Development and evaluation of a multiplex serodiagnostic bead assay (BurkPx) for accurate melioidosis diagnosis

Abstract

Burkholderia pseudomallei, the causative agent of melioidosis, is a gram-negative soil bacterium well recognized in Southeast Asia and northern Australia. However, wider and expanding global distribution of B. pseudomallei has been elucidated. Early diagnosis is critical for commencing the specific therapy required to optimize outcome. Serological testing using the indirect hemagglutination (IHA) antibody assay has long been used to augment diagnosis of melioidosis and to monitor progress. However, cross reactivity and prior exposure may complicate the diagnosis of current clinical disease (melioidosis). The goal of our study was to develop and initially evaluate a serology assay (BurkPx) that capitalized upon host response to multiple antigens. Antigens were selected from previous studies for expression/purification and conjugation to microspheres for multiantigen analysis. Selected serum samples from non-melioidosis controls and serial samples from culture-confirmed melioidosis patients were used to characterize the diagnostic power of individual and combined antigens at two times post admission. Multiple variable models were developed to evaluate multivariate antigen reactivity, identify important antigens, and determine sensitivity and specificity for the diagnosis of melioidosis. The final multiplex assay had a diagnostic sensitivity of 90% and specificity of 93%, which was superior to any single antigen in side-by-side comparisons. The sensitivity of the assay started at >85% for the initial serum sample after admission and increased to 94% 21 days later. Weighting antigen contribution to each model indicated that certain antigen contributed to diagnosis more than others, which suggests that the number of antigens in the assay can be decreased. In summation, the BurkPx assay can facilitate the diagnosis of melioidosis and potentially improve on currently available serology assays. Further evaluation is now required in both melioidosis-endemic and non-endemic settings.

Fig 1. Serum reactivity of melioidosis-confirmed patients and non-melioidosis individuals to B. pseudomallei whole cell lysate ELISA.

Serum IgG (A) or IgM (B) reactivity as determined by B. pseudomallei 1026b whole cell lysate ELISA. Triangles represent ELISA analyzed serum (n = 585) from 436 individuals (36 melioidosis and 400 non-melioidosis individuals) and squares represent selected serum samples to evaluate potential diagnostic antigens. The overlap between the negative and the positive samples are shown in grey.

Fig 2. Heat map of MAGPIX multiplexed assay for IgG (A) and IgM (B).

Each row represents an individual antigen, and each column represents an individual serum sample (n = 188). The sera are stratified by non-melioidosis negative and melioidosis samples. The negative group is further stratified by the 1026b whole cell lysate (WCL) reactivity (overlap and non-reactive IgG and IgM). The melioidosis serum is grouped by the sample’s collection time after initial sample collection. For the heat map, blue indicates 0 median fluorescent intensity (MFI) for both panels. For IgG panel A, yellow indicates 5,000 MFI and red indicates >20,000 MFI. For IgM panel B, yellow indicates 4,000 MFI and red indicates >8,000 MFI. Black indicates samples where data was not collected.

Fig 3. ROC analysis of B. pseudomallei lysate (multiple antigens—A) and single antigens (B-D) selected by day 21 area under the curve (AUC) analysis performance.

Responses to 1026b WCL (A), CPS (B), GroEL (C), and LPS B (D) antigens are shown using non-melioidosis human sera samples compared with day 0 (black), day 21 (purple), or all melioidosis-confirmed patient samples (dashed). The area under the curve (AUC) value for each data set is shown.

Fig 4. p-hat distribution using the LASSO (A and C) and Ridge Regression (RR, B and D) models.

Samples were stratified based on the patient status and collection time. The models were developed using IgG (A and B) and IgM (C and D) reactivity. A $\widehat{\text{p}}$ (p-hat) value of 0 is unlikely to be melioidosis and a value of 1 is likely to be melioidosis. A cutoff to call sensitivity and specificity is shown by the dotted line and corresponds to Table 4. $\widehat{\text{p}}$ value significance was determined by a Kruskal-Wallis test followed by a Dunns pairwise comparison. The Day 0 vs Day 21 was not significant for all antibody isotype models.

Fig 5. ROC analysis of the LASSO and Ridge Regression (RR) IgG and IgM models using human patient samples.

A (p-hat) $\widehat{\text{p}}$ score were generated from the multiple antigen reactivity (median fluorescent intensity shown in Fig 2) for IgG or IgM. n = 188 patient samples with 112 paired samples from 56 individuals.

Fig 6. Comparison of the IHA results to the multivariate serology assay.

IHA results (X-axis) and the multivariate p-hat probability score (Y-axis) are shown for the same sample collected from culture confirmed melioidosis patients. The different multivariate models are shown in separate panels (A-D). The day 0 (red circle) and day 21 (white squares) are shown. The cutoff for the IHA assay is 1:40 and the multivariate model cut-offs are consistent as shown in Table 5. The interpretation of the cutoff panels and which assay would correctly diagnose melioidosis is shown in panel A. A total of 40 samples that have both IHA and multivariate model values were compared.