Emergence of decreased susceptibility and resistance to extended-spectrum cephalosporins in Neisseria gonorrhoeae in Korea

Abstract

Objectives

Antimicrobial resistance (AMR) in Neisseria gonorrhoeae is a major concern globally; however, no comprehensive AMR data for gonococcal isolates cultured after 2006 in Korea have been published internationally. We determined the susceptibility of N. gonorrhoeae isolates cultured in 2011–13, the mechanism of extended-spectrum cephalosporin (ESC) resistance and the molecular epidemiology of gonococcal strains in Korea.

Methods

In 2011–13, 210 gonococcal isolates were collected in Korea and their AMR profiles were examined by the agar dilution method. The penA, mtrR, penB, ponA and pilQ genes were sequenced in 25 isolates that were resistant to ESCs and 70 randomly selected isolates stratified by year. For molecular epidemiology, N. gonorrhoeae multiantigen sequence typing and MLST were performed.

Results

None of the N. gonorrhoeae isolates was susceptible to penicillin G and most were resistant to tetracycline (50%) and ciprofloxacin (97%). The rates of resistance to ceftriaxone, azithromycin, cefpodoxime and cefixime were 3%, 5%, 8% and 9%, respectively. However, all isolates were susceptible to spectinomycin. Twenty-one (84%) of the 25 ESC-resistant isolates contained the non-mosaic PBP2 XIII allele; however, the remaining 4 (16%) possessed the mosaic PBP2 X allele, which has been previously associated with ESC resistance including treatment failures.

Conclusions

In Korea, susceptibility to spectinomycin remains high. However, the recent emergence of ESC-resistant N. gonorrhoeae strains, including strains possessing the PBP2 mosaic X and non-mosaic XIII alleles, is a major concern and enhanced AMR surveillance is necessary to prevent transmission of these strains.

Introduction

Neisseria gonorrhoeae is the causative agent of various infections ranging from uncomplicated urethritis or cervicitis to ascending infections, resulting in pelvic inflammatory disease, ectopic pregnancy and infertility, or disseminated infections. According to the WHO, in 2008 the burden of gonococcal infections was estimated as 106 million new cases among adults globally.¹ In Korea, it is difficult to measure the exact number of gonococcal infections because gonorrhoea cases are under-reported to protect patients following the introduction of antiprostitution legislation in 2004. However, ∼50 000 cases (incidence of ∼100 cases per 100 000 inhabitants) of gonococcal infection remain reported to the Korea Health Insurance Review and Assessment Service annually.²

N. gonorrhoeae has developed resistance to all previously used first-line antimicrobials for treatment of gonorrhoea and in many countries the extended-spectrum cephalosporins (ESCs) are the only remaining options for empirical antimicrobial monotherapy.^3,4 Worryingly, the ESCs are now also threatened due to evolving resistance. Treatment failures with cefixime have been reported in several countries^3,5–7 and rare cases of treatment failures as well as high-level in vitro resistance to the more potent ceftriaxone have been verified in a few countries.^3,5,8–12 Due to this developing situation, dual antimicrobial therapy with ceftriaxone and azithromycin is now recommended in Europe and the USA.^13,14 In Korea, antimicrobial resistance in N. gonorrhoeae has been a serious problem for many years. However, no comprehensive antimicrobial resistance data for isolates cultured after 2006 have been published internationally.² In recent years, spectinomycin and ceftriaxone have been used mainly as monotherapy to treat gonococcal infections. However, in the 2011 Korean guideline, dual antimicrobial therapy (250 mg of ceftriaxone plus 1 g of azithromycin or 250 mg of ceftriaxone plus 100 mg of doxycycline twice daily for 7 days) has been introduced as the recommended first-line therapy for uncomplicated gonococcal infections.¹⁵ Nevertheless, the high selective pressure resulting from widespread use can lead to the development of resistance. Accordingly, national and international quality-assured surveillance of antimicrobial resistance are essential to identify the emerging resistance, trends in resistance and provide appropriate data supporting timely revisions of treatment guidelines.

In this study, the antimicrobial susceptibility of N. gonorrhoeae isolated between 2011 and 2013 to eight relevant antimicrobials and the antimicrobials used for treatment of gonorrhoea in Korea from 2009 to 2012 were investigated. The genetic determinants of resistance to ESCs and the molecular epidemiological relatedness of the N. gonorrhoeae isolates were also elucidated.

