Clinical evaluation of commercial SARS-CoV-2 serological assays in a malaria endemic setting

Abstract

The levels of immune response to SARS-CoV-2 infection or vaccination are poorly understood in African populations and is complicated by cross-reactivity to endemic pathogens as well as differences in host responsiveness. To begin to determine the best approach to minimize false positive antibody levels to SARS-CoV-2 in an African population, we evaluated three commercial assays, namely Bio-Rad Platelia SARS-CoV-2 Total Antibody (Platelia), Quanterix Simoa Semi-Quantitative SARS-CoV-2 IgG Antibody Test (anti-Spike), and the GenScript cPass™ SARS-CoV-2 Neutralization Antibody Detection Kit (cPass) using samples collected in Mali in West Africa prior to the emergence of SARS-CoV-2. A total of one hundred samples were assayed. The samples were categorized in two groups based on the presence or absence of clinical malaria. Overall, thirteen out of one hundred (13/100) samples were false positives with the Bio-Rad Platelia assay and one of the same one hundred (1/100) was a false positive with the anti-Spike IgG Quanterix assay. None of the samples tested with the GenScript cPass assay were positive. False positives were more common in the clinical malaria group, 10/50 (20%) vs. the non-malaria group 3/50 (6%); p = 0.0374 using the Bio-Rad Platelia assay. Association between false positive results and parasitemia by Bio-Rad remained evident, after adjusting for age and sex in multivariate analyses. In summary, the impact of clinical malaria on assay performance appears to depend on the assay and/or antigen being used. A careful evaluation of any given assay in the local context is a prerequisite for reliable serological assessment of anti-SARS-CoV-2 humoral immunity.

Keywords: Antibodies; COVID-19; Mali; Plasmodium falciparum; Validation.

Fig. 1

Seropositivity rate in pre-pandemic samples by three different commercial serological assays and by malaria status. Archived pre-COVID-19 plasma samples (N = 100) were tested using three commercial assays, namely Bio-Rad Platelia SARS-CoV-2 Total Antibody, Quanterix Simoa Semi-Quantitative SARS-CoV-2 IgG Antibody Test (anti-Spike), and the GenScript cPass™ SARS-CoV-2 Neutralization Antibody Detection Kit, as shown on the X-axis of each panel. Green and yellow bars in Fig. 1A represent percentage of negative and positive samples, respectively per assay type. Similarly, numbers on the top of the bar graphs indicate the percentage of positive and negative samples identified by each assay. In Fig. 1B, seropositivity was disaggregated based on the assay being used (on X-axis) and presence or absence of clinical malaria. Bar graphs show number of seropositive samples (on Y-axis) in malaria cases (yellow) and healthy individuals (green). Data labels on top of each graph represent number of seropositive samples as indicated on the Y-axis. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Relationship between seropositivity and malaria status by Bio-Rad Platelia in pre-pandemic samples. In this Figure, samples tested with the Bio-Rad Platelia assay were stratified based on malaria status, namely 50 Healthy (green bar) and 50 Malaria cases (yellow bar), as presented on the X-axis. The Y-axis shows the proportion of seropositive samples in each group, which is also indicated on top of each bar graph. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Relationship between specimen ratio by Bio-Rad and parasitemia or age. Each dot indicates a given sample analyzed with Bio-Rad. We plotted parasitemia (Fig. 3A) or age (Fig. 3B) on the X-axis against the specimen ratio (mean OD value of the sample/ cut-off value) on the Y-axis. Dotted vertical lines indicate parasitemia or age categories. Values in percentage in each category represent the proportion of false positive samples detected within each parasitemia or age category.