As a suffer of hay fever from grass pollen allergies, I think you may be onto something omportant--you body is already reacting to one alien "invader", so that could that "tune up" the body to be on the alert for others, or at least throw out the virus along with the other garbage?

Dear Craig,

Thank you for commenting.

Yes, that's the idea. "Non-specific" stimulation of the immune system to be better prepared against an infection is not an uncommon medical technique. For example, prophylactic treatment with interferons can be done, or, as in the case of COVID-19 is now being tried, infection with a live mycobacterium using the BCG vaccine. However, "non-specific" is an oversimplification, since different agents provoke different immune strategies, and those different strategies can block each other.

If you have hay fever, for a few weeks or months your body thinks that it is infected with parasites. That definitely is a strong stimulation of the immune system (as you as hay fever patient very well know), and this should be considered in regard to COVID-19, especially since it concerns the lungs.

Hay fever (and parasite infections) mounts a "Type 2" immune response. A Type 2 immune response is relatively protective of tissues, since most parasites are extracellular, whereas a Type 1 immune response eliminates infected cells together with the virus. Type 2 immunity and Type 1 immunity have a suppressive effect on each other.

In the case of COVID-19, fatalities are importantly caused by inflammation caused by the immune system. Among the different immune responses, we can assume that luminal IgA, which is present in the lungs and often described as "non-inflammatory", helps protect against the COVID-19 virus (SARS-CoV-2) without adding much to the inflammation-related death. Thus, the more someone is protected by IgA, the less he will depend on more inflammatory immune responses, and, probably, the less likely he will get severe symptoms.

IgA production is stimulated by Type 2 immunity, especially by the cytokine IL-5, of which the concentration is quite high in hay fever patients. So, although this needs evidence, a logical hypothesis is that hay fever patients have a stronger IgA response against COVID-19 and may thus better be protected.

While my above idea is relatively simple, it suffers from a thorough knowledge of luminal IgA in the lung. Most studies on IgA have been dedicated to the intestine instead.

So I welcome any comments that can elaborate on the above idea.

By the way, I just found an interesting article showing that luminal (mucosal) IgA can help protect against the SARS virus (SARS-CoV-1, which is quite similar to the COVID-19 virus). The article is: Du et al. J Immunol 2008; 180:948-956;

http://www.jimmunol.org/content/180/2/948

There might bepossibility of protection due to hay fever

I think at least three different conditions in relationship to the virus could be investigated: 1.) Hay fever sufferers, 2.) Chronic asthma, and 3.) Chronic gum disease. The Hay fever sufferers may be actively battling the pollen and eliminating it, whereas asthma and gum disease are chronic inflammation sites where the body is not winning the battle, so could be areas where the virus could get through the weakened defenses?

Dear Craig,

You are right that disease conditions will affect susceptibility to COVID-19 both by immune priming and pathology, possibly in opposing directions, and that this may be most favorable in the case of hay fever.

Nevertheless, also asthma may counteract COVID-19 susceptibility, as an interesting article from China statesArticle Risk factors for severity and mortality in adult COVID-19 in...

"Interestingly, the prevalence of asthma in COVID-19 patients 365 (0.9%) in our study was markedly lower than that reported in the adult population of Wuhan 366 (6.4%). We thus speculate that Th2 immune response in asthmatic patients may counter the 367 inflammation process induced by SARS-CoV-2 infection. Further studies are required to 368 characterize the immune response and inflammation features of COVID-19."

As you see, they also speculate that type 2 immunity ("Th2 immune response") is involved. However, of the only five asthma patients in their study of SARS-CoV-2 positive patients, three showed severe symptoms, which did not significantly deviate from the overall division into severe and non-severe cases in their study. So, rather than being better protected by immunity, asthma patients may be better in not catching COVID-19, possibly by their behavior. Nevertheless, the observation in their study is very interesting and their model may be true.

It would be nice if in Japan a similar study could be done as this one in China, but then including hay fever as well (although the season is coming to an end). In Japan, maybe 60-70% of people is affected to some degree by the pollen (just an estimation by me and some friends), with more medical estimations of "hay fever" being around 25% of the population.

One way or the other, positive (by immune priming) or negative (by immune priming or pathology), such pre-existing immune condition in the respiratory tracts of such a large part of the population should affect the reproduction rate of the virus.

Also in Western countries asthma patients are underrepresented, meaning that asthma might have a protective component

https://medicalxpress.com/news/2020-04-asthma-common-covid-patients-died.html

and

https://www.thelancet.com/journals/lanres/article/PIIS2213-2600(20)30167-3/fulltext

Maybe if a vaccine is not available for a while, whatever primes the body to fight hay fever and/or tries to fight asthma, could be enlisted to at least help the fight against the virus, with a minimal of side effects? I think the hay fever physical action of mucus production and sneezing, could be helped by frequent physical removal using a Neti-pot for example from the nose and sinuses? Since it seems that it takes from a few days to a few weeks for the virus to get established, then that period could be where we help the body defenses kick it out, like it does with invading pollen allergens?

If hay fever and/or asthma have a protective element against COVID-19, which still needs to be proven, that most likely is the higher level of luminal IgA (secreted IgA or sIgA) . See the levels in this paper https://www.sciencedirect.com/science/article/abs/pii/0022175994002752

I wonder, if a group of people are confirmed to have the virus, and the symptoms are not life threatening, and they do not have allergies or asthma, would there be a way to prime the body to start up those defenses? And also add the twice-daily Neti-pot saline rinse of the sinuses, to help move any particles to move out more quickly? The slowness of the virus to "take root" means there may be a few days to get the defenses started?

What if the virus could be tagged, to add a component that is very allergic to normal people, so that their normal defenses could help to expel it, like they would do pollen grains for example?