Materials and methods

The work was performed at the Department of Laboratory Medicine, Research Institute of Bacterial Resistance, Yonsei University College of Medicine, Seoul, Korea.

Gonorrhoea patients and N. gonorrhoeae isolates

A total of 210 N. gonorrhoeae isolates were collected from patients with urethritis (136 males and 47 females) and female commercial sex workers (n = 27) in 2011 (n = 60), 2012 (n = 91) and 2013 (n = 59) in Korea, through a national surveillance programme for gonococcal resistance supported by the Korean Centers for Disease Control and Prevention. Most (71.4%) of the N. gonorrhoeae isolates were cultured from specimens obtained from 35 primary urological clinics, which were widely distributed across Korea. However, N. gonorrhoeae isolates were also collected from public health centres, secondary care hospitals and tertiary care hospitals.

Transgrow media and BD CultureSwab MaxV (+) (Becton Dickinson, Cockeysville, MD, USA) were used for specimen transportation. Transgrow media was prepared in-house with GC agar base (Becton Dickinson, Sparks, MD, USA), haemoglobin, IsoVitaleX (Becton Dickinson) and VCNT inhibitor (Becton Dickinson), which contains vancomycin, colistin, nystatin and trimethoprim. This medium was distributed into screw-cap tubes, subsequently solidified in a horizontal position and, finally, a CO₂-enriched atmosphere was introduced into the tubes.

Modified Thayer–Martin agar plates (Becton Dickinson) were used for culture isolation and conventional biochemical tests including Gram's stain and the Vitek Neisseria-Haemophilus Identification (NHI) system (bioMérieux, Marcy-l'Étoile, France) were used for species identification. All isolates were stored in 20% skimmed milk (Difco, Detroit, MI, USA) at −70°C prior to analysis.

Antimicrobial susceptibility testing

Antimicrobial susceptibility testing was performed using the agar dilution method according to guidelines published by the CLSI.¹⁶ MICs of ceftriaxone (Hanmi, Seoul, Korea), cefixime (Dong-A, Seoul, Korea), cefpodoxime (Pfizer Korea, Seoul, Korea), spectinomycin (Kuk Je, Seoul, Korea), azithromycin (Pfizer Korea), penicillin G (Sigma Chemical, St Louis, MO, USA), tetracycline (Pfizer Korea) and ciprofloxacin (Bayer Korea, Seoul, Korea) were determined using a GC II agar base supplemented with 1% IsoVitaleX (Becton Dickinson). Briefly, ∼10⁴ cfu of each isolate were applied to the agar plates with a Steer's replicator (Craft Machine, Chester, PA, USA) and the inoculated plates were then incubated in a 5% CO₂-enriched atmosphere at 35°C for 24 h. A cefinase disc test was performed to determine β-lactamase production. The resistance breakpoints stated by EUCAST (www.eucast.org) were applied. However, EUCAST does not state any resistance breakpoint for cefpodoxime and, accordingly, the cefpodoxime breakpoint from the CLSI was applied. N. gonorrhoeae reference strains ATCC 49226, WHO A, WHO B, WHO G, WHO J, WHO K and WHO L¹⁷ were used for quality control.

Identification of genetic determinants associated with resistance to β-lactam antimicrobials

The entire penA gene was sequenced in all isolates that were resistant to at least one of the ESCs (n = 25) and for 70 randomly selected isolates stratified by year using a statistically random method. The PBP2 sequences, encoded by the penA gene sequences, were assigned allele numbers according to previously published nomenclature.⁸ Other genetic determinants responsible for resistance to ESCs (ceftriaxone, cefixime or cefpodoxime) and/or penicillin G were also sequenced, including the mtrR, penB (in the porB1b gene), ponA and pilQ genes. The primers and conditions used for amplifications were described in a previous study.¹⁸ The amplification process consisted of 35 cycles of denaturation at 94°C for 30 s, annealing at 50°C (mtrR), 46°C (penB), 56°C (ponA) or 52°C (pilQ) for 30 s and elongation at 72°C for 1 min. The size of the amplified product was confirmed by electrophoresis and DNA was extracted from the gel and sequenced at a commercial laboratory (Macrogen, Seoul, Korea).

