DIAGNOSIS OF GENITAL CHLAMYDIAL INFECTIONS IN THE ERA OF AMPLIFICATION TECHNOLOGIES
A. Stary
Outpatients'Center for Diagnosis of Infectious Venerodermatological Diseases, Vienna, Austria Diagnostic procedures making use of
amplification of specific nucleic acid sequences have provided laboratories with powerful tools with particular impact on the detection of bacterial genital infections such as Chlamydia trachomatis and Neisseria gonorrhoeae.
Compared with cell culture methodologies nucleic acid amplification (NAA) assays such as polymerase chain reaction (PCR), ligase chain reaction (LCR), strand displacement amplification (SDA), and transcription mediated amplification (TMA) have shown to be highly sensitive and specific for the detection of genital chlamydial infections in symptomatic and asymptomatic men and women, detecting up to one third more infected individuals. NAA technologies have already replaced other diagnostic methods for routine diagnosis of genital chlamydial infections although the commonest test still used in many countries is the enzyme immunosorbent assay (EIA) and only a minority of diagnostic centres use DNA amplification methods for daily routine. In several studies antigen detection methods available on the market have shown to perform with a remarkable lower sensitivity on invasive genital samples than NAA techniques on the noninvasive specimen type of first voided urine (FVU) in both, men and women and have been even declared as unethical in respect of the patient.
The high sensitivity of NAA techniques enables the detection of amplified chlamydia specific DNA or RNA even in specimens contaminated with microorganisms such as FVU and introital vulvovaginal specimen types with similar sensitivities when compared to
cervical and urethral specimens in both, men and women. This offers screening possibilities for the detection of a low number of organisms present in asymptomatic patients and their contact persons without signs of inflammation.
Screening abilities are even increased by using self administered vulvovaginal samples for the diagnosis of both, N. gonorrhoeae and C.
trachomatis which permits testing without a speculum examination and even without clinical inspection. Community-based urine testing by using NAA assays successfully identified chlamydial infections in different risk groups and was well accepted by community-members. Studies have been performed in female and male military recruits, in asymptomatic men and women, in high school students, in pregnant women, and in other population groups of special relevance for chlamydial diagnosis. It has been shown that screening for C. trachomatis can significantly impact STD epidemiology by facilitating early detection, treatment, and interruption of transmission.
The amplification tests focusing on the diagnosis of lower genital tract infection have long been restricted to two commercially available techniques, the Amplicor PCR, a DNA target amplification test for the diagnosis of C. trachomatis, and the ligase chain reaction LCx, a DNA probe amplification assay for the diagnosis of both, C. trachomatis and N.
gonorrhoeae. In 1996, the Amplicor PCR for C. trachomatis was compared with other diagnostic methods such as culture and Gen Probe by testing cervical and urine samples and showed a sensitivity similar but not higher than chlamydial culture. Since then, the performance procedure and handling recommendations of Amplicor PCR were
improved e.g. by changing the transport medium and by including an internal control for the amplification procedure used in the COB AS Amplicor. Since 2003, the LCR has been retracted from the market and is not available but will be replaced by another amplification technique in the near future.
The Gen-Probe APTIMA Combo 2 assay is a relatively new second generation of NAA technology after the AMP CT test and combines the technologies of target capture, transcription mediated amplification (TMA) and Dual Kinetic Assay. It is the only NAA which is already FDA approved for the use of vaginal specimen types in women. According to the results of a study in men, even penile specimens were recommended as alternative noninvasive sample types for men in addition to FVU for chlamydial and gonococcal coamplification when tested "by the APTIMA Combo 2 assay.
The strand displacement amplification technique ProbeTec, is based on the simultaneous amplification and detection of target DNA by the use of amplification primers along with a fluorescein-labelled detector probe and offers real-time amplification and detection. It has been approved for practical use for the diagnosis of chlamydial as well as gonococcal infections in genital und urine specimens for both, men and women and is widely used for routine diagnosis.
Disadvantages of NAA assays may still be the high costs of the performance in laboratories. Furthermore, problems with inhibition of amplification and contamination under certain conditions may lead to reduced reproducibility of positive and negative results.
