Science News

T cell cross-reactivity and the herd immunity threshold

Reposted by Dr. Judith Currys Climate Etc.

Posted on October 14th, 2020 by niclewis | 5 comments

By Nic Lewis

An interesting new paper was recently published by Marc Lipsitch and his co-authors “Cross-reactive memory T cells and herd immunity to SARS-CoV-2”. (1) It discusses immunological and epidemiological aspects and implications of pre-existing conditions: Cross-reactive adaptive immune system memory resulting from previous exposure to circulating cold coronaviruses. They argue that important potential effects of cross-reactive T-cell memory are already contained in epidemiological models based on data on transmission dynamics, particularly with regard to its effects on herd immunity. I believe they are wrong about herd immunity as I will show in this article.

The first point is that cross-reactive T cells have never been seen as a major cause of herd immunity threshold (HIT) (2) for COVID-19 being lower than the often cited {1 – 1 / R0} level, which is usually applies to vaccination. It is typically estimated that the heterogeneity of social connectivity (contact rates) lowers HIT much more than the heterogeneity of biological susceptibility to the causative SARS-CoV-2 virus. (3)

Possible effects of cross-reactive T cells on infection progression

Lipsitch and co-authors note that recent reports have shown that SARS-CoV-2 cross-reactive memory T cells, mostly CD4 + T cells derived from previous exposures to circulating cold coronavirus, occur at ~ 28 –50% of the people demonstrably are not exposed to SARS-CoV-2. They say that only tissue-resident memory T cells (TRM cells) can produce a rapid response, with TCM and TEM T cells taking several days to recirculate to fight infection. They suggest that CD4 + T cells generally limit disease severity, decrease viral load, and / or limit disease duration rather than preventing initial infection.

While I do not intend here to question any of the above, it should be noted that they treat "infection" to include a case where so few cells have been infected that a (RT-) PCR Test would be negative for the virus.

The paper states that CD4 + T-cell-mediated memory responses to a virus include some or all CD4 + TFH cell types (required for the help of B cells and therefore almost all neutralizing antibody responses), TH1 and CTL cell types (with direct antiviral activities in) may include infected tissue) and that the CD4 + T cells involved may be TRM cells or respond more slowly to recirculating TCM / TEM cells.

The authors propose four immunological scenarios for the influence of cross-reactive CD4 + storage T cells on the severity of COVID-19 and virus transmission. The four model scenarios presented are:

Lung Load Reduction: CD4 + T cells reduce COVID-19 symptoms and viral load of the lungs, but have minimal effects on upper respiratory tract viral load (URT). TFH Cell Accelerated Antibodies: CD4 + TFH cells trigger a faster and better antibody response, resulting in an accelerated control of the virus in the URT and in the lungs. TRM cells in the URT: CD4 + TRM cells at the infection site allow rapid control of the virus in the URT and the lungs. Transient infection: The TRM cell immunity "flashes" virus replication in the URT to eliminate all infected cells within one day of the initial infection at the entry portal.

The first three scenarios, together with the case in which no cross-reactive T cells exist, are shown in Figure 1 of the paper reproduced below.

Which models do the data fit?

The authors argue that biological evidence implies that model scenario 4 is very unlikely when only CD4 + T cells are involved. They point out that seroconversion (i.e. a de novo antibody response to SARS-CoV-2) would not occur if the pre-existing immunity of the CD4 + TRM cells were so extreme that significant virus replication would be excluded. Such individuals would not be detectable by virological (e.g., PCR) or serological diagnostic tests and would not shed virus; effectively, these people would be immune to infection and would not be reported as cases. The authors say evidence of other human coronaviruses makes this implausible and that scenario 4 is highly unlikely when epidemiological evidence of very high rates of attack in some ship-based outbreaks is added.

In the most frequently examined ship escape, however, the proportion of infection was below 20%. (4) In addition, the results of a study by Lipsitch et al. not citing (5) show that in households where a person was confirmed to have COVID-19, a significant proportion of the other household members had negative PCR test results, meaning they were not infectious, even though most of them were typical of COVID – had 19 symptoms. In addition, these individuals did not develop detectable SARS-CoV-2-specific antibodies, but rather SARS-CoV-2-specific (as opposed to cross-reactive) T-cell responses, suggesting that they may be related to SARS to some extent. Were infected with CoV. 2.