Molecular epidemiological typing

To determine the molecular epidemiological relatedness, N. gonorrhoeae multiantigen sequence typing (NG-MAST) was carried out for all isolates and MLST was performed for the 25 isolates that showed resistance to at least one of the ESCs examined. NG-MAST and MLST were performed according to the guidelines outlined on the respective database web site (NG-MAST, http://www.ng-mast.net;¹⁹ MLST, http://pubmlst.org/neisseria/) with minor modifications. Phylogenetic tree analysis for the NG-MAST porB and tbpB alleles was performed by MEGA6 software using the neighbour-joining method.

Antimicrobial agents used to treat gonorrhoea in Korea

Given that all people in Korea are covered by the National Health Insurance System and antimicrobial agents are not available without prescription, the prescription data for treatment of gonorrhoea were acquired from the Korea Health Insurance Review and Assessment Service to analyse the recent selective antimicrobial pressure for development of resistance in N. gonorrhoeae. The percentage of patients (diagnosed with gonococcal infection without any accompanying additional sexually transmitted infection) treated with each specific antimicrobial from 2009 to 2012 was analysed.

Ethics statement

All examined gonococcal isolates were cultured and stored as part of routine diagnostics (standard care) and no patient identification information was available in the study. This study was also granted an exemption from requirement for ethics approval because it was performed as one of the projects of the national surveillance programme supported by the Korean Centers for Disease Control and Prevention (2013-E44003-00).

Results

Antimicrobial susceptibility in N. gonorrhoeae isolates from Korea during 2011–13 (n = 210)

The results of the antimicrobial resistance testing for all isolates are summarized in Table 1. From 2011 to 2013, the MIC distributions and susceptibility to the surveyed antimicrobials did not differ greatly by year. All N. gonorrhoeae isolates were susceptible to spectinomycin and the MIC range, MIC₅₀ and MIC₉₀ were ≤16–32, 32 and 32 mg/L, respectively. The rates of resistance to ceftriaxone, cefpodoxime and cefixime were 3.3%, 8.1% and 9.0%, respectively. The MIC₉₀ (MIC range) of ceftriaxone, cefpodoxime and cefixime was 0.12 mg/L (≤0.008–0.25 mg/L), 0.5 mg/L (≤0.008–4 mg/L) and 0.12 mg/L (≤0.008–0.5 mg/L), respectively. The rate of azithromycin-resistant isolates was 4.8% and an additional 33% of the isolates showed intermediate resistance. However, the MICs for the azithromycin-resistant isolates ranged from 1 to 8 mg/L and, accordingly, no isolates with high-level resistance to azithromycin (MIC ≥256 mg/L) were identified. All isolates were resistant (29%) or intermediate (71%) to penicillin G. The proportion of penicillinase-producing N. gonorrhoeae (PPNG) was 6.2% (13/210). Most of the isolates were resistant to tetracycline (50%), including 17% with high-level resistance, and ciprofloxacin (97%). Furthermore, 27% of isolates were resistant to penicillin G, ciprofloxacin and tetracycline and 12% of these MDR isolates were also resistant to azithromycin. Coresistance to ESCs and azithromycin was noted in one isolate that was resistant to ceftriaxone (MIC = 0.25 mg/L), cefpodoxime (MIC = 2 mg/L) and azithromycin (MIC = 2 mg/L).

Table 1.

Antimicrobial susceptibility of N. gonorrhoeae isolates collected from 2011 to 2013 in Korea (n = 210)

Antimicrobial	MIC (mg/L)			Susceptibility/resistance (percentage of isolates)
	range	50%	90%	S	I	R
Ceftriaxone	≤0.008–0.25	0.03	0.12	97		3
Cefpodoxime	≤0.008–4	0.12	0.5	92		8
Cefixime	≤0.008–0.5	0.03	0.12	91		9
Spectinomycin	≤16–32	32	32	100	0	0
Azithromycin	≤0.06–8	0.25	0.5	62	33	5
Penicillin G	0.12 to >128	1	4	0	71	29
Tetracycline	≤0.12–64	2	32	26	24	50
Ciprofloxacin	≤0.008–64	4	16	3	0	97

MICs were determined using the agar dilution technique.

