Article Text
Abstract
Background: The new variant of Chlamydia trachomatis (nvCT), discovered in Sweden in 2006, contains a 377-bp cryptic plasmid deletion, which includes the targets for the COBAS Amplicor/TaqMan C trachomatis/Neisseria gonorrhoea and Abbott m2000rt C trachomatis/N gonorrhoea tests.
Objectives: To evaluate the new real-time COBAS TaqMan CT test v2.0 (CTM CT v2.0) for C trachomatis diagnostics and to investigate whether the proportion of nvCT was affected by the introduction of genetic diagnostics detecting nvCT (LightMix 480HT) in Örebro county, Sweden.
Methods: CTM CT v2.0 compared with LightMix 480 HT PCR for the diagnosis of C trachomatis was evaluated. Discrepant samples were analysed using BD ProbeTec ET and Abbott m2000rt RealTime CT II. All previously LightMix and cell culture-positive samples were analysed using an nvCT-specific PCR.
Results: The sensitivity, specificity, negative predictive value and positive predictive value of CTM CT v2.0 for examined samples (n = 1058) was 100%, 99.8%, 100% and 98.2%, respectively. Of 11 577 consecutive PCR samples, 9.4% (n = 1084) were positive and 34.3% (n = 372) of these were nvCT. Of 2306 consecutive culture samples, 5.0% (n = 116) were C trachomatis positive and 38.8% (n = 45) of these were nvCT.
Conclusions: CTM CT v2.0 is a sensitive and specific method for C trachomatis detection. Studies including larger numbers of symptomatic and asymptomatic patients as well as genital and extragenital samples, and in comparison with other internationally validated and, ideally, US Food and Drug Administration-approved C trachomatis nucleic acid amplification tests are imperative. The proportion of nvCT remains high in Örebro county, Sweden, despite the introduction of genetic diagnostics to detect the mutant.
Statistics from Altmetric.com
In 2006 a new variant of Chlamydia trachomatis (nvCT), containing a 377-bp cryptic plasmid deletion including the targets for the COBAS Amplicor/TaqMan C trachomatis/Neisseria gonorrhoea (Roche Diagnostics, Branchburg, New Jersey, USA) and Abbott m2000rt RealTime C trachomatis/N gonorrhoea (Abbott Molecular Diagnostics, Des Plaines, Illinois, USA) assays, was discovered in Sweden.1 2 These tests were used in two-thirds of Swedish counties.3 The remaining counties utilised BD ProbeTec ET (Becton Dickinson Diagnostic Systems, Sparks, Maryland, USA), which targets a plasmid sequence outside the deletion. After nvCT samples were discovered, they were reported all over Sweden, in proportions of 10–66% in different counties, resulting in huge numbers of false-negative results.4 5 This emergent situation forced the affected companies to develop new nucleic acid amplification tests (NAAT) for C trachomatis diagnostics. Most affected laboratories introduced, ie, pending new diagnostic solutions, the real-time LightMix 480HT PCR (TIB MOLBIOL, Berlin, Germany) targeting the omp1 gene. In Örebro county, Sweden, cell culture was continuously used for diagnosis and LightMix was introduced. Before this introduction, the nvCT proportions were 38% and 25% in culture-positive and LightMix-positive samples, respectively.4 In June 2008, Roche Diagnostics received CE Mark Certification in the European Union for their new real-time COBAS TaqMan CT Test v2.0 (CTM CT v2.0).
After nvCT was detected, the incidence of chlamydia has displayed a continuous increase in Sweden. In Örebro county, in 2007 the incidence was 469.1 per 100 000 inhabitants, which represented an increase of over 51% compared with 2006.6
Our aims were to evaluate CTM CT v2.0 (Roche Diagnostics) for C trachomatis diagnostics and to investigate whether the proportion of nvCT was affected by the introduction of genetic diagnostics detecting nvCT (LightMix) in Örebro county, Sweden.
MATERIALS AND METHODS
See web-only repository.