Despite the fact that the sample size in this study was small, there seems to be some doubt about the Lipsitch et al. To admit that scenario 4 is highly implausible. It also casts doubt on their later claim that almost all people infected with SARS-CoV-2 seroconverts (develop antibodies to it), although the test used may not have been sensitive enough to detect low levels of antibodies. In this context, Lipsitch et al. say a recent study (6) observed (only) about 3 cases of non-PCR, confirmed potentially asymptomatic COVID-19 cases with T-cell responses without seroconversion, but their interpretation of the results of this study has been questioned . (7)

A significant proportion of PCR test positive people – for some localized outbreaks, the vast majority of them – have asymptomatic infections. In the most studied ship escape4, almost half of the infected people remained consistently asymptomatic. (8) If this is due to the cross-reactivity of T cells in combination with innate immune responses, only model scenarios 3 or 4 would fit since then. Model scenarios 1 and 2 imply significant symptoms.

In the remainder of this article, I will not pursue the possibility that model scenario 4 is relevant. Rather, I will focus on showing that the effects of Model Scenario 3 (possibly also Model Scenario 2) on herd immunity, which were varied due to the variation in virus dose and the strength of the innate immune system, most likely have not yet been taken into account in simple epidemiological models based on data of the transmission dynamics. In this context, it should be noted that the volume and quality of the biological and epidemiological data available do not provide high quality evidence, so it is difficult to draw firm conclusions in both cases.

The low level of asymptomatic transmission

Importantly, there is pretty strong evidence that infected people transmit SARS-CoV-2 much less frequently when they are asymptomatic (and not just presymptomatic). Biological evidence neither proves nor disproves that a positive PCR test for SARS-CoV-2 implies significant infectivity (although a negative PCR test can be viewed as an indication of a lack of significant infectivity). (9) Epidemiological evidence, however, strongly suggests that transmission rates from asymptomatic individuals are far lower than those from symptomatic or presymptomatic individuals.

A number of studies have examined the transmission of index cases that remained asymptomatic during their infections. In a review study (10) it was estimated that the average secondary attack rate in the household in asymptomatic cases was only 3.5% of that in symptomatic cases. As found in this study, the household secondary attack rate provides a useful estimate of both the susceptibility to contact and the infectivity of index cases. However, both in this study and in another review study (11) which estimated a much higher proportion than 3.5%, the statistical analysis appears to be seriously flawed. (12) It is therefore necessary to take into account the actual results of the relevant original studies that they have reviewed. Two of these studies (13) (14) found no cases of asymptomatic transmission, although the number of affected contacts was very low in one case. Two further studies each found a case of asymptomatic transmission of 305 and 119 contacts (15) (16) with the respective relative risks of 6% and 19%. Averaging the risk ratios of all four studies in a manner that gives appropriate weight to the evidence presented in each case gives an estimated overall risk ratio of 8% for asymptomatic cases compared to symptomatic and presymptomatic cases. (17)

That is, an asymptomatically infected person, as in Lipsitch et al. Model scenario 3 appears to be only one-tenth or less likely to transmit the virus as a symptomatic or presymptomatic person does. This conclusion need not depend on asymptomatic infected people whose URT has a much lower viral load, which may not be the case. For example, you might transmit less because there is no cough (especially with sputum15), less deep breathing, or a similar factor; because a greater part of their PCR-measured viral load is not a viable virus; and / or because their viral load remains at infectious levels for a shorter period of time.

The likely importance of the virus dose (inoculum) and innate immune responses

Two factors that Lipsitch et al. Not including in their model scenarios, but which likely seem very relevant, are the size of the virus dose and the strength of a person's fast-reacting innate immune system. It is expected that the likelihood and severity of symptoms, as well as the infectiousness of a person exposed to SARS-CoV-2, will depend on the dose of virus involved in their exposure (18) (19) and the severity of theirs innate immune system as well as on any cross-reactive T-cell and / or antibody-adaptive immune system memory. I'm going to focus here on differences that result from the interaction between virus dose and cross-reactive T cells, but in reality, differences in the strength of the innate immune system, as well as general health and other factors, also affect how one affects virus doses of different strengths reacts and the extent to which they become sick and / or infectious.