I have allergies, but this month a new and never before symptom has been added of a drippy nose component, since the virus has been in the area, that I have interpreted as the virus trying to find a new home?

So, once I feel the tiniest drip, it is saline rinse-time, until that stops. And if the throat feels a little scratchy, which never happens normally when I have my regular allergies, it is gargle-time with a hydrogen peroxide 1.5% solution--not necessarily to kill the virus, but to help dislodge it, also if it is trying to find a new home?

My previous comment was incomplete. Type 2 immunity, induced by hay fever or asthma, may protect against COVID-19 by:

1. Direct anti-viral activity by IgA antibodies.

2. Suppression of Type 1 immunity inflammation, thereby reducing the risk of acute respiratory distress syndrome.

Production of IgA takes some time and stimulation of Type 2 immunity with such target may be best if done prophylactically.

As for downregulation of inflammation in established COVID-19 cases, many doctors seem to use corticosteroids.

To give some insights about the (crazy high) prevalence of hay fever in Japan.

Here https://guidable.co/move_to_japan/how-can-we-prevent-hay-fever-in-japan/ a percentage of 70% is mentioned. Here https://prtimes.jp/main/html/rd/p/000000274.000013377.html , which may be a more reliable source but in Japanese, it appears to be said (my Japanese is not very good) that by self-estimation 50% of Japanese consider themselves as suffering from hay fever.

In the USA, April is National Poetry Month.. so a lot of people around the country, are having fun writing some Coronavirus Haiku, including myself with 73 offerings so far, at http://www.ecoseeds.com/virus-haiku.html --Some serious ones and hopefully a few funny ones, to give everyone some belly-laughs?

Coronavirus: Hay fever symptoms could mimic Covid-19, GPs warn. Doctors warn those allergic to pollen to consider if their reaction has changed from previous years.

https://www.bbc.com/news/amp/health-52349794

Hay Fever or Coronavirus? For Allergy Sufferers, a Pollen Season of Extra Worries Is Starting Up

https://time.com/5812628/hay-fever-allergies-pollen-coronavirus/

I do recommend Johannes' comment.

People with hay fever should not confuse their reaction to pollen with the symptoms of coronavirus, GPs say.

While many symptoms - such as a runny nose - are different, hay fever can also prompt a cough that can alarm both sufferers and those around them.

That has prompted many of those suffering with the allergy to contact family doctors for advice.

The Royal College of GPs said sufferers should consider whether their symptoms are the same as in previous years.

But it has also expressed concern that people may leave the house thinking they have just got the seasonal illness when they have actually contracted the deadly virus.

The main coronavirus symptoms are a fever or a new continuous dry cough, which can sometimes lead to breathing problems at a later stage of the illness.

Dr Jonathan Leach, of the RCGP, said: "For most people who have hay fever it is the same symptoms as they have each year.

"What we are finding is that some patients are saying 'look this is a different thing to what I had last year, could this be coronavirus?' and in that case it might be."

Smoking and COVID-19. A very interesting study in France found significant underrepresentation of smokers in SARS-CoV-2 positive patients, suggesting a protective effect (Miyara et al. 2020 https://www.qeios.com/read/WPP19W.3 ).

This probably needs confirmation, and this overall finding does not mean that smoking can not have an exacerbating effect on the worst cases (Gilmore, 2020 Article Review of: "Low incidence of daily active tobacco smoking in...

Article The Relationship Between Environmental Factors and the Profi...

I'll give some more suggestive evidence for Type 2 immunity being protective against COVID-19. Infections with helminth parasites (worms) are common in many non-rich countries, and helminths induce Type 2 immunity. If you compare frequencies of helminth infections and COVID-19 cases, there is an inverse relationship ( https://figshare.com/articles/World_maps_of_heminthiases_and_COVID-19_pdf/12227963 ). There are many confounding factors, like age build-up of the population, differences in testing for COVID-19, etc. Nevertheless, it is a notable observation.

How to tell the difference between Covid-19 and hay fever

https://www.kilkennypeople.ie/news/kilkenny-news/537680/explained-how-to-tell-the-difference-between-covid-19-and-hay-fever.html

Women may also be better protected against COVID-19 than men because of a higher ratio of Type-2/Type-1 immunity. In countries as diverse as China https://ichef.bbci.co.uk/news/624/cpsprodpb/F197/production/_111074816_death_ratio-nc.png and the USA

https://www.worldometers.info/coronavirus/coronavirus-age-sex-demographics/, the death rate among men is >1.5x that of women. It is well established that pregnant women have an increased Type-2/Type-1 immunity ratio to protect the fetus Article Pregnancy and pregnancy-associated hormones alter immune res...

Article Lung Cancer Incidence in Never Smokers

I like to express that this is an important discussion.

At this momenthttps://www.worldometers.info/coronavirus/ , the official COVID-19 death rates (fatalities per million inhabitants) in Western countries as for example Italy (467 deaths/Million), Spain (531), Holland (286), UK (405) and USA (199) are 50x or more (!!!!!!!!) that of Japan (4). However, as far as I can judge, the government measures in Japan are less strict than in these Western countries, and it doesn't feel strict here at all.

So it would be smart to find out what explains these differences in mortalities. Hay fever is just one of the possible explanations, but one that in my opinion should be taken very seriously. Because, if it is, Japan is in for a rough ride if the virus hits outside the pollen season.

The not so strict situation in Japan is explained here https://mainichi.jp/english/articles/20200430/p2g/00m/0na/048000c

But, as said, the number of casualties is comparably very low.

Speaking of mortality, Japan seems to be putting higher priorities of PCR tests to those who have more severe symptoms. This policy is quite different from the one in Korea. (Korea is very actively carrying out the PCR tests.)

