The frontiers of addressing antibiotic resistance in Neisseria gonorrhoeae

Abstract

The sexually transmitted infection gonorrhea, caused by the Gram-negative bacterium Neisseria gonorrhoeae, can cause urethritis, cervicitis, and systemic disease, among other manifestations. N. gonorrhoeae has rapidly rising incidence along with increasing levels of antibiotic resistance to a broad range of drugs including first-line treatments. The rise in resistance has led to fears of untreatable gonorrhea causing substantial disease globally. In this review, we will describe multiple approaches being undertaken to slow and control this spread of resistance. First, a number of old drugs have been repurposed and new drugs are being developed with activity against Neisseria gonorrhoeae. Second, vaccine development, long an important goal, is advancing. Third, new diagnostics promise rapid detection of antibiotic resistance and a shift from empiric to tailored treatment. The deployment of these new tools for addressing the challenge of antibiotic resistance will require careful consideration to provide optimal care for all patients while extending the lifespan of treatment regimens.

INTRODUCTION AND CLINICAL MANIFESTATIONS

Neisseria gonorrhoeae, also known as the gonococcus, is a global pathogen with significant morbidity. The WHO estimates that there were 87 million new cases of gonorrhea, the disease caused by Neisseria gonorrhoeae, worldwide in 2016,¹ an increase from the estimated 78 million new cases in 2012.² The burden of disease appears to be greatest in resource-limited settings, with an estimated incidence of 50 cases per 1000 men and 42 cases per 1000 women in sub-Saharan Africa, as compared to 9 cases per 1000 men and 6 cases per 1000 women in Northern America and Europe.¹ In the US, almost 600,000 new cases were identified in 2018 - up more than 80% from a historic low in 2009, with similar numbers from the UK (up 26% in 2018 compared to 2017 and at highest levels since 1978) and Australia (an increase of 63% over 5 years).^3–5 This rise has been driven by increases predominantly in young individuals and men who have sex with men (MSM).^6,7

N. gonorrhoeae infects and causes inflammation in mucosal tissues, including the urethra, cervix, rectum, pharynx, and conjunctiva and can lead to disseminated disease.⁸ Infection of the urethra is most common and often results in discharge and dysuria. In women, cervical infection is often asymptomatic.⁹ Asymptomatic and symptomatic infection with N. gonorrhoeae can progress to pelvic inflammatory disease (PID),¹⁰ which can result in infertility, ectopic pregnancy, and chronic pelvic pain—sequelae that impact a substantial fraction of those infected (18%, 0.6%, and 29%, respectively)¹¹ with worse outcomes associated with delayed care and repeated PID episodes. The rate of untreated N. gonorrhoeae infections that result in PID has been estimated as high, or higher than, the 15% seen with chlamydial infection.¹⁰ In one study, PID remained a risk even in treated infections, with an overall risk in the year after treatment for gonorrhea or chlamydia of 9%, which may be due to untreated reinfections or treatment resistance.¹² While N. gonorrhoeae can lead to pharyngitis and proctitis, pharyngeal and rectal infections are commonly asymptomatic.^13,14 Babies born to women with gonorrhea as well as adults can develop conjunctivitis caused by N. gonorrhoeae which, if left untreated, can lead to blindness. Disseminated gonococcal infection is rare in the antibiotic era^15,16 and can present with rash, polyarthralgia, and tenosynovitis. Transmission is associated with insertive intercourse, though recent epidemiological and modeling studies have also suggested that oral-oral contact may also enable transmission¹⁷.

HISTORICAL TREATMENTS OF GONORRHEA

The classic symptoms of gonorrhea have been reported in humans for millennia. Gonorrhea takes its name from Greek, combining gonos for “semen” and rhoea for “flow.”¹⁸ Gonorrhea first appears in the historical record in the first and second century CE, though scholars have also identified venereal diseases mentioned in ancient Egyptian papyri and the Hebrew Bible that bear similarity to gonorrhea. Greek writers mistakenly believed gonorrhea represented an increased amount of seed that is produced without sexual intercourse, though there was debate about whether the disease could affect both men and women. Even in ancient times, medical writers were concerned about the effect of gonorrhea on fertility in both sexes.¹⁹ Treatments for gonorrhea in the Medieval world included honey with water and milk, goat or breast milk, extract from the Spanish fly, and mercury.²⁰ Some of the earliest animal experiments were carried out by Abu Bakr Mohammad Ibn Zakariya Razi (known as Rhazes), who investigated the pharmacology of mercury as a treatment for venereal diseases, including gonorrhea.^20,21

Recent genomics-based work postulated a more recent origin of gonorrhea. Sánchez-Busó et al. (2019) use population genomics methods to date the origin of gonococcus to the late 16^th and early 17^th century, with spread to Asia within a century. This suggests that ‘gonorrhea’ in the ancient world may have been a general term for venereal disease.²²