Box 1: Could a low virus dose explain the COVID-19 epidemic in Tokyo? The results of a study (20) of 1,877 asymptomatic (at the time of the test) company employees from 11 different locations in Tokyo are consistent with the importance of the virus dose. The study showed that the seroprevalence increased from 6% to 47% between late May and late August. If the sample is representative of the Tokyo metropolitan area, which the authors suspect, this implies a seroconversion of about 5.7 million people during the study period.
Since the corresponding number of deaths attributed to COVID-19 appears to be little more than 30, it means that the infection rate in Tokyo could be as low as 0.0006% – about a thousand times lower than commonly thought. In Japan, the very high rate of mask-wearing (and generally high personal hygiene standards) may have reduced virus doses enough that the vast majority of infections are asymptomatic and almost all symptomatic cases are mild, regardless of the presence of cross-reactive T cells. (21) (22)

It is quite possible that a person with cross-reactive CD4 + T cells is either infected so little that – regardless of whether a PCR test with a sufficiently large cycle threshold (high sensitivity) is positive or not – you are infected with a sufficiently low virus dose not only remain asymptomatic, but also have negligible infectivity. With a sufficiently low virus dose, the model scenario 3 by Lipsitch et al. Have similar effects as the model scenario 4 for a high virus dose. For most purposes, it seems inappropriate to view such a non-infectious, healthy person as a COVID-19 case or even infected at all. (23) These people are accordingly treated here as not infected. However, it seems possible that a low viral load, without leading to symptoms or a non-negligible infectivity, still induces effective immunity through the development of SARS-CoV-2-specific antibodies and / or T cells.

In contrast, Lipsitch et al. seem to consider someone infected even if only a single cell in their body has been infected by a virus. Although this definition may be logical from a technical-biological point of view, it does not seem appropriate from an epidemiological point of view. For epidemiological purposes it is relevant whether and to what extent a person is or will become sick, infectious and / or immune.

If the virus dose is quite high, even if a person has cross-reactive CD4 + T cells, they would almost certainly test PCR positive and be more likely to develop symptoms. In these cases, model scenarios 2 and 3 of the paper can be relevant. While in such a case the affected person would be infectious, albeit much less if they were asymptomatic, if they had cross-reactive CD4 + T cells, they would likely be significantly less infectious (and much more likely asymptomatic or have mild symptoms).

If infected people are asymptomatic and have a low (but not negligible) infectivity, it may be that, in cases in which they transmit an infection, the virus dose is usually sufficiently low that the person thus infected is also asymptomatic and of low infectivity In this case, such asymptomatic transmission contributes to the gradual spread of immunity without leading to disease. (24)

Modeling the effects of different susceptibility and infectivity from cross-reactive T cells

I have created a greatly simplified toy model that illustrates the potential impact on epidemic progression and herd immunity of cross-reactive T cells having the effects discussed in this article. The model divides the population into two equal parts, one of which has cross-reactive T cells and the other does not. It differentiates between symptomatic and asymptomatic infections, the latter only having a ninth of the probability of causing an infection as the former.

The detailed assumptions made in the model are listed in an appendix. While these assumptions are illustrative only, they are intended to be largely consistent with the evidence and discussion above in this article. The most important assumptions about the effects of cross-reactive T cells are that their presence halves the risk of infection through potentially infectious contact, quarters the likelihood of symptomatic infection, and can lead to the development of immunity in a significant proportion of immunity occurrences Infection does not occur.

The epidemic modeled is triggered by the symptomatic infection of a naive individual (a person without cross-reactive T cells). The number of close contacts per generation is then adjusted to produce a reproduction number of 2.4 at the beginning of the epidemic, after the epidemic has adjusted from the initial seeding pattern to its natural pattern, which therefore approaches R0.

The projections of the toy model show that after the initial exponential growth, new infections decrease, indicating that herd immunity has been achieved. At this point, 41% of the population was infected, with approximately 43% of the infections being asymptomatic. At 41%, the HIT is 58%, a little over two thirds of the classic HIT level for a homogeneous population. (25)

Another 20% of the population will have become immune with virtually no infection. On the other hand, if no exposed but uninfected (i.e., asymptomatic and non-infectious) individuals develop immunity, the HIT is closer to the classical level, but is still more than 10% below it. If the likelihood of infection is reduced by 85% in the presence of cross-reactive T-cell memory, HIT could be a third less than classic HIT, even if no exposed but uninfected person develops immunity. It is not believed that such a large differential is likely. However, it appears that cross-reactive T-cell memory in combination with a varying dose of virus (and the strength of the innate immune system) can lead to a much lower herd immunity threshold than that obtained from data earlier in the epidemic using a homogeneous population compartment. SIR / was estimated. SEIR models as routinely done.

