Cefixime Plus Doxycycline Compared to Ceftriaxone Plus Azithromycin for Treatment of Gonorrhoea

Non-inferiority, open-label, multicentre randomised clinical trial (RCT) compared two regimens for the treatment of uncomplicated urogenital, rectal or pharyngeal gonorrhoea in men and women between 18 and 65 years of age. Patients were enrolled and treated from April 2021 to June 2022 at the Dermatovenerology Department, University Hospital Bulovka, Prague, and the Venereology Prague, Medicentrum Beroun, Prague, Czech Republic.

Laboratory testing The diagnosis of gonorrhoea and C. trachomatis infection was established from urogenital, rectal and pharyngeal swab specimens. Laboratory testing for inclusion in the study and at 3 weeks after initiation of treatment was performed using the Cobas 4800 CT/NG NAAT assay (Roche Diagnostics, Indianapolis, IN, USA). At test of cure visits 1 and 3 weeks after the initiation of treatment, swabs for cultivation of N. gonorrhoeae were also sampled and immediately inoculated onto non-selective and selective Modified Thayer-Martin agar plates. Inoculated agar plates were directly transferred to the hospital laboratory for incubation at 36°C in a humid 5% CO2-enriched atmosphere for 48 hours. Suspected gonococcal colonies were species verified using the biochemical NEISSERIAtest® (LACHEMA, Brno, Czech Republic) or the PolyViteX VCAT3® medium (Biomérieux, Marcy l'Etoile, France).

Statistical analyses Sample size was based on assessing non-inferiority of the cefixime plus doxycycline arm compared to the ceftriaxone plus azithromycin arm with a one-sided 5% type I error rate. We estimated the sample size based on estimates from other gonorrhoea treatment studies and expert opinions. The target sample size of 152 participants (76 per group) was based on an assumed treatment failure of 0% in the ceftriaxone-azithromycin arm and 2% in the cefixime-doxycycline arm, which would provide at least 90% power for the comparison of the primary endpoint. Intention-to-treat (ITT) and per-protocol analyses were done; per protocol results are reported here and ITT results in the online supplement. The ITT population was composed of all randomised patients; the per-protocol population consisted of patients who were randomised, received the allocated regimen (for doxycycline, at least ≥85% (12/14) of the prescribed dose) underwent follow-up examinations, and abstained from any sexual activities, including protected intercourse, during the study period. For each study group, the proportion of patients with urogenital, rectal or pharyngeal gonorrhoea who achieved microbiological cure was calculated. The Clopper-Pearson exact method was used to estimate confidence intervals (CIs). We used chi-square tests (Pearson's or Fisher's test when less than five observations) for binomial outcome measures and Student's t test for continuous outcome measures, as appropriate.