As the understanding of infectious disease shifted to germ theory, so did the stratification of gonorrhea symptoms and treatments. By the 1840s, it was believed that both syphilis and gonorrhea were caused by a venereal ‘virus’ that could be passed through infectious pus.²³ The symptoms identified with this ‘gonorrheal virus’ are similar to the modern conception of canonical gonorrhea, but despite these advances in knowledge, opportunities for treatment were limited. There was no known prophylaxis, though cleanliness and diet were implicated as being preventive. Abortive treatment included leeches as well as injection of lead and sulfur or silver nitrate directly into the urethra.²³

Following the differentiation of gonorrhea from syphilis in the late 18^th century²⁴ and the discovery of the bacterium by Albert Neisser in 1879, efforts to develop treatment abounded. One of the first attempts was a killed bacterial vaccine. Though the authors saw some therapeutic effect when the vaccine was injected into infected patients, a similar effect was also seen in injections with meningococci and colonic bacilli.²⁵ Early antibiotic treatment included topical mercurochrome-220, a mercury-containing compound with therapeutic anti-gonococcal activity. ²⁶ Based on observations that high temperatures killed gonococcal strains, multiple clinical studies suggested that heat lamps could treat gonococcal urethritis and cervicitis, and other pyretic interventions, such as infection with malaria, were also studied.^27,28

THE ANTIBIOTIC ERA

The arrival of sulfa antibiotics heralded the antibiotic era. But resistance emerged quickly,²⁹ and sulfa was eclipsed by the arrival of penicillin in the 1940s. Here, too, the broad use of penicillin resulted in increasing gonococcal resistance^30,31 and a corresponding steady rise in the treatment dose, until the acquisition by N. gonorrhoeae of a plasmid carrying a beta-lactamase and its global spread resulted in high-level resistance that could not be overcome by increased penicillin dose.^32–34 These examples reflect the trend of the emergence of resistance seen with each of the antibiotics subsequently used for treatment, including tetracyclines, quinolones, macrolides, and cephalosporins (Table 1).

Table 1.

Mechanisms of antibiotic resistance in Neisseria gonorrhoeae and global phenotypic resistance.

Drug (class)	Drug Target (gene)	Target function	Resistance mechanisms 164,168	% isolates with phenotype in GISP dataset from U.S., 20185	% isolates with phenotype in EURO GASP dataset from Europe, 2017169	% isolates with resistance in AGSP dataset from Australia, 2017170
Sulfonamides (sulfonamides)	Dihydropterate synthase (folP) 168	Folate metabolism	Target modification	N/A	N/A	N/A
Penicillin (penicillins)	Penicillin-binding protein (PBP) 1 (ponA) and PBP2 (penA) 171	Cell wall biosynthesis	Target modification Permeability Efflux Inactivation	13.7%	14.9%*	26.1%**
Tetracycline (tetracyclines)	30S ribosomal subunit 172	Protein synthesis	Target modification Permeability Efflux	25.6%	N/A	10.2%
Ciprofloxacin (quinolones)	DNA gyrase (gyrA) and topoisomerase IV 173	DNA replication, unwinding, and topology	Target modification Efflux	31.2%	46.5%	27.5%
Spectinomycin (aminocyclitol)	30S ribosomal subunit 174	Protein synthesis	Target modification	N/A	N/A	0%
Azithromycin (macrolide)	50S ribosomal subunit 175	Protein synthesis	Target modification Efflux Inactivation	4.6%	7.5%	9.3%
Cefixime (cephalosporin)	PBP2 (penA) 47	Cell wall biosynthesis	Target modification	0.3%	1.9%	N/A
Ceftriaxone (cephalosporin)	PBP2 (penA) 47	Cell wall biosynthesis	Target modification Permeability	0.2%	0%	1.06%

Bold = % resistant, Underlined = % with elevated MIC, N/A = not tested.

^*= data from 2016 and measured by plasmid-mediated penicillinase production ¹⁷⁶.

^**= data represent both chromosomally mediated and penicillinase mediated resistance.

GISP = Gonococcal Isolate Surveillance Project, GASP = Gonococcal Antimicrobial Surveillance Programme, AGSP = Australian Gonococcal Surveillance Programme. GISP samples were assessed using the Clinical and Laboratory Standards Institute (CLSI) resistance breakpoints (penicillin resistance MIC ≥ 2.0 μg/mL, tetracycline resistance MIC ≥ 2.0 μg/mL, ciprofloxacin resistance MIC ≥ 1.0 μg/mL, azithromycin elevated MIC ≥ 2.0 μg/mL, cefixime elevated MIC ≥ 0.25 μg/mL, ceftriaxone elevated MIC ≥ 0.125 μg/mL).⁵ EURO GASP samples were assessed using European Committee on Antimicrobial Susceptibility Testing (EUCAST) resistance breakpoints (ciprofloxacin resistance MIC > 0.06 mg/L, azithromycin resistance MIC > 0.5 mg/L, cefixime/ceftriaxone resistance MICs > 0.125 mg/L).¹⁶⁹ AGSP samples were assessed based on AGSP resistance breakpoints (chromosomal penicillin resistance MIC ≥ 1 mg/L, tetracycline resistance MIC ≥ 16 mg/L, ciprofloxacin resistance MIC ≥ 1 mg/L, spectinomycin resistance MIC ≥ 128 mg/L, azithromycin resistance MIC ≥ 1 mg/L, ceftriaxone decreased susceptibility MIC ≥ 0.06 mg/L).¹⁷⁷

