Detecting anti-SARS-CoV-2 antibodies in urine samples: A noninvasive and sensitive way to assay COVID-19 immune conversion

Abstract

Serum-based ELISA (enzyme-linked immunosorbent assay) has been widely used to detect anti-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibodies. However, to date, no study has investigated patient urine as a biological sample to detect SARS-CoV-2 virus-specific antibodies. An in-house urine-based ELISA was developed using recombinant SARS-CoV-2 nucleocapsid protein. The presence of SARS-CoV-2 antibodies in urine was established, with 94% sensitivity and 100% specificity for the detection of anti-SARS-CoV-2 antibodies with the urine-based ELISA and 88% sensitivity and 100% specificity with a paired serum-based ELISA. The urine-based ELISA that detects anti-SARS-CoV-2 antibodies is a noninvasive method with potential application as a facile COVID-19 immunodiagnostic platform, which can be used to report the extent of exposure at the population level and/or to assess the risk of infection at the individual level.

Fig. 1.. IgG immunological conversion to SARS-CoV-2 N protein in patient urine samples.

Patients are represented by the lines interconnected by the points (sample collected in different days PSO). The immunological conversion observed for each patient varied in days PSO. The plotted index values (I) are related to the absorbance ratio of the cutoff values. Positive index value, above 1.1; indeterminate index value, between 0.8 and 1.1 (gray); and negative index value, below 0.8.

Fig. 2.. Urine and serum index values (I) of each patient according to the days PSO.

The index values obtained from urine and serum samples for each patient are represented by circles and diamonds, respectively. Both samples, from the last day before immune conversion (index < 1.1) and from the first day of immune conversion (index > 1.1), collected from the same patient were plotted and are indicated by the matching symbol below the patient number. Only the first sample collected was plotted for those patients with all samples with positive index. Individual data were divided according to the PSO days of the collection date: <10 (green), 11 to 15 (blue), 16 to 20 (yellow), and >20 days (red). Positive index value, above 1.1; indeterminate index value, between 0.8 and 1.1; and negative index value, below 0.8.

Fig. 3.. Evaluation for SARS-COV-2 diagnosis by using rSARS-CoV-2 N protein against patient urine and serum samples.

ELISA was done using urine and serum samples (n = 209 and n = 187, respectively) from COVID-19 patients with positive qRT-PCR. Urine and unpaired serum samples from healthy subjects (n = 30 and n = 37, respectively) were also used. The mean of each group is shown, and the dashed line indicates the cutoff value determined for each type of biological sample (urine = 0.123 and serum = 0.323). The cutoff values were determined as the mean plus three times the SD of negative samples. Bottom: Positive sample groups are divided according to the PSO days of the collection date: <10 (green), 11 to 15 (blue), 16 to 20 (yellow), and >20 days (red).

Fig. 4.. ROC curve for comparative diagnostic performance of urine- and serum-based ELISA for COVID-19.

ROC curves were constructed using the individual OD values for each sample to obtain the sensitivity, specificity, and area under the curve.