Problems of reproducibility were reported for both, Amplicor PCR Chlamydia trachomatis and for LCR. In a study on cervical specimens tested by Amplicor PCR a broad variation for discrepant or equivocal PCR results could be observed on several repeat testing in both ways with false positive or negative results and wide
variations in the OD values. As possible causes of nonreproducibility false-positive-hybridisation during the detection assay, nonspecific priming in the amplification phase, amplicon contamination, inhibition of the amplification, or a very low concentration of chlamydia were discussed. It has to be considered, that the study has been performed before the internal control for inhibition was available. Reproducibility problems were also reported for the performance of the LCR and occur also for the SDA. They have raised the question whether the confirmation of positive results should be demanded for routine diagnosis which is controversially valued.
Comparison studies with different NAA assays and specimen types have demostrated that certain substances in clinical specimens may inhibit amplification in a certain extense. For PCR Amplicor, inhibitors were observed in 19% of cervical specimens, and could be reduced by heat treatment at 95 °C, freeze-thawing, by a 10-fold dilution of the samples, and at a higher degree by combined methods. Genital samples tested by NAA assays have more often a higher sensitivity than urine samples for detection of C. trachomatis as well as N. gonorrhoeae.
Although this may be partly due to the lower number of target nucleic acids, it has been shown that inhibition occurs more often in urine than in genital samples. Urinary substances associated with inhibition and removal of inhibitory activity were described in pregnant and nonpregnant women in different NAA assays due to different chemical substances. The complete removal of urine in the sediment decreases also the number of inhibition. It seems that the frequency of inhibition for NAA assays in urine specimens varies among different populations, among individuals in the same area at different times, and among different NAA assays with variations between 3 and 20% and is more often observed in chlamydia negative individuals.
Despite some limitations the NAA assays offer an important approach for chlamydia diagnosis in men and women at STI risk and in the general population acting as an unrecognized large reservoir of chlamydia infected persons, capable to transmit the microorganism to their partner or to newborns during perinatal exposure.
COMPUTATIONAL ANALYSIS OF INCA STRUCTURE FOR THE RATIONAL DESIGN OF NOVEL ANTIBODY-BASED CHLAMYDIA IMMUNODIAGNOSTICS
Larcombe, LD„ 2Karim, K„ Goodman, AC.
t 9 Cranfield Biomedical Centre & Cranfield Centre for Supramolecular Technology, Cranfield University, Silsoe UK
This work set out to apply computational analytical tools and biophysics principles to facilitate the development towards a novel chlamydia-specific immunodiagnostic. It aims to demonstrate a systematic approach to enable the rational design of antibodies from annotated genome database sequences, which would be relevant to a wider application in diagnostic development.
The amino acid sequence of IncA, translated from genomic data, was analysed using secondary structure prediction algorithms, and the results combined to identify target regions suitable for the preparation of synthetic haptens.
An initial Kyte-Doolittle hydropathy plot was carried out to establish the presence of transmembrane helices. Subsequent analysis with nnPredict, and the Sybyl™
Qian/Sejnowski, Maxfield/Scheraga and Garnier/Osguthorpe/Robson algorithms from the Tripos Inc. software package was used to identify non-transmembrane helical domains.
Transmambrane domains shown by KD plot were corroborated by use of the PHDhtm algorithm.
Non-helical regions were then divided into peptide sequences of suitable length for use as hepten for generating antibodies. Each 15mer peptide was then analysed for net charge, hydrophobicity, and other sequence-specific characteristics in order to select the most suitable peptide. A candidate was chosen from each non-helical domain, and subjected to BLAST searches to identify homology with other proteins likely to be present in any clinical samples.
Finally the most suitable peptide was synthesised and used to raise a polyclonal antibody.
After amalgamation of data from the structure prediction algorithms, a representation of the secondary helical structure of IncA was produced sufficient to identify three candidate non-helical regions.
35 single amino acid frame-shifted sequences were analysed resulting in a 15mer peptide from the cytoplasmic terminal end of the protein being chosen.
The peptide had suitable net charge for aqueous solubility, appropriately distributed charged, uncharged and hydrophobic residues for synthesis, and a minimal probability of homology with any relevant alternative proteins.
It has been demonstrated that specific antibodies can be produced which could be used for diagnostic development solely based on data from genomic databases.
Sufficient tools and theoretical knowledge can now be applied so that with only minimal experimental evidence of protein expression, structure and function, useful diagnostics could be developed by identifying relevant genomic candidates.
ARRAYTUBE™ MICROARRAY HYBRIDISATION ASSAY FOR IDENTIFICATION AND DETECTION OF ALL CHLAMYDIAL SPECIES
Konrad Sachse1*, Helmut Hotzel1, Peter Slickers2, Thomas Ellinger2, and Ralf Ehricht2 'Federal Research Centre for Virus Diseases of Animals (BFAV), Institute of Bacterial Infections and Zoonoses, Jena, 2Clondiag Chip Technologies GmbH, Jena, Germany
To design, produce and test both in situ synthesized and spotted microarrays to differentiate among all species of Chlamydiaceae.
