The ResistancePlus MG kits were provided by SpeeDx Pty Ltd.
This work was supported by The University of Queensland strategic funding, as well as research funding from SpeeDx Pty Ltd.
Note that SpeeDx had no role in the design of this study.
E.L.S. contributed to the design of the study and the acquisition, analysis, and interpretation of data and drafted the manuscript. E.T. assisted in the acquisition and analysis of data. C.B. assisted in the recruitment of patient samples and reviewed and provided feedback on the manuscript. C.S.B. provided critical feedback and clinical perspectives on the data within the paper and assisted in drafting the manuscript. A.M. provided significant input into the clinical perspectives of M. genitalium infections and provided feedback on the manuscript. F.F. assisted in sample collection and reviewed the final version of the manuscript. J.L.-L. provided clinical insights into M. genitalium infections and reviewed/critically evaluated the manuscript. G.R.N. was instrumental in the provision of patient samples and critically reviewed and assisted in the drafting of the manuscript. D.M.W. contributed to the design of the study and analysis and interpretation of the data and critically reviewed and assisted in the drafting of the manuscript. All authors approved the final submitted manuscript.
REFERENCES
- 1.Tully JG, Taylor-Robinson D, Cole RM, Rose DL. 1981. A newly discovered mycoplasma in the human urogenital tract. Lancet i:1288–1291. doi: 10.1016/S0140-6736(81)92461-2. [DOI] [PubMed] [Google Scholar]
- 2.Taylor-Robinson D, Jensen JS. 2011. Mycoplasma genitalium: from Chrysalis to multicolored butterfly. Clin Microbiol Rev 24:498–514. doi: 10.1128/CMR.00006-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cohen CR, Nosek M, Meier A, Astete SG, Iverson-Cabral S, Mugo NR, Totten PA. 2007. Mycoplasma genitalium infection and persistence in a cohort of female sex workers in Nairobi, Kenya. Sex Transm Dis 34:274–279. [DOI] [PubMed] [Google Scholar]
- 4.Workowski KA, Bolan GA, Centers for Disease Control and Prevention. 2015. Sexually transmitted diseases treatment guidelines, 2015. MMWR Recomm Rep 64(RR-03):1–137. [PMC free article] [PubMed] [Google Scholar]
- 5.Soni S, Horner P, Rayment M, Pinto-Sander N, Naous N, Parkhouse A, Bancroft D, Patterson C, Fifer H. 2018. BASHH UK national guideline for the management of infection with Mycoplasma genitalium. https://www.bashhguidelines.org/media/1182/bashh-mgen-guideline-2018_draft-for-consultation.pdf. [DOI] [PubMed]
- 6.Australian Government Department of Health. 2018. Australian STI management guidelines for use in primary care. Mycoplasma genitalium. Australian Government Department of Health, Canberra, Australia. [Google Scholar]
- 7.Read TRH, Fairley CK, Murray GL, Jensen JS, Danielewski J, Worthington K, Doyle M, Mokany E, Tan L, Chow EPF, Garland SM, Bradshaw CS. 5 June 2018. Outcomes of resistance-guided sequential treatment of Mycoplasma genitalium infections: a prospective evaluation. Clin Infect Dis doi: 10.1093/cid/ciy477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jensen JS, Bradshaw CS, Tabrizi SN, Fairley CK, Hamasuna R. 2008. Azithromycin treatment failure in Mycoplasma genitalium-positive patients with nongonococcal urethritis is associated with induced macrolide resistance. Clin Infect Dis 47:1546–1553. doi: 10.1086/593188. [DOI] [PubMed] [Google Scholar]
- 9.Li Y, Le WJ, Li S, Cao YP, Su XH. 2017. Meta-analysis of the efficacy of moxifloxacin in treating Mycoplasma genitalium infection. Int J STD AIDS 28:1106–1114. doi: 10.1177/0956462416688562. [DOI] [PubMed] [Google Scholar]