The two drugs most commonly recommended in current treatment guidelines are a macrolide, azithromycin, and a third-generation cephalosporin, ceftriaxone. CDC guidelines for most gonorrheal infections recommend dual one-time therapy with 250mg intramuscular (IM) ceftriaxone and 1g oral azithromycin. Prophylaxis in neonates for conjunctival gonorrhea include treatment with the macrolide erythromycin.³⁵ EU and Australian recommendations similarly emphasize dual therapy.^36,37 The British Association for Sexual Health and HIV recently changed its recommendations from dual therapy to a single IM dose of 1g of ceftriaxone, with an option for oral ciprofloxacin if the isolate is susceptible.³⁸ WHO guidelines recommend either dual therapy or monotherapy depending on local resistance.⁶ Treatment guidelines also vary by country-specific regulation and antibiotic production. The aminocyclitol spectinomycin, which is not currently available in the United States,³⁹ is used as first-line monotherapy in China.⁴⁰ Though spectinomycin has poor pharyngeal efficacy⁴¹ and resistance to spectinomycin appears to correlate with increased spectinomycin use in the population,⁴² lack of use in countries such as the United States as well as promising mouse infection data⁴³ suggest that spectinomycin may be an effective treatment for urogenital multidrug resistant Neisseria gonorrhoeae in certain populations.

Resistance to both macrolides and cephalosporins is on the increase. Azithromycin resistant strains have been identified around the world,⁴⁴ including outbreaks of highly azithromycin resistant gonococcus in the UK⁴⁵ and Hawaii.⁴⁶ One Japanese sampling study in 2001 found substantial cefixime and ceftriaxone resistance,⁴⁷ and recent cases of infections with ceftriaxone resistant isolates, and some also resistant to azithromycin, have appeared internationally.^46,48–50 One study of isolates in China found that 3.3% had both resistance to azithromycin (measured as a minimum inhibitory concentration (MIC) ≥ 1.0 mg/L) and decreased susceptibility to ceftriaxone (MIC ≥ 0.125 mg/L).⁵¹

Though data about antibiotic resistance are limited in many low- and middle-income countries (LMIC), it is thought that the prevalence of resistance is likely to be greater in LMICs due to limited access to new or alternative antibiotics and an increased burden of disease. Antibiotic resistance has a disproportionate human and economic cost in LMICs through a decreased capacity for infection control surveillance and a lack of access to expensive second-line drugs.⁵² A study of N. gonorrhoeae isolates from Kisumu, Kenya found that resistance to ciprofloxacin (MIC ≥ 1.0 mg/L) increased from 9.5% of isolates in 2007 to 50% of isolates in 2009. No cephalosporin resistance was observed. Nonetheless, despite the rare use of cephalosporins in this population, the mean MIC to ceftriaxone increased from 2002–2009.⁵³ In KwaZulu Natal, South Africa, 68% and 70% of clinical N. gonorrhoeae isolates from a study were found to be resistant to azithromycin (MIC > 0.5 mg/L) and ciprofloxacin (MIC > 0.06 mg/L) respectively.⁵⁴ Quinolone resistance appears to be even more prevalent in Bangkok, Thailand, as 92.4% of isolates in the Enhanced Gonococcal Antimicrobial Surveillance Programme Thailand were ciprofloxacin resistant (MIC ≥ 1.0 mg/L).⁵⁵ Combating antibiotic resistance in LMIC will require a global effort and has global implications,⁵² as strains that are resistant to first-line drugs can cross borders and continents.⁵⁶

A PERFECT DRUG

What constitutes a perfect drug for gonorrhea? It should have efficacy in multiple tissues and especially the pharynx, ideally in a single dose and via oral administration. One-time dosing allows for observed administration, though further follow-up may still be required. Given the global burden of disease and the disproportionate impact on economically disadvantaged populations, a new drug for gonorrhea should be inexpensive. A new drug should also be safe, especially in pregnancy. Finally, the drug should be stable across a broad range of temperatures to enable global distribution without requiring refrigeration. Other considerations that would be helpful but not necessary for a globally useful drug include safety in children, allowing for treatment of neonatal conjunctivitis, and few drug-drug interactions, allowing for combination therapy and concomitant treatment for other sexually transmitted infections.