S (susceptible), I (intermediate resistant) and R (resistant) were interpreted in accordance with the EUCAST breakpoints (http://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/Breakpoint_tables/Breakpoint_table_v_4.0.pdf). However, EUCAST (www.eucast.org) does not state any resistance breakpoints for cefpodoxime and, accordingly, the breakpoint from the CLSI (www.clsi.org) was applied.

Antimicrobial agents used to treat gonorrhoea in Korea, 2009–12

During 2009–12, more than half of the patients with gonorrhoea in Korea were treated with spectinomycin (51.7%–72.8%). However, the usage rate decreased from 72.8% in 2009 to 51.7% in 2012. By contrast, the proportion of patients treated with ceftriaxone significantly increased from 23.8% in 2009 to 45.3% in 2012. Cefixime, penicillin G and ciprofloxacin were also used; however, each was prescribed in only <1% of cases. Other cephalosporins including cefotaxime, cefpodoxime and ceftizoxime were used to treat 0.9%–2.1% of gonococcal infections (Table 2).

Table 2.

Antimicrobials used for treatment of gonococcal infections in Korea from 2009 to 2012

Antimicrobial	Percentage of patients treated in
	2009	2010	2011	2012
Spectinomycin	72.8	68.9	64.8	51.7
Ceftriaxone	23.8	27.6	32.5	45.3
Cefixime	0.3	0.4	0.4	0.5
Other cephalosporina	1.8	2.1	0.9	1.5
Ciprofloxacin	0.9	0.7	0.9	0.5
Penicillin G	0.5	0.4	0.4	0.4
Total	100.0	100.0	100.0	100.0

^aCefotaxime, cefpodoxime and ceftizoxime.

Mechanisms of resistance to β-lactam antimicrobials

The most common PBP2 allele in the 25 isolates that were resistant to at least one of the ESCs (ceftriaxone 7, cefixime 19 and cefpodoxime 17) and 70 randomly selected ESC-susceptible isolates was the non-mosaic PBP2 XIII allele (56/95, 59%; Table 3), which contains the A501V and P552S alterations that both might increase the MICs of ESCs.^3,4,20,21 Twenty-one (84%) of the ESC-resistant isolates contained this non-mosaic PBP2 XIII allele; however, the remaining 4 (16%) of the ESC-resistant isolates possessed the mosaic PBP2 X allele, which has been previously associated with ESC treatment failures and high MICs of ESCs.^3,8 Many ESC-susceptible isolates that had MICs exactly at the ESC-resistance breakpoints also contained the XIII allele, while the non-mosaic IV, V, XII, XVI, XVII and XVIII alleles were mostly found in isolates with full susceptibility to these ESCs. The isolates (n = 4) with the mosaic PBP2 X allele showed an MIC range of 0.06–0.25 mg/L for ceftriaxone, 0.25–0.5 mg/L for cefixime and 2–4 mg/L for cefpodoxime (Table 3). Additional genetic resistance determinants for β-lactam antimicrobials were found in most of the N. gonorrhoeae isolates, e.g. mtrR (A deletion in promoter: 99%; MtrR A39T mutation: 1%), penB (71%) and ponA (L421P, 94%), the last one mainly involved in penicillin G resistance.^3,22–24 In the pilQ gene, an N648S mutation was found in three isolates and one isolate had both an N648S mutation and the previously described E666K mutation,^23,25,26 but all of these isolates were susceptible to the ESCs examined.

Table 3.

MIC distributions of ceftriaxone, cefixime and cefpodoxime for 95 N. gonorrhoeae isolates according to PBP2 allele⁸

Antimicrobial	PBP2 allele8	MIC (mg/L) (number of isolates)
		≤0.008	0.015	0.03	0.06	0.12	0.25	0.5	1	2	4
Ceftriaxone	IVa		2	5	1
	Va	6	1	2	2	1
	Xb				1	2	1
	XIIa			1		1
	XIIIa	1	1	7	14	27	6
	XVIa	1
	XVIIa				1	1
	XVIIIa	1		3	1	5
Cefixime	IV		7		1
	V	4	4	1	2	1
	X						3	1
	XII			1		1
	XIII	1		2	15	23	14	1
	XVI	1
	XVII			1	1
	XVIII		1	4	4	1
Cefpodoxime	IV				1	6	1
	V			6	2	2	1	1
	X									3	1
	XII						1	1
	XIII			1		4	13	25	10	2	1
	XVI			1
	XVII						1	1
	XVIII			1	1	2	5	1

Bold formatting indicates resistant N. gonorrhoeae isolates.

