One of the critical and key uncertainty about the COVID-19, among so many, is the immunity a patient may have after recovery from the COVID-19. In other words, can someone who recovered from the COVID-19 catch the disease again and infect other people again?
As long as we have neither vaccine nor fully efficient antiviral treatment, specific acquired immunity, i.e. the immunity developed as the body fights then recovers from the disease, is one of the key variables at the center of the few solutions we have to handle the pandemic. Because, as we saw, we shall not be able to use vaccination for immunisation before at the very best winter 2022, and considering the uncertainty considering treatments against the SARS-CoV-2, specific acquired immunity becomes even more important.
This immunity is also key to determine exit strategies to isolation and lockdown. Indeed, one of the components of the exit strategy that may be designed is to allow people who have developed an acquired immunity to return to normal life (e.g. Ran Balicer, “Coronavirus: Two Things Must Happen Before Initiating Exit Strategy“, Haaretz, 2 April 2020).
Thus what do we know or not, so far, about this immunity? How can we handle the uncertainty? Finally, what does that imply for an exit strategy? This is what we shall see in this article.
Many questions and few answers, yet.
In a nutshell and schematically, when a pathogen such as the SARS-CoV-2 enters the body, the immune system develops a range of reactions to fight against the intruder and attacker (for a very interesting clear, detailed biological and medical explanation see, for example, “Features of an Immune Response“, in Immune System Research, National Institute of Allergy and Infectious Diseases). The creation of antibodies is one of these responses. Antibodies will attack the intruder. If the immune system is victorious against the SARS-CoV-2, then the patient recovers. His or her body keeps traces of war that took place. The patient will now also have an acquired immunity (e.g. Encyclopaedia Britannica, “Immune System“).
However, as Morgane Bomsel, virologist and immunologist underlines:
“The question is to know if it [the acquired immunity] will be protective of nor, and how long it will last” (La question est de savoir si elle va être protectrice ou pas, et combien de temps elle va durer) .in Camille Gaubert, Interview with Morgane Bomsel, “Covid-19 : l’immunisation pourrait, chez certains, ne pas protéger d’une deuxième infection“, Sciences et Avenir, 1 April 2020)
Protective acquired immunity after recovery from the COVID-19?
First, thus, one needs to find the various components of the acquired immunity in the body. For example, antibodies need to be present in such a quantity that they are sufficient to prevent infection again (Wu, IBId., Callow, K A et al., ibid.). Such antibodies were detected in a patient with mild-to-moderate symptoms “before symptomatic recovery. These immunological changes persisted for at least 7 d following full resolution of symptoms” (Thevarajan, I., Nguyen, T.H.O., Koutsakos, M. et al., “Breadth of concomitant immune responses prior to patient recovery: a case report of non-severe COVID-19“, Nat Med; 2020).
Then, Linlin Bao, et al., in a not yet peer-reviewed article, showed on rhesus macaques that those could not be reinfected, “after the symptoms were alleviated and the specific antibody tested positively”, at 5 days post-infection (“Reinfection could not occur in SARS-CoV-2 infected rhesus macaques“, bioRxiv, 14, March 2020.
On 27 March 2020, the Helmholtz Centre for Infection Research (HZI) in Germany announced the start of a much larger study, on 100.000 individuals. The donors’ “blood will be regularly tested for antibodies against the Covid-19 pathogen. The study will provide a more accurate picture of immunity and pandemic development.” The center goes on highlighting that following this study, one may imagine giving a kind of immunity certificate to people who developed an immunity, which would allow them to return to normal life (Ibid.). The tests should start in April 2020 and first results should be available at the end of the same month (Veronika Hackenbroch, “Large Antibody Study to Determine Germans’ Immunity to Covid-19“, der Spiegel, 27 March 2020). Improvements in the test procedure – thus reliability of the study – should take place between end of May 2020 and end of June 2020 (Ibid,).
Thus, it would seem, according to what we now know, that we indeed obtain a protective acquired immunity. Utmost caution, however, must still be exerted waiting for other studies’ results, such as the German study.
Furthermore, we must also account for the possibility, that, for some individuals, a different immune response develops. In two other coronaviruses, the SARS and the MERS, for some people, the antibodies facilitated infection rather than preventing it Camille Gaubert, Interview with Morgane Bomsel, “Covid-19 : l’immunisation pourrait, chez certains, ne pas protéger d’une deuxième infection“, Sciences et Avenir, 1 April 2020). Favourable results for experiments in vitro gave opposite, negative results in vivo experiments (ibid.). If such were the case for the SARS-CoV-2, however, possible negative impacts of antibodies could then be blocked with adequate treatment (ibid.). However, this would be again more pharmaceutical efforts to endeavour.
The possible existence of such individuals that would then possibly be more fragile after infection needs to be thoroughly deepen and then checked before general measures are applied to the population.