In silico analysis of a multiple sequence alignment of chlamydial ribosomal rRNA genes by the "most variable window approach"
resulted in the identification of a highly discriminatory region for all species of the family Chlamydiaceae in domain I of the 23 S rRNA gene. This led us to design in situ synthesized microarrays carrying 512 distinct, but sequence-related 26-nt oligonucleotides in two-fold redundancy. Spot size was 32x32 pm.
For low-density spotted microarrays, species-specific probes from the most variable window were spotted as 3'-amino-modified 26-nt oligos (9-fold redundancy, spot size 80 pm).
Internally biotinylated products of a consensus PCR amplifying a 1-kbp segment comprising the 3'-end of the 16S rRNA gene, the intergenic spacer and the 5'-terminal region of the 23 S rRNA gene were hybridized to the microarrays using the ArrayTube™ system (Clondiag Chip Technologies, Jena, Germany).
The hybridization reaction was monitored in the ATR-01 array tube reader at 25 °C for 40 min, recording 41 images. Signal amplification was achieved by enzyme-catalyzed silver precipitation, and signal intensity data with local background correction were obtained for each spot using the Iconoclust software, version 2.2 (Clondiag Chip Technologies).
Hybridization of all consensus PCR products of chlamydial strains to in situ synthesized combinatorial microarrays revealed clearly distinct hybridization patterns for all nine chlamydial species. The high
stringency of hybridization conditions ensured that target strands annealed only to probes having the exactly matching complementary sequence, whereas no hybridization signals at all were seen on derived non-matching but sequence-related probes, even including those with a single-base variation, unless at the 3'-end. Likewise, unambiguous differentiation among the species Chlamydia (C.) trachomatis, C. suis, C. muridarum, Chlamydophila (Cp.) pneumoniae, Cp.
psittaci, Cp. abortus, Cp. felis, Cp. caviae, and Cp. pecorum was achieved on spotted arrays as well. To assess the suitability of the present system for clinical specimens, a series of porcine lung tissue samples, two milk samples and spiked cell culture were examined. The results were in agreement with those from diagnostic PCR.
The present ArrayTube™ platform, which involves plastic tube-integrated microchips, is comparatively inexpensive and easy to handle. The finding that PCR-amplified bacterial DNA could be detected and identified by the present microarray assay directly from clinical material clearly illustrates its potential and opens up interesting possibilities for chip-based genotyping of chlamydiae in the absence of cultural isolates.
CHARACTERISTICS OF A NEW ABBOTT M2000 AUTOMATED MAGNETIC SAMPLER COMBINED WITH MULTIPLEX HOMOGENOUS REAL TIME PCR AND THE ABILITY TO DETECT CHLAMYDIA TRACHOMATIS IN RESIDUAL CLINICAL SWAB SAMPLES
Chernesky MA1*, Jang D1, Marshall R2, Yu J2, Howell-Adams B2, Ho S2, Welk J2, Lai-Zhang J2, Brashear R2, Diedrich B2, Otis K2 and Webb E2
McMaster University St. Joseph's Healthcare, Hamilton, Ontario, Canada1, Abbott Laboratories, Abbott Park, IL, USA2
Although nucleic acid amplification (NAA) tests are sensitive and specific for detection of Chlamydia trachomatis and Neisseria gonorrhoeae, most are sensitive to amplification inhibitors and lack sufficient throughput. The second generation NAA assay from Abbott Molecular Diagnostics is comprised of a rapid automated magnetic sample prep system to be used in combination with homogeneous Real Time PCR (HRT-PCR) DNA amplification. The objectives were to determine various operating parameters of the HRT-PCR and M2000 including throughput, analytical sensitivity and target specificity. This prototype assay was then used to detect Chlamydia DNA in residual swab samples, which had been collected and tested by the LCx LCR test (Abbott), the Amplicor PCR assay (Roche) or the ProbeTec ET multiplex strand displacement amplification assay (Becton Dickinson).