IN WITH THE OLD…

While a number of antibiotics already approved by the FDA or other regulatory agencies have activity against N. gonorrhoeae, they are often less efficacious in human clinical trials than existing first-line agents (Table 2). Gentamicin, an aminoglycoside that targets the 30S ribosome, appeared to achieve 100% cure in combination with azithromycin, though very few isolates from this trial were azithromycin resistant or from extra-genital sites.⁵⁷ Gentamicin was also compared as a single injection against ceftriaxone in a large randomized controlled trial (RCT) in which patients with genital, pharyngeal, or rectal gonorrhea were randomized to drug treatment and assayed for cure by nucleic acid amplification test (NAAT). Gentamicin was not non-inferior to ceftriaxone, particularly with lower pharyngeal cure rates, though gentamicin did appear to lead to microbiological cure in 94% of cases of genital gonorrhea.⁵⁸ Recent studies have also indicated that gentamicin has poor pharyngeal efficacy,^58,59 possibly due to the poor tissue penetration common to aminoglycosides.⁶⁰ As with every antibiotic, the extent of resistance in the population also undermines effectiveness. While gentamicin does not currently have a defined breakpoint under Clinical and Laboratory Standards Institute (CLSI) criteria, the identification of isolates in the United States with an “Intermediate Susceptibility” phenotype of an MIC ≥ 8 μg/mL raises the concern that isolates with reduced susceptibility may already be circulating.⁶¹

Table 2:

Novel or repurposed therapies for treatment of N. gonorrhoeae

Antibiotic (Class) Route of administration	Protein target or mechanism of action	Stage of development and current status
Gentamicin (macrolide) Intramuscular	Peptidyl transferase of the 50S ribosome178	Approved by FDA Failed to show noninferiority to ceftriaxone in a recent RCT with 94% clearance at genital sites in the intention-to-treat group vs. 98% for ceftriaxone58 Large-scale clinical trial currently underway comparing gentamicin, ertapenem, and ceftriaxone alone66
Delafloxacin (quinolone) Oral	DNA gyrase A subunit (gyrA) and, to a lesser extent, topoisomerase IV (parC)179	Approved by FDA in 201767 Failed to show noninferiority to ceftriaxone in an RCT with 85% clearance at genital sites in the microbiological intention-to-treat group vs. 91% for ceftriaxone68
Fosfomycin (fosfomycin) Oral	UDP-GlcNAc enolpyruvyl transferase (murA)180	Approved by FDA Small-scale clinical trial showed efficacy in vivo with 96.8% clearance at genital sites in the per-protocol group vs. 95.3% for ceftriaxone plus azithromycin65
Sitafloxacin (quinolone) Oral	DNA gyrase A subunit (gyrA) and, to a lesser extent, topoisomerase IV (parC)179	Approved in Japan74 Effective in vitro69
Solithromycin (fluoroketolide) Oral	50S ribosomal subunit, activity in macrolide resistant strains78	Phase 3 RCT did not show noninferiority to ceftriaxone plus azithromycin80
Gepotidacin (triazaacenaphthylene) Oral	DNA gyrase A subunit (gyrA) and topoisomerase IV (parC) at a different site than quinolones82	Promising phase 2 clinical trial 83 Phase 3 clinical trial is currently underway85
Zoliflodacin (spipyrimidenitrone) Oral	DNA gyrase B subunit86,87	Promising phase 2 clinical trial89 Phase 3 clinical is currently underway90
Lefamulin (pleuromutilin) Oral	50S ribosomal subunit at the acceptor (A) and peptidyl (P) sites91	Approved by FDA in 201994 Effective in vitro95
Fatty Acids Topical	Possibly disrupts the bacterial lipid bilayer98	Effective in vitro97,98
Antiseptic Mouthwash Oral	Multiple, including disruption of the bacterial cell membrane181	Effective in vitro and in a small RCT103 Large, double-blind RCT is currently underway104

Ertapenem, a carbapenem antibiotic, has been shown to treat extensively drug resistant gonorrheal infections in a clinical setting,⁶² but has not been studied in a clinical trial.⁶³ Fosfomycin, a drug that targets peptidoglycan synthesis, has been shown in small-scale clinical trials to be effective in treatment of urogenital gonorrhea.^64,65 A new clinical trial, NABOGO (New AntiBiotic Treatment Options for Uncomplicated Anogenital GOnorrhea), was recently undertaken to investigate the first-line efficacy of IM ertapenem, IM gentamicin, oral fosfomycin, and IM ceftriaxone as single agents in patients with anogenital gonorrhea, though the fosfomycin arm was dropped in late 2018.⁶⁶

Delafloxacin, a fluoroquinolone approved by the FDA in 2017,⁶⁷ was examined as a single oral dose against ceftriaxone in genital gonorrhea. A total of 430 patients were randomized and assessed for microbiological cure by negative culture. The trial showed that delafloxacin was not noninferior to ceftriaxone, with treatment failure correlating with a higher delafloxacin MIC.⁶⁸

As early as 2000, the fourth-generation quinolone sitafloxacin was shown to have activity against Neisseria gonorrhoeae in an in vitro setting.⁶⁹ Further studies showed that sitafloxacin was far more potent than other quinolones and that mutations in DNA gyrase (gyrA) or topoisomerase IV (parC) that caused high ciprofloxacin resistance led to only moderate sitafloxacin resistance.^70,71 Sitafloxacin may thus be a valuable tool in either dual or monotherapy as a more potent fluoroquinolone.^72,73 However, it faces major hurdles before it could be considered as a treatment for gonorrhea on a global scale. As of 2017, sitafloxacin was only approved in Japan,⁷⁴ with one major barrier to approval outside of Asia being the UVA phototoxicity in Caucasian but not Asian subjects.^75,76 An additional concern is that resistance will emerge quickly, building on the existing wide dissemination of strains with resistance to other quinolones.⁷⁷

