Research progress, current and future course of COVID-19: Revisiting the basics I Part 1
07 May 2020
As we sail through the COVID-19 pandemic, we all would acknowledge the speed with which things are progressing, including the research on COVID19. The onslaught of information – numbers, graphs, predictions, etc. – has informed as well as overwhelmed many of us. It is with this view that the 'Webinar on COVID-19: Ask The Experts – Part 6’ hosted by DBT/Wellcome Trust India Alliance (India Alliance), THSTI, IAVI and Nature India on 1 May 2020 aimed to take stock of recent scientific advances, public health actions and the future course of this pandemic.
Our expert guests for the webinar were Dr. Gagandeep Kang, Dr. Shahid Jameel and Dr. Jacob John. This is the first part of edited excerpts from the webinar.
On the learnings from the course of the pandemic
Gagandeep Kang: There's been a lot of advancement that has happened in the recognition of the illness as well as in the science that informs our understanding of disease and of potential interventions. When we started out, we thought that this was fever and cough, and that this would result in an interstitial pneumonia. That made sense because we knew that the virus bound to a receptor that was expressed on numerous sites. We now know that that receptor is very widely expressed, including endothelial cells, gut and potentially even other organs. And some of what we are seeing in terms of disease is paralleling the distribution of that receptor. So, our understanding of the clinical picture is changing quite rapidly.
Shahid Jameel: We have learned over the past few weeks that this is neither like the first edition of SARS that appeared in 2003 nor like the common flu that everyone's been talking about. Now we are beginning to understand that, unlike SARS, here virus shedding is happening before symptoms even appear. 2020 is very different from 2003 – it's different in terms of technology – it has taken only two months for the first COVID19 vaccine to roll out from a company into human trials. So that's the power of technology that gives us immense hope that we will be able to overcome this.
But there is also a darker side to technology - the infodemic or the fake news that has spread now, which is much worse than what we saw in 2003. What we have also learned is that while the virus may be very inclusive, in the sense that it affects the royals and commoners, it doesn't care what your religion is, it doesn't care whether you are rich or poor, the pandemic certainly isn't inclusive. It has hit the poor and vulnerable much harder than it has hit those of us who are privileged. I think that is something that has not been addressed too much at the level of scientists, and I hope scientists are also going to be addressing the thin future.
Jacob John: I think it's been a very interesting two months and there's been a lot of information to sift through. We're learning a lot about how the disease spreads, and that the asymptomatic people have a lot more role to play in this pandemic. There's still a lot to be learned and part of the reason is that, while we have a lot of data accessible, we're not very sure about quality of the information that's coming out. So, I think the next couple of weeks are going to be very important for us, both in terms of learning about the disease and its treatment and its transmission, and also to prepare ourselves to be more equitable about how we handle these problems.
Suggested reading: Opinion | In India, COVID-19 Pandemic Has Painfully Laid Bare Our Societal Faultlines. Shahid Jameel, Outlook India.
On the measures in research and diagnostics space to detect and handle spread from asymptomatic people
Shahid Jameel: You only test when somebody comes up with symptoms because the PCR test, which is the confirmatory diagnostic test, is a cumbersome test to do. It's an expensive test and you can't really deploy it on large populations; what you can deploy on large populations are antibody tests. However, antibody tests don't become positive for about 7 to 10 days into symptoms. So, for the asymptomatic phase, we really have no diagnostic right now.
Gagandeep Kang: The only way that we know that there are asymptomatic infections is when contacts are tested and frequently you will find that contacts will test positive even without having symptoms. So, we have an estimate of the proportion of contacts that test positive and then you can follow them up to see what proportion develop symptoms. There are clearly people who do not develop symptoms. A large proportion of the infections that people are getting now are acquired from asymptomatic individuals.