A more realistic model would include continuous probability distributions for all key parameters. However, the basic point illustrated by the very simple model would still hold true. Homogeneous population-based compartment models imply that epidemic growth proportionally slows down the shrinking pool of uninfected people. However, if there are differences in susceptibility to infection within the population, so that more susceptible individuals are, on average, infected earlier, then epidemic growth will inevitably decline faster. As a result, the herd immunity threshold is lower than that of a homogeneous population, with the reduction in HIT being greater when less biologically susceptible individuals are also less biologically infectious when infected.

Conclusion

I have shown that the Lipsitch et al. That the possible effects on the herd immunity threshold of cross-reactive T-cell memory are already contained in epidemiological models based on data on transmission dynamics is wrong, even if they rightly argue that their model scenario 4 is highly implausible.

In this article, I've only looked at the potential effects of cross-reactive T cells. Even when combined with other causes of interpersonal differences in biological susceptibility, including age, this heterogeneity is not considered to be the main reason why the herd immunity threshold is below classical levels for a homogeneous population. In practice, the interpersonal variation in contact rates (social connectivity) is usually seen as a much more important reason. 3

Appendix – Assumptions made in the toy model about the effects of cross-reactive T cells

The population is one million and is homogeneous, except that only 50% of people have cross-reactive memory T cells. The generation interval is fixed, infected persons are only infectious in the generation interval after infection and then not infectious and immune. Infections are only possible through close contact. The number of close contacts an infected person has made is independent of their T-cell status and whether their infection is asymptomatic (never symptomatic), and each infected person has only had one contact with one during the generation interval that they became infected infectious person you become infected. Close contact between a symptomatic (including presymptomatic) infected and a naive individual (one without cross-reactive T cells) leads to infection in 90% of cases, with 80% of these infections being symptomatic due to a high value of close contact between an asymptomatic Infected and a naive individual leads to infection in 10% of cases, 20% of these infections being symptomatic and the virus dose being lower. Close contact between a symptomatic infected person and a resistant individual (one with cross-reactive T cells) leads to infection in 45% of cases, with 20% of these infections being symptomatic. Close contact between an asymptomatic infected person and a resistant individual leads to infection in 5% of cases, with 5% of these infections being symptomatic. If such a low dose of virus is transmitted in close contact that a resistant recipient not only has no symptoms but is completely non-infectious they will be treated as not infected, but (unless otherwise stated) they will still become immune in 60% of these cases.

Nicholas Lewis October 14, 2020

(1) Marc Lipsitch, Yonatan H. Grad, Alessandro Sette and Shane Crotty: Cross-reactive memory T cells and herd immunity to SARS-CoV-2. Nature Reviews Immunology October 6, 2020 https://doi.org/10.1038/s41577-020-00460-4

(2) The herd immunity threshold is the proportion of the population that has become infected at the point where each new infection does not cause more than one more infection on average. For an epidemic in a homogeneous population, it is {1 – 1 / R0}, where R0 is the basic (initial) reproduction number.

(3) e.g. Tkachenko, A.V. et al .: Persistent heterogeneity, not short-term over-dispersion, determines herd immunity to COVID-19. medRxiv July 29, 2020 https://doi.org/10.1101/2020.07.26.20162420

(4) The diamond princess. 712 of 3,711 on board tested positive for PCR, with at least 295 and likely closer to 334 cases (295 returning from Tokyo plus 40 on charter flights, one of which died of COVID-19) remaining asymptomatic during their infection. https://www.mhlw.go.jp/stf/newpage_11441.html

(5) Gallais, F., Velay, A., Wendling, MJ, Nazon, C., Partisani, M., Sibilia, J., Candon, S. and Fafi-Kremer, S., 2020. Intrafamilial exposure to SARS -CoV-2 induces a cellular immune response without seroconversion. MedRxiv. https://www.medrxiv.org/content/medrxiv/early/2020/06/22/2020.06.21.20132449.full.pdf

(6) Sekine, T. et al. Robust T-cell immunity in convalescents with asymptomatic or mild COVID-19. Cell https://doi.org/10.1016/j.cell.2020.08.017 (2020).