…AS WELL AS THE NEW

A number of drugs in development show promise, though challenges have arisen as several have moved through to clinical trials (Table 2). Solithromycin is a fluoroketolide, a derivative of macrolide antibiotics, that has an alkyl-aryl side chain that allows for activity against strains with macrolide resistant modifications to the 50S ribosomal subunit.⁷⁸ A phase 2 trial of single oral dose solithromycin showed positive results, with apparent efficacy at urogenital, rectal, and pharyngeal sites of infection as measured by negative bacterial culture. Patients experienced GI toxicity manifesting as nausea, vomiting and loose stool, but these adverse events decreased at a lower and still efficacious, dose.⁷⁹ Phase 3 findings, however, were less promising. The authors compared single dose PO solithromycin to the standard dual therapy of IM ceftriaxone and oral azithromycin. A total of 262 patients with genital gonorrhea were randomized to treatment with a primary outcome based on negative genital culture after one week. In an evaluation of patients who were not lost to follow-up and did not take other antibiotic therapy, 100% of patients in the dual therapy had negative genital cultures, compared to 92% in the solithromycin group (difference −7.6%, 95%CI −14.3 to −3.9). The solithromycin group also had higher rates of gastrointestinal adverse events, limiting the possibility of increasing dosage. Based on these results, solithromycin cannot be considered noninferior to dual therapy with ceftriaxone plus azithromycin,⁸⁰ and this, together with questions around liver hepatotoxicity, have stalled efforts to gain approval.⁷³

Gepotidacin is the founding member of a class of antibiotics known as the triazaacenaphthylenes and has broad in vitro activity against Gram-positive and Gram-negative organisms,⁸¹ Although gepotidacin, like quinolones, targets type IIA topoisomerases, including DNA gyrase (gyrA) and topoisomerase IV (parC), it binds at a distinct target.⁸² In clinical studies, gepotidacin was relatively safe and effective against gonococcus. In a phase 2 trial with gepotidacin, 106 patients with urogenital gonorrhea were randomized to two different doses of oral gepotidacin in a single administration. At both concentrations, gepotidacin cured ~95% of patients as measured by negative culture at follow-up. This high cure rate occurred in a background of 33% of strains phenotypically resistant to ciprofloxacin. Treatment failures arose on therapy, due to a GyrA A92T mutation and thought to be a second step mutation in the presence of a ParC D86N mutation.^83. Adverse events from gepotidacin treatment appeared to be primarily related to gastrointestinal symptoms and were not treatment-limiting.⁸⁴ A phase 3 clinical trial was commenced in October of 2019 comparing two dose oral gepotidacin (one dose at baseline followed by self-administration 6–12 hours later) to ceftriaxone plus azithromycin.⁸⁵

Zoliflodacin, a member of a new class of antibiotics known as spiropyrimidenitriones, is another novel antibiotic that binds bacterial topoiomerases. Zoliflodacin also appears to have a novel mechanism of action to the fluoroquinolones despite targeting the same proteins. This is borne out by in vitro resistant mutants that carry GyrB, rather than GyrA, mutations and the fact that there is little cross-resistance with fluoroquinolones.^86,87 In vitro, zoliflodacin appears to have substantial activity against multidrug resistant strains of gonococcus.⁸⁸ As an oral antibiotic for gonorrhea, zoliflodacin has received an FDA “fast track” designation.⁸⁹ From a clinical standpoint, zoliflodacin has been relatively successful. In a phase 2 trial, 179 patients with urogenital gonorrhea were randomized to single dose oral zoliflodacin or single dose IM ceftriaxone. Clearance was assessed by negative culture at a follow-up visit. Microbiological cure rates were >95% for two different doses of zoliflodacin, as well as for ceftriaxone. Ceftriaxone was found to be superior for treatment of pharyngeal gonorrhea. Adverse events were minimal and were mainly gastrointestinal.⁸⁹ A phase 3 clinical trial is currently recruiting for single dose oral zolifodacin as compared to ceftriaxone plus azithromycin.⁹⁰

Lefamulin is a semisynthetic pleuromutilin that appears to target the 50S ribosome via the A and P sites.⁹¹ Multiple phase 3 clinical trials have shown that lefamulin is noninferior to the standard of care in community acquired pneumonia^92,93 and the drug was approved by the FDA in August, 2019.⁹⁴ In vitro studies show that lefamulin appears to be highly active against N. gonorrhoeae with little cross-resistance to other antibiotics,^95,96 suggesting that it may be a suitable treatment for gonorrhea.