Length of the protective acquired immunity
However, antibodies remain in the body, for a while (e.g. interviews with virologists and immunologists in Katherine J. Wu, “What Scientists Know About Immunity to the Novel Coronavirus“, Smithsonian Magazine, 30 march 2020; Callow, K A et al. “The time course of the immune response to experimental coronavirus infection of man.” Epidemiology and infection vol. 105,2 1990; Gaubert, Ibid.).
But how long is this while? This is the first unknown we face. Antibodies usually diminish with time then disappear (Wu, ibid.). Thus, how long do we keep these antibodies? For how long will the acquired immunity be protective?
Then, another related question is about the immune memory: will antibodies be able to remember the attacker well enough to generate the proper response (Wu, Ibid.)?
Thus, to summarise, for our purpose the key question is: for how long will the acquired immunity be protective?
Currently, although we do not know with certainty, most scientists seem to consider as likely hypothesis that, in general, patients who have recovered from the COVID-19 will be sufficiently immunised, for a while.
The possible length of the naturally acquired immunity considered varies.
Indeed, our knowledge of the SARS-CoV-2 is extremely recent. It started with recorded data mainly in January 2020. Thus at the beginning of April 2020, we cannot know with certainty the possible length of the immunity beyond 2 to 3 months. This is one more reason why monitoring what is happening in China, where the first patients recovered, is so important.
Various hypotheses are considered.
If the SARS-CoV-2 is similar to the coronavirus giving the common cold, then some scientists state that the immunity could last “years” (Interview with Angela Rasmussen, a virologist at Columbia University in Brian Resnick, “The 9 most important unanswered questions about Covid-19“, Vox, 20 March 2020). However, other results, obtained with the coronavirus 229E, show a more complex picture, as some individuals could also, in experiment, become reinfected a year later (Callow, K A et al. “The time course of the immune response to experimental coronavirus infection of man”, Epidemiology and infection, vol. 105,2, 1990).
If the coronavirus behaves as the seasonal flu, the hypothesis the Imperial College COVID-19 Response Team used, then re-infection is considered as “highly unlikely in the same or following season” (Impact of non-pharmaceutical interventions (NPIs) to reduce COVID19 mortality and healthcare demand, 16 March 2020, p. 4). However, the seasonal flu is not a coronavirus.
The condition and age of the patient, as well as genetics, may also impact the response of the immune system (Wu, Ibid.).
Finally, mutations may occur as the virus duplicates itself, leading to new strains the body cannot recognise, as with the seasonal flu (Wu, Ibid.). This is however less likely for coronaviruses than for flu viruses (Ibid.). But coronaviruses can also “trade segments of their genetic code with each other”, which allows them to trick the immune system. (Ibid.). In that case, the acquired immunity would be useless. Note that this would also be true for a vaccine.
Virologists and immunologists most probably have other much more specific questions to which they need to find answers.
Thus, with such a new disease, we are still faced with many uncertainties. How can we handle them?
Impacts on the architecture of scenarios
Again, scenarios are a crucial tool to handle these uncertainties.
Our scenario structure is currently as follows. The main scenario we consider as most likely is that we shall have to wait for a vaccine until winter 2022 (at best) (see Hélène Lavoix, The COVID-19 Pandemic – Surviving and Reconstructing, The Red (Team) Analysis Society, 24 March 2020, last updated 3 April 2020). Then we need to account for the possibility to see the emergence of treatments impacting the disease (see Hélène Lavoix, Covid-19 – Scenarios – Making Sense of Antiviral Treatment, The Red (Team) Analysis Society, 30 March 2020).
Now, ideally, we would need to have another epidemiological layer of models and scenarios that vary to include various possibilities for the acquired immune response. We would build the next layer of our scenarios out of these.
Until such detailed epidemiological models are available, if ever, we need to handle the variable “immunity” as correctly as possible, through different sub-scenarios useful for our purpose. The best way at this stage is to consider a first batch of sub-scenarios where a fully protective immunity is developed upon recovery and to have this immunity vary according to time.
Considering that the detailed epidemiological model that many governments use is the model the Imperial College COVID-19 response team developed (Ibid.), it is interesting for our purpose to look at a scenario that is less optimistic than the “same season and next one” immunity they used, for example less than one year, one that is the Imperial College scenario and one that is more optimistic, for example, an immunity that lasts from one and a half to two years.
That said, the Imperial College’s model shows that “temporary suppression” (with social distancing of the entire population, case isolation, household quarantine and school and university closure) is the only way forward not to overwhelm the health system and to prevent massive fatalities. It also shows that because this suppression is successful, then only a small number of individuals will develop immunity. Hence, for a collective approach focused necessarily on health, fatalities and not overwhelming the health system, variations on the acquired immunity, because they play on small numbers, may not be a key variable.
Things are, however, more challenging for the second objective all polities must fulfil, i.e. ensuring the fundamental security a society needs to survive as well as not to break down (see The COVID-19 Pandemic – Surviving and Reconstructing, and Summary of previous findings in Covid-19 – Scenarios – Making Sense of Antiviral Treatment). Indeed, critical functions must continue, and, as much as possible, a new economy must start emerging. As a reminder, the first objective is to reduce as much as possible the fatalities resulting from the disease (see Summary of previous findings, ibid.).