(7) https://twitter.com/WesPegden/status/1313649435642077184

(8) This is in line with the results of Sakurai et al., Natural History of Asymptomatic SARS-CoV-2 Infection. New England Journal of Medicine. 2020 June 12th https://www.nejm.org/doi/full/10.1056/NEJMc2013020

(9) https://www.cebm.net/covid-19/infectious-positive-pcr-test-result-covid-19/

(10) Madewell, Z. J., Yang, Y., Longini Jr., I. M., Halloran, M. E., and Dean, N. E., 2020. Household transfer of SARS-CoV-2: a systematic review and meta-analysis of secondary attack rate. medRxiv. (Version of August 1)

(11) Buitrago-Garcia D. et al. (2020) Occurrence and Transmission Potential of Asymptomatic and Presymptomatic SARS-CoV-2 Infections: A Lively Systematic Review and Meta-Analysis. PLoS Med 17 (9): e1003346. https://doi.org/10.1371/journal.pmed.1003346

(12) Madewell et al. included a study where the secondary attack case was asymptomatic instead of the index case. Buitrago-Garcia D, et al., Who estimated a relative risk of 0.35 for asymptomatic transmission, included in this estimate a study on presymptomatic transmission (their reference 111) and an estimate based on the combined asymptomatic and presymptomatic transmission ( from their reference 65). and they mistakenly estimated a very high risk of asymptomatic transmission from studies that did not identify cases.

(13) Cheng HY, Jian SW, Liu DP, Ng TC, Huang WT, Lin HH et al. Assessment of contact tracing of Covid-19 transmission dynamics in Taiwan and the risk at different exposure periods before and after the onset of symptoms. JAMA Intern Med. 2020. Epub 2020/05/02. https://doi.org/10.1001/jamainternmed.2020.

(14) Park SY, Kim YM, Yi S., Lee S., Na BJ, Kim CB, et al. Coronavirus disease outbreak in the call center,

South Korea. Emerg Infect Dis. 2020. https://doi.org/10.3201/eid2608.201274

(15) Luo L, Liu D, Liao X-1, Wu X-B, Jing Q-1, Zheng J-Z et al. Types of contact and risk of transmission in Covid-19 under close contacts. bioRxiv. 2020 (version of March 26th) https://www.medrxiv.org/content/10.1101/2020.03.24.20042606v1

(16) Zhang W., Cheng W., Luo L., Ma Y, Xu C., Qin P. et al. Secondary transmission of coronavirus disease from presymptomatic individuals, China. Emerg Infect Dis. 2020. https://doi.org/10.3201/eid2608.201142

(17) The averaging over the four studies, weighted according to the number of contacts according to asymptomatic index cases, results in a pooled relative risk estimate of 8.2%. A more differentiated meta-analysis using the “Fixed Effects” function (Mantel-Haenszel) of the R software in the rmeta package estimates the relative risk of a pooled study at 7.9%, with a 75% probability that it will not exceed 13% , and a 90% chance that you won't exceed 20% (assuming the confidence intervals are symmetric).

(18) Steinmeyer, Shelby H., Claus O. Wilke, and Kim M. Pepin. "Methods for Modeling Viral Disease Dynamics on the Scale Within and Between Hosts: The Influence of Virus Dose on Host Population Immunity." Philosophical Transactions of the Royal Society B: Biological Sciences 365.1548 (2010): 1931-1941.

(19) Goyal, Ashish et al. "Wrong person, place and time: viral load and contact network structure predict SARS-CoV-2 transmission and super-spreading events." medRxiv (2020) (version of August 7th).

(20) Hibino, Sawako, et al. "Dynamic Change in COVID-19 Seroprevalence in the Asymptomatic Tokyo Population During Wave Two." medRxiv (2020). (September 23 release)

(21) Gandhi, Monica, Chris Beyrer, and Eric Goosby. "Masks Protect More Than Others During COVID-19: Reduce the Inoculum of SARS-CoV-2 to Protect the Wearer." Journal of General Internal Medicine (2020): 1-4.

(22) Gandhi, Monica, and George W. Rutherford. "Face Masking For Covid-19 – Potential For 'Variolation' While We Wait For A Vaccine." New England Journal of Medicine (2020).

(23) While such cases could lead to a positive result in a PCR test at some point, such a test result could represent evidence of only non-viable virus fragments or a viable viral load that is too low to be infectious.

(24) Unfortunately, there are too few cases of asymptomatic transmission to provide reliable evidence on this point, but what evidence appears to exist agrees with this argument. Zhang et al. identified one case of asymptomatic transmission leading to asymptomatic infection, while 73% of the 11 cases of symptomatic or presymptomatic transmission they identified resulted in symptomatic infection. (Luo et al. Unfortunately did not provide the symptom status of the asymptomatic transmission case they identified.)

(25) For an R0 of 2.4, the classic herd immunity threshold is {1 – 1 / 2.4} = 0.583

Like this:

To like Loading…

Related Articles