There are also a number of promising therapies that have been shown to have activity against gonorrhea in vitro. One of the most interesting focuses on the use of fatty acids as topical treatment for neonatal gonorrheal conjunctivitis. As gonococcus is known to be sensitive to multiple fatty acids,⁹⁷ a number of fatty acids have been probed in cell culture infection models, as well as in an eye irritant model.⁹⁸ Fatty acids may therefore not only be significant for prophylaxis and treatment of the conjunctiva, but also mucosal surfaces in the pharynx and cervix. Other novel drug strategies include targeting efflux pumps that mediate multidrug resistance and thereby re-sensitizing N. gonorrhoeae to other antibiotics. MtrCDE, an efflux pump in N. gonorrhoeae that is known to confer resistance to many antimicrobial compounds⁹⁹ and protect from immune destruction,¹⁰⁰ is a particularly tempting target, as decreasing or knocking out the MtrCDE pump leads to increased susceptibility to multiple antibiotics.^101,102

Mouthwash may be an important tool to prevent or treat pharyngeal gonorrhea. In vitro, the common mouthwash brand Listerine Cool Mint was shown to inhibit N. gonorrhoeae.¹⁰³ From a clinical standpoint, men with pharyngeal gonorrhea that were randomized to use Listerine Cool Mint were significantly less likely to be culture positive at the end of a trial than men treated with saline.¹⁰³ A large, double-blind RCT called the OMEGA (Oral Mouthwash use to Eradicate GonorrhoeA) trial was recently undertaken to compare two different mouthwash regimens, one of which has been shown to have in vitro efficacy.¹⁰⁴ By one model, an efficacious mouthwash could halve the prevalence of oropharyngeal, rectal, and urethral gonorrhea in the MSM community.¹⁰⁵ There are many questions surrounding mouthwash use and gonorrhea, including what mouthwash composition would be most effective and whether mouthwash should be used as an intervention after unprotected oral sex.¹⁰⁶

VACCINOLOGY

Vaccines, in theory, surpass even the ideal drug for gonorrhea. A vaccine can be safe, cheap, stable, and efficacious against a broad range of strains, but provides the added benefit of preventing disease morbidity and reducing transmission.

Though vaccines for gonococcus have been studied since the early 20^th century, there has been little success in generating immune protection. Through the early 1990s, only four vaccines were introduced into clinical trials,¹⁰⁷ ranging from a whole cell vaccine¹⁰⁸ to a vaccine that attempted to elicit an immune response against the gonococcal pilus.⁴² None of these vaccines conferred protection against infection.¹⁰⁷

Recent evidence suggests that the Neisseria meningitidis serogroup B vaccine, however, may offer some protection against gonorrhea. In contrast with the polysaccharide vaccines for N. meningitidis serogroups A, C, Y, and W-135,¹⁰⁹ the cross-reactivity between the meningococcal serogroup B capsule polysialic acid and human neural tissues required alternative vaccine targets.¹¹⁰ The 4CMenB vaccine, a rationally designed vaccine containing three recombinant proteins, two fusion proteins, and outer membrane vesicles (OMVs) from a pathogenic strain of meningococcus¹¹¹ induced protective antibodies in large-scale phase 3 trials in infants.¹¹² In further studies, it appeared to be effective in preventing meningococcal disease in infants in England,¹¹³ in Canada,¹¹⁴ and on college campuses in the United States.¹¹⁵ Intriguingly, early work on the meningococcal vaccine antigens indicated that the protein targets were conserved in N. gonorrhoeae,¹¹⁶ though the variants were hypothesized to be dissimilar enough to prevent substantial cross-protection.¹¹⁷

An OMV meningococcal vaccine appears to decrease the incidence of clinical diagnosis with gonorrhea (adjusted odds ratio 0.69, 95% CI 0.61–0.79) – the first evidence of clinical protection from gonorrheal infection.¹¹⁸ Similar cohort studies found that the same vaccination decreased the number of hospitalizations for gonorrhea.¹¹⁹ A vaccine consisting of gonococcal OMVs combined with an IL-12 adjuvant appeared to provide protection in a mouse model of cervical infection.¹²⁰ Other candidate vaccine approaches include an oligosaccharide epitope¹²¹ that decreased the time to clearance in a mouse genital model of infection¹²² and reduced colonization.¹²³ A number of other possible vaccine antigens have also been identified.¹²⁴

In sum, the spillover success of the meningococcal vaccines has invigorated the search for a broadly effective gonococcal vaccine. The findings of cross-reactivity with the meningococcal vaccine has led to excitement that development of an effective vaccine for gonorrhea may be imminent.^125,126 Substantial work will be necessary to validate that candidate vaccines protect against a diverse set of pathogenic gonorrhea.

DIAGNOSTICS

An ideal diagnostic for gonorrhea shares many of the same characteristics as a vaccine or an antibiotic. It should be inexpensive, both in terms of start-up costs and daily use. A diagnostic test should be easily interpretable by a non-expert and should be suitable for diagnosing gonorrhea at multiple body sites, allowing for effectiveness in resource-limited settings. The greatest utility would be gained from a point-of-care diagnostic that reliably and quickly determined antibiotic resistance, enabling a medical provider to provide a tailored antibiotic therapy.