By considering this, the mortality in Japan may be biased to higher than those in the other countries. How do you think about this possibility?

By the way, have you compared the global maps of Covid-19 frequency and pollen concentration in the air?

In people (and mice), lung lymphoid tissue (Bronchus-associated lymphoid tissue, or BALT) is only present if induced (iBALT). Thus, in people where the lungs are immune-stimulated (e.g. asthma, hay fever, smoking), there is organized immune tissue which is absent in pristine healthy lungs. https://en.Article Inducible Bronchus-Associated Lymphoid Tissue: Taming Inflam...

I doubt that nature truly opted for lymphoid-tissue-free lungs, but rather, in some species, chose for a level of plasticity and that under natural conditions there are always some immune stimulants (e.g. helminths) inducing the formation of lymphoid tissue.

So, people with continuous immune stimulation of the lung receive COVID-19 virus with more organized immune tissue than people without such stimulation. This may be extra important because most people are expected to have some immune memory (cross-)recognizing COVID-19 virus from previous infections with common coronaviruses Article Expected immune recognition of COVID-19 virus by memory from...

I did not realize that I could upload images. A few posts ago, I referred to the inverse relationship between prevalences of helminthiases (worm infections that stimulate Type 2 immunity) and COVID-19 cases. Here I include an image of worldmaps with (indications for) both.

Dear Dr. Nakajima,

My apologies for not earlier answering your question "By the way, have you compared the global maps of Covid-19 frequency and pollen concentration in the air?"

A more relevant rephrasing of the question may concern the prevalence of hay fever across the globe. However, even that is quite hard to answer for me as a non-specialist, since it is not very clear to me what different numbers mean as they may not necessarily catch the important differences in severity etc. What I can say is that I am originally from Europe and know Western Europe rather well, and that there I have never observed a hay fever problem in such a large part of the population as here in Japan or even something remotely similar. Personally, in Europe neither I or any of my family experienced hay fever problems, but after living seven years in Japan my immune system was sensitized and since then every year I experience the symptoms. Among Westerners living for a long time in Japan, that is quite common. From landscape point of view, I also doubt that any other country as a whole is more suitable for creating hay fever problems. A lot of mountainous mainland Japan is monotonous monoculture of planted cedars, trees that are growing overly old because they are never cut because wood became cheaper elsewhere. These old trees drop massive amounts of pollen in a limited time, and the problem has worsened since I came in Japan >20 years ago. In my personal case, I speculate that my hay fever is exacerbated by air pollution coming from China (another characteristic of the Japan air space) and, possibly because of that, this year my symptoms are only mild because of the Wuhan lockdown.

As for Northern America, I have no year-round living experience there, but would assume that hay-fever would be much more reflected in tv series etc. if the problem was anything like in Japan.

As for numbers, I found the following (see that the numbers in Europe and the US are lower than in Japan) https://emedicine.medscape.com/article/134825-overview#a6

Epidemiology

Frequency

United States

The prevalence of allergic rhinitis in the United States ranges from 3% to 19%, and 30 to 60 million people are affected each year. The development of allergic rhinitis before 20 years of age occurs in 80% of cases. [14] In 2012, 9% of children younger than 18 years and 7.5% of adults reported allergic rhinitis in the past 12 months. [15]

International

Throughout the world, the prevalence of allergic rhinitis has slightly escalated. [16] Currently, approximately 10 to 30% of adults and 40% of children are affected. [14] The European Community Respiratory Health survey recorded a prevalence of 10 to 41% in adults with allergic rhinitis. [17] Scandinavian studies have demonstrated a cumulative prevalence rate of 15% in men and 14% in women. [18] The prevalence of allergic rhinitis may vary within and among countries. [19, 20, 21, 22] Highest prevalence of severe allergic rhinitis symptoms in children were observed in Africa and Latin America. [23] This may be due to geographic differences in the types and potency of different allergens and the overall aeroallergen burden.

A recent study provides anecdotal evidence for antibody IgA being important for protection from COVID-19. Dahlke et al. "Distinct early IgA profile may determine severity of COVID-19 symptoms: an immunological case series" https://www.medrxiv.org/content/10.1101/2020.04.14.20059733v1.full.pdf

The study investigated a husband and wife who both were infected, and found that the wife who mounted a very fast strong IgA response had only mild symptoms whereas the husband who had a delayed antibody response suffered severe symptoms. Thus, IgA production, which is enhanced in hay fever patients (and asthma patients and smokers) may be critical indeed.

That is a very intriguing question but I confess I could not offer any data to support or refute your contention. One thing, however, may endanger hay fever suffers from the viral infection.

At the peak of the hay fever season, most people would suffer from severe nasal congestion. As a result, they would more often use mouth breathing. Breathing through nose is safer because of the structure inside the nasal tract and the presence of hairs could block intrusion of any particles including the virus. In contrast, mouth breathing would directly send the air down to the lung with no interference. Because of the change of how they breathe, those hay fever patients may inhale more viral particles down to the lung. If the high dose volume of viral particles dictates what symptoms may evolve, as professor Rabinowitz suggested in his article in New York Times on April 1, inhalation through mouth is not a safe choice.