- 10.Bebear CM, Renaudin J, Charron A, Renaudin H, de Barbeyrac B, Schaeverbeke T, Bebear C. 1999. Mutations in the gyrA, parC, and parE genes associated with fluoroquinolone resistance in clinical isolates of Mycoplasma hominis. Antimicrob Agents Chemother 43:954–956. doi: 10.1128/AAC.43.4.954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Beeton ML, Chalker VJ, Kotecha S, Spiller OB. 2009. Comparison of full gyrA, gyrB, parC and parE gene sequences between all Ureaplasma parvum and Ureaplasma urealyticum serovars to separate true fluoroquinolone antibiotic resistance mutations from non-resistance polymorphism. J Antimicrob Chemother 64:529–538. doi: 10.1093/jac/dkp218. [DOI] [PubMed] [Google Scholar]
- 12.Deguchi T, Maeda S, Tamaki M, Yoshida T, Ishiko H, Ito M, Yokoi S, Takahashi Y, Ishihara S. 2001. Analysis of the gyrA and parC genes of Mycoplasma genitalium detected in first-pass urine of men with non-gonococcal urethritis before and after fluoroquinolone treatment. J Antimicrob Chemother 48:742–744. doi: 10.1093/jac/48.5.742. [DOI] [PubMed] [Google Scholar]
- 13.Gruson D, Pereyre S, Renaudin H, Charron A, Bebear C, Bebear CM. 2005. In vitro development of resistance to six and four fluoroquinolones in Mycoplasma pneumoniae and Mycoplasma hominis, respectively. Antimicrob Agents Chemother 49:1190–1193. doi: 10.1128/AAC.49.3.1190-1193.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tagg KA, Jeoffreys NJ, Couldwell DL, Donald JA, Gilbert GL. 2013. Fluoroquinolone and macrolide resistance-associated mutations in Mycoplasma genitalium. J Clin Microbiol 51:2245–2249. doi: 10.1128/JCM.00495-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Trembizki E, Buckley C, Bletchly C, Nimmo GR, Whiley DM. 2017. High levels of macrolide-resistant Mycoplasma genitalium in Queensland, Australia. J Med Microbiol 66:1451–1453. doi: 10.1099/jmm.0.000584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Read TR, Fairley CK, Tabrizi SN, Bissessor M, Vodstrcil L, Chow EP, Grant M, Danielewski J, Garland SM, Hocking JS, Chen MY, Bradshaw CS. 2017. Azithromycin 1.5g over 5 days compared to 1g single dose in urethral Mycoplasma genitalium: impact on treatment outcome and resistance. Clin Infect Dis 64:250–256. doi: 10.1093/cid/ciw719. [DOI] [PubMed] [Google Scholar]
- 17.Read T, Bradshaw C. 2016. Managing Mycoplasma genitalium infections during a rapid upsurge in antibiotic resistance. MLO Med Lab Obs 48:28, 30. [PubMed] [Google Scholar]
- 18.Trevis T, Gosse M, Santarossa N, Tabrizi S, Russell D, McBride WJ. 2018. Mycoplasma genitalium in the far north Queensland backpacker population: an observational study of prevalence and azithromycin resistance. PLoS One 13:e0202428. doi: 10.1371/journal.pone.0202428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bissessor M, Tabrizi SN, Twin J, Abdo H, Fairley CK, Chen MY, Vodstrcil LA, Jensen JS, Hocking JS, Garland SM, Bradshaw CS. 2015. Macrolide resistance and azithromycin failure in a Mycoplasma genitalium-infected cohort and response of azithromycin failures to alternative antibiotic regimens. Clin Infect Dis 60:1228–1236. doi: 10.1093/cid/ciu1162. [DOI] [PubMed] [Google Scholar]
- 20.Murray GL, Bradshaw CS, Bissessor M, Danielewski J, Garland SM, Jensen JS, Fairley CK, Tabrizi SN. 2017. Increasing macrolide and fluoroquinolone resistance in Mycoplasma genitalium. Emerg Infect Dis 23:809–812. doi: 10.3201/eid2305.161745. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pitt R, Cole MJ, Fifer H, Woodford N. 2017. Evaluation of the Mycoplasma genitalium Resistance Plus kit for the detection of M. genitalium and mutations associated with macrolide resistance. Sex Transm Infect 94:565–567. doi: 10.1136/sextrans-2017-053366. [DOI] [PubMed] [Google Scholar]