Currently used diagnostic tests fall short in many of these categories. The classic diagnostic is microscopic identification of Gram-negative diplococci within polymorphonuclear leukocytes. Though this has a sensitivity and specificity greater than 90% for urethral specimens, the sensitivity falls to less than 70% for cervical specimens. Furthermore, the test relies on the specific skill of the microscopist.¹²⁷ While the gold-standard of diagnosis is bacterial culture, this test also has limitations. This method allows for testing of antibiotic resistance and sampling multiple sites, but often requires overnight growth or longer to determine definitive susceptibilities. In addition, culturing methods depend greatly on the method of sample transport¹²⁸ and have lower sensitivity as compared to nucleic acid amplification tests (NAATs) for detection of rectal and pharyngeal gonorrhea.¹²⁹

NAATs have become the most commonly used diagnostic in higher income settings.¹³⁰ NAATs provide rapid detection of Neisseria gonorrhoeae at multiple sites including the urogenital region,¹³¹ the rectum,¹³² and the oropharynx.¹³³ NAAT methods, however, do not detect antibiotic susceptibilities, necessitating bacterial culture to determine phenotypic resistance.

There is an urgent need for point-of-care (POC) diagnostic tests that detect antimicrobial resistance (POC-AMR).¹³⁴ Current approaches have focused on developing an amplification-based approach to predicting resistance phenotype from genotype. As ciprofloxacin susceptibility is predicted by the amino acid at GyrA 91 with high sensitivity and specificity, it is an ideal target. A diagnostic called SpeeDx ResistancePlus GC assay, targeting this site, was recently approved in Europe for detection of ciprofloxacin resistance and has received an FDA breakthrough designation.¹³⁵ Numerous other start-up and established diagnostics companies have explored similar approaches. GeneXpert is a modular diagnostic platform that uses real-time PCR to assay for specific sequences in bacteria and was approved by the FDA in 2012. In total, the diagnostic process takes less than two hours.¹³⁶ GeneXpert could detect gonococcus in urogenital and urine samples¹³⁶ as well as rectal samples,¹³⁷ and could distinguish between Neisseria gonorrhoeae and other Neisseriae species.¹³⁸ GeneXpert protocols have been developed to assay for rifampin resistance in Mycobacterium tuberculosis,¹³⁹ but there have not yet been assays developed to look at gonococcal resistance markers.¹³⁴ Another similar method being investigated has been developed by Binx Health (formerly Atlas Genetics). The platform, known as Velox, can utilise both molecular and immunoassay tests – as one example, the chlamydia diagnostic is effectively a NAAT that produces results in less than 25 minutes.¹⁴⁰ The company is currently investigating the development of an assay for fluoroquinolone resistance.¹⁴¹

However, there are substantial challenges to genotype-based resistance prediction in gonococcus. For most antibiotics other than quinolones, resistance arises through multiple pathways, which may have additive or synergistic effects. This poses methodological challenges,¹³⁴ including amplification strategies that have low positive predictive values.¹⁴² Fundamentally, genomic-based tools will only be as good as the understanding of the underlying genetic causes of antibiotic resistance and the prevalence of these known genes and variants. Current POC platforms may also be of limited utility in resource-poor settings, as they require a start-up cost, regular maintenance, and trained staff.

A tantalizing approach to characterizing the genetic basis of resistance is the use of machine learning. Over the past several years, a number of papers have attempted to use machine learning to characterize antibiotic resistance in a vast diversity of bacteria, including Mycobacterium tuberculosis,¹⁴³ Pseudomonas,¹⁴⁴ nontyphoidal Salmonella,¹⁴⁵ and species that make up the microbiome.¹⁴⁶ Despite this promise, machine-learning methods face important limitations, including variability in model success between strains, between antibiotics within a strain, and within an antibiotic based on the definition of resistance.¹⁴⁷

One notable avenue for future work is in rapid phenotypic testing. A recent paper suggests that transcriptional signatures based on either whole transcriptome analysis or RT-qPCR would be an effective method for identifying azithromycin susceptibility across diverse strains of N. gonorrhoeae.¹⁴⁸ This method could also be efficient, as clinical strains were able to be characterized as susceptible or resistant after only ten minutes of exposure to ciprofloxacin.¹⁴⁹ In other bacteria, assays that combine genomic data with RNA-seq have been shown to identify antibiotic resistant strains from blood culture within four hours.¹⁵⁰ RNA may prove to be a key component of any future phenotypic rapid POC diagnostic.

Substantial work is therefore still needed for a rapid POC diagnostic that can also diagnose resistant isolates. Advances in engineering, biological understanding, and medical infrastructure must happen in parallel for such a diagnostic to be effective in combating drug resistance.