Shih-Wen Huang,MD

Dear Dr. Huang,

Thank you for your contribution. You are definitely right that reaching the lung is a critical determinant for the disease progress, and that inhaling the virus through the mouth probably contributes to that, although it probably is not sure yet how the virus reaches the lung https://www.nature.com/articles/d41586-020-01315-7?utm_source=twt_nnc&utm_medium=social&utm_campaign=naturenews

The immune system is not very well understood (except for a large number of individual mechanisms/responses), the immune system of the lung probably should be considered understudied, and coronaviruses are not very well understood. So, any discussion on differences between resistant and sensitive cases will be largely based on speculation. Nevertheless, we know that:

1. In most COVID-19 cases the immune system can deal with the virus.

2. However, in the fatal cases, it probably is exactly the immune system (cytokine storm) that does the killing.

3. Thus, we have to find out the differences in immune responses between the resistant and sensitive cases.

What I am trying with this thread is to create the awareness that hay fever or other immune conditions with a Type 2 polarization seem to be protective and that this agrees with basic immunology theory. I hope that this will lead to a more intensive investigation of this possible correlation, which at this stage, as you correctly indicate, should only be considered as suggestive (e.g. Article Clinical characteristics of 140 patients infected by SARS-Co...

For interpretation of the data from immunology point of view, differences in behavior between hay fever sufferers and other individuals are irritating confounding factors. You are probably right that hay fever leads to more mouth breathing. On the other hand, I know from myself that I become quite moody if having hay fever, so hay fever patients may communicate less with other people. They will also better protect their mouths with mouth caps and stay more inside, so overall I would guess their behavior has a protective effect. Having said this, if the assumed protection by hay fever would come from iBALT (inducible bronchus-associated lymphoid tissue), that protection may last several months past the actual hay fever as here is summarized for iBALT in generalArticle Inducible Bronchus-Associated Lymphoid Tissue: Taming Inflam...

So I hope that in Japan the COVID-19 patients will be questioned about their hay fever history.

In order for this thread to be complete, also a recent study by Jackson et al. should be mentioned https://www.jacionline.org/article/S0091-6749(20)30551-0/fulltext

Inspired by the data suggesting that asthma/allergies may have a protective effect against COVID-19 (also mentioned above in this thread), Jackson and co-workers investigated the effect of asthma/allergies on the expression of the receptor for the SARS-CoV-2 virus which is ACE2. They found that allergic/asthmatic airway stimulation led to a decrease in ACE2 expression as measured by PCR analysis of nasal and bronchial epithelial brush samples. Furthermore, they found that in vitro the Type 2 immunity cytokine IL-13 had a reducing effect on ACE2 expression in nasal and bronchial epithelial cells. Thus, besides that the immune response induced by asthma and other allergies may have a direct anti-COVID-19 effect (e.g. by an increased IgA response), it may also interfere with viral replication by lowering the expression of its receptor.

Although not scientific, here some other words that express the general surprise that Japan stays quite free from COVID-19 despite not implementing strict measures. https://asiatimes.com/2020/04/why-japan-gets-no-covid-19-respect/

—Al Johnson, an ex-US serviceman and long-term Asia watcher has written: “Japan is the girl that stayed skinny despite eating junk food and not exercising.”—

By the way, considering weight, other than abundant hay fever, the slim bodies of the Japanese may also be an important part of the explanation for COVID-19 resistance.

Could BALT (bronchus-associated lymphoid tissue) also explain the resistance in children? Would the answer to explain the resistance in children really be that simple?

It hereArticle What is the clinical relevance of different lung compartments?

The study refers to an earlier study Article Bronchus-associated lymphoid tissue (BALT) in the lungs of c...

It seems to be all about the lungs (lower respiratory tract). A German study found that infected children and adults have similar amounts of viruses (viral load) in the upper respiratory tract https://zoonosen.charite.de/fileadmin/user_upload/microsites/m_cc05/virologie-ccm/dateien_upload/Weitere_Dateien/analysis-of-SARS-CoV-2-viral-load-by-patient-age.pdf (unfortunately that study does not describe the methods carefully), just as another study found when comparing symptomatic and asymptomatic (adult) COVID-19 cases in an Italian townPreprint Suppression of COVID-19 outbreak in the municipality of Vo, Italy

Thus, what happens in the upper respiratory tract seems to resemble what happens if people are infected with one of the innocent human coronaviruses (OC43, HKU1, 229E, and NL63) and not directly relevant to the infected person. In this article https://www.nature.com/articles/d41586-020-01315-7?utm_source=twt_nnc&utm_medium=social&utm_campaign=naturenews Cyranoski presents a nice model explaining that SARS-CoV-2 is a "double-phenotype" virus behaving as an innocent OC43/HKU1/229E/NL63-like virus in the upper respiratory tract and a dangerous SARS-CoV-1/MERS-like virus in the lower respiratory tract. Resistance and sensitivity to the virus appears to be determined by what happens in the lower respiratory tract, and the presence of BALT may be an important factor for resistance.

By the way, if this model is true, the virus probably will establish mutants without the lung-destroying properties that spread faster than the symptom-creating variants (since we hunt the virus based on the disease symptoms it induces) and so may immunize people against its more dangerous brother.

How about the observed resistance against COVID-19 in babies, many of which have no BALT?

The presence of BALT may explain resistance in children and, as I speculate, in adults with lungs chronically stimulated by immunogenic agents such as pollen (see the above posts). Challenging this model is the fact that small babies are also resistant against COVID-19 but tend to lack BALT (Fig. 3 in https://thorax.bmj.com/content/thoraxjnl/50/6/658.full.pdf ). Maybe this can be explained by small babies having relatively strong immunosuppressive functions (a high concentration of cytokine IL-10) whereby they avoid damage through inflammation https://www.nature.com/articles/s41390-019-0383-y .

Dear Johannes M Dijkstra

Your discussion is very interesting

your idea is based on increased Th2 polarization and the subsequent decreased Th1 polarization protect from COVID-19 severity, However, theoretically, decreased Th1 differentiation increase the susceptibility to/persistent/severity of viral infection, which are supported by delayed viral clearance, persistent virus-induced inflammation and amplification of the allergic inflammation in asthmatic patients (https://erj.ersjournals.com/content/18/6/1013).