- 22.Le Roy C, Henin N, Bebear C, Pereyre S. 2017. Evaluation of a commercial multiplex quantitative PCR (qPCR) assay for simultaneous detection of Mycoplasma genitalium and macrolide resistance-associated mutations in clinical specimens. J Clin Microbiol 55:978–979. doi: 10.1128/JCM.02168-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Su JP, Tan LY, Garland SM, Tabrizi SN, Mokany E, Walker S, Bradshaw CS, Read T, Murray GL. 2018. Evaluation of the SpeeDx ResistancePlus MG diagnostic test for Mycoplasma genitalium on the Applied Biosystems 7500 Fast quantitative PCR platform. J Clin Microbiol 56:e01245-17. doi: 10.1128/JCM.01245-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hamasuna R, Le PT, Kutsuna S, Furubayashi K, Matsumoto M, Ohmagari N, Fujimoto N, Matsumoto T, Jensen JS. 2018. Mutations in ParC and GyrA of moxifloxacin-resistant and susceptible Mycoplasma genitalium strains. PLoS One 13:e0198355. doi: 10.1371/journal.pone.0198355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Couldwell DL, Tagg KA, Jeoffreys NJ, Gilbert GL. 2013. Failure of moxifloxacin treatment in Mycoplasma genitalium infections due to macrolide and fluoroquinolone resistance. Int J STD AIDS 24:822–828. doi: 10.1177/0956462413502008. [DOI] [PubMed] [Google Scholar]
- 26.Basu I, Roberts SA, Bower JE, Henderson G, Reid M. 2017. High macrolide resistance in Mycoplasma genitalium strains causing infection in Auckland, New Zealand. J Clin Microbiol 55:2280–2282. doi: 10.1128/JCM.00370-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kikuchi M, Ito S, Yasuda M, Tsuchiya T, Hatazaki K, Takanashi M, Ezaki T, Deguchi T. 2014. Remarkable increase in fluoroquinolone-resistant Mycoplasma genitalium in Japan. J Antimicrob Chemother 69:2376–2382. doi: 10.1093/jac/dku164. [DOI] [PubMed] [Google Scholar]
- 28.Shimada Y, Deguchi T, Nakane K, Masue T, Yasuda M, Yokoi S, Ito S, Nakano M, Ito S, Ishiko H. 2010. Emergence of clinical strains of Mycoplasma genitalium harbouring alterations in ParC associated with fluoroquinolone resistance. Int J Antimicrob Agents 36:255–258. doi: 10.1016/j.ijantimicag.2010.05.011. [DOI] [PubMed] [Google Scholar]
- 29.Dionne-Odom J, Geisler WM, Aaron KJ, Waites KB, Westfall AO, Van Der Pol B, Xiao L. 2018. High prevalence of multidrug-resistant Mycoplasma genitalium in human immunodeficiency virus-infected men who have sex with men in Alabama. Clin Infect Dis 66:796–798. doi: 10.1093/cid/cix853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Jensen JS, Fernandes P, Unemo M. 2014. In vitro activity of the new fluoroketolide solithromycin (CEM-101) against macrolide-resistant and -susceptible Mycoplasma genitalium strains. Antimicrob Agents Chemother 58:3151–3156. doi: 10.1128/AAC.02411-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Chrisment D, Charron A, Cazanave C, Pereyre S, Bebear C. 2012. Detection of macrolide resistance in Mycoplasma genitalium in France. J Antimicrob Chemother 67:2598–2601. doi: 10.1093/jac/dks263. [DOI] [PubMed] [Google Scholar]
- 32.Couldwell DL, Lewis DA. 2015. Mycoplasma genitalium infection: current treatment options, therapeutic failure, and resistance-associated mutations. Infect Drug Resist 8:147–161. doi: 10.2147/IDR.S48813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Whiley DM, Trembizki E, Buckley C, Freeman K, Baird RW, Beaman M, Chen M, Donovan B, Kundu RL, Fairley CK, Guy R, Hogan T, Kaldor JM, Karimi M, Limnios A, Regan DG, Ryder N, Su JY, Ward J, Lahra MM. 2017. Molecular antimicrobial resistance surveillance for Neisseria gonorrhoeae, Northern Territory, Australia. Emerg Infect Dis 23:1478–1485. doi: 10.3201/eid2309.170427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Barbera MJ, Fernandez-Huerta M, Jensen JS, Caballero E, Andreu A. 2017. Mycoplasma genitalium macrolide and fluoroquinolone resistance: prevalence and risk factors among a 2013-2014 cohort of patients in Barcelona, Spain. Sex Transm Dis 44:457–462. doi: 10.1097/OLQ.0000000000000631. [DOI] [PubMed] [Google Scholar]