A luciferase immunoprecipitation systems assay enhances the sensitivity and specificity of diagnosis of Strongyloides stercoralis infection

Abstract

Background: We investigated whether luciferase immunoprecipitation systems (LIPS) can be the basis for a more rapid, specific, and standardized assay for the diagnosis of Strongyloides stercoralis infection.

Methods: A LIPS assay was developed based on immunoglobulin (Ig) G or IgG4 antibody to a recombinant Strongyloides antigen (NIE) and was compared with an NIE enzyme-linked immunosorbent assay (ELISA). A second antigen, S. stercoralis immunoreactive antigen (SsIR), was tested alone and in combination with NIE. The assays were tested using serum samples from patients with parasitologically proven S. stercoralis or filarial infections and from healthy, uninfected control subjects.

Results: The NIE LIPS assay based on IgG antibody easily differentiated between S. stercoralis-infected and uninfected patients (P< .0001) and demonstrated improved specificity compared with the NIE ELISA (100% vs. 95%). Serum from filaria-infected patients did not cross-react when tested with the NIE LIPS assay. When SsIR was used in combination with NIE in the LIPS format, sensitivity and specificity improved to 100%, with a 7-fold difference between positive and negative values. No advantage was found in using a LIPS assay based on IgG4. At posttreatment follow-up, a significant decline in antibody titers was detected using the NIE ELISA (P< .0017) and the NIE LIPS assay (P< .0001).

Conclusions: LIPS addresses several limitations of current ELISAs and represents a major advance in the diagnosis of S. stercoralis infection.

Figure 1

Distribution of levels of antibody to NIE, as determined by ELISA (A) and luciferase immunoprecipitation systems (LIPS) assay (B), in patients infected with Strongyloides stercoralis, filaria, or helminths and in healthy, uninfected control subjects (normals). The 6 helminth-infected patients were a subset of the filaria-infected group. C, Levels of antibody to S. stercoralis immunoreactive antigen, as determined by LIPS assay. A significant difference between S. stercoralis–infected and uninfected control subjects was found for all 3 assays (P < .0001; Mann-Whitney U test). No significant difference was found between the control subjects and the filaria-infected patients for the NIE LIPS assay (P = .6909; Mann-Whitney U test).

Figure 2

Anti-NIE titers determined by ELISA vs. luciferase immunoprecipitation systems assay for Strongyloides stercoralis–infected patients and healthy, uninfected control subjects (R ² = 0.8101; P < .0001). Ruc, Renilla luciferase.

Figure 3

Distribution of levels of antibodies to Ruc-NIE (A), Ruc–SsIR (B), and Ruc-SsIR/Ruc-NIE (C) in 33 patients with inconsistent results between the Ruc-NIE– and Ruc-SsIR– based tests or with borderline results close to the designated cutoff (Ruc, Renilla luciferase; SsIR, Strongyloides stercoralis immunoreactive antigen). The solid line in panel B indicates the cutoff for the SsIR luciferase immunoprecipitation systems assay (23,565 luminometer units [LU]); the gray bars in panels A and C highlight the separation between positive and negative values. The difference between the lowest antibody level for an S. stercoralis–infected patient and the highest for a healthy, uninfected control subject (normals) was 71,229 LU for the Ruc-NIE/Ruc-SsIR combination and 15,919 LU for Ruc-NIE alone. P values were calculated using the Mann-Whitney U test.

Figure 4

Antibody titers before and after treatment for Strongyloides stercoralis infection, as determined by NIE ELISA (A), the NIE luciferase immunoprecipitation systems (LIPS) assay (B), the S. stercoralis immunoreactive antigen (SsIR) LIPS assay (C), and the SsIR/NIE LIPS assay (D). A significant difference was found between antibody levels at baseline and at the last follow-up time point (for the NIE ELISA, P < .0017; for the NIE LIPS assay, P < .0001; for the SsIR LIPS assay, P < .0001; for the SsIR/NIE LIPS assay, P < .0001; Mann-Whitney U test). Dotted lines indicate arbitrary cutoff values determined for each assay. Seroreversion was seen for 58% (21/36) of patients by the NIE LIPS assay, on the basis of a cutoff of 24,772 LU; 17% (6/35) of patients by the NIE ELISA, on the basis of a cutoff of 156 U mL⁻¹; and 69% (25/36) of patients by the Ruc-SsIR assay, on the basis of a cutoff of 23,565 LU. The mean duration of follow-up for 36 patients was 17.47 months (range, 6 –32 months). A combination assay, Ruc-SsIR/Ruc-NIE, was tested in 6 patients who did not serorevert by either the NIE LIPS or SsIR LIPS assay (D). Four of 6 selected patients tested by the Ruc-SsIR/Ruc-NIE assay demonstrated seroreversion on the basis of a cutoff value of 20,000 LU. Pre- and posttreatment titers for each patient are shown in the inset in panel D. Asterisks indicate samples demonstrating seroreversion.