Gepotidacin for the Treatment of Uncomplicated Urogenital Gonorrhea: A Phase 2, Randomized, Dose-Ranging, Single-Oral Dose Evaluation

1Gonorrhea is a sexually transmitted infection caused by Neisseria gonorrhoeae (NG), with 78 million infections reported globally in 2012. In the United States, approximately 468500 gonococcal infections were reported in 2016, an 18.5% increase from 2015, while in Europe, approximately 66000 gonococcal infections were reported in 2014, a 25% increase from 2013. Consequences of untreated gonococcal infections include pelvic inflammatory disease, infertility in women and men, ectopic pregnancy, tubo-ovarian abscess, neonatal conjunctivitis, and disseminated gonorrhea.

2Over the past few decades, NG has demonstrated the ability to develop resistance to most antibiotics recommended or used for treatment, suggesting the possibility of untreatable gonorrhea in the future. The Centers for Disease Control and Prevention and the World Health Organization have labeled drug-resistant NG with threat levels of urgent and high, respectively, and identified a critical need for new antibiotic treatments. Current guidelines recommend first-line dual antibiotic therapy that consists of intramuscular ceftriaxone combined with oral azithromycin. Although this approach may delay the emergence of cephalosporin-resistant NG, the impending threat remains. Surveillance data demonstrate increasing minimum inhibitory concentrations (MICs) for NG to third-generation extended-spectrum cephalosporins and azithromycin in the United States, Canada, and Europe. Several therapeutic options are being evaluated for the treatment of gonorrhea, including dual combinations of established and novel antibiotics.

3Gepotidacin (GSK2140944) is a novel triazaacenaphthylene bacterial type II topoisomerase inhibitor that is currently in development. Gepotidacin selectively inhibits bacterial DNA replication by interacting in a unique way on the GyrA subunit of bacterial DNA gyrase and the ParC subunit of bacterial topoisomerase IV. Gepotidacin has in vitro activity against pathogens resistant to established antibacterials, including ciprofloxacin-resistant and -susceptible strains of NG. In this study, we evaluated the efficacy of single, oral doses of gepotidacin for treatment of uncomplicated urogenital gonorrhea.

4This phase 2, randomized, multicenter, open-label, dose-ranging study was conducted at 12 sites (11 US sites and 1 UK site) from April 2015 to August 2016 to evaluate the efficacy, safety, tolerability, and plasma exposures of 2 single oral doses of gepotidacin in uncomplicated urogenital gonorrhea. Throughout the study, real-time sponsor-blinded reviews of safety data and NG isolate identification and susceptibilities by GlaxoSmithKline review teams were conducted. Additionally, an independent review team had the authority to recommend discontinuation of a dose, continue randomization as planned, or end the trial due to success or futility (Figure 1). The study was conducted in accordance with the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use, Good Clinical Practice, and applicable country-specific requirements, including institutional review board and ethics committee approvals. Participants provided written, informed consent before any study procedures were performed.

5Men and nonpregnant, nonlactating women aged ≥18 years who had suspected uncomplicated urogenital gonorrhea, defined as the presence of purulent urethral or cervical discharge upon physical examination and a prior culture or nucleic acid amplification test (NAAT) positive for NG, a Gram stain positive for gram-negative diplococci from male urethral specimens, or reported sexual contact with a partner diagnosed with gonorrhea within the past 14 days were enrolled.

6Exclusion criteria included a body mass index ≥40.0 kg/m2; hysterectomized women without a cervix; men diagnosed with epididymitis or orchitis; a medical condition or required medication that may have been aggravated by acetylcholinesterase inhibition; diagnosis of Clostridium difficile (CD) infection; liver disease; prespecified baseline electrocardiogram (ECG) abnormalities and cardiovascular conditions; disseminated gonococcal infections; antibiotic treatment within 14 days; or taking medications with a known risk of torsades de pointes. Participants agreed to abstain from or to use a male condom for any sexual activity from baseline through test-of-cure (TOC) to prevent reinfection.