Moreover, Th1 might mediates a protective immune response against COVID-19, which is supported by the probability of decreased mortality rate and increased recovery rates accompanied by BCG vaccines (https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31025-4/fulltext). It is well known that BCG induce strong Th1 response, which might mediates a non-specific immune response against COVID-19.

Interestingly, COVID-19 mortality rates in countries that have BCG in their routine vaccine programs, are obviously less than other countries, for example, the total death in Africa and Asia is about 23,000, while in Europe and North america is 237,000 (the rate is 1:10).

The link for BCG distribution (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3062527/), The link for death rate by countries (https://www.worldometers.info/coronavirus/).

Dear Mohamed Hamed Abdelaziz,

Thank you for your contribution to this important discussion.

In this and my next two posts I will reply to your comments. I hope that I summarized your comments correctly.

1. You state that, generally, Type 1 immunity (Th1 polarization) is important for virus clearance, and that, therefore, you are uncomfortable with the model that the Type 2 immunity polarization induced by Asthma may be protective against virus by suppression of Type 1 immunity.

Your chosen reference https://erj.ersjournals.com/content/18/6/1013 may not be the best to make your point, since it (also) says "One recent attempt to provide such a model utilized RV infection in subjects with allergic rhinitis. Individuals received three high-dose allergen challenges in the week prior to innoculation to mimic combined allergen exposure and virus infection. Interestingly, prior allergen challenge in this model, somewhat unexpectedly, appeared to protect against a RV cold with delayed nasal leukocytosis, increased generation of the pro-inflammatory cytokines IL-6 and IL-8 and a delayed, less severe clinical course." However, I do agree with you that, usually, Type 1 immune reactions, involving cell-mediated killing of virus-infected cells, are important for virus clearance. However, the relative importance of cell-mediated killing versus antibody-mediated killing differs per viral disease. I am confident that also against the SARS-CoV-2 virus the Type 1 immunity helps to fight the virus, but, unfortunately, inflammation probably including this type of immunity appears to be risky for a subset of patients. I speculate that Type 2 immune diseases in the lungs, such as asthma and hay fever, are so common because this tissue has a constitutive Type 2 immune polarization in order to protect from damaging Type-1/3 (Th1/Th17) inflammation. Apart from this, mucosal IgA appears to protect against SARS-CoV-1 https://www.jimmunol.org/content/180/2/948 and mucosal IgA concentrations are enhanced by Type 2 immunity in hay fever and asthma patients Article IgA, IgG and IgM quantification in bronchoalveolar lavage fl...

Although I added hay fever in Japan to the discussion, the model of asthma-induced Type 2 immunity protecting from lethal COVID-19 inflammation had been proposed before (also mentioned earlier in this thread) Article Risk factors for severity and mortality in adult COVID-19 in...

"Interestingly, the prevalence of asthma in COVID-19 patients 365 (0.9%) in our study was markedly lower than that reported in the adult population of Wuhan 366 (6.4%). We thus speculate that Th2 immune response in asthmatic patients may counter the 367 inflammation process induced by SARS-CoV-2 infection. Further studies are required to 368 characterize the immune response and inflammation features of COVID-19."

Dear Mohamed Hamed Abdelaziz,

2. As to your second point "Moreover, Th1 might mediates a protective immune response against COVID-19, which is supported by the probability of decreased mortality rate and increased recovery rates accompanied by BCG vaccines (https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31025-4/fulltext). It is well known that BCG induce strong Th1 response, which might mediates a non-specific immune response against COVID-19."

As you know, the article reference that you provide does not provide evidence of BCG vaccines protecting against COVID-19. You are right that BCG vaccines have been shown to have a protective effect against several other viruses, probably through the non-specific enhancement of Type 1 immunity.

Personally, I do not like the experiment in the Netherlands and Australia to immunize health care workers with BCG vaccine. Most of them should be rather well protected by their immune system anyway because of their age, and I feel that giving them sufficient rest and protection wear should be enough (and should be done!!!!!). My expectation is that prophylactic administration of the BCG vaccine in these health care workers will, if they later get infected with SARS-CoV-2, on average lead to a faster clearance of the virus; I do not debate that Type 1 immunity is one of the ways to beat a virus. However, I am afraid that if the SARS-CoV-2 virus breaks through the first line of defense, the Type 1 immune polarization in these health care workers may prohibit them from properly dealing with the inflammation, and the overall death rate may be higher than without the BCG vaccination.

Some other point of consideration:

One of the most important comorbidity factors for COVID-19 is obesity. That may have a variety of reasons. However, one of the possible reasons is that obesity induces Type 1 immunity Article The Macrophage Switch in Obesity Development

Dear Mohamed Hamed Abdelaziz,

3. As to your third point "Interestingly, COVID-19 mortality rates in countries that have BCG in their routine vaccine programs, are obviously less than other countries, for example, the total death in Africa and Asia is about 23,000, while in Europe and North america is 237,000 (the rate is 1:10)."

Many of these countries are poor, and I speculate (see earlier posts) that they are protected by a high incidence of helminthiasis by which lungs will have associated BALT lymphoid tissue with Type 2 immune polarization. The major weakness of the BCG vaccine theory is that non-specific immunity is not expected to last for decades. It says about non-specific "trained immunity" (Netea et al., 2020, in Nature Reviews Immunology) https://www.nature.com/articles/s41577-020-0285-6 : "The immunological phenotype of trained immunity has been proven to last at least 3 months and up to 1 year, although heterologous protection against infections induced by live vaccines can last for up to 5 years." Like discussed in this thread about BALT tissue, the creation of "tertiary" lymphoid tissue that can last for months is commonly known. However, a non-specific immune memory that can last for years is, to my knowledge, not known. The statement ".. up to 5 years" by Netea et al. appears to be unfounded, as far as I can see, as it refers to a study in Uganda Article Child survival and BCG vaccination: a community based prospe...

