Editorial Commentary: When to Perform a Test of Cure for Gonorrhea: Controversies and Evolving Data

(See the Major Article by Wind et al on pages 1348–55.)

Gonorrhea is on the rise. The rate of reported infection in the United States increased 13% between 2009 and 2014, largely as a consequence of a dramatic increase in diagnoses among men who have sex with men (MSM). Although national data on trends in MSM are limited, the proportion of all urethral gonorrhea cases occurring in MSM attending sexually transmitted disease (STD) clinics increased from 5% to 37% between 1990 and 2014, and an estimated 31% of all gonorrhea cases in 2013 occurred in the 2% of the US population who are MSM [1] (Mark Stenger, Centers for Disease Control and Prevention [CDC], personal communication). Although some of this increase is a result of more testing, the rise in urethral infections cannot be explained by increased screening as these infections are almost always symptomatic [2].

The threat posed by increased gonorrhea incidence is magnified by rising concern related to gonococcal resistance. Gonorrhea was recently named one of the CDC's top 3 antimicrobial-resistant (AMR) threats [3]. Since the inception of antimicrobial therapy in the 1930s, the gonococcus has developed resistance to all first-line therapies beginning with the sulfonamides, followed by penicillin, tetracycline, fluoroquinolones, and most recently cefixime, an oral third-generation cephalosporin. The CDC removed cefixime from its list of first-line recommended gonococcal therapies in 2012 when the prevalence of isolates with “alert value” minimum inhibitory concentration (MIC) rose 17-fold over a 5-year period [4]. (The Clinical and Laboratory Standards Institute has not defined breakpoints for cefixime resistance for Neisseria gonorrhoeae; however, the CDC defines an alert value MIC as ≥0.25 µg/mL). Although the percentage of gonococcal isolates in the United States with elevated MICs to cefixime has declined since 2010–2011 and is currently <1% [1], this short-term trend should not be interpreted as a triumph; AMR gonorrhea remains a threat and requires action.

Part of that action needs to be more screening. Case finding and treatment are mainstays of public health gonorrhea control. The United States instituted widespread gonorrhea screening in women in the 1970s, and that effort was temporally associated with a decline in gonorrhea rates [1, 5]. Over the last 2 decades, nucleic acid amplification tests (NAATs) have largely replaced culture to detect N. gonorrhoeae and Chlamydia trachomatis, and the advent of NAATs has altered our understanding of the epidemiology of gonorrhea at the pharynx and the rectum. Among MSM attending STD clinics in the United States, upward of 10% are diagnosed with gonorrhea at the pharynx and/or rectum, and >10% are diagnosed with chlamydia at the rectum [6, 7]. The overwhelming majority of these infections (>85%) are asymptomatic [6]. Although data are more limited, extragenital gonorrhea and chlamydia may also be common infections among women [8–10].

Extragenital infections seldom result in significant morbidity, but they are important for several reasons. First, these infections serve as a reservoir promoting sustained transmission. Among MSM, at least 30% of symptomatic gonococcal urethritis is acquired from the pharynx [11]. Second, the preponderance of evidence suggests that rectal infection with gonorrhea and/or chlamydia increases the risk of human immunodeficiency virus (HIV) acquisition [12–15]. Because HIV, gonorrhea, and chlamydia are all sexually transmitted, it is difficult to know with certainty if and how large a direct biological role these bacterial infections play in promoting HIV transmission. However, the risk of HIV associated with rectal infections appears to be independent of sexual behavior [14, 16–18], suggesting that these sexually transmitted infections (STIs) facilitate HIV transmission, a hypothesis that is supported by biological plausibility [19, 20]. Last, infection with gonorrhea in the pharynx is thought to contribute to the evolution of antimicrobial resistance. The gonococcus mingles with commensal Neisseria species in the pharynx, acquiring genetic material through transformation [21–25]. Pharyngeal gonococcal infections are more difficult to eradicate than infections at other sites—presumably due to the low levels of drug achieved at the site of infection—meaning that the gonococcus is exposed to suboptimal concentrations of antibiotic, allowing for the selection of subpopulations with elevated MICs [22, 26].

Recent gonorrhea trends and observations related to extragenital infections have 2 central clinical implications. First, clinicians need to screen MSM for extragenital infections. The CDC recommends that medical providers screen sexually active MSM for STIs and HIV annually, and for STIs every 3–6 months if they have multiple sex partners or partners who are nonmonogamous or anonymous; such screening should include gonorrhea and chlamydial testing at all exposed anatomical sites. Guidelines from other countries and some local guidelines in the United States are more explicit in defining risk factors meriting more frequent HIV/STI screening (Table 1).