7The total study duration was approximately 1 week with 2 study visits: baseline (day 1) and TOC (days 4–8). At baseline, a physical examination, assessment of vital signs, clinical laboratory tests, and ECGs were performed. Baseline microbiology procedures included collection of pretreatment urogenital swab specimens for NG culture from all participants; a Gram stain was performed on male urogenital specimens. From areas of known exposure, pretreatment pharyngeal and rectal swab specimens for NG culture were collected. Pretreatment urogenital swab or urine specimens for detection of NG and Chlamydia trachomatis (CT) by NAAT assay were obtained at baseline only. Eligible participants were stratified by gender and randomized 1:1 to receive a 1500-mg (3 × 500-mg capsules) or 3000-mg (6 × 500-mg capsules) single oral dose of gepotidacin administered open label with food at the clinic (Figure 1). Dose selection was based on in vitro and in vivo data and pharmacokinetic modeling, which indicated the low and high doses would provide systemic exposures to cover urogenital NG isolates with gepotidacin MICs of 0.5 µg/mL and 1 µg/mL, respectively. At 2 hours postdose, ECG and blood pharmacokinetic sample collection were performed. Based on blinded, real-time assessment of systemic exposures and ECG data from the first 37 participants, the sponsor safety review team determined postdose ECG and pharmacokinetic requirements could be safely removed.

8Participants returned to the clinic at TOC for safety procedures similar to baseline and adverse event (AE) and concomitant medication data collection. At TOC, posttreatment urogenital swab specimens were collected for NG culture (and Gram stain for male urogenital specimens) from all participants who were positive for NG at baseline. Posttreatment pharyngeal and rectal swab specimens were also collected from anatomical sites that had a positive baseline NG culture. Standard-of-care treatment for CT was allowed after all TOC procedures were complete.

9Gram stain, NG culture, and presumptive organism identification were performed at local laboratories. Recovered isolates of presumptive NG were sent to central laboratories (Q2 Solutions, Valencia, California, and University of Alabama, Birmingham) for confirmation of identification and susceptibility testing by agar dilution based on Clinical and Laboratory Standards Institute guidelines. Microbiological outcome (bacterial eradication, bacterial persistence, or unable to determine) and response (microbiological success or failure) for NG at TOC were determined based on predefined criteria. NG and CT NAATs were performed using the US Food and Drug Administration–cleared commercially available methods at local laboratories.

10Participants were randomized into the study until approximately 30 microbiologically evaluable (ME) participants were included in each treatment arm (defined as all randomized participants who had NG isolated from baseline cultures of urogenital swab specimens, received either dose of gepotidacin, and returned for TOC). Assuming the true cure rate was 95%, a sample size of 30 ME participants was expected to provide at least 80% power to detect a difference of 15% (ie, cure rate of 95% under the alternative hypothesis and ≤80% under the null hypothesis) using a 1-sided binomial test at the 0.05 significance level.

11The primary efficacy endpoint was the culture-confirmed bacterial eradication of urogenital NG at TOC (microbiological success). The null hypothesis (cure rate ≤80%) was tested using the binomial test for each dose level. To control the type I error rate of the final analysis at <0.05, a closed-testing procedure was used that required establishing significance at the higher dose before assessing the significance at the lower dose. The 2-sided 90% exact confidence interval (CI) was produced to reflect the 1-sided α = 0.05. If the 95% 1-sided CI for a dose level included 80%, then the null hypothesis was not rejected for that dose level. Subgroup analyses for age (18–64, 65–74, and ≥ 65 years) and gender were also performed. All analyses were conducted using SAS software, version 9.2 (SAS Institute, Inc., Cary, North Carolina).

12Safety was assessed in all randomized participants who received either dose of gepotidacin. The safety secondary endpoint was evaluated by the analysis of AEs, including predefined cardiovascular and gastrointestinal AEs of special interest, vital sign assessments, laboratory values, physical examinations, and ECG parameters. The exploratory endpoints were bacterial eradication of pharyngeal and rectal NG when positive at baseline, microbiological characteristics of the isolates recovered, and a descriptive statistical summary of gepotidacin plasma concentrations. Gepotidacin plasma concentrations were also compared with phase 1 data from healthy volunteers.

13A total of 106 participants were randomly assigned to 1500 mg (n = 53) or 3000 mg (n = 53) gepotidacin (Figure 2). One randomized participant (1500 mg) did not receive study treatment and 37 participants did not have NG isolated at baseline; thus, the ME population consisted of 69 participants (30 participants in the 1500-mg group and 39 in the 3000-mg treatment group).

14The majority of participants were men (95%) and black (48%) with a mean age of 33.3 years; only 5 women were enrolled (Table 1). Based on NAAT results, 68 of 106 participants (64%) were positive for NG, 9 participants (8%) were positive for CT, and 4 (4%) were positive for both. The majority of Gram stain results for male urogenital specimens revealed the presence of white blood cells with gram-negative diplococci. In the ME population, NG isolates were identified in 69/69 (100%) urogenital, 2/69 (3%) pharyngeal, and 3/69 (4%) rectal specimens (Table 2). Of the urogenital NG isolates, 97% (67/69) were from men and 3% (2/69) were from women. Pharyngeal and rectal baseline NG isolates were recovered from men only.