Thanks Johannes M Dijkstra for this continuing discussion

- Your suggestion as the protection against COVID-19 is mediated by IgA induced by IL-5 in allergic rhinitis patients. But, in this case, the non-specific IgA doesn’t contain the enough affinity that is required for viral neuralization, and as you know the high affinity antibodies are needed for efficient neutralization.

- The second suggestion of decreased Th1 polarization will weaken the inflammatory response accompanied by COVID-19. In this case the patient will be vulnerable for various types of infection (bacterial and viral) that require an efficient Th1 response.

Dear Mohamed Hamed Abdelaziz,

Type 2 immunity promotes IgA production, including against SARS-CoV-2 viurs after infection.

I agree that Type 2 immune polarization, in many cases, reduces resistance against viral infection, as is known in pregnant women. However, in the lungs, where IgA is an important part of the immune response, the 1/2 discussion may be less simple.

So far, I have avoided a detailed discussion of the inflammation caused by SARS-CoV-2 in the lung because there are not enough data yet. However, it seems that SARS-CoV-2 prohibits the mounting of a typical Type 1 response by actively interfering with interferon production. If this is true, the IgA production may be even more important. I’ll try to add a more detailed discussion on this when more data are available.

Do you mean the protective IgA is not allergen specific and is COVID-19 specific?

-In case of IgA is allergen specific; it will be low affinity

-In case of IgA is COVID-19 specific; also, it will be low affinity, due to high viral mutation rate.

Secondly, what is the possible pathway for a virus to activate Th2 response?

Normally, the virus is recognized by pattern recognition receptor in cytoplasm of innate immune cells, then it is processed and presented with MHCII to Th0 inducing Th1, or presented by any cell with MHCI to CD8+ cell

Mohamed Hamed Abdelaziz

The polarization of cells like macrophages or T helper cells, or the type of antibody class switch, depends (also) on the local cytokine concentrations. Thus, in an immune milieu that enhances IgA production, also the production of specific IgA antibodies against SARS-CoV-2 will be enhanced.

So far, I did not discuss the possibility of Th2 cells directed against SARS-CoV-2, but indeed this is possible. For example, whereas helminths usually generate Th2 cells, in an immune milieu polarized towards Type 1 immunity they generate Th1(-like) cells https://www.frontiersin.org/articles/10.3389/fcimb.2017.00341/full . Thus, if in the presence of abundant agents that induce a Type 2 immune response, such as pollen during hay fever, Th2 cells may be generated against SARS-CoV-2 antigens.

IgA is very well known to be important for controlling viral infection. For example, the presence versus absence of IgA induction is thought to explain the success of the OPV versus IPV polio vaccines ( https://science.sciencemag.org/content/368/6489/362/tab-pdf ).

The expression of IgA may be even more important for controlling large viruses like SARS-CoV-2 that appear to have tricks up their sleeves to inhibit interferon production and do not induce a classical Type 1 response (e.g. https://www.medrxiv.org/content/10.1101/2020.04.19.20068015v1.full.pdf ).

SARS-CoV-2 viruses mutate remarkably slow for an RNA virus, presumable because of their possession of an exonuclease.

The emerging model in this thread is that patients predisposed to a rapid IgA response by having BALT tissue and Type 2 immune polarization (e.g. hay fever patients, asthmatics, smokers, people suffering parasite infections) are better protected from COVID-19 pneumonia than others.

In theory, mucosa immunity of which IgA is a major player protects us from viral infection through mucosal surface like GI tract or respiratory tract. It should be reminded, however, one in 80 to 200 normal individuals have selective IgA deficiency.

Shih-Wen Huang,MD

Dear Dr. Huang,

Thank you for raising attention to this important issue. In general, IgA deficiency leads to an increased susceptibility to pneumonia Article Long-term follow-up of health in blood donors with primary s...

Article Two X‐linked agammaglobulinemia patients develop pneumonia a...

Article A possible role for B cells in COVID-19?: Lesson from patien...

Great! You are good!

Shih-Wen Huang

Johannes M Dijkstra There is no doubt that specific IgA mediate antiviral immunity against stable viruses (consider the mutation rate of COVID-19, which still not clear), and further studies are required to elucidate the pathway of COVID-19 IgA generation.

A recent big population study in the UK by The OpenSAFELY Collaborative (Williamson et al.) should be mentioned here https://www.medrxiv.org/content/10.1101/2020.05.06.20092999v1.full.pdf

In contrast to previous studies (already discussed in this thread), they found asthma patients to be overrepresented among COVID-19 deaths, concluding it as a risk factor.

However, as in other studies (also already discussed in this thread), they found that current smokers were underrepresented. On the other hand, they found ex-smokers to be overrepresented. Their study suggests that smoking induces permanent health problems (lung tissue damage?) that increase the sensitivity to COVID-19, but simultaneously has a protective effect from which current smokers benefit. As discussed earlier in this thread, I suspect that the protection comes from their immune status which is known to include enhanced IgA production, but others speculate that it derives from nicotine action.

Oxford University is making an important comparison between governments in regard to stringencies of anti-COVID-19 measures, providing a convenient on-line tracker system ( https://www.bsg.ox.ac.uk/research/publications/variation-government-responses-covid-19 and https://www.bsg.ox.ac.uk/sites/default/files/2020-05/BSG-WP-2020-032-v5.0_0.pdf ). They agree that the measures in Japan are not very strict compared to other countries. I tried to summarize their data in this figure, in which I added blocks with the stringency-color of Japan to other parts of the world for easier comparison. I also added the stringency numbers of Japan before, showing that although not very stringent the current stringency is the highest for Japan yet.

