“Any definitive conclusion on SARS-CoV-2 immunity needs more evidence”

November 9, 2020 0 By S Harachand

Dr Chandrakant Lahariya is a medical doctor and a leading public policy and health systems expert. He has received advanced training in epidemiology, vaccinology, immunology and public health. In the last 15 years, he has worked closely with India’s topmost health-policy makers to develop and implement several health policies and programmes at both the union and state levels. He is the youngest fellow ever elected to the Indian Public Health Association (IPHA).

In an interaction with FM, Dr Lahariya points out that while serosurveys remain the best tool to fathom the level of population immunity due to past infection, it is important that the results of these studies be interpreted carefully, factoring in their context and limitations.

Edited excerpts:

Data from metros like Mumbai and Delhi show that seroprevalence is declining in consecutive surveys across population groups. Is the disparity a result of different methodological approaches or is it indicative of a deeper problem specific to SARS-CoV-2, such as quickly vanishing antibodies?

We really cannot say that the trend is declining till we see such a pattern in comparable surveys and we have the confidence intervals (CI) along with the point prevalence, which are being reported. The findings of the third serosurvey in Delhi are not comparable with the previous two, and definitely, the methods are much more robust. As for serosurveys in Mumbai, we don’t really know the CI. However, at least the difference from 16% to 18% is unlikely to be statistically significant.

It is too early to become concerned about the level of antibodies. There is evidence that even when there are no detectable antibodies, the person could be protected. In addition, the antibody response after vaccination is far superior to antibody response after natural infection. Therefore, any definitive conclusion needs more evidence, which is emerging with every passing day.

Critics argue that the serosurvey reports conducted in India may not be representative because they are not conducted under a uniform guideline nor are the sampling methods/sizes statistically significant. What is your comment?

For a country of the size of India, no matter what sample size and approach is used, it will always have some limitations. That is true for serosurvey at the national, state and district levels. An ideal, not necessarily the best, approach would be to test the entire population, which is neither feasible and nor required. Clearly, serosurveys are the best alternative. The role of the scientific community is to minimise those limitations. It is important that the limitations are acknowledged, and we keep improving the methods adopted. Most important is that interpretations are done carefully, factoring in the context and limitations.

The case-to-infection ratio has also fallen significantly; according to most recent serosurveys. What does this suggest? Is the community spread of the virus slower than predicted?

The falling case-to-infection ratio is a reflection of the improved implementation of the test, trace and isolate strategy. COVID-19 testing services have increased across the country. Tracing efforts have improved and people are also more aware, and volunteer for testing. For the spread of the virus, there is no expected speed or rate, and the only direction we want is downward.

Virologists question the sensitivity and specificity of most of the widely deployed antibody tests. How does a high percentage of false positives or negatives impact the results of a survey?

The antibody kits which have been used for serosurvey have fairly high sensitivity and specificity. Specific approaches have been followed to further enhance these parameters within the serosurvey. In any case, these reflect past infections, and therefore, while sensitivity and specificity are important, these are not everything. While it results in limitations in stand-alone surveys; repeat surveys provide very useful guidance on trends.

Can we make serological testing the only way to assess the immunity level of an individual, particularly at this point of time when the herd immunity threshold for SARS-CoV-2 is still being debated?

Everything in COVID-19 provides a part of the answer and no single approach is enough for a complete answer. There are many unknowns, such as what is the protective level of antibodies, how long the immunity lasts and what proportion of infected individuals develop detectable levels of antibodies. Our knowledge and understanding are getting better with every passing day. In this backdrop, we should use all the approaches at hand – serosurveys, antibody levels, immunity research and epidemiological data – to measure immunity levels. One of the impacts of the pandemic has been that we have learnt to use various tools across specialties — including clinical research, epidemiological tools and scientific data — in coherence.