The Abbott M2000 magnetic sample prep technology removes both inhibitors and fluorophores. The DNA is captured generically using a silica based magnetic particle, washed, and eluted with water. The HRT-PCR amplification and detection method uses DNA amplification primers with fluorescence-quenched DNA detection probes. The homogeneous assay format allows for detection of Chlamydia trachomatis, Neisseria gonorrhoeae and internal control without the necessity of opening the reaction vessel. The system will provide for automation of primary barcoded tubes with caps that are pierceable by the pipetting system. Both the purified DNA and the PCR mastermix are automatically added to the PCR tray. Serial dilutions of DNA were tested to determine the analytical sensitivity of the new test and a panel of 58 different organisms were used to test specificity. A time study was conducted to determine throughput for varying numbers of clinical samples. Swab samples processed in the
LCx (n=64) ProbeTec (n=496) and Amplicor (n=67) collection systems and assay kits were stored up to 12 weeks at -20°C. Residual material containing the swab was processed into the HRT-PCR assay via the M2000. Positive, negative and overall concordance was calculated comparing the results from the new assay to the FDA-cleared tests.
The dilution experiments demonstrated uniform curve separation from 107 to 20 genome equivalents of C. trachomatis. None of the organisms in the specificity panel organisms were positive in the HRT-PCR test when tested at biologically normal concentrations. The throughput studies showed the following: 96 samples yielded 192 results in 4.5 hours; 192 gave 384 results in 7 hours and 288 provided 576 results in 9.5 hours. For clinical samples already tested for C. trachomatis in 3 first generation NAA tests, the concordance of M2000 prepared samples and Abbott HRT-PCR with the ProbeTec Chlamydia test was 99.1% (325/328) for positives, 98.2%
(165/168) for negatives and 988% (490/496) overall; with the Amplicor Chlamydia results the values were 96.3 % (26/27) for positives, 100%
(40/40) for negatives and 98.5% (66/67) overall;
with the LCx assay the values were 100% (32/32), 100% (32/32) and 100% (64/64) overall.
The M2000 magnetic sample preparation combined with homogeneous PCR technology provides a rapid, sensitive, and specific methodology for detection of C. trachomatis with negligible risk of PCR product contamination. The ability of the HRT-PCR to detect positives and negatives from residual samples collected into a variety of collection systems suggests versatility and robustness for testing clinical specimens.
USING PATIENT- & CLINICIAN-COLLECTED VAGINAL SWABS IN THE APTIMA® CT ASSAY FOR CHLAMYDIA TRACHOMATIS & APTIMA® GC ASSAY FOR NEISSERIA GONORRHOEAE
Schachter J1, Chernesky M2, Willis D3, Fine P4, Hook EW5, Martin DH6, Fuller D7, Jordan J8, Janda W9
'University of California, San Francisco, SF, CA, 2St. Joseph's Health Care Regional Virology and Chlamydiology Laboratory, Hamilton Ontario, Canada, 3Florida State Department of Health, Jacksonville, FL, 4Planned Parenthood of Houston and Southeast Texas, Houston, TX, 5University of Alabama, Birmingham, AL, 6Louisiana State University Medical Center, New Orleans, LA,
7Wishard Memorial Hospital, Indianapolis, IN, 8MGees Women's Research Institute, Pittsburgh, PA, 9University of Illinois, Chicago, IL
The literature suggests vaginal swabs (VS) may be useful specimens for diagnosing genital infections. A kit for the collection of VS for screening for Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (GC) using Gen-Probe Incorporated's commercially available APTIMA Combo 2 Assay (AC2)has recently received FDA clearance. We assessed the performance of the APTIMA CT Assay (ACT) and APTIMA GC Assay (AGC) using patient- and clinician-collected vaginal swab (PVS and CVS) specimens. ACT and AGC target rRNA sequences different from those used in AC2. but the three assays use the same procedures.
Females were enrolled at eight geographically diverse, high and low prevalence sites. First catch urine (FCU), one PVS, one CVS, and two endocervical swabs (ES) were collected from each subject. ACT and AGC were done on PVS, CVS, FCU, and one ES. AC2 was done on the same FCU and ES tested by ACT and AGC. The second ES and the FCU were tested using BDProbeTec (Becton Dickinson and Company) for CT and GC. Subjects were defined as infected if two FDA-cleared tests on FCU or ES were positive.
There were 180 CT and 78 GC infected subjects among the 1,464 subjects (respective prevalences of 12.5% and 5.4%). ACT Assay sensitivities and specificities were 98.3%
(175/178) and 96.5% (1,202/1,245) for PVS and 97.2% (175/180) and 95.2% (1,208/1,269) for CVS. AGC Assay sensitivities and specificities were 96.1% (74/77) and 99.3%
(1,336/1,345) for PVS and 96.2% (75/78) and 99.3% (1,359/1,368) for CVS. The ACT Assay VS results were in >95% agreement with FCU and ES results. With the AGC Assay, agreement was >98%. Further testing was performed to determine whether the apparent false positives (FP) were true positives (TP).
