Validation of an Aptima-format Finnish new variant of Chlamydia trachomatis (FI-nvCT) surveillance assay, 2019

The Finnish new variant of Chlamydia trachomatis (FI-nvCT) is escaping diagnostics in Finland, Norway and Sweden. We have developed and validated an Aptima-format nucleic acid amplification test (NAAT) designed specifically to detect the FI-nvCT. This NAAT has high sensitivity (100%) and specificity (100%) for the FI-nvCT strain, enabling further investigation of the geographic distribution, prevalence and transmission of this diagnostic-escape mutant in screening populations in Europe.

The Finnish new variant of Chlamydia trachomatis (FI-nvCT) is escaping diagnostics in Finland, Norway and Sweden. We have developed and validated an Aptima-format nucleic acid amplification test (NAAT) designed specifically to detect the FI-nvCT. This NAAT has high sensitivity (100%) and specificity (100%) for the FI-nvCT strain, enabling further investigation of the geographic distribution, prevalence and transmission of this diagnostic-escape mutant in screening populations in Europe.
In February 2019, investigation of discrepant nucleic acid amplification test (NAAT) results led to the discovery of a new genetic variant strain of Chlamydia trachomatis in south-western Finland [1]. This strain, designated Finnish new variant of C. trachomatis (FI-nvCT), harbours a single nt base mutation (C1515T; Escherichia coli numbering) in 23S rRNA. This mutation was determined to be the root cause for compromised detection of the organism by the Aptima Combo 2 (AC2) diagnostic test (Hologic Inc., San Diego, California, United States (US)), which targets C. trachomatis 23S rRNA [2,3]. Archived and newly received clinical specimens in Finland and Sweden with equivocal results or high-negative relative light units (RLU) from 15 to 99 in the AC2 test, and discrepant results between AC2 (target: 23S rRNA) and the separate Aptima Chlamydia trachomatis (ACT) test (target: 16S rRNA) (i.e. AC2 negative/equivocal-ACT positive) signalled the presence of the new strain. Data collected in subsequent investigations suggested a recent emergence of FI-nvCT in south-western Finland, Norway and Sweden with relatively low prevalence (ca 1-6%) among patients with C. trachomatis infections [1,[3][4][5]. We have developed and validated a research-use only Aptima-format surveillance NAAT designed specifically to detect the FI-nvCT strain. This assay should facilitate further investigations to determine the geographic distribution, prevalence and transmission of this diagnostic-escape mutant.

New Aptima-format surveillance assay for the Finnish new variant of Chlamydia trachomatis
To understand the geographic distribution, prevalence and transmission dynamics of the FI-nvCT strain nationally and internationally, we have developed an Aptima-format NAAT that sensitively and specifically detects FI-nvCT (23S rRNA gene C1515T mutation) but not wild-type (C1515) C. trachomatis strains. This research-use only test uses target capture and transcription-mediated amplification chemistries on the automated Panther instrument (Hologic Inc.) to isolate and amplify C. trachomatis 23S rRNA, and an acridinium ester probe to selectively detect the C→T mutation at position 1515.

Validation and verification results for the Aptima-format surveillance assay for the Finnish new variant of C. trachomatis
All validation and verification studies of the new FI-nvCT assay were conducted according to Clinical Laboratory Standards Institute (CLSI) standard protocols used for in vitro diagnostic assay validation [6]. The cut-off value of the test (100,000 RLU) was initially established by testing 454 de-identified remnant clinical specimens obtained from a large reference laboratory in the US. The collection consisted of 266 specimens (endocervical and vaginal swabs, male and female urine, PreservCyt ThinPrep liquid cytology) positive for C. trachomatis by the AC2 assay, and 188 specimens (mixed types) that were AC2-negative ( Figure 1). Signal-to-cut-off (S/CO) ratio values for all clinical specimens tested ranged from 0 to 0.45. Mean S/CO ratio values ranged from 0.02 to 0.071 for C. trachomatis C1515-positive (wild type) specimens, and equalled 0.046 for specimens negative for C. trachomatis. Negative controls included in this analysis (n = 6), consisting of 10,000 copies/mL of wild type (C1515) C. trachomatis RNA, yielded S/CO ratio values ranging from 0.016 to 0.112 (mean: 0.044); positive controls (n = 6) consisting of 1,000 copies/mL of IVT RNA corresponding to FI-nvCT (C1515T) rRNA yielded S/ CO ratio values ranging from 3.60 to 9.02 (mean: 6.38) (data not shown).
The analytical sensitivity of the new Aptima-format FI-nvCT surveillance assay for detecting in vitro transcript (IVT) RNA containing the C. trachomatis 23S rRNA C1515T mutation using the 100,000 RLU cut-off was established by Probit analysis of a panel of six positive and one negative FI-nvCT RNA samples (n = 20 replicates each panel member) ( Figure 2). This analysis established the 95% limit of detection (LOD) for the assay is 231 copies/mL in Aptima sample transport medium and 210 copies/mL in urine sample matrix, equivalent to ca 0.072 infectious units per mL (IFU/mL) and 0.065 IFU/mL, respectively (copies to IFU conversion data not shown).
Analytical specificity (cross-reactivity) and interference (spike and recovery) of the assay against nontarget organisms potentially encountered in urogenital or extra-genital specimens was assessed by testing a minimum of ten replicates of each of 22 pools consisting of 3 to 4 organisms per pool, encompassing 59 bacterial, 2 yeast, 4 protozoal and 19 viral species at high titre ( Table 1). All non-target organisms tested yielded negative results in the assay. To assess nontarget organism interference, 23S rRNA C1515T IVT RNA was spiked to a concentration of three times the assay 95% LOD (660 copies/mL) and tested in the presence of the same high titre pools. All results were positive for the 23S rRNA C1515T-spiked specimens (Table 1) except for Panel 20, which was 95% positive. In addition, culture isolates of C. trachomatis serotypes A, B, Ba, C, D, E, F, G, H, I, J, K, L1, L2 and L3 (all containing the wild type base C1515 in the 23S rRNA) were tested at 1x10 5 IFU/mL each (except serotype G, which was tested at 3x10 4 IFU/mL). All serotypes tested (n = 5 replicates each) yielded negative results in the assay with S/CO ratio values ranging from 0.129 to 0.248 (data not shown).