How about the possibility that lung physiological conditions cause the difference between mild and severe COVID-19 cases? The lung surfactant concentration in old people may be different.

As an immunologist, I would like to explain the differences between resistant and sensitive patients by differences in immune systems. However, immunology-based theories have trouble explaining (discussed earlier): (a) Why are both babies and young adults resistant, despite big differences in their immune systems? (b) Why aren’t the viral titers in the upper respiratory tract related to the severity of the disease, since an immunological explanation would expect at least some level of correlation between what happens in the upper and lower respiratory tracts.

Then how about a theory that COVID-19 basically is an upper respiratory tract disease which can only infect the lungs under atypical physiological conditions? And what would those conditions, which should be more prevalent in the elderly, be?

As for the virus, notable physiological features are that: (1) It has at least two virion proteins, M and E, that effect membrane curvature (e.g., https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4486061/ ); (2) It doesn’t easily leave the respiratory tract in droplets or aerosols https://www.nature.com/articles/s41591-020-0843-2.pdf , suggesting it binds well to cells or mucus layer without fusing/opening/disintegrating. So, besides finding cells with the appropriate receptors and co-receptors, the virus may need special physiological conditions for optimal fusion with cells, which would be supported by the virus not readily infecting other cells in the body outside the respiratory tract.

In the lungs, BALF surfactants affect membrane and protein properties. For example, the major surfactant phosphatidylethanolamine (PE) affects membrane curvature, and it is said (Woods et al. 2016) “PE modulates membrane curvature and may play an important role in both membrane fusion and alveolar surfactant architecture.” https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5206402/ At least in horses, the concentration of BALF phospholipids decreases with age (Christmann et al. 2009 https://onlinelibrary.wiley.com/doi/full/10.1111/j.1939-1676.2009.0298.x ), and Christmann et al. speculate that the reduced requirement for surfactant in old animals is caused by a reduction in pulmonary elastic recoil (increased compliance).

Another interesting physiological parameter for this discussion thread would be the pH, because airway acidification is observed in patients with asthma and allergic rhinitis https://onlinelibrary.wiley.com/doi/full/10.1111/j.1399-3038.2006.00426.x and in smokers https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2528202/ . However, airway pH seems not to be different in old people https://link.springer.com/article/10.1007/s00408-005-2580-1 , so pH variation may not be an important factor for determining COVID-19 resistance.

The host response is always important.

However, the amount of virus exposure also counts. Prof. Rabinowitz wrote in NY Times on April 1 that the exposure of virus could be high or low. High dose exposure like you sit next to a virus carrier in a jammed subway car is quite different from siting in a spacious office with your co-workers feet away. The high dose exposure almost led to severe lung infection that could mean a death or long recovery compared to low exposure.

Shih-Wen Huang, MD

Dear Dr. Huang,

Thank you for mentioning this article https://www.nytimes.com/2020/04/01/opinion/coronavirus-viral-dose.html . It explains why heavily exposed (and exhausted) health care workers are at a much higher risk since a larger number of viruses (stringent challenge) gives the individual less time to mount an immune response. In COVID-19 that may mean that if the virus needs amplification in the upper respiratory tract before reaching the lung in sufficient numbers, a normal young person has time enough to generate an immune response [maybe (including) IgA] for blocking productive infection of the lung. However, if the lung is reached directly by large numbers of virus, the outcome may be worse.

Nevertheless, the immunological puzzle remains as to why the virus titer in the upper respiratory tract seems to be completely independent of disease severity (see my previous post).

It could be due to inaccuracy of measuring, but apart from the earlier mentioned studies also He et al. https://www.nature.com/articles/s41591-020-0869-5.pdf who measured virus titers in throat swabs state “There was no obvious difference in viral loads across sex, age groups and disease severity.” So let’s assume those statements are correct.

An important conclusion by He et al. is that the maximum virus titer in the upper respiratory tract is reached 1 day before symptom onset, which makes it easier to accept that symptomatic and asymptomatic patients can have similar viral loads. On average, observable lung pathologies start later than 2 days after symptom onset https://pubs.rsna.org/doi/10.1148/radiol.2020200463 , and the puzzle is why that is not related to the viral load in the upper respiratory tract.

He et al. estimate the incubation period, the time between infection and first symptoms, to be around 5 days. That would give the immune system a week or so to block the first pathologies in the lung, which for the adaptive immune system is suboptimal for recognizing new antigens. So, any prior sensitization to an enhanced immune response, possibly the propensity of hay fever patients to mount an IgA response, or maybe the actual recognition of SARS-CoV-2 epitopes by MHC-I restricted memory T cells derived from previous infections with other coronaviruses https://f1000research.com/articles/9-285 , could be important.

Still, the puzzle remains (anybody please comment), and it seems that either:

1. There is little correlation between the immune responses necessary in the upper and lower respiratory tracts

2. The damage done by the virus in the lower respiratory tracts is not primarily related to immune system differences between individuals, but to physiological differences in their lungs

We have still a lot to learn about this virus and the consequence

after the infection. Your speculation may turn out to be correct.

Shih-Wen Huang, MD

The situation in experimental COVID-19 rhesus macaques seems to be similar.(see Fig. 1 in Article Age-related rhesus macaque models of COVID-19

Does this mean my kids (5 yr and 3 yr olds) who have allergic Rhinitis@ are at least safe from Covid-19?