Previous evaluations of the ACT Assay have found that its exquisite sensitivity results in apparent FPs that are shown to be TPs by repeat testing. That was the case here as with CVS 57% (24/42) of the apparent ACT FP, and67% (6/9) of apparent AGC FP were positive by AC2.
The ACT Assay and the AGC Assay using VS are sensitive and specific for the detection of both CT and GC. Patients are are equally adept as clinicians in collecting VS.
VS are well suited for screening for CT and GC with the APTIMA CT and GC assays.
SCREENING FOR CHLAMYDIA TRACHOMATIS USING SELF-COLLECTED SPECIMENS: PERFORMANCE, EASE OF COLLECTION AND PREFERENCE
Max A. Chernesky*
McMaster University and St. Joseph's Healthcare. Hamilton, Ontario, Canada With high rates of asymptomatic infection,
Chlamydia trachomatis is easily transmitted. To prevent complications of upper genital tract infection, screening programs are needed to facilitate screening, non-invasive samples (NIS) such as first catch urine (FCU) and patient collected vaginal swabs (PVS) will need to be easy to collect and preferred. Assays used to test NIS will have to be as accurate as testing traditional cervical swabs (CS). The objective was to review studies examining preference and/or ease of collection of NIS from different female populations and to present the results of a recent study in North America examining questions regarding self-sampling.
Four recently published papers were selected for review and examined for evidence of women's preference, ease of collection and performance characteristics on FCU and PVS compared to CS in nucleic acid amplification (NAA) assays. In an unpublished study, women from 8 North American cities collected a PVS and FCU which were compared to a clinician-collected vaginal swab and CS by the APTEMA® Combo 2 assay (Gen-Probe Inc.). The CS and FCU were also tested in the ProbeTec i ET assay (BD Biosciences) to determine infected patients. A total of 1090 women completed a questionnaire for ease of collection and preference for a PVS or FCU compared to the CS. Illustrated instructions concerning handling the swab and tube and collecting the sample were provided.
A study of California adolescents1 showed that an FCU was ranked slightly higher than PVS as a preferred NIS. Non-participation was due to a lack of trust for collection and feeling that a doctor's expertise was required. Female prisoners2
and army recruits3 were studied in Baltimore. Most of the prisoners found both samples easy to collect and preferred a PVS over CS. The army recruits found the FCU easier and preferred it slightly over PVS. They would prefer to use a PVS at home and FCU in the field. Young women in Hamilton4 preferred self-collection over physician collection unanimously in student health, family doctor, birth control, STD and street health settings. Greatest preference for PVS was shown in student health
and street health clinics. When testing was done in these studies the PVS detected as many or more infected women as FCU and CS, using the LCx (Abbott) assay. In the current study, using the APTIMA® Combo 2 assay for C. trachomatis, the PVS and CS detected 95.5%, and the FCU 91.6%, of the infected women. Across all of the testing sites, 90.4% of the women found that a PVS was very easy to collect and virtually none found it difficult. The ease of collection was not affected by race, age, education, tampon or diaphragm use, or STD experience. Almost 80% preferred a PVS over a pelvic examination and 60% over an FCU.
Over 90% would get tested more often if a PVS was available.
When tested in an effective NAA assay PVS can identify most women infected with C.
trachomatis, so they can be treated to prevent further transmission and ascending genital infection. PVS and FCU are easy to collect and preferred over CS. Preferences may be influenced by instructions provided to enable easy self-collection. Findings from these studies should be applied to female screening programs for C.
trachomatis.
References:
1. Serlin et al. What sexually transmitted disease screening method does the adolescent prefer?
Adolescents' attitude toward first-void urine, self-collected vaginal swab, and pelvic examination.
Arch. Pediatr. Adolesc. Med. 2002; 156(6):588-591.
2. Neromau et al. Female Prisoners' Preferences of Collection Methods for Testing for C. trachomatis and N. gonorrhoeae Infection. Sex. Transm. Dis.
2003;30:306-309.
3. Hsieh et al. Preferences Among Female Army Recruits for Use of Self-Administrated Vaginal Swabs or Urine to Screen for C. trachomatis Genital Infections. Sex. Transm. Dis. 2003;30:769-773.