15The primary protocol-defined endpoint was met overall and in both treatment groups (Table 3).

16Microbiological success for urogenital gonorrhea was achieved by 97%, 95%, and 96% of participants (lower 1-sided exact 95% CI bound, 85.1%, 84.7%, and 89.1%, respectively) for the 1500-mg, 3000-mg, and combined dose groups, respectively. The ME population included 2 women, 1 in each treatment group, both of whom were microbiological successes for urogenital NG. Three participants (4%; 1 participant who received 1500 mg and 2 participants who received 3000 mg) were microbiological failures.

17Analysis of subgroups was limited because the majority of study participants were in the age group of 18 to 64 years and were men. Only 2 participants had pharyngeal NG at baseline, both in the 3000-mg treatment group; 1 participant was a microbiological success and the other participant was a microbiological failure (Table 3). All 3 participants with rectal NG at baseline were microbiological successes, with 1 participant in the 1500-mg treatment group and 2 participants in the 3000-mg treatment group (Table 3). All microbiological failures received standard-of-care ceftriaxone 250 mg intramuscularly and azithromycin 1 g orally per national treatment guidelines.

18In the ME population, of the 69 baseline urogenital NG isolates, 23/69 (33%), 19/69 (28%), and 14/69 (20%) were resistant to ciprofloxacin, penicillin, and tetracycline, respectively (Table 4). All of the baseline urogenital NG isolates were susceptible to cefixime, ceftriaxone, and spectinomycin.

19In the ME population, gepotidacin was active in vitro against the 69 urogenital NG isolates recovered, with overall MIC 50% and 90% values of 0.12 µg/mL and 0.5 µg/mL, respectively (Table 5), and MICs ranging from ≤0.06 to 1 µg/mL. Gepotidacin MICs ranged from ≤0.06 to 0.25 µg/mL against the pharyngeal (n = 2) and rectal (n = 3) NG isolates recovered.

20All 3 participants who were urogenital microbiological failures had baseline NG isolates with a gepotidacin MIC of 1 µg/mL, which was the highest baseline gepotidacin MIC value observed in this study. Subsequent sequencing of the quinolone resistance-determining region (QRDR) of GyrA and ParC in these isolates revealed that all 3 isolates were quinolone resistant and had a preexisting D86 substitution due to a mutation in the parC gene, which is known to affect gepotidacin binding. Resistance also emerged between baseline and TOC for 2 of these urogenital microbiological failures, both in the high-dose 3000-mg treatment group. The gepotidacin MIC for both NG isolates from these 2 failures increased from 1 µg/mL to ≥32 µg/mL between baseline and TOC, and both TOC isolates were found to have an additional A92T substitution due to a mutation in the gyrA gene. For the 1 pharyngeal microbiological failure, the NG isolate had no change in gepotidacin MIC from baseline to TOC (0.12 µg/mL at both time points) and no mutations were observed in the QRDR of GyrA or ParC.

21There were no treatment-limiting AEs for either gepotidacin dose. Adverse events occurred in 27 of 52 (52%) and 34 of 53 (64%) participants in the 1500-mg and 3000-mg treatment groups, respectively (Table 6). No AEs led to study withdrawal and no deaths or serious AEs were reported. The most frequently reported AEs (>10% total) were diarrhea (27%), flatulence (23%), abdominal pain (15%), and nausea (13%). For nausea and diarrhea, the comparative relative risk 95% CIs were >1, suggesting a higher risk in the 3000-mg treatment group for these 2 AEs. The majority of AEs were mild to moderate in intensity, with 2 severe AEs, 1 each of flatulence (1500 mg) and dizziness (3000 mg).

22Cardiovascular AEs were reported in 2 participants (ECG ST segment elevation and palpitations) in the 1500-mg treatment group and 1 participant (tachycardia) in the 3000-mg treatment group. Investigators considered only the mild tachycardia as treatment related in a participant who had no underlying cardiovascular history. There were no dose-related trends observed for clinical laboratory evaluations, vital signs, or ECGs.

23The mean gepotidacin plasma concentrations (coefficient of variations) for the 2-hour postdose time point were 2.89 µg/mL (62.2%) and 6.35 µg/mL (42.9%) for the 1500-mg (n = 34) and 3000-mg (n = 36) single oral doses of gepotidacin, respectively, which was a dose proportional increase. A comparison of gepotidacin plasma concentrations between participants in the present study and phase 1 healthy volunteers showed comparable 2-hour postdose concentrations (Figure 3).

