SARS-CoV-2 lateral flow assays for possible use in national covid-19 seroprevalence surveys (React 2): diagnostic accuracy study
BMJ 2021; 372 doi: https://doi.org/10.1136/bmj.n423 (Published 02 March 2021) Cite this as: BMJ 2021;372:n423Read our latest coverage of the coronavirus outbreak
All rapid responses
Dear Editor,
SARS-CoV-2 lateral flow assays are used in national covid-19 seroprevalence surveys (React 2) and a number of lateral flow immunoassays have been evaluated for possible use.[1] Any lack of sensitivity of the assays can be compensated for mathematically if the sensitivity and specificity are known. Seroprevalence estimates can be adjusted taking into account the sensitivity and specificity of the antibody tests used, according to the method of Rogan and Gladen (1978) [2], for instance. If it is the frequency of positive tests, then rather than use this value directly to estimate seroprevalence, the sensitivity, denoted by a, and the specificity, denoted by b, can be corrected for to give the seroprevalence as (t-b-1)/(a-b-1). For lateral flow assays with very high specificity (approximately 1), the seroprevalence can be adjusted to t/a. If, for instance, the sensitivity of the tests is only 80%, missing a fifth of the actual positives, then the adjusted seroprevalance is found by dividing the positive frequency by 80% (equivalent to adding 25%). In the case of 8% positive tests, the adjusted seroprevalence estimate would be 10%.
In order to make these adjustments, all that is required is that the sensitivity and specificity are known. It has been noted that there can be a difference in sensitivity for tests using blood from a finger prick compared with serum assayed in a laboratory.[1] AbC-19 has sensitivity of only 69% using finger prick blood, compared to 92% for serum tested in the laboratory. The reasons for differences in assay sensitivity between finger prick tests and serum assayed in the laboratory could be due to assay inhibitors in blood that have been removed in serum, and perhaps also the expertise of those performing the assays. In such cases using the sensitivity relevant to the particular assay method would give the most accurate results. If, in the AbC-19 example, assays were done using finger prick blood but the assay sensitivity for lab tests on serum was used in calculations then the adjusted seroprevalance would be estimated to be only 75% (92%/69%=75%) of the correct value, 6% rather than 8%, for example. When seroprevalence is properly adjusted it is the accuracy in determining sensitivity that is critical rather than how high the sensitivity is. If sensitivity criteria could be relaxed by making the appropriate adjustments to seroprevalence then other criteria such as cost could be considered.
[1] BMJ 2021; 372 doi: https://doi.org/10.1136/bmj.n423
[2] Rogan W and Gladen B 1978 “Estimating prevalence from the results of a screening test” Am J of Epidemiology Vol. 107 pp 71 – 76
Competing interests: No competing interests
Use of Soberana 02 (Cuban Vaccine) in Iran.
Dear Editor
The vaccine candidate against COVID-19 Soberana 02 complements phase III of its clinical trials in Iran with the arrival in that country of 100,000 doses (Pasteur Institute). The study of this project was planned to be carried out abroad, with this objective, at the beginning of the year, and a collaboration agreement was signed with the Pasteur Institute of Iran. This synergy would allow faster progress in immunization in both countries against the SARS-CoV-2 virus, which causes COVID-19. In the event that the final efficacy results of Soberana 02 are confirmed, the mass production of this product will be launched both in Iran and in Cuba. For this, the technical methodology will be transferred to Tehran through such collaboration.
Last Monday, Cuba began phase III of clinical trials of Soberana 02 in eight municipalities of Havana with more than 44,000 volunteers. In this way, it became the first candidate from Latin America to reach that stage. Among the objectives of the study are to evaluate the efficacy of the product to prevent symptomatic disease and, in addition, that the individual does not develop serious forms of the disease or die.
Competing interests: No competing interests