^aNon-mosaic PBP2 allele.⁸

^bMosaic PBP2 allele.⁸

Molecular epidemiology of N. gonorrhoeae isolates resistant to ESCs

According to the epidemiological analysis of the 25 ESC-resistant N. gonorrhoeae isolates, 21 (84%) were collected at clinics and hospitals that were located in geographically closely related regions (Seoul, Incheon and Gyeonnggi province). The four isolates containing the mosaic PBP2 X allele were isolated in 2011 (one isolate) and 2013 (three isolates). All of these four isolates were resistant to cefixime and cefpodoxime; however, only one of them was also resistant to ceftriaxone. All four isolates belonged to the MLST ST1901, which has been spreading globally and accounted for the majority of the ESC resistance internationally.^3,4 Three of them were assigned to NG-MAST ST2958 and one to NG-MAST ST9669. Among the 21 ESC-resistant isolates containing the non-mosaic PBP2 XIII allele, 3 isolates were resistant to all three ESCs, i.e. ceftriaxone, cefixime and cefpodoxime (Table 4). The NG-MAST STs of the 25 ESC-resistant N. gonorrhoeae isolates varied, but most (76%) of them had the NG-MAST tbpB 21 allele and the majority of the isolates were also genetically related by phylogenetic analysis of the porB gene (Figure 1). In NG-MAST analysis of the 10 azithromycin-resistant N. gonorrhoeae isolates, 5 had the identical tbpB 328 allele (3 ST2066, 1 ST5699 and 1 ST8162) and their porB gene sequences were genetically related (data not shown).

Table 4.

N. gonorrhoeae isolates resistant to any ESC from 2011 to 2013 in Korea

PBP28	Year	Region	MIC (mg/L)				NG-MAST			Other genetic determinants
			CRO	CFM	CPD	PEN	ST	porB	tbpB	mtrR	penB	ponA	pilQ
X (n = 4)a	2011	Incheon	0.06	0.25	2	2	2958	1785	110	deletion of A	G120K, A121D	L421P	—
	2013	Seoul	0.25	0.5	4	8	2958	1785	110	deletion of A	G120K, A121D	L421P	—
	2013	Gyeonnggi	0.12	0.25	2	4	2958	1785	110	deletion of A	G120K, A121D	L421P	—
	2013	Seoul	0.12	0.25	2	4	9669	4529	110	deletion of A	G120K, A121D	L421P	—
XIII (n = 21)	2011	Daejeon	0.12	0.25	0.25	2	3968	2409	21	deletion of A	G120K, A121D	L421P	—
	2011	Seoul	0.12	0.25	1	1	8044	3	21	deletion of A		L421P	—
	2011	Seoul	0.12	0.5	4	4	8867	5219	21	deletion of A		L421P	—
	2011	Incheon	0.12	0.12	1	2	7069	3989	21	deletion of A	G120K, A121D	L421P	—
	2011	Seoul	0.12	0.12	1	1	8869	4620	1611	deletion of A	G120N, A121D	L421P	—
	2011	Incheon	0.25	0.12	0.5	2	7548	4530	21	deletion of A	G120K, A121D	L421P	—
	2012	Incheon	0.25	0.25	2	2	7069	3989	21	deletion of A	G120K, A121D	L421P	—
	2012	Incheon	0.12	0.25	1	1	7686	1036	21	deletion of A		L421P	—
	2012	Seoul	0.25	0.25	1	2	7548	4530	21	deletion of A	G120K, A121D	L421P	—
	2012	Incheon	0.12	0.25	1	1	7687	90	620	deletion of A	D120K, A121G	L421P	N27S
	2012	Cheonbuk	0.12	0.25	0.5	1	7688	23	21	deletion of A		L421P	—
	2012	Seoul	0.12	0.25	1	1	8041	2125	21	deletion of A		L421P	—
	2012	Incheon	0.12	0.25	1	1	8193	4917	21	deletion of A	G120K, A121D	L421P	—
	2012	Incheon	0.12	0.12	1	2	7684	2389	21	deletion of A	G120K, A121N	L421P	—
	2013	Cheonbuk	0.25	0.25	1	4	7069	3989	21	deletion of A	G120K, A121D	L421P	—
	2013	Cheonbuk	0.12	0.25	0.5	1	8044	3	21	deletion of A		L421P	—
	2013	Seoul	0.12	0.25	0.5	1	8046	4797	21	deletion of A		L421P	—
	2013	Seoul	0.12	0.25	0.5	1	8052	4801	21	deletion of A		L421P	—
	2013	Seoul	0.03	0.25	0.25	0.5	8164	4895	33	deletion of A	G120N, A121V	L421P	—
	2013	Seoul	0.25	0.12	2	1	7688	23	21	deletion of A		L421P	—
	2013	Incheon	0.25	0.12	0.5	2	4502	1785	21	deletion of A	G120K, A121D	L421P	—