24As resistance to established antibiotics continues to rise for NG, new therapeutic options are needed to treat gonococcal infections. Gepotidacin is in development for the treatment of uncomplicated urogenital gonorrhea and is a first-in-class triazaacenaphthylene antibiotic with a novel mechanism of action. Gepotidacin given alone as a single oral dose (3 or 6 capsules) was ≥95% effective for bacterial eradication of culture-proven uncomplicated urogenital gonorrhea in this study. Gepotidacin may prove to be a therapeutic alternative for the treatment of uncomplicated gonorrhea. NG can demonstrate various resistance mechanisms. To maintain the efficacy of gepotidacin and to help preserve the efficacy of other antibacterials, different dosing strategies for gepotidacin, including dual-agent therapy, may ultimately be needed.

25For baseline urogenital NG isolates with gepotidacin MICs ≤0.5 µg/mL, the microbiological success rate was 100% in both treatment groups (n = 28, 1500 mg; n = 36, 3000 mg). Three urogenital microbiological failures were observed; all had isolates with the highest observed gepotidacin MIC of 1 µg/mL together with a common gene mutation known to affect gepotidacin binding; and all were men who have sex with men. Emergence of resistance to gepotidacin was observed for 2 of the urogenital microbiological failures. Further clinical development of gepotidacin for gonorrhea will need to consider dose optimization strategies and options to address NG isolates with higher gepotidacin MICs to minimize the potential for emergence of resistance.

26With the single gepotidacin doses administered, there were no observed dose-limiting AEs. Blinded real-time reviews of safety data supported the randomization of participants to both the low- and high-dose groups throughout the study. The most frequently reported AEs were gastrointestinal in nature. Gepotidacin was administered with food in this study to minimize these effects. Such AEs have been observed previously for gepotidacin and are well known to be associated with some antibacterial therapies. CD infections were not reported in this study. Adverse events potentially associated with acetylcholinesterase inhibition, a known effect of gepotidacin based on in vitro competitive and reversible inhibition of the enzyme at clinically relevant concentrations, include dizziness, hyperhidrosis, and headache and were experienced by ≤6 participants in either treatment group. Potential dose-related trends were observed for some gastrointestinal and acetylcholinesterase inhibition AEs, but statistical significance could not be declared due to the small number of participants.

27The effect of gepotidacin on cardiac conduction has been thoroughly evaluated, with results demonstrating a mild increased heart rate effect and a predicted mean QT prolongation of <15 milliseconds at plasma concentrations of 9 µg/mL, which is the highest plasma exposure expected clinically. With the high dose of 3000 mg in this study, mean 2-hour postdose plasma concentrations were approximately 6 µg/mL and were below the predicted QT threshold (<10 milliseconds), with no significant cardiac safety observations.

28A comparison of gepotidacin plasma concentrations between participants in the present study and phase 1 healthy volunteers showed comparable 2-hour postdose concentrations (Figure 3). This phase 2 population consisted primarily of otherwise healthy participants; therefore, systemic exposures similar to those observed in phase 1 populations were expected.

29One limitation of this study was the sample size, and a larger trial will be needed to confirm these results. Additionally, enrollment focused on participants with uncomplicated urogenital gonorrhea and few participants with pharyngeal or rectal gonorrhea were enrolled. Of these, 1 of 2 pharyngeal and all 3 rectal gonococcal infections were eradicated, indicating potential efficacy. Both men and women were enrolled; however, only 3% of the ME population were women. Future investigations will aim for a higher enrollment of women, but there will be limitations due to required reproductive health exclusions. Future studies will also collect sexual orientation for all participants to allow further assessment of resistance patterns. The ME population consisted of only US participants; thus, these study results may not reflect global NG epidemiology and resistance patterns. Last, coinfection with CT is common in individuals with gonorrhea. Gepotidacin has not been shown to have activity against CT; thus, standard-of-care therapy was allowed after completing TOC assessments for participants who tested positive at baseline.

30In conclusion, these results demonstrated that 1500-mg and 3000-mg single oral doses of gepotidacin were ≥95% effective for bacterial eradication of NG in adult participants with uncomplicated urogenital gonorrhea. Emergence of high-level resistance in 2 of 3 treatment failures is a concern. However, new therapies for drug-resistant NG are urgently needed. With additional clinical evaluation, gepotidacin may provide a much needed oral therapeutic option for uncomplicated urogenital gonorrhea as part of a new combination regimen.