Dear Dr. Kowsar,

I am not a medical doctor, so I shouldn't give good advice on this. However, I can give my opinion. The possible risk for severe COVID-19 is all about percentages (chances), and nobody can be sure to be really safe from COVID-19 until actually getting infected. However, it is so rare that small children are seriously affected by COVID-19 that one probably could say that children are "basically safe". That is the reason why in many countries after lockdown the schools are opening again (and many of those children have allergies). There are no indications that allergic rhinitis increases the risk of COVID-19, and there are some indications/reasons that suggest that allergic rhinitis may decrease the risk of COVID-19 (the topic of this discussion thread).

Dear Johannes,

Yes you are absolutely right nobody can be really sure to be safe from Covid-19 but if there will be research which results allergic rhinitis may decrease the risk of this virus then it will be a good news for mothers like me

best of all stay home stay safe

We just learned several children in NY and other states reported several case of developing significant multi-system inflammatory conditions that mimic Kawasaki disease in children. Some even had cardiac complications.

Like I said, we are still learning this mysterious covid19 virus.At this point, we should keep our mind open to deal with any surprises related to this viral illness.

Shih-Wen Huang, MD

Dear Dr. Huang,

Thank you for your contribution. In my understanding, COVID-19 may not be exceptional among viral diseases to induce some cases of Kawasaki disease, but you are right that we should be careful.

Despite difficulties of explaining differences in COVID-19 severity by differences in immune systems (see previous posts), there are studies that indicate that if the virus gets a foothold in the lungs, differences in immune responses do determine the outcome. I’ll summarize two of those studies below.

1. A study in humans by Liao et al. https://www.nature.com/articles/s41591-020-0901-9.pdf

This study compared the immune cells within the bronchoalveolar lavage fluid (BALF) in healthy donors and moderate and sever COVID-19 cases; the moderate cases also suffered from pneumonia. Some pronounced differences between the three moderate cases versus the six severe cases were:

- Strong expansion of ZNF683+ CD8 T cell clones (their Figs. 1b, 2c, and Extended Fig. 4c)

- Lack of neutrophil expansion (their Figs. 1b and Extended Fig. 1d)

- Lack of cytokine storm (their Extended Fig. 5a)

The ZNF683 (aka HOBIT) marker seems to be associated with T cell persistence https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6825592/#R31 and tissue-resident T cell memory cells https://science.sciencemag.org/content/352/6284/459.long . Could it really be that MHC-I restricted immune memory from previous common coronavirus infections helps protect against COVID-19 https://f1000research.com/articles/9-285 ?

2. A study in rhesus macaques (and mice) by van Dooremalen et al. https://www.biorxiv.org/content/10.1101/2020.05.13.093195v1 i

In this study, adenovirus-vectored vaccine ChAdOx1 nCoV-19, encoding the spike protein of SARS-CoV-2, was injected intramuscularly into six monkeys. Control animals were mock vaccinated. Four weeks later, the animals were challenged by 4 mL intratracheally, 1 mL intranasally, 1 mL orally and 0.5 mL ocularly of 4 x 105 TCID50/mL virus suspension. Notable observations comparing the vaccinated animals with the controls were:

1. Viral gRNA was detected in nose swabs from all animals and no difference in viral load in nose swabs was found on any days between vaccinated and control animals (their Fig. 3c).

2. However, only in the mock-vaccinated monkeys virus replication in the lungs was observed, as also some lung pathology at day 7 (their Figs. 3 and 4).

3. Although the experiment was stopped at day 7, it seems that also in the controls the virus infection was self-limiting (their Figs. 1a and 1b).

Thus, also in this monkey model, there is no obvious correlation between what happens in the upper and lower respiratory tracts. Importantly, the immune system affected what happened in the lung, as only in the vaccinated animals no virus replication in the lung was observed. Furthermore, despite that considerable virus numbers were directly injected into the trachea, and virus replication was established in the lung, even in the non-vaccinated healthy adults the virus infection was self-limiting.

An important paper just appeared in Cell

https://www.cell.com/cell/fulltext/S0092-8674(20)30610-3 . Grifoni and co-workers collected convalescent blood samples at 20-35 days post-symptoms onset from nonhospitalized adult COVID-19 patients who were no longer symptomatic. These samples revealed efficient adaptive immune responses including cytotoxic T cells, Th1 helper T cells, and IgM, IgG, and IgA antibody production. Importantly, no other Th polarizations were clearly observed. The very good news, of course, from the Grifoni et al. study, is that people after recovering from COVID-19 do not only have antibody memory (already shown by other groups) but also memory for cell-mediated cytotoxicity.

Grifoni and co-workers also tested a few-year old blood samples of healthy donors that had, as is common among most people, antibodies against common human coronaviruses. Also in these blood samples they found some reactivity by CD4 T and CD8 T cells upon incubation with SARS-CoV-2 peptides, which they interpret as crossreactivity by immune memory generated by infections by common coronaviruses (OC43, HKU1, 229E, or NL63). This beautifully agrees with expectations that we expressed in a paper https://f1000research.com/articles/9-285#ref-49 (Dijkstra and Hashimoto, Expected immune recognition of COVID-19 virus by memory from earlier infections with common coronaviruses in a large part of the world population, F1000Research preprint; mentioned in this thread before), except that for the S protein we do not expect crossreactive T cell memory because of absence of shared linear epitopes for MHC binding (the protein sequence is not so well conserved). However, the S protein is highlighted by Grifoni et al. as one of the proteins associated with this pre-existing T cell memory. I have now added a question to the Cell paper asking the authors to elaborate on this matter.

The Grifoni et al. paper is celebrated in Science https://www.sciencemag.org/news/2020/05/t-cells-found-covid-19-patients-bode-well-long-term-immunity# by Leslie, together with a paper by Braun et al. https://www.medrxiv.org/content/10.1101/2020.04.17.20061440v1 that also found that CD4 T cells of healthy patients never experiencing COVID-19 were reactive upon incubation SARS-CoV-2 S protein peