Clinician‐collected urethra swab: A good alternative sample type using cobas 4800 system for Chlamydia trachomatis and Neisseria gonorrhoeae in men

Abstract Background Nucleic acid amplification tests (NAATs) are being used increasing to detection of CT (Chlamydia trachomatis) and NG (Neisseria gonorrhoeae) infections for superior sensitivity and specificity than other tests. Male first‐void urine (FVU) sample is the optimal sample type for detection of CT and NG by NAATs. Although not being the recommended by NAATs, clinician‐collected urethra swab (CCUS) is perhaps a good alternative sample type compared with the FVU sample in men. Methods Paired samples (FVU and CCUS) from one hundred male outpatients were simultaneously detected by urine pattern and swab pattern using cobas 4800 CT/NG assay on cobas 4800 system for the detection of CT and NG, respectively. And twenty‐one positive controls were also detected on cobas 4800 system. Results The CT/NG cycle thresholds (Ct) value of urine pattern is lower than that of swab pattern for the same positive samples (clinical samples and positive controls) on the cobas 4800 CT/NG assay. The final CT/NG results of two sample patterns from patients were highly consistent except for four discordant results. Conclusion CCUS is validated for a good alternative sample type for the CT/NG detection on the cobas 4800 system in this study.


| INTRODUC TI ON
Chlamydia trachomatis and NG are the most prevalent bacterial sexually transmitted infections globally, about 131 million and 78 million new cases each year, respectively. 1 In 2017, a total of 17 million CT and 555 608 cases were reported by centers for Disease Control and Prevention (CDC) in the United States, 22% and 67% increasing since 2013. 2 If not treated early, CT/NG can cause serious health problems, painful complications, emotional injury, and so on. NAATs are being used increasing to detection of CT/NG infections for superior sensitivity and specificity than other tests. Male FVU sample is currently the optimal sample type for detection of CT/NG by NAATs. 3 Adopted widely in numerous settings, cobas CT/NG assay for use on the cobas 4800 system provides automated solutions for the detection of CT/NG with excellent sensitivity and specificity. 4 But in practice, the CCUS sample type should be validated by the laboratories for the male CT/NG screening using cobas 4800 system. Because clinician workers could choose the CCUS sample type for male patients in some situation (offering urine/re-sampling difficulty, multiple items simultaneously, not likely to return for test reports, and so on). So we compared the consistent of CCUS and FVU samples using cobas CT/NG assay on cobas 4800 system (Roche Molecular Systems).

| MATERIAL S AND ME THODS
We retrospectively the CT/NG results of one hundred paired samples (100 FVU and 100 CCUS) from male outpatients attending urology department in the hospital (mostly symptomatic) from

| RE SULTS
The overall prevalence of CT was higher than that of NG infection (6.1% vs. 2.5%) for male patients attending the hospital during the past two years, and the median age of the positive patients was 32.7 years old. Then, the prevalence of CT (18% vs. 15%) was also higher than that of NG (9% vs. 8%) infection for CCUS and FVU of the recruited male patients in the study, respectively. And the me-  Then, as shown in Figure 1, the mean C t value of CT from positive con- In this study, the mean C t value of CT and NG from positive controls/samples detected on urine pattern was 0.7/2.5 and 0.5/0.8 lower than those on swab pattern, respectively. C t values versus DNA load are inverse, indicating that urine pattern is better than swab pattern for the same sample on cobas 4800 CT/ NG assay. Obtaining one hundred paired samples is perhaps difficult for doctor workers due to expensive process and patient agreement in this study. For many years, CCUS collected from men may have posed a barrier for screening men. 8 Cobas 4800 system offers high sensitivity, specificity, negative, and positive predictive values for the detection of CT/NG in urine specimens. 9 And the detail preparation for CCUS in the study was not provided in the manufacturer's instructions of the cobas 4800 system. We described good concordance of CCUS with FVU on cobas 4800 CT/NG assay in this study. Four discordances of CT or NG results (CCUS samples showed C t ＞ 39 cycles but the corresponding FVU samples were all negative) showed in this study.
Unfortunately, we could not use more methods to confirm the real CT/NG infection of the discordance samples due to the retrospective study. But we believed that the inhibition of amplification in urine specimens for PCR and a lack of monitoring self-collection of FVU could provide false-negative report in some cases, and a follow-up study for that would be performed in future. Then, the supplementary testing of NG weak-positive samples with C t ＞ 38 has limited utility 10 ; and the detection of NG from rectal and urogenital samples using cobas 4800 system require no routine confirmatory testing. 11 So being a good alternative sample type, the results from CCUS samples perhaps strengthen the identify and treatment of CT/NG infections using cobas 4800 system when applicable in the future.
One limitation of this study was the specimen size. Another limitation was not rating the collection of CCUS as easy, difficult, or neither easy nor difficult. But we believe that date presented here may be also useful for clinician workers to choose the CCUS sample type for male patients in some situation, for example, offering urine/ re-sampling difficulty, multiple items simultaneously, not likely to return for test reports, and so on.

| CON CLUS ION
Effective laboratory diagnosis of CT/NG infection is essential, the study showed that CCUS was a good alternative male specimen type for the detection of CT/NG using cobas 4800 system. So clinical workers could select both male sample types (FVU and CCUS) detected by NAATs if impossible.

ACK N OWLED G M ENTS
None.