Objectives Global concerns regarding the prevalence, asymptomatic nature and burden of disease associated with Trichomonas vaginalis (TV) continue. The lack of a portable molecular point-of-care assay to detect this infectious disease has meant that many remote or low-resource settings still need to rely on delayed results from central laboratories and/or syndromic management as treatment strategies. We evaluated the new GeneXpert (Gx) TV nucleic acid amplification test (NAAT) compared with an in-house laboratory NAAT to determine whether it would be suitable for use at the point of care.
Methods In a state-based laboratory and using their in-house NAAT, we selected the first 60 urine samples that were positive and the first 60 that were negative (n=120) in the study period for Gx TV testing in order to reduce collection delays and avoid the freezing of samples.
Results Positive percentage agreement between the Gx TV and NAAT was 95.0% (95% CI 86.1% to 99.0%), negative percentage agreement was 100.0% (95% CI 93.5% to 100.0%) and overall percentage agreement was 97.4% (95% CI 92.5% to 99.5%). Three discordant results were detected with each being close to the cycle threshold of detection using the in-house NAAT assay.
Conclusions Findings suggest the Gx TV assay is easy to use and has suitable overall agreement for sexually transmissible infection (STI) testing at the point of care. It may be used in combination with the Gx CT/NG assay to test for all three STIs simultaneously using this portable and modular-based NAAT platform.
- MOLECULAR TECHNIQUES
- CLINICAL STI CARE
- INFECTIOUS DISEASES
Statistics from Altmetric.com
Trichomonas vaginalis (TV) is the most common curable sexually transmissible infection (STI) globally.1 In Australia, TV is most common in rural locations and remote Aboriginal communities where the incidence in 16–19 year olds was reported to be 19.8 and 3.6 per 100 person-years in young women and men, respectively.2 A large proportion of TV infections are asymptomatic, and if left undiagnosed and untreated, TV increases the risk of perinatal morbidity such as premature delivery and low birth weight.
Traditionally, diagnosis of TV has relied on microscopy and culture.3 More recently, nucleic acid amplification tests (NAATs) using either urine or swab specimens have become the preferred method of diagnosis in Australia and other countries due to greater sensitivity and specificity. In many remote Aboriginal communities, annual screening for TV, chlamydia and gonorrhoea is recommended.4 However, due to distances from centralised laboratories and difficulties recalling patients, 11–25% of people with a positive STI test are left untreated, and the average time to treatment is 21 days.5 The availability of an accurate and portable molecular TV point-of-care (POC) test may improve TV screening and management in these settings and other low-resource settings where there is limited laboratory infrastructure.
To date, preliminary data from only one prototype molecular TV POC assay are available.6 The GeneXpert (Gx) platform (Cepheid, Sunnyvale, California, USA) is a molecular system for the diagnosis of infectious diseases that is suited for POC testing, and in this paper, we provide a brief evaluation of the new Gx TV assay.
We undertook a laboratory-based assessment of the Gx TV test in September 2014 at the Pathology Queensland laboratory in Townsville, which provides diagnostic services to many remote Aboriginal communities in far north Queensland. The evaluation was undertaken to assess the suitability of the Gx TV assay for a larger field study called TTANGO, which involved the dual Gx chlamydia and gonorrhoea test.7
We used 120 urine samples (from 42 males and 72 females; mean age 27.1 years) collected by primary healthcare centres during routine community screening in remote settings for the evaluation as (1) the prevalence of TV is high in these areas, ensuring the sample size of positive and negative tests would be met in a short time period; (2) relatively new samples would be tested (not processed or frozen) and (3) the large volume of urine would provide adequate original sample for dual testing (Gx and in-house NAAT).
We selected the in-house NAAT assay as the comparator test rather than a commercially available assay as we were interested to compare the performance of the Gx TV test against what is being used routinely in the participating laboratory. The in-house TV NAAT used in this evaluation is based on a previously published method8 and was shown in our laboratory to have 98% sensitivity (95% CI 93.2% to 99.5%) and 100% specificity (95% CI 97.1% to 100%) for testing urine samples compared with a second in-house NAAT (unpublished data). In-house TV NAAT involved sample extraction using the QIAxtractor as per manufacturer's instructions (Qiagen, Doncaster, Australia). Each in-house NAAT reaction mix consisted of 12.5 µL of QuantiTect Probe PCR master mix (Qiagen), 0.4 µM primers (TV4C and TV5 targeting the β-tubulin gene)8 and 0.2 µM of probe (TV7P).8 From 200 µL of urine, 5.0 µL of sample was extracted and made up to 25.0 µL using DNase-free water. Cycling was then performed on a Rotagene 6000 real-time PCR instrument using the following cycle conditions; and initial hold at 95°C for 15 min, followed by 45 cycles at 95°C for 15 s and 60°C for 60 s.
