Robustness of serologic investigations for Chikungunya and Mayaro viruses following coemergence

Abstract

Since 2013, the arthropod-borne Chikungunya virus (CHIKV) has cocirculated with the autochthonous Mayaro virus (MAYV) in Latin America. Both belong to the same alphavirus serocomplex, termed the Semliki Forest serocomplex. The extent of antibody cross-reactivity due to the antigenic relatedness of CHIKV and MAYV in commonly used serologic tests remains unclear. By testing 64 CHIKV- and 37 MAYV-specific sera from cohort studies conducted in Peru and Brazil, we demonstrate about 50% false-positive test results using commercially available enzyme-linked immunosorbent assays (ELISAs) based on structural antigens. In contrast, combining ELISAs for CHIKV and MAYV significantly increased positive predictive values (PPV) among all cohorts from 35.3% to 88.2% for IgM and from 61.3% to 96.8% for IgG (P < 0.0001). Testing of longitudinally collected CHIKV-specific patient sera indicated that ELISA specificity is highest for IgM testing at 5 to 9 days post-onset of symptoms (dpo) and for IgG testing at 10 to 14 dpo. IgG cross-reactivity in ELISA was asymmetric, occurring in 57.9% of MAYV-specific sera compared to 29.5% of CHIKV-specific sera. Parallel plaque reduction neutralization testing (PRNT) for CHIKV and MAYV increased the PPV from 80.0% to 100% (P = 0.0053). However, labor-intense procedures and delayed seroconversion limit PRNT for patient diagnostics. In sum, individual testing for CHIKV or MAYV only is prone to misclassifications that dramatically impact patient diagnostics and sero-epidemiologic investigation. Parallel ELISAs for both CHIKV and MAYV provide an easy and efficient solution to differentiate CHIKV from MAYV infections. This approach may provide a template globally for settings in which alphavirus coemergence imposes similar problems. IMPORTANCE: Geographically overlapping transmission of Chikungunya virus (CHIKV) and Mayaro virus (MAYV) in Latin America challenges serologic diagnostics and epidemiologic surveillance, as antibodies against the antigenically related viruses can be cross-reactive, potentially causing false-positive test results. We examined whether widely used ELISAs and plaque reduction neutralization testing allow specific antibody detection in the scenario of CHIKV and MAYV coemergence. For this purpose, we used 37 patient-derived MAYV-specific sera from Peru and 64 patient-derived CHIKV-specific sera from Brazil, including longitudinally collected samples. Extensive testing of those samples revealed strong antibody cross-reactivity in ELISAs, particularly for IgM, which is commonly used for patient diagnostics. Cross-neutralization was also observed, albeit at lower frequencies. Parallel testing for both viruses and comparison of ELISA reactivities and neutralizing antibody titers significantly increased diagnostic specificity. Our data provide a convenient and practicable solution to ensure robust differentiation of CHIKV- and MAYV-specific antibodies.