CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

of each infection at 6 and 12 months after PrEP initiation using Kaplan-Meier survival analysis. We also describe the frequency and percentage of Ct/Ng detection per anatomical site and calculate the percentage of missed diagnosis if molecular testing for Ct/Ng were applied only for symptomatic patients, or if screening is done in urine only. Results: 386 PrEP users under follow-up in a single institution in Sao Paulo, Brazil, were included in the study. Most (94%) were men who have sex with men, with median age of 31 years old (interquartile range [IQR] 27-37). At baseline, active syphilis was detected in 23 participants (7%; 3 symptomatic and 20 latent or unknown stage), whereas Ct and Ng were detected in 9 patients each (8% and 9%) of whom only one Ng-positive patient had symptoms. After a median follow-up of 278 days (IQR 180-370), incident syphilis was detected in 24 PrEP users, with a cumulative incidence of 12% at 12 months; of those, 10 were symptomatic (3 in primary stage and 7 in secondary stage). Ct and Ng were detected in 13 patients and 10 patients, with a cumulative incidence of 12% and 10% at 12 months respectively. Had Ct/Ng molecular testing been used for symptomatic patients only, 15/16 (94%, 95% CI 70-100) cases would have been missed at baseline and 18/20 (90%; 95% CI 68-99) incident cases would have been missed. Had screening been performed in urine only, 12/16 (75%; 95%CI 48-93) cases would have been missed at baseline and 14/20 (70%; 95%CI 46-88) incident cases would have been missed. Conclusion: Multiple anatomic site sampling is a powerful strategy to increment the diagnostic sensitivity of Ct/Ng molecular screening. This approach should be applied in high-risk PrEP users as to improve the capacity of accurate diagnosis and treatment.

1052 GYRASE A SERINE 91 GENOTYPING PREDICTS GONORRHEA CIPROFLOXACIN TREATMENT OUTCOME Jeffrey D. Klausner 1 , Sheldon Morris 2 , Claire C. Bristow 2 , for the NIH/NIAID STAR STI CTG Ciprofloxacin Gyrase A Serine 91 Neisseria Gonorrhoeae Study Group 1 University of California Los Angeles, Los Angeles, CA, USA, 2 University of California San Diego, La Jolla, CA, USA Background: Neisseria (N). gonorrhoeae infections are rapidly increasing and among the most common co-infections in human immunodeficiency virus-infected patients. With great public health concern, there have been cases of N. gonorrhoeae resistant to all available antibiotics. In order to slow the continued emergence of antimicrobial resistance, new treatment strategies are urgently needed. The use of resistance-guided therapy—treatment based on the antimicrobial susceptibility of the infection—is one such promising strategy. We developed an assay to predict the susceptibility of N. gonorrhoeae to ciprofloxacin based on the gyrase A (gyrA) serine 91 codon, a locus previously shown to be highly predictive of in vitro resistance. In this study, we tested the efficacy of that assay in predicting clinical outcomes. Methods: We conducted a single armmulti-site clinical study of the efficacy of ciprofloxacin 500 mg by mouth for the treatment of wild-type gyrase A N. gonorrhoeae infections. We recruited and enrolled study participants from sexually transmitted disease clinics across the United States. We determined N. gonorrhoeae gyrA serine 91 wild type status using a previously Clinical Laboratory Improvement Act-verified laboratory-developed PCR assay with high-resolution melt analysis. We report outcomes in participants who were N. gonorrhoeae culture positive for gyrA serine 91 wild type infection at enrollment and had culture assessment 5-10 days after treatment. We also report treatment outcomes in cases with non-wild type gyrA serine 91 N. gonorrhoeae infections at enrollment. Results: Among 106 patients with 117 urogenital, rectal or pharyngeal infections across 6 clinics, the frequency of microbiological cure was 100% (95% one-sided confidence interval 97.5-100%). The cure frequency did not vary by anatomic site of infection, sex or age of the study participant. Two cases with mutated gyrase A N. gonorrhoeae infection failed therapy (0% cure). Conclusion: GyrA serine 91 N. gonorrhoeae genotyping was highly predictive of clinical outcomes in patients with gonorrhea treated with ciprofloxacin. 1053 IMPROVING DIAGNOSIS OF CT/NG AMONG PrEP USERS WITH MULTIPLE SITE SCREENING Jorge S. Moreira 1 , Ricardo d. Vasconcelos 2 , Vivian I.Avelino-Silva 2 1 Hospital das Clínicas da Universidade de Sao Paulo, Sao Paulo, Brazil, 2 University of Sao Paulo, Sao Paulo, Brazil Background: PrEP users are under high risk of bacterial sexually transmitted infections. Sensitive and timely diagnostic strategies are crucial to allow rapid prescription of antimicrobial treatment. Several studies have shown that Chlamydia trachomatis (Ct) and Neisseria gonorrhea (Ng) screening at multiple anatomic sites may improve the diagnostic yield in high-risk populations. Methods: In this retrospective cohort study, HIV-uninfected patients referred for PrEP were followed with periodic serologic testing of Syphilis (every 3 months) and culture/molecular testing of Ct/Ng (approximately every 6 months in asymptomatic patients; as needed for those with symptoms). We describe the baseline prevalence of Syphilis, Ct and Ng as well as the cumulative incidence

Poster Abstracts

1054 THE PERFORMANCE OF POOLED 3-ANATOMIC-SITE CHLAMYDIA AND GONORRHEA TESTING Claire C. Bristow 1 , Sanjay R. Mehta 1 , Martin Hoenigl 1 , Susan J. Little 1 1 University of California San Diego, La Jolla, CA, USA Background: While molecular testing for Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) is much more sensitive than traditional culture and immunostaining approaches, the cost can be more than 20 times higher per test. These costs are amplified further, as optimal testing requires specimens from 3 anatomic sites (rectal, pharyngeal and urogenital [urine or vaginal swab]), tested individually. While individual testing of samples from all three sites is currently recommended, pooled testing may offer a cost-saving alternative. We assessed the performance of routine versus pooled 3 anatomic site testing (1 test per person versus 3) for CT and NG. Methods: Using the Xpert® CT/NG assay (Cepheid, Sunnyvale, CA) we tested urine, rectal and pharyngeal swabs for CT and NG. Remnant specimens (0.34 mL from each anatomic site specimen) were combined to perform a single ‘pooled’ test. We calculated positive and negative percent agreement between the pooled testing results with the single specimen Xpert CT/NG test results as the reference. Results: We conducted 403 pooled tests. Of those, 366 (90.8%) gave valid results. Of the 37 pooled tests for which a valid result was not obtained, 3 were positive for CT, 3 were positive for NG and 1 was positive for both CT and NG on individual tests. The CT positive and negative percent agreement were 95.8% (95% CI: 85.7%, 99.5%) and 99.1% (97.3%, 99.8%), respectively. The NG positive and negative percent agreement were 96.9% (95% CI: 83.8%, 99.9%) and 99.7% (95% CI: 98.3%, 100%), respectively. Pooled testing identified 3 CT and 1 NG infections that were negative at all anatomic sites by individual testing. Conclusion: Three-site pooled CT and NG testing performs similarly to single anatomic site testing among tests providing a valid result. Optimizing the

CROI 2020 396