Quinolone resistance testing in Mycoplasma genitalium infection: A call for caution

Dear Editor,

We read with interest the scientific article by Moreno-Flores et al. entitled “Macrolide and fluoroquinolone resistance in Mycoplasma genitalium in the Northwestern region of Spain, 2019–2022”.¹ In their manuscript, the authors present recent data on macrolide and quinolone resistance in M. genitalium obtained using two commercially-available CE-IVD marked qPCR assays developed by Seegene. Antibiotic resistance estimates in this sexually transmitted infection are valuable, and we commend the authors for this very important contribution. However, we would like to offer a few comments regarding antimicrobial resistance in M. genitalium.

First, the reported rate of macrolide resistance in M. genitalium within this cohort was 21%,¹ which aligns with previous findings from nearby Spanish regions.^2,3 Since macrolide resistance in M. genitalium is particularly prevalent among men who have sex with men (MSM),^2–4 stratifying this estimate by sexual behavior would be very insightful. While the authors acknowledge the lack of epidemiological information as a major limitation in this cross-sectional study,¹ we suggest disaggregating macrolide resistance data for women and men, if possible. Notably, transmission-dynamic studies in M. genitalium indicate that the prevalence of macrolide resistance in men who have sex with women may mirror that of women.⁴ Therefore, this additional information, if available, could potentially be used to infer antibiotic resistance among MSM.

On the other hand, the authors report that the prevalence of fluoroquinolone resistance in this series was 7% (5/73).¹ This estimate is based on the identification of four infections harboring mutation G259A in the parC gene and one presenting mutation G248T,¹ using the Allplex™ MG & MoxiR qPCR assay (Seegene, South Korea). This assay detects six mutations in parC that result in amino acid changes in the quinolone resistance-determining region of the topoisomerase ParC (Table 1): this is, residues serine 83 (S83) and aspartic acid 87 (D87), which mediate antibiotic-enzyme interactions.⁵ However, unlike macrolide resistance, the correlation between mutations in parC and treatment failure is less definitive.^6,7 In 2013, Couldwell et al. first reported moxifloxacin treatment failure associated with mutations S83I in the ParC protein.⁸ Subsequently, Murray et al. observed similar failure events linked to S83I and S83R mutations.⁹In vitro susceptibility studies have associated the S83I mutation (and, to a lesser extent, D87Y) with increased minimum inhibitory concentrations to the fourth-generation fluoroquinolone moxifloxacin.⁷ In conclusion, the clinical relevance of the parC genetic markers detected by the Allplex™ MG & MoxiR qPCR assay is uncertain except for mutation G248T (S83I), which is consistently associated with fluoroquinolone resistance (Table 1). Therefore, limiting quinolone resistance estimates in M. genitalium to the presence of the S83I amino acid change in ParC may be appropriate. Consequently, the prevalence of fluoroquinolone resistance in the cohort described by Moreno-Flores et al. would decrease to 1.4% (1/73; 95% confidence interval, 0.0–7.4%).

Table 1.

Mycoplasma genitalium mutations in the parC gene detected by the Allplex™ MG & MoxiR qPCR assay.

parC gene mutation^a	ParC amino acid change^a	Quinolone resistance^b
A247C	Ser83Arg (S83R)	Possible
G248A	Ser83Asn (S83N)	Uncertain^c
G248T	Ser83Ile (S83I)	Likely
G259A	Asp87Asn (D87N)	Uncertain^d
G259C	Asp87His (D87H)	Uncertain
G259T	Asp87Tyr (D87Y)	Uncertain^e

a

Mycoplasma genitalium numbering.

b

The clinical significance provided do not correspond to a systematic classification but to the opinion of the authors according to the current scientific evidence.

c

Conservative amino acid replacement.

d

Some moxifloxacin treatment failures have been apparently associated with mutation D87N in ParC (unpublished data).¹¹

e

Despite mutation D87Y in ParC has been associated with an in vitro increase of the minimum inhibitory concentration (MIC) to fluoroquinolones,⁷ the clinical relevance of this phenomenon remains unclear.

Finally, we would like to caution against the routine use of molecular assays detecting quinolone resistance-associated mutations in M. genitalium (such as the Allplex™ MG & MoxiR) in clinical settings. As demonstrated in the study led by Moreno-Flores, the prevalence of potential fluoroquinolone resistance in M. genitalium is low.¹ Furthermore, the correlation between parC mutations and moxifloxacin treatment failure is suboptimal, and antimicrobial alternatives are very scarce. As a result, fluoroquinolone resistance testing is not currently indicated on a routine basis in Europe.¹⁰ However, despite evidence is limited, international guidelines strongly recommend the use of molecular assays for detecting quinolone resistance-associated mutations in M. genitalium (primarily in parC) among patients experiencing moxifloxacin treatment failure.¹⁰ This approach may restrict the use of the very few third (and last)-line antimicrobials (such as pristinamycin) for patients with genetically-documented moxifloxacin resistance.^10,11

Funding

None declared.

Conflicts of interest

None declared.