Comparing lateral flow testing with RT-PCR to detect SARS-CoV-2

During the early stages of the coronavirus disease 2019 (COVID-19), rapid testing was imperative for public health policymakers and health care workers in order to determine where and what measures would be necessary to slow the spread of the disease, and to ensure infected individuals were isolated and treated correctly.

The first tests available were probe-based assays reliant on reverse transcription-polymerase chain reaction (RT-PCR) to amplify certain areas of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2 nucleocapsid (N) gene. While RT-PCR was not designed for this purpose, the protocols became more optimized and several reliable assays were approved over time. However, this method of detecting SARS-CoV-2 requires a laboratory and trained staff – making it relatively expensive and slow, as each sample has to be sent away for testing. Lateral flow devices (LFDs) are point-of-care tests that require significantly more time to develop than RT-PCR assays but can be used locally with minimal to no training, some of which can provide results in as little as 30 minutes. However, there have been some worries about the sensitivity of such kits. As a result, researchers from Micropathology Ltd have been comparing the two methods.

A preprint version of the group’s study is available on the medRxiv* server, while the article undergoes peer review.

The basic premise behind most lateral flow devices (or lateral flow immunoassays) is simple. The finished product is a ~5mm wide strip of backing card, upon which is placed a sample pad, filter pad, conjugate pad, a membrane that can bind to antibodies (normally nitrocellulose), and an absorbent pad.

The conjugate pad contains an antibody that can bind to the target molecule conjugated to a marker, such as colloidal gold. The membrane has two lines of antibodies bound to it: one of which will only bind to the target, while the other will bind to the antibody found in the conjugate pad. The rest of the membrane will be blocked to prevent non-specific binding, or a blocking agent will be present in the buffer. The sample is loaded onto the sample pad alongside a buffer.

Agents in the buffer release the conjugated antibody and marker from the pad, which mix with the sample and flow along the test together. If the target is present, the antibody conjugated to the marker will bind to one epitope, and then the antibody immobilized on the membrane will bind to another. This causes the marker to accumulate on the line sprayed on the membrane if the target is present. The control line will always show positive, as long as the sample was loaded correctly.

Comparing LFD with RT-PCR. Samples that were positive in at least 2 out of the 3 replicates were classified as positive. A green data point represents a sample where at least 2 out of 3 replicate tests were positive. A red data point indicates a sample where 0 or 1 of the 3 replicates were positive. Of the 52 positive samples tested, 51 were positive in all 3 RT-PCR replicates. One sample was positive in 2 out of 3 RT-PCR replicates. Photographs of the LFD device show examples of positive (top) and negative (bottom) test results. The 95% confidence interval was calculated using a Probit regression analysis

The researchers compared the government-supplied Innova LFD with a known RT-PCR protocol with a wide range of samples known to be positive for SARS-CoV-2. In total, they used 62 stored samples, including ten negatives. Both tests successfully identified all ten negatives on all three replicates used.

Of the 52 positive samples, PCR identified 51 as positive in all three replicates and the final sample as positive in two out of three replicates. LFD testing, unfortunately, revealed significantly fewer positives samples, only identifying 24 as positive. Generally, samples above 1 million copies per mL showed positive results on LFD testing. Two samples at 3 million and 1.5 million showed negative results on all three LFD replicates – this may be due to the samples being positive for a different strain of SARS-CoV-2 than the tests were designed to show.

The authors make conclude that LFD testing is of limited merit for reducing SARS-CoV-2 infection in the community. The authors point to the reasonably high risk of false negatives and argue that these could provide a false sense of security, leading to further transmission. However, government advice has never been merely to take one LFD, but to return to a testing data three days later or to get tested weekly.

As well as this, the testing in this study only accounts for one brand, and LFD testing for COVID-19 has been active significantly less time than PCR testing, with less time for optimization. The reduced cost and extra convenience of LFDs allow much more regular testing, and while they cannot detect SARS-CoV-2 to the same sensitivity as PCR testing, multiple experts have suggested that they can detect SARS-CoV-2  as long as it is of transmissible levels.

*Important notice

medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information

Written by

Sam Hancock

Sam completed his MSci in Genetics at the University of Nottingham in 2019, fuelled initially by an interest in genetic ageing. As part of his degree, he also investigated the role of rnh genes in originless replication in archaea.