The serodiagnositic value of Chlamydia trachomatis antigens in antibody detection using luciferase immunosorbent assay

Abstract

Introduction: Among the different antigens used in the detection of anti-Chlamydia trachomatis antibodies, significant differences in sensitivity and specificity have been observed. Further evaluation of C. trachomatis antigens in antibody detection is urgently needed for the development and application of C. trachomatis serologic assays.

Methods: Chlamydia trachomatis antigens Pgp3, TmeA, InaC, and HSP60 were selected and used in luciferase immunosorbent assay (LISA). The detection results obtained from well-defined C. trachomatis positive and negative samples were compared with the commercial C. trachomatis ELISA (Mikrogen) for performance evaluation.

Results: Pgp3, TmeA, InaC, and HSP60-based LISA showed sensitivity of 92.8, 88.8, 90.4, and 94.4%, and specificity of 99.2, 99.2, 99.2, and 92%, respectively. ROC analysis indicated that Pgp3-based LISA showed similar performance to Mikrogen ELISA (AUC 0.986 vs. 0.993, p = 0.207). Furthermore, four C. trachomatis antigens achieved strong diagnostic efficiency, i.e., positive likelihood ratios [+LR] ≥ 10 in C. trachomatis-infected women and negative likelihood ratios [-LR] ≤ 0.1 in C. trachomatis negative low exposure risk children, but only Pgp3 and TmeA showed strong diagnostic value in general adults. In addition, Pgp3, TmeA, and InaC, but not HSP60, achieved high performance, i.e., both positive predictive value (PPV) and negative predictive value (NPV) ≥ 90.9%, and showed no significant cross-reactivity with anti-Chlamydiapneumoniae.

Conclusion: Three C. trachomatis species-specific antigens Pgp3, TmeA, and InaC show superior performance in the detection of anti-C. trachomatis antibody, indicating the potential to be used in developing C. trachomatis serologic tests.

Keywords: Chlamydia trachomatis; heat shock protein 60 (HSP60); inclusion membrane protein for actin assembly (InaC); luciferase immunosorbent assay (LISA); plasmid-encoded protein 3 (Pgp3); translocated membrane-associated effector A (TmeA).

Figure 1

ROC curves of different Chlamydia trachomatis antigen-based LISA and commercial ELISA in detecting anti-C. trachomatis antibody. 125 C. trachomatis NAAT-positive and 125 C. trachomatis negative from low exposure risk children sera were used. The average Log2RLU signals of LISAs and the OD values of commercial ELISA were used as predictor variables. The solid color lines indicate the maximum likelihood-fitted ROC curves.

Figure 2

Diagnostic utility of anti-Chlamydia trachomatis antibody assays by likelihood ratios in control sera (A) and general adults (B). Sensitivities were calculated at specificities ranging from 70% (left) to 99% (right). Using sensitivity and specificity data of ROC curves which C. trachomatis exposure status was set as the gold standard for control sera and the determination of Mikrogen ELISA was set as the gold standard for general adults, positive-likelihood ratios (+LR) and negative-likelihood ratios (−LR) were calculated. The three-gray shaded areas indicate the zones of strong (+LR ≥ 10, −LR ≤ 0.1), moderate (+LR ≥ 5, −LR ≤ 0.15), and poor (+LR ≥ 2.5, −LR ≤ 0.25) performance of the assays, corresponding to the strong, moderate and poor diagnostic efficiencies.

Figure 3

Diagnostic suitability of anti-Chlamydia trachomatis antibody assays by predictive values in control sera (A–D) and general adults (E–F). Positive and negative predictive values are dependent on the prevalence of populations of anti-C. trachomatis antibodies. Four specificity cut-off values and resultant sensitivities in ROC curve analysis were selected for panels (A,E; 90% specificity), (B,F; 95%), (C,G; 98%), and (D,H; 99%). The three-gray shaded areas indicate the zones of high (PPV and NPV ≥ 90.9%), moderate (PPV and NPV ≥ 83.3%), and poor (PPV and NPV ≥ 71.4%) diagnostic efficiency of assays.