European validation of the Aptima-format surveillance assay for the Finnish new variant of C. trachomatis
Additional validation studies of the Aptima-format FI-nvCT surveillance assay using spiked specimens and clinical specimens was performed at the World Health Organization (WHO) Collaborating Centre for Gonorrhoea and other Sexually Transmitted Infections in Örebro, Sweden. The results of this validation are summarised in Table 2. Briefly, all 10 WHO Neisseria gonorrhoeae reference strains; 153 isolates of 14 non-N. gonorrhoeae Neisseria species and three Moraxella species; 10 Trichomonas vaginalis (TV) samples, including the two TV ATCC strains 30001 and 50140; five reference strains of four non-TV Trichomonas species; nine C. trachomatis strains of different genotypes and the Swedish new variant of C. trachomatis (SE-nvCT); four non-C. trachomatis Chlamydia species; 428 clinical C. trachomatis wild type positive AC2 specimens; 23 clinical specimens positive for N. gonorrhoeae, T. vaginalis, bacterial vaginosis and Candida species; and 443 clinical AC2-negative specimens, were negative in the FI-nvCT assay. However, all 69 FI-nvCT positive clinical AC2 specimens (verified by 23S rRNA gene sequencing) were positive in the new Aptima-format FI-nvCT surveillance assay ( Table 2).

Discussion
This study presents validation and verification data for a new research-use Aptima-format NAAT for detection of the FI-nvCT strain. The assay selectively detects with high sensitivity (100%) and specificity (100%) C. trachomatis strains containing 23S rRNA with the C1515T mutation found to be the root cause of compromised detection in the AC2 assay and does not detect C. trachomatis strains lacking this nt variance.
Genetic variation in C. trachomatis, a highly prevalent bacterial STI [7], has historically been considered uncommon. However, a recent large-scale genomic analysis of geographically and temporally diverse C. trachomatis strains revealed a remarkable degree of homologous recombination and mutations that have occurred in the organism in the recent millennia [8].
This new understanding provides additional insight into the previous discovery in 2006 of a diagnostic-escape mutant of C. trachomatis that spread undetected in Sweden for 3 to 4 years before its discovery [9][10][11][12]. This SE-nvCT has a deletion in the plasmid DNA targeted by two widely used diagnostic NAATs in use at that time [9], which allowed the SE-nvCT strain to reach high prevalence levels (> 60%) in some segments of the infected population [10].
The automated research-use only Aptima-format FI-nvCT surveillance test described here should facilitate the rapid completion of epidemiologic surveillance studies to estimate FI-nvCT prevalence and to track the geographic and temporal spread of this new diagnosticescape variant in people screened for STIs in Europe. Probit 95% LOD values for detection of RNA containing the C1515T sequence are 210 copies/mL for urine specimens, and 231 copies/mL for sample transport medium (STM) specimens.

Table 1b
Cross reactivity and interference testing of non-target organisms with the Aptima-format Finnish new variant of Chlamydia trachomatis (FI-nvCT) surveillance assay on the Panther instrument, United States, 2019 Table 2 Cross reactivity of non-target organisms (spiked samples) and detection in clinical specimens of the Aptima-format Finnish new variant of Chlamydia trachomatis (FI-nvCT) surveillance assay on the Panther instrument, Europe, 2019 Bacterial species or specimen Number tested FI-nvCT positive FI-nvCT negative