RESULTS
Performance characteristics of CTM CT v2.0
Of 1059 consecutive PCR samples, 100 (9.4%) were positive in both LightMix and CTM CT v2.0. Fourteen additional samples, urine (n = 10), cervical (n = 1), cervical/urine (n = 1), pharyngeal (n = 1) and vaginal specimens (n = 1), were positive in CTM CT v2.0 only. One of these samples was lost and excluded from further analysis. Ten of the remaining 13 discrepant samples were confirmed using BD ProbeTec ET (eight verified) and/or Abbott m2000rt RealTime CT II (10 verified) as C trachomatis positive. The three remaining low positive samples were not confirmed. However, one of these specimens was from a positive patient (C trachomatis detected in another specimen) and was considered as a true positive.
The concordance with LightMix and CTM CT v2.0 for positive and negative results was 87.7% and 98.5%, respectively. The sensitivity, specificity, negative predictive value and positive predictive value of CTM CT v2.0 was 100%, 99.8%, 100% and 98.2%, respectively. The detection limit of CTM CT v2.0 was one or less C trachomatis genome/chromosome copy per reaction. The sensitivity, specificity, negative predictive value and positive predictive value of LightMix was 90.1%, 100%, 98.9 and 100%, respectively.
Proportion of nvCT in Örebro county, Sweden
From February to December 2007, the prevalence of C trachomatis was 9.4% (1084 positives of 11 577 consecutive samples) and 34.3% (n = 372) were nvCT positive. The nvCT proportion ranged from 23.3% to 48.3% during the different months (fig 1). With the exception of the increasing proportions in October and December the nvCT proportion seemed to decrease from April 2007.
As a result of the unexpected increase in the nvCT proportion among NAAT-positive samples in December 2007 (35.6%), a one-month sentinel surveillance was performed in June 2008, which identified a proportion of 30.9%.
From August to December 2007, 5.0% (116/2306) of cultured samples were positive and 38.8% (n = 45) were nvCT positive. The nvCT proportion ranged from 31.3% to 46.7% during the different months (fig 2).
A one-month sentinel surveillance in June 2008 identified an nvCT proportion among culture-positive samples of 37.5%.
The mean and median age of the patients infected with nvCT was 21.2 years and 20.0 years, respectively. Patients infected with wild-type C trachomatis displayed a mean and median age of 22.3 years and 20.9 years, respectively. For women, the highest proportions of both nvCT and wild-type C trachomatis were identified in the 0–19 year age group followed by the 20–24 year age group. In men, both nvCT and wild-type C trachomatis were most prevalent in the 20–24 year age group followed by the 0–19 year age group and the 25–29 year age group. Notably, for both sexes nvCT was slightly preponderant over wild-type C trachomatis in the 0–19 year age group and for women also in the 20–24 year age group.
DISCUSSION
The CTM CT v2.0, targeting the cryptic plasmid (typically 7–10 copies per bacterial cell) and the chromosomal omp1 gene (single-copy gene), comprised a high sensitivity (100%) and specificity (99.8%) for the detection of C trachomatis. However, these calculations are not ideal because the method used for comparison (LightMix) only detects the omp1 gene. Furthermore, the CE Mark-certified protocol was slightly modified; however, these modifications were previously evaluated and may even represent optimisations.7–12 Three CTM CT v2.0 low positive samples were not confirmed using BD ProbeTec ET or the new Abbott m2000rt CT II assay. However, this does not necessarily mean that those samples were false positive, eg, ProbeTec can sometimes comprise a lower sensitivity than COBAS Amplicor.13 Furthermore, these discordant samples can be caused by, for ProbeTec and Abbott m2000rt CT II, suboptimal sampling tubes, DNA isolation, storage and transport of samples, etc. Most important, because of the dual-target approach of CTM CT v2.0, all serovars of wild-type CT, nvCT and also the rare plasmid-free strains are effectively detected. Overall, dual-target C trachomatis NAAT are certainly preferable for the avoidance of similar emergent situations as the nvCT created in Sweden. Ideally, NAAT for most infectious agents would be provided with several targets, multicopy and/or essential genes, resulting in high sensitivity, specificity and the prevention of false-negative results caused by mutations.