IMPLEMENTATION

How should novel diagnostics, vaccines, and antibiotics (Table 3) be implemented to achieve the dual goals of reducing the burden of disease and limiting the emergence and spread of antibiotic resistance? The complexities of modeling implementation are multifold¹⁵¹ and are especially evident in determining the population level effects of a novel diagnostic. Development of a diagnostic that provides information about antibiotic resistance appears to decrease overall resistance in the population based on mathematical modeling¹⁵² and has the potential to extend the lifetime of current drugs.¹⁵³ In contrast, implementation of a rapid POC diagnostic without information about resistance appears to increase overall resistance in a model. This possibly counterintuitive result is explained by fewer asymptomatic patients being lost to follow-up, leading to increased antibiotic treatment and therefore increased undetected resistance.¹⁵² In another modeling study, use of a POC diagnostic that only reports on susceptibility to one antibiotic may slow the spread of resistance in the short term but result in more multidrug resistance in the long term.¹⁵⁴ This suggests that POC diagnostics should aim to define susceptibilities for more than one antibiotic. Additional modeling studies will be needed in a shifting landscape of diagnostic capabilities and available therapies. As we gain further understanding of the dynamics and patterns of gonorrhea in the large parts of the world where little surveillance data exist currently, the applicability of the models’ conclusions will also need to be evaluated.

Table 3:

Approaches to combating antibiotic resistance

New therapies

Antibiotics approved for other indications

Novel antibiotics with new mechanisms of action

Topical antibiotic treatment

New preventive vaccines

OMV-based vaccines

Oligosaccharide-based vaccines

Peptide antigen-based vaccines

New diagnostics

DNA-based approaches

SpeeDx ResistancePlus GC assay

GeneXpert

Velox

Machine learning predictive approaches

RNA-based approaches

Vaccine implementation for gonorrhea has also been investigated through mathematical modeling. One model suggests that a non-waning vaccine providing 20% efficacy would decrease gonorrheal prevalence by 40% in a population of 100,000 heterosexual individuals.¹⁵⁵ Though this model is limited by only evaluating the heterosexual population, it does suggest that even a weakly efficacious vaccine could prevent a substantial amount of disease. A workshop of academics, members of industry, and government agencies reached the consensus that universal vaccination of adolescents would be the most likely real-world strategy for vaccination, though members also noted that far more work was needed to show cost-effectiveness.¹⁵⁶ Other strategies suggested have included a targeted vaccination of high risk individuals, particularly in areas with low disease prevalence.¹²⁴ Vaccines would also face the challenge of social acceptance in some settings, with objections likely to echo those raised in response to HPV vaccination.^157–159 Work on how to promote the acceptance and uptake of STI vaccination is critical.

In the case of a new drug, how should it best be used and deployed? Any new drug will generate resistance, as seen with all previous antibiotics. N. gonorrhoeae has many traits that are conducive for the development of antibiotic resistance, including asymptomatic infection and high transmissibility.¹⁶⁰ On a population-level, antibiotic consumption correlates with resistant isolates of gonococcus, with contributions from both direct treatment of gonorrhea and asymptomatic gonorrhea being exposed to antibiotics used for other indications.^161,162 If the former is predominant, then introduction of a new drug in combination with other drugs may slow this development. If the latter is the primary mechanism, then the new drug should be used strictly in gonorrheal cases to decrease the likelihood of resistance.¹⁶²

All three of these strategies will require a systematic program of phenotypic and genotypic characterization of gonococcal strains. Recent work has modeled genomic sampling strategies for discovering novel mechanisms of resistance.¹⁶³ Such a surveillance strategy has implications for genotype-dependent diagnostics, clinical care, and epidemiology through mapping of sexual networks.¹⁶⁴

In conjunction with improved surveillance, prevention and control interventions are critical tools for combating antibiotic resistance. Partner notification may prove to be an important method of preventing onward transmission, and a randomized trial (“Safetxt”) is currently underway investigating the use of digital tools in preventing STD spread.¹⁶⁵ Technological advancements and improved health access are essential components to a behavioral approach, as even in the developed world, it is difficult to properly implement behavioral interventions.¹⁶⁶ Behavioral-cognitive approaches decrease risky behavior and rates of infection with gonorrhea in a targeted population,¹⁶⁷ and provide another option to reduce disease prevalence and thereby antibiotic exposure on a population-level. Finally, further research and modeling is needed to measure the economic cost of resistance and inform priorities for future investment.

Conclusion

Advances in multiple strategies for addressing the threat of antibiotic resistance in N. gonorrhoeae offer hope for a new era in clinical and public health gonorrhea control. New, rapid, and cheap diagnostics should allow for point-of-care resistance testing and enable individually tailored therapy. Novel antibiotics will expand our repertoire of anti-gonococcal therapies and have the potential to delay the development of resistance to other antibiotics. A vaccine could prevent outbreaks of gonorrhea or limit the severity of disease; though disease eradication seems unlikely, an effective vaccine will be a critical factor in the effort to reduce gonococcal morbidity.

This promising future is inextricably connected to the harsh realities of fighting a global disease. A substantial portion of scientific focus has been on the developed world, where drugs, diagnostics, and vaccines may be profitable. The greatest need, however, lies in parts of the world where the disease – genotype, phenotype, and pathogenesis – is understudied. Researchers must ensure that their work is relevant and available to affected populations around the world. As resistance spreads, it is our responsibility to ensure that basic science and methodology come together to combat gonorrhea efficiently, effectively, and equitably.