Jacob John: There has been a lot of conflicting literature about what proportion of population is asymptomatic. The initial estimates were about 37% would be symptomatic and the rest would be symptomatic, but that's changing quite dramatically. What we saw with SARS in 2003 is that people became infectious really when they became symptomatic, which is not the case for COVID-19. As for research and diagnostics for asymptomatic individuals, I think serology has huge value. If a lot of people get a symptomatically infected and if they mount an immune response that protects them, then that leads us towards having a population which has better levels of immunity. If that doesn't happen, then all of us are still susceptible to the infection.
On predicting the course of COVID-19 in India
Shahid Jameel: Models are all based on assumptions. Any model is only as good as the assumptions you make, and usually early in any outbreak the data is very limiting, so you make a lot of assumptions. As time goes on, you refine your models and a lot of that is happening now. The other point is that we can't make a general model on an entire country because conditions vary. For example, the population density of Delhi or Mumbai is very different from the population density of Arunachal Pradesh or the desert area of Rajasthan. The disease is not going to spread at the same rate in these places. So, what is going to be more useful in future would be state-specific models, region-specific models and as we go on, if decisions are going to be based on data.
Jacob John: Using doubling time of COVID-19 as a measure to make projections is a good idea, in the sense that you can use doubling time to tell how fast the outbreak is going. However, doubling time is a great tool if testing is uniform or probability-based and is consistent over a time period. Unfortunately, in an epidemic, it's difficult to do that. One, because we have shortage of kids; second, we have strategies which test only contacts and so on. So, there are caveats – if it’s only being used to see we are on the upward slope of the epidemic or we are plateauing, that's fine. But, if you try to do more about comparing India's trajectory against Italy’s trajectory or something like that, that becomes more difficult.
On how long till we achieve herd immunity
Gagandeep Kang: The idea behind herd immunity is that, as the population gets larger and larger some members of the population move from a susceptible to immune group. They tend to afford some kind of protection to those who are not yet infected and who are still susceptible. Herd immunity is a fundamental concept of all infectious diseases. As the population reaches a certain level of immunity, the transmission of a disease depends on three parameters. It depends on probability of transmitting an infection to a susceptible person, the number of contacts that you have with susceptible people, and the duration of contact. If we have to have immunity, some of this effect can also be generated through vaccination.
Herd immunity is generally the idea that the population which is susceptible is protected by people who have already become immune to the past. About 60% of the population needs to be infected before you can really think of breaking transmission to a level at which the epidemic is no longer active.
On differences or similarities between animals and humans, in immune response to the virus
Gagandeep Kang: All viruses induce an immune response; whether that response is protective or not, is something that needs to be determined in the case of coronavirus. We know from seasonal coronavirus (common cold) that the antibody response tends to be quite short-lived, that is, there's a short-term protection of a few months and then you become susceptible again. In the case of MERS and SARS, which are the two other coronaviruses, there is an antibody response where not only do the antibodies bind to antigen on the virus, but also show some level of functionality. Functional antibodies or functional immune responses are measured in viruses mostly by neutralization; this means that if you have an antibody, it binds the virus and stops the virus from infecting cells.
In the case of SARS coronavirus to we know that antibodies are made, and that they are neutralizing antibodies. We are seeing that people who have severe disease tend to have a stronger immune response than people who are asymptomatic. There are studies that have been done in animal models, one of the most urgent tasks was to try and establish good animal models.
Shahid Jameel: Besides the adaptive response, the first line of defense is the innate response. How it is able to control the virus early on decides the future course of action. In people who died of coronavirus disease, there was a hyperactive immune response called the cytokine storm that includes a lot of inflammation and accumulation of pus in the lung, which is generated by the host fighting the disease. It’s interesting that other mammals that harbor many of these viruses never really showed signs of this disease. Data suggest that bats don't get that hyperactivation or cytokine storm, which is seen in humans.
On whether the mutations in this virus will affect vaccine and therapeutics development
Shahid Jameel: By now, there are close to 5,000 virus sequences in public databases, and they show that while the virus is evolving it's not changing enough to become a different virus. There are very confusing terms that are being used, for example, different strains of the virus. These are not different strains of the virus, but different groups of viruses. To give you an example, HIV has two types of viruses HIV1 and HIV2 – those are different strains of HIV.