Hence the need for sub-scenarios that consider acquired immunity and its length.
Finally, to make sure we cover the whole range of possible futures, we may create a “complex immunity” scenario that would actually cover all other cases. This scenario would, for example, include a situation where the acquired immunity varies so much according to diverse criteria that it becomes difficult, rapidly, to create adequate understanding and thus policies. It could also be used if our knowledge is so uncertain and the involved risks are so high that similarly, no policy can be created easily. With time, or according to the decision-makers for whom the actionable scenarios are created, this “cluster scenario” would need to be developed adequately.
This “complex scenario” would be the least favourable.
Immunity and exit strategy
We must highlight first that the theories and models created to handle the exit of the “suppression/isolation” period need to account for the immunity uncertainty.
Thus, considering the high cost in lives and sufferings, as well as the impacts across domains, we must consider all scenarios. We cannot consider only the most likely and most preferable scenario. Actually, we need either to make sure that policies will be correct across scenarios or that they are flexible enough to switch in a timely way from one scenario to another. In that case, this demands precise monitoring and warning that will allow to steer policies, again in a timely manner. This flexibility should also allow to fully integrate new understanding and new results on the length and protection of the acquired immunity, as they become known.
Policies also need to be correct at both individual and collective level, considering the high stakes in terms of legitimacy for political authorities. For example, the policies should try to consider the possibility of individual variations in terms of acquired immunity.
In terms of exit strategy, for example, the current assumption, considering the early results (see above) is that people who were positive to the COVID-19 and recovered, now have a protective immunity to the SARS-CoV-2. However, it does not seem that the length of the immunity is, so far, taken into account.
The challenge thus becomes, in terms of handling of the pandemic and exit of the isolation/suppression phase to identify who has antibodies. If we wanted to also make sure the length of the immunity is considered, then we would need to make sure that a possible fading of the immunity can be identified.
The answer to this need will be in the serological tests, which are currently developed worldwide (Chad Terhune, Allison Martell, Julie Steenhuysen, “U.S. companies, labs rush to produce blood test for coronavirus immunity“, Reuters, 25 March 2020; Gretchen Vogel, “New blood tests for antibodies could show true scale of coronavirus pandemic“, Science, 19 March 2020; Hugo Jalinière, “Les tests de sérologie, clé du déconfinement“, Sciences et Avenir, 30 mars 2020; Lauren Chadwick, “Coronavirus: Antibody tests ‘will be crucial’ in determining when to lift lockdowns“, Euronews, 6 April 2020; for a list of commercially developed tests all categories, not only serological, see Find, Covid-19 Diagnostics resource centre).
Assuming that the tests are reliable, we nonetheless find again the familiar problem of quantities. The ongoing “war for face masks” is very likely to be again reproduced, this time, with tests. Masks as well as serological tests become crucial stakes to fulfil the two objectives of societies faced with the COVID-19 pandemic. Those who will be able to develop and secure for their populations as much and as many of the necessary tools – including smart strategies – to both survive and ensure the fundamentals of security, will survive best. Furthermore, they are also likely to be earlier and better able to interact again with each other.
To account for the length of the immunity, if the tests which are developed cannot detect early enough a fading of immunity, then testing many times subjects may become necessary. However, here the problem of test quantity – and test operationalisation – increases. Continuing imperatively protective gestures as well as a generalisation of face masks is thus likely to be necessary to compensate for insufficient serological testing.
With the next articles we shall continue exploring the factors which are key to build the general architecture of our scenarios.
Some detailed references and bibliography
Callow, K A et al. “The time course of the immune response to experimental coronavirus infection of man.” Epidemiology and infection vol. 105,2 (1990): 435-46. doi:10.1017/s0950268800048019
Linlin Bao, Wei Deng, Hong Gao, Chong Xiao, Jiayi Liu, Jing Xue, Qi Lv, Jiangning Liu, Pin Yu, Yanfeng Xu, Feifei Qi, Yajin Qu, Fengdi Li, Zhiguang Xiang, Haisheng Yu, Shuran Gong, Mingya Liu, Guanpeng Wang, Shunyi Wang, Zhiqi Song, Wenjie Zhao, Yunlin Han, Linna Zhao, Xing Liu, Qiang Wei, Chuan Qin, “Reinfection could not occur in SARS-CoV-2 infected rhesus macaques“, bioRxiv, 14, March 2020, 2020.03.13.990226; doi: https://doi.org/10.1101/2020.03.13.990226
Thevarajan, I., Nguyen, T.H.O., Koutsakos, M. et al. Breadth of concomitant immune responses prior to patient recovery: a case report of non-severe COVID-19. Nat Med (2020).https://doi.org/10.1038/s41591-020-0819-2
Shi, Y., Wang, Y., Shao, C. et al. COVID-19 infection: the perspectives on immune responses. Cell Death Differ (2020). https://doi.org/10.1038/s41418-020-0530-3