CRO, ceftriaxone; CFM, cefixime; CPD, cefpodoxime; PEN, penicillin G.

^aAll isolates assigned to the MLST ST1901.

Figure 1.

Phylogenetic tree of the porB gene sequences, tbpB alleles and NG-MAST STs of N. gonorrhoeae isolates with resistance to at least one ESC from 2011 to 2013 in Korea.

Discussion

Internationally, as well as in Korea, the threat of untreatable or difficult-to-treat gonorrhoea is real. None of the Korean isolates collected between 2011 and 2013 was susceptible to penicillin G, although the PPNG rate remained low (6%). Accordingly, resistance due to chromosomal mutations was common. All isolates contained a PBP2 mosaic allele or a PBP2 allele with an aspartate insertion at position 346⁸ and mtrR mutation, which collaboratively result in increased penicillin G MICs.^4,22,23 Most isolates also had resistance mutations in the porB1b (penB resistance determinant) and ponA genes. In porB1b, the G120K and A121D mutations (38% of isolates) were the most common, which is in agreement with previous reports.^3,22,23 The ponA L421P mutation was found in 94% of isolates. Interestingly, three isolates had an N648S mutation and one isolate had both an N648S mutation and the previously described pilQ E666K mutation;^23,25 their penicillin G MICs ranged between 0.5 and 1.0 mg/L. SNPs in the pilQ gene, e.g. E666K, have been reported in laboratory-derived isolates and associated with high-level penicillin G resistance when accompanied with penA, mtrR and penB mutations.²³ However, these mutations were stated to inhibit proper pili formation, which is essential for virulence of gonococcal strains.²⁶ Nevertheless, we found the pilQ E666K SNP in one clinical isolate, but the penicillin G MIC was only 1 mg/L despite concomitant penA, mtrR, penB and ponA mutations. This might indicate that this isolate was lacking the still unknown non-transformable resistance determinant ‘Factor X’.^{3–5,8,21–23} The significance of pilQ mutations, e.g. E666K, in clinical isolates requires further investigation. In Korea, the use of tetracyclines and fluoroquinolones in gonorrhoea treatment was abandoned in the mid-1980s and mid-2000s, respectively. In this study, many isolates were resistant to tetracycline (50%) and ciprofloxacin (97%), which is in agreement with earlier Korean studies.² The rate of plasmid-encoded high-level resistance to tetracycline (TRNG) was stable at 0%–4% until the mid-2000s, but then started to increase and reached 17% during 2011–13. Due to the low tetracycline selective pressure in Korea, foreign import of TRNG might be the most plausible explanation for this increase.

Despite spectinomycin having been commonly used in Korea, gonococcal susceptibility to spectinomycin remains exceedingly high. In Korea, the majority of gonorrhoea patients from 2002 to 2006 were treated with spectinomycin² and the situation was similar in 2009–12. In many countries, spectinomycin is not produced any more and its use in gonorrhoea treatment was abandoned or at least significantly decreased after the first reports of resistance, e.g. among US military in Korea;²⁷ however, internationally, spectinomycin resistance has been exceedingly rare in recent decades.^3,4 Even in Korea, spectinomycin resistance has been very rare, i.e. not reported after 1993. The spread of spectinomycin resistance in earlier decades might represent the spread of a few successful gonococcal strains and additional research regarding possible fitness cost of spectinomycin resistance would be valuable, which might explain the reasons for the limited spread of spectinomycin resistance in Korea and globally in modern times.