The Gx instrument system is automated and integrates sample processing, cell lysis, purification, nucleic acid amplification and detection of the target sequence. The system requires the use of a single-use self-contained Gx cartridge and each test includes a sample processing control (SPC), sample adequacy control (SAC) and probe check control.9 The Gx TV testing involved adding 500 µL of urine specimen to the test cartridge, entering patient details into the Gx computer software, and loading the cartridge into the machine module. Negative Gx TV results were available in 53 min. Positive Gx TV results were available between 45 and 52 min depending on the sample organism load.
Samples were collected from a range of remote locations and sent to the laboratory using routine air or road freight methods. Time from collection to arrival was no more than two days, which is standard clinical practice in this region. Among the urine samples routinely received at the laboratory, we selected the first 60 samples that were in-house NAAT TV positive and the first 60 samples that were in-house NAAT TV negative for the evaluation. All samples were tested by the Gx TV assay within 48 h of arriving at the laboratory and stored at 4°C until testing. Both the in-house NAAT and Gx TV POC tests were conducted by two qualified laboratory staff. Neither were fully blinded to the testing process given the sample selection process used. A 10-fold dilution series (10E-1 to 10E-7) was also prepared using a frozen in-house NAAT TV-positive urine sample. The dilutions were tested using the in-house NAAT and Gx TV assays. The relative detection of each assay was determined as the lowest concentration returning a positive result and results compared between the two assays.
Gx TV assay and in-house NAAT data were collated, and we calculated positive, negative and overall percentage agreement between the two assays using standard methods10 and 95% CIs using the binomial approximation method. Based on a sample size of 60 and assuming at least 95% agreement between the tests, we estimated the 95% CIs would be ±10%.
Of the 120 samples tested by Gx TV, 5 (4%) gave invalid results on initial and repeat testing and were excluded from the analysis (4 Gx tests indicated internal SPC failure and 1 SAC failure). The positive percentage agreement between the Gx TV assay and in-house NAAT was 95.0% (95% CI 86.1% to 99.0%, 57/60), the negative percentage agreement was 100.0% (95% CI 93.5% to 100.0%, 55/55) and overall percentage agreement was 97.4% (95% CI 92.9% to 99.5%, 112/115) (table 1).
The three discordant (Gx TV negative, in-house NAAT positive) samples had in-house NAAT cycle threshold values of 38, 39 and 40 cycles. The dilution series was prepared from a positive urine sample and showed the in-house NAAT detected to 10E-4 dilution, whereas the Gx TV detected to the 10E-3 dilution.
The evaluation suggests the Gx TV assay is suitable for testing at the POC. The three discordant samples provided relatively high crossing points in the in-house NAAT results. Such high crossing points are typically associated with low DNA loads and suggests these discordant results may be due to low organism loads that were close to the Gx assay detection threshold.
The Gx TV assay combined with the chlamydia and gonorrhoea Gx assay would enable health services in remote and low-resource settings to test and treat for all three STIs at the POC using the same Gx platform within a 90 min timeframe. Moreover, simultaneous testing and detection may then improve time to treatment and reduce loss to follow-up rates, which continues to be an issue in these settings.
The authors thank Cepheid for providing the Research Use Only TV kits for this study.
Handling editor Jackie A Cassell
Contributors SGB was responsible for the construction of this short report and its final form in preparing the manuscript for publication. Revisions, comments and feedback were provided by LMC, SNT and BD. FF and DW were responsible for the setting up and implementation of the comparative laboratory evaluation of the GeneXpert TV assay. DW also reviewed this manuscript for molecular accuracy. SGB also conducted the data analysis and statistical calculations for test agreement. He also generated the table of results for this manuscript. SGB is the guarantor and considers himself responsible for the overall content of this manuscript.
Funding TTANGO is funded by a National Health and Medical Research Council (NHMRC) project grant #109902. The Kirby Institute receives funding from the Australian Government Department of Health and Ageing. TTANGO is a collaboration between researchers named in the authorship list. Additional TTANGO investigators not named as authors include John Kaldor, David Wilson, David Regan, Handan Wand, from the Kirby Institute; Mark Shephard, Flinders University; James Ward, South Australia Heath and Research Institute; Kit Fairley, Monash University; Belinda Hengel, Apunipima Cape York Health Council; Annie Tangey, Ngaanyatjarra Health Service; David Anderson and Lisa Natoli, Burnet Institute); the Queensland Aboriginal and Islander Health Council; Queensland Health; Aboriginal Health Council of Western Australia; West Australia Department of Health; West Australian Country Health Service; Aboriginal Health Council of South Australia; South Australia Health; National Reference Laboratory (NRL); Medical Communication Associates (MCA); Western Diagnostic Pathology, WA; Pathwest Laboratory Medicine WA; Clinipath Pathology, WA; Queensland Health Pathology and Scientific Services; Sullivan Nicolaides Pathology, QLD; and SA Pathology. The authors also acknowledge the important contributions of the TTANGO Reference Group.
Competing interests None declared.
Ethics approval Children's Health Services Queensland Human Research Ethics Committee (HREC/12/QRCH/246).
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.