So, it depends on how much of the virus has evolved from the parental strain to be called a separate strain. There is no evidence so far that we have multiple strains circulating around; the mutation one sees is not enough to suspect at this time that diagnostic test or vaccine won’t work.
What can we learn from the genomes of the novel coronavirus? Chitra Pattabiraman, covidgyan.in
When will a COVID-19 vaccine be ready? Shahid Jameel, 22 March 2020, The Hindu.
On the possible reasons behind slower spread and low number of deaths in India
Jacob John: I can think of three points, at least. First, we went into a lockdown early in and out. It's a good thing that we have decreased the contact between individuals and that might have had an impact and that's the biologically possible bit. But I think what we are really seeing is just a tip of the iceberg. We are later in the epidemic than most other countries, and our testing algorithms have been very restrictive to say the least. India is a very large country proportionate to the rest of the world, and we've only been testing if you had history of contact with the case for very long time. We don't see enough of the severe acute respiratory infections being tested.
Deaths come much later after the infection than we earlier anticipated, so deaths will come. I think we are too early in this game to talk about how many deaths or cases are going to be there. We've done some things well such as being able to decrease contacts. But decreasing contacts has only bought us time, it's not changed the entire story yet. The story is going to play out over a few months and that's what we’ll see. Hopefully there are things that might work in our favor – one is the extremely high temperatures, but there's no evidence yet to suggest that that's going to happen.
On comparison between different testing methods
Gagandeep Kang: RT-PCR is essentially detecting the presence of the virus – whether the virus is capable of replication or not – you're looking for the nucleic acid inside the virus. Genome of the virus is a long string of RNA which is wrapped in a ball of protein; you can either look for the RNA or you can look for the protein. Nucleic acid amplification technologies look for the RNA. So, RT-PCR is a very sensitive test – it detects the really tiny amounts of the virus. People say that the sensitivity of RT-PCR is 70%, which means that out of hundred infected people at the time that you test them, you will get only 70 RT-PCR positive. How is that sensitive? It's because a lot of the problems with the tests come in how samples are collected for the tests. Also, it’s not easy to do a nasopharyngeal swab or a deep oropharyngeal swab, so your test is going to be a false negative. So, RT-PCR assays need to be done well to be sensitive. They are also very specific, so you don't usually get false positives on an RT-PCR.
An antigen test looking for the protein would be extremely useful, because most of the formats that are used for antigen detection are ones that are easily available in many parts of the country. These can include ELISA formats that many laboratories are familiar with. You could also do the antigen testing in a variety of different ways. You could have approaches that are aptamers which is short strings of nucleic acids that can bind to various components. You can have approaches that make it into a really rapid test, such as the lateral flow formats where you put a drop of your sample on a piece of paper and then it gives you a signal that is detected as lines on the paper. These kinds of lateral flow formats have already been adapted for the identification of antibodies such as IgM, which comes up first in infection followed by IGG. These are also the rapid tests that have just been withdrawn from India because of their poor performance.
Antibody tests measure how you respond to the viral infection, so they don't become positive early in the infection. Your body needs time to recognize that the virus is there and then react to it by making antibodies. So, you don't see antibodies till at least a week to two weeks after you have the infection. Usually the first antibody you'll see is IgM and that comes up 7 to 10 days after you have started to have symptoms. IgG antibody comes up usually in the third week of infection. By the time the antibodies start coming up, your RT-PCR is going to start to miss cases because there will be less and less virus since the body is now handling the infection.
For every test, you have to identify the users of that test as well as its limitations. So, if you have to do testing in the first week, it has to be RT-PCR or antigen test, and and from the second week onwards, you can think about antibody test. Be aware that antibody tests have limitations in that their sensitivity and specificity is not perfect; you can have both false positives as well as false negatives with his test.
Suggested reading: What is the rapid antibody blood test for COVID-19? Shahid Jameel, Fit (The Quint), 28 April 2020
* * * * * * * * * * * * * *
Watch the webinar HERE.