CRISPR-Cas-based diagnostics for point-of-care detection of sexually transmitted infections: a laboratory development and evaluation study.

Abstract

Timely, point-of-care diagnosis of sexually transmitted infections (STIs) is crucial for enabling prompt treatment and reducing transmission. We aimed to develop a portable, multiplexed, CRISPR-based assay panel for the detection of Neisseria gonorrhoeae (including the ciprofloxacin resistance marker gyrA S91F), Chlamydia trachomatis, Treponema pallidum, and herpes simplex virus (HSV). In this laboratory development and evaluation study, we developed and optimised four multiplexed, CRISPR-based, diagnostic STI assays for point-of-care use. The complete assay panel comprised a CRISPR TP-HSV (cTP-HSV) panel for the detection of T pallidum and pan-HSV, with reflex testing to distinguish HSV-1 from HSV-2, and a CRISPR NG-CT (cNG-CT) panel for the detection of N gonorrhoeae and C trachomatis, with reflex testing to detect N gonorrhoeae using two additional genome regions and to identify the gyrA S91F mutation. Each pathogen was targeted at two independent genomic regions by isothermal amplification and CRISPR-Cas reaction using Cas12a and Cas13a, each with distinct fluorescent reporters. Analytical specificity and limits of detection (LODs) were determined, and a retrospective, masked concordance study was conducted on genomic DNA from 900 clinical samples (400 for cTP-HSV and reflex testing and 500 for cNG-CT and reflex testing), using quantitative PCR as the reference standard. The diagnostic accuracy of the test was assessed by analysis of receiver operating characteristic curves. The overall sensitivity of the TP-HSV CRISPR assay was 82·5% (95% CI 74·0-88·7) for T pallidum and 94·4% (90·2-97·0) for pan-HSV; LODs were 6·2 copies per μL for T pallidum and 7·8 copies per μL for HSV. Reflex testing gave sensitivities of 97·0% (91·1-99·3) for HSV-1 and 96·0% (89·7-98·7) for HSV-2. The NG-CT CRISPR assay had an overall sensitivity of 80·0% (74·0-84·9) for N gonorrhoeae and 73·0% (65·5-79·3) for C trachomatis, with a LOD of 3·9 copies per μL for both pathogens. Reflex testing for the detection of the gyrA S91F mutation in N gonorrhoeae showed an overall sensitivity of 63·1% (55·1-70·4); however, this was dependent on sample type, with a sensitivity of 85·7% (46·7-99·5) in genital samples and 61·2% (52·8-68·9) in extragenital samples. For all pathogens, assay sensitivity was positively correlated with pathogen load. Area under the curve (AUC) values were 0·90 for T pallidum and 0·99 for pan-HSV in the TP-HSV assay, with values of 0·99 for HSV-1 and 0·97 for HSV-2 obtained in the reflex HSV-1-HSV-2 assay. For the cNG-CT assay, AUC values were 0·90 for N gonorrhoeae and 0·85 for C trachomatis, with a value of 0·72 obtained for gyrA S91F in the reflex cNG-gyrA assay. Our multiplexed, CRISPR-based, point-of-care platform achieved performance consistent with WHO target product profiles for N gonorrhoeae and T pallidum. Proof-of-concept detection of the gyrA S91F resistance marker highlights its potential for resistance-guided therapy. Although optimisation is required before large-scale deployment, this suite offers a promising approach for rapid, decentralised, and resistance-informed STI diagnosis, particularly in resource-limited settings. Victorian Government Department of Health, Australian Government Department of Health, Disability and Ageing and Aged Care, and Australian Research Council.