Key messages
-
The CTM CT v2.0 is a sensitive and specific method for C trachomatis diagnostics. However, additional large and well-designed studies are crucial.
-
LightMix 480HT PCR comprises a suboptimal sensitivity (90.1%), which may have caused many false-negative results in Sweden.
-
The nvCT proportion remained high (NAAT, 34.3%; culture, 38.8%) in Örebro county in 2007.
The suboptimal sensitivity of LightMix (90.1%) may be explained by the single-copy target gene (omp1) and the 10-fold lower volume of DNA template. This is most alarming because many Swedish laboratories temporarily shifted to LightMix, pending the development of new assays, and consequently in Sweden many false-negative results may also have been provided using LightMix during recent 1–2 years.
The nvCT proportion remains high in Örebro county, Sweden, despite the introduction of NAAT detecting the mutant. The nvCT proportion in NAAT-positive samples (34.3%) was even higher than in our previous study (25%).4 However, during 2007 the nvCT proportion in NAAT-positive samples decreased mainly monthly from April, ie, 2 months after the introduction of effective nvCT diagnostics. Nevertheless, according to the one-month sentinel surveillance (June 2008), the nvCT proportion has remained high (30.9%). The nvCT was identified in all ages. However, in comparison with wild-type C trachomatis, nvCT was slightly more prevalent in younger ages, which already have the highest incidences of chlamydia. In culture-positive samples, the nvCT proportion was similar to our previous study (38%),4 and thus has not been affected by the introduction of effective nvCT diagnostics. This may indicate that the nvCT/wild-type C trachomatis proportion was already stabilised before the discovery of nvCT, perhaps at least partly because culture diagnostics were also performed before the discovery of nvCT (a smaller proportion of especially cervical samples due to request). Concerning symptoms and signs, sequelae, growth characteristics, cells/DNA load in NAAT samples etc, no significant differences between nvCT and wild-type C trachomatis have yet been identified.4 14 However, overall, our results may imply that nvCT will not (at least in the near future) be eradicated from the Swedish C trachomatis population, comprises strong survival capabilities and biological advantages over wild-type C trachomatis can still not be excluded. Further studies are in progress.
nvCT is of genotype E and a unique multilocus sequence type (21 (hctB), 19 (CT058), 1 (CT144), 2 (CT172), 1 (pbpB)).4 15 16 Extraordinarily, this single clone has been identified all over Sweden but not abroad.3–5 16–24 Only exceedingly rare nvCT cases, mainly Swedish contacts, have been found in Denmark, Norway, Ireland and France.25–28 However, enhanced surveillance, because nvCT may still be in early transmission in other countries, and evaluations of diagnostic methods worldwide are crucial.
In conclusion, CTM CT v2.0 is a sensitive and specific method for C trachomatis detection. However, it would benefit from automation regarding sample preparation and possibilities to analyse more samples per run. Furthermore, studies including higher numbers of patients (symptomatic and asymptomatic) and samples (genital and extragenital), and in comparison with other internationally validated and, ideally, US Food and Drug Administration-approved C trachomatis NAAT are imperative. The nvCT proportion remains high in Örebro county, Sweden, despite the introduction of genetic diagnostics to detect the mutant.
Acknowledgments
The authors thank Björn Herrmann, Uppsala University Hospital and Peter Nilsson, Halland County Hospital, for performing the discrepancy analysis.
REFERENCES
Supplementary materials
Web only appendix 85:3;190-3
Files in this Data Supplement:
Footnotes
-
Additional information on methods published online only at http://sti.bmj.com/content/vol85/issue3
-
Funding: This study was funded by the Research Committee of Örebro County and the Örebro University Hospital Research Foundation, Sweden.
-
Competing interests: None.
-
Contributors: RH, HF and MU designed and initiated the present study as well as analysed all the data; RH performed all the laboratory work; RH and MU wrote the manuscript in collaboration with HF.