The recent emergence and spread of ESC-resistant N. gonorrhoeae has become a global concern.^3–14 In Korea, the resistance rates from 2011 to 2013 were 3.3% to ceftriaxone, 8.1% to cefpodoxime and 9% to cefpodoxime. The MIC₉₀ of ceftriaxone increased 2-fold from 0.06 to 0.12 mg/L compared with a previous report from the mid-2000s.² In infections caused by gonococci with a ceftriaxone MIC of 0.12 mg/L, even when 1 g of ceftriaxone is administered, the time period that the concentration of free ceftriaxone exceeds the MIC (fT_>MIC) may be <20 h in the lower 95% CI.²⁸ Furthermore, gonococcal isolates with a ceftriaxone MIC ≤0.12 mg/L have resulted in failures to treat pharyngeal gonorrhoea with ceftriaxone.^3,9,10,12 Regarding cefixime, already in 2004 one resistant isolate was reported in Korea.² Cefixime resistance has been closely associated with mosaic penA alleles worldwide and both the isolate from 2004 and four cefixime-resistant isolates from 2013 had the mosaic PBP2 X allele, which resulted in early cefixime resistance in, e.g. Japan.^3,8,18 Interestingly, the 15 additional cefixime-resistant isolates instead had the non-mosaic PBP2 XIII allele, which also was the most common PBP2 type in Korea. This non-mosaic PBP2 XIII allele has five specific amino acid alterations (A501V, F504L, A510V, A516G and P551S),⁸ among which A501V has been verified (and P551S statistically associated) to be important in ESC resistance.^3,4,20,21 However, most of the Korean N. gonorrhoeae isolates with the PBP2 XIII allele showed susceptibility to all the ESCs. Further investigation is crucial to determine the differences between these ESC-resistant and -susceptible isolates. The mosaic PBP2 X allele and some additional mosaic PBP2 alleles, e.g. mosaic PBP2 XXXIV allele, have been reported to be also involved in ceftriaxone resistance.^3–5,8 Accordingly, antimicrobial resistance surveillance or screening is very important nationally and internationally. However, in many geographical settings, diagnosis based on culture has been replaced by molecular diagnostics and, consequently, a highly sensitive and specific molecular antimicrobial resistance screening test is essential. Our results indicate that a silent dissemination of N. gonorrhoeae with the mosaic PBP2 X allele is initiated and genetic screening and/or surveillance should be performed to prevent the spread of ESC resistance.

Azithromycin has been suggested to have several advantages in the treatment of gonococcal infections. However, selection of azithromycin resistance in N. gonorrhoeae and Mycoplasma genitalium, which correlates with azithromycin use in monotherapy, has been described internationally.^3,4,29 In Korea, only 62% of isolates were susceptible to azithromycin and the resistance rate was 5%. However, no isolates with high-level resistance to azithromycin (MIC ≥256 mg/L), which have been identified in many other countries worldwide,⁴ were identified or have been previously found in Korea.² All except one of the azithromycin-resistant N. gonorrhoeae isolates were collected from the capital area (Seoul, Incheon and Gyeonnggi). Three of the 10 azithromycin-resistant isolates belonged to NG-MAST ST2066 and two additional isolates possessed the identical tbpB 328 allele and showed genetic relatedness with ST2066 in the porB phylogenetic tree analysis. This indicates that clonal dissemination of azithromycin-resistant N. gonorrhoeae occurred in the capital area. The dissemination not only reduced the clinical usefulness of monotherapy with azithromycin, which should be avoided, but might also reduce the efficacy of the newly introduced combination treatment with ceftriaxone.

In conclusion, the susceptibility of N. gonorrhoeae to spectinomycin remains high. However, in vitro resistance to the ESCs ceftriaxone, cefixime and cefpodoxime is identified in Korea. The recent emergence of ESC-resistant N. gonorrhoeae strains containing the mosaic PBP2 X allele is a major concern and enhanced AMR surveillance is essential to prevent transmission of these strains. However, in Korea, as in many other countries globally, molecular detection of N. gonorrhoeae has rapidly been replacing culture and, accordingly, a highly sensitive and specific genetic screening for ESC-resistant isolates would be exceedingly valuable.

Funding

The present work was supported by grants from the National Institute of Health, Korea Centers for Disease Control and Prevention, Ministry of Health and Welfare, Republic of Korea.

Transparency declarations

None to declare.