The second implication relates to performing a test of cure (TOC). TOC has been justified primarily as a means to identify persons with persistent infections that could have serious health consequences—an issue of particular importance in women—and as a means to identify relatively rare cases of AMR N. gonorrhoeae. The subject is controversial and involves 2 main questions: In what clinical circumstances should clinicians perform a TOC, and how long after treatment should they perform that test? The CDC only recommends TOC in the following circumstances: (1) persons with pharyngeal infections treated with a regimen other than the first-line recommended regimen of ceftriaxone 250 mg plus azithromycin 1 g; (2) persons with urogenital or rectal infection treated with drugs other than ceftriaxone or cefixime plus azithromycin; and (3) persons with persistent signs or symptoms of infection [28, 29]. The World Health Organization likewise recommends against routine TOC [30]. In contrast, the most recent European (2012) [31], British (2011) [32], and Canadian (2013) [33] treatment guidelines all recommend routine TOC in all persons with gonorrhea. At least in part, these different recommendations reflect differences in the epidemiology of resistant gonorrhea and timing of publication of the guidelines. At the time that the European guidelines were written and using the European breakpoint to define cefixime resistance (≥0.125 µg/mL), which is 1 dilution lower than the US definition, 7.6% [34] of European surveillance isolates were cefixime resistant (2011), with wide variance by country (2.8% in the United Kingdom, 10% in Germany, and 37% in Slovenia). European clinicians and public health officials also identified several instances of multidrug-resistant (MDR) gonorrhea, organisms that were resistant to ceftriaxone and cefixime, penicillin, azithromycin, doxycycline, and ciprofloxacin [35, 36]. In contrast to Europe, at the peak of cefixime resistance in the United States in 2011, only 1.4% of tested isolates met US “alert value” criteria, and the United States has yet to identify a case of MDR N. gonorrhoeae [1].

Controversy about whether to perform TOC has been compounded by uncertainty about how to interpret NAAT TOC results. Because NAATs detect genetic material regardless of organism viability, it has been difficult to know if a positive NAAT TOC represents treatment failure or just the persistence of nonviable genetic material. The study by Wind and colleagues in this issue of Clinical Infectious Diseases provides new data on this important topic. In a well-designed and well-executed study, the investigators had 61 patients collect daily specimens from the rectum and genital tract for 28 consecutive days following treatment. Although time to clearance varied somewhat by site of infection, regardless of anatomic site of infection >95% of samples tested using RNA tests (eg, Aptima Combo 2) had cleared by 7 days, and >95% of specimens tested using DNA tests (eg, Cobas polymerase chain reaction) had cleared by 14 days. Blips—isolated positive tests following 3 negative tests—occurred in only 0.8% and 1.5% of postclearance samples with RNA testing and DNA testing, respectively, suggesting that a positive RNA test after 7 days or a positive DNA test after 14 days is usually a treatment failure or reinfection. Of note, none of the patients followed in the study had a treatment failure.

Superficially, these findings appear to diverge somewhat from those reported by Beymer and colleagues in Los Angeles [37]. Using an RNA-based NAAT (Aptima Combo 2), they found that none of 95 men with urethral gonorrhea had a positive TOC 6–21 days after treatment, but that 10 of 135 (7%) men with rectal gonorrhea had a positive test >7 days after treatment and that 7 of 134 (5%) persons with pharyngeal gonorrhea had a positive NAAT TOC 3–14 days following treatment. These positive tests were probably all or almost all false positives, as the majority of persons with a positive NAAT were culture negative, and among 12 of 14 without an identified reexposure to gonorrhea retested >21 days following treatment, all were NAAT negative. However, several of the men who tested positive were potentially reexposed to gonorrhea. Assuming these reexposures were reinfections, approximately 4% of men with rectal gonorrhea and <2% of persons with pharyngeal infection had a false-positive NAAT >7 days following treatment, a finding that is largely consistent with Wind et al's data. The findings of Wind et al related to the clearance of DNA NAATs are also consistent with Bachmann et al's older data demonstrating that ligase chain reaction tests of urine and vaginal swab specimens were consistently negative among persons tested >8 days following treatment [38]. Of note, these findings are somewhat at odds with studies assessing the clearance of Chlamydia trachomatis, which suggest that RNA NAATs can be positive for >2 weeks [39].

The Wind et al study leaves some questions related to the timing of TOC unanswered. Future studies need to better address the time to N. gonorrhoeae clearance from the pharynx and clearance of C. trachomatis from the rectum. Treatment failure appears to be common in persons with rectal chlamydial infection treated with azithromycin [40]. Clinicians should consider performing a TOC when treating rectal chlamydial infection with azithromycin or, better still, should routinely use a 7-day course of doxycycline in treating rectal chlamydia.

That brings us back to the question of whether to perform TOC. In the current climate of AMR gonorrhea and given current treatment recommendations, it remains unclear if the costs of routine TOC outweigh the benefits. Pharmacokinetic/pharmacodynamic modeling [41] suggests that a single 250-mg dose of ceftriaxone is sufficient to eradicate gonococci with MICs ≤ 0.125 µg/mL from the oropharynx; in 2013, no N. gonorrhoeae isolates tested through CDC surveillance had an MIC higher than this. Assuming that at least 95% of these infections would be cured using CDC guideline–recommended or alternative therapies, clinicians would need to perform 2000 TOCs to identify a resistant isolate that had failed therapy. In a place like King County, Washington, with a population of 2 million and approximately 3000 gonorrhea cases per year, we might find 1 resistant isolate every 1–3 years through such an effort. (TOC rates will never be 100%.) Given that reality, we do not believe that routine TOC for all persons with gonorrhea is indicated and that current CDC guidelines focusing on persons receiving nonstandard treatments and those with persistent signs or symptoms are reasonable. Clinicians should attempt to obtain gonorrhea cultures from persons who fail therapy to allow for antimicrobial susceptibility testing. In places where AMR N. gonorrhoeae is more common, or as resistance becomes more widespread, routine TOC may be worthwhile. Because at least some European nations are now performing significant numbers of TOCs, public health authorities in those areas should investigate how often those tests are positive and not likely to be a consequence of reinfection. Fortunately, we now have good data to guide us in interpreting the NAAT TOC results for anogenital gonorrhea.

Note

Potential conflict of interest. L. A. B. reports receiving in-kind research materials from Hologics (Aptima test kits). M. R. G. reports no potential conflicts. Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References