4. Richardson et al. Prevalence of C. trachomatis Infections and Specimen Collection Preference Among Women, Using Self-Collected Vaginal Swabs in Community Settings. Sex Transm. Dis.
2003;30:880-885.
DETECTION OF CHLAMYDIA TRACHOMATIS AND NEISSERIA GONORRHOEAE INFECTIONS IN MEN BY TESTING FIRST CATCH URINE IN NEW APTIMA® CT AND APTIMA GC ASSAYS
Chernesky MA1*, Martin DH2, Hook EW3, Willis D4, Jordan J5, Wang S6, Lane J6' Fuller D7 and Schachter J8
^t. Joseph's Healthcare, Hamilton Ontario, Canada, Louisiana State University Health Sciences Center, New Orleans, LA, 3University of Alabama, Birmingham, AL, 4Florida State Department of Health, Jacksonville, FL, 5Magee Womens Research Institute, Pittsburgh, PA, 6 Gen-Probe Incorporated, San Diego, CA, 7Wishard Memorial Hospital, Indianapolis, IN, 8University of California, San Francisco, CA, USA
To compare the APTIMA CT (ACT) Assay for Chlamydia trachomatis and APTIMA GC (AGC) Assay for Neisseria gonorrhoeae to the APTIMA Combo 2 (AC2) Assay (Gen-Probe Incorporated) and the BD ProbeTec® (PT) energy transfer (ET) amplified DNA assays for CT and GC (Becton Dickinson Bioscience, Sparks, Md.) performed on urethral swabs (US) and first catch urine (FCU) from men.
From October 2002 to January 2003, we enrolled 1,322 men between 15 and 77 years of age (mean 28.5) from 6 sites in North America.
Subjects were classified as symptomatic if discharge or dysuria were reported by the subject.
Two US and one FCU specimen were collected from each patient. One of the swabs and an aliquot of FCU was tested by ACT, AGC and AC2. The other US and an aliquot of FCU were tested by the PT assay. For ACT and AGC, swab or urine specimens were transferred into AC2 Assay transport media. The transport solutions in these tubes released the rRNA targets and protected them from degradation during storage. The target rRNA molecules were isolated from the urine or swab samples by the use of a target capture method using capture oligomers on magnetic microparticles. The ACT and AGC use oligonucleotides that target rRNA sequences different from those of the commercially available AC2. The procedures for the three assays, which include target capture, transcription mediated amplification (TMA) and a hybridization protection assay (HPA) detection technique, are the same. The AC2 and PT assay protocols were followed from their package inserts. ACT and AGC assay results from US and FCU were compared to infected patient status (IPS) for determining sensitivity, specificity and predictive values with 95% confidence intervals. A patient was considered infected if both of the specimen types were positive in at least one of the
FDA-cleared tests (AC2 or PT) or if at least one of the 2 specimen types was positive in both AC2 and PT.
A total of 236 patients were infected with CT and 183 with GC according to the IPS algorithm. Symptoms were reported in 59.7%
(141/236) of infected and 40% (435/1086) of uninfected men. There were 236 men with CT (prevalence 17.9%) and 183 men with GC (prevalence 13.9%) infections. Hamilton and San Francisco were low prevalence sites for CT and GC. The other sites were of high prevalence, except for Pittsburgh where the number of subjects enrolled was too small to allow meaningful calculations. The ACT and AGC assays performed similarly as sensitivity and specificity ranges were similar across sites and from one specimen type to the other in symptomatic and asymptomatic patients. The respective ACT and AGC sensitivities were 96.2% and 98.9% for FCU and 97.5% and 99.5% for US. Specificities ranged from 96.9% to 99.3% for both assays and specimen types. An equal or higher percent was achieved in asymptomatic patients infected with CT by testing US or FCU. In patients infected with GC, the FCU specimen appeared to be more effective in symptomatic men. Positive and negative agreements between the assays and for both specimen types ranged from 94.3% to 100%
and 93.9% to 99.3%, respectively. These ranges were not significantly different. When the data were stratified by symptoms, the agreements remained similar for CT but GC positive agreement was 10% less between AGC and PT in asymptomatic subjects
The ACT and AGC assays demonstrated excellent sensitivity and specificity to identify infected men using US and FCU when compared to the FDA-cleared assays, AC2 and PT. This high performance on a non-invasive specimen such as FCU from asymptomatic men indicated that these tests could be used for screening urine from men.