Viral Shedding and Antibody Response in 37 Patients With Middle East Respiratory Syndrome Coronavirus Infection

Abstract

Background: The Middle East respiratory syndrome (MERS) coronavirus causes isolated cases and outbreaks of severe respiratory disease. Essential features of the natural history of disease are poorly understood.

Methods: We studied 37 adult patients infected with MERS coronavirus for viral load in the lower and upper respiratory tracts (LRT and URT, respectively), blood, stool, and urine. Antibodies and serum neutralizing activities were determined over the course of disease.

Results: One hundred ninety-nine LRT samples collected during the 3 weeks following diagnosis yielded virus RNA in 93% of tests. Average (maximum) viral loads were 5 × 10(6) (6 × 10(10)) copies/mL. Viral loads (positive detection frequencies) in 84 URT samples were 1.9 × 10(4) copies/mL (47.6%). Thirty-three percent of all 108 serum samples tested yielded viral RNA. Only 14.6% of stool and 2.4% of urine samples yielded viral RNA. All seroconversions occurred during the first 2 weeks after diagnosis, which corresponds to the second and third week after symptom onset. Immunoglobulin M detection provided no advantage in sensitivity over immunoglobulin G (IgG) detection. All surviving patients, but only slightly more than half of all fatal cases, produced IgG and neutralizing antibodies. The levels of IgG and neutralizing antibodies were weakly and inversely correlated with LRT viral loads. Presence of antibodies did not lead to the elimination of virus from LRT.

Conclusions: The timing and intensity of respiratory viral shedding in patients with MERS closely matches that of those with severe acute respiratory syndrome. Blood viral RNA does not seem to be infectious. Extrapulmonary loci of virus replication seem possible. Neutralizing antibodies do not suffice to clear the infection.

Keywords: MERS; antibodies; clearance; shedding; viral load.

Figure 1.

Viral loads in patients with Middle East respiratory syndrome coronavirus (MERS-CoV). Mean viral loads in positive-testing samples per day and specimen type. Maximum and minimum viral loads are shown as purple and cyan lines, respectively. Error bars represent standard deviation. Sample numbers and proportion of positive samples are summarized in Supplementary Figure 2.

Figure 2.

Correlation of serum viral RNA detection with neutralizing antibodies and viral RNA concentration in respiratory samples. A, Neutralizing antibodies; B, Viral RNA concentration in lower respiratory tract samples. Columns in both panels show serum viral load. Empty spaces represent serum samples that tested negative for viral RNA.

Figure 3.

Distribution of RNA viral loads in lower respiratory tract Middle East respiratory syndrome coronavirus (MERS-CoV) samples in 3 time windows. Columns show viral loads for each patient averaged over the time windows indicated to the right of each panel. Curves represent ideal normal distributions based on sample means and variance.

Figure 4.

Kinetics of antibody production. The red line shows mean immunoglobulin G (IgG) titer, represented as optical density (OD) ratios obtained from S1 enzyme-linked immunosorbent assay (ELISA). The orange line shows mean immunoglobulin M (IgM) titer from an immunofluorescence assay (IFA). The cyan line shows virus microneutralization titer (NT). Titers from each patient are averaged over successive 3-day time intervals.

Figure 5.

Effect of serum antibodies on lower respiratory tract (LRT) viral loads. This analysis is based on paired serum and LRT samples taken from the same patient on the same day. Antibodies are shown as line graphs. Viral loads in the corresponding LRT samples are shown as columns. The panels show samples sorted according to increasing levels of enzyme-linked immunosorbent assay (ELISA) optical density (OD) ratios (A) or neutralizing antibody titers (B). Sample numbers for this analysis are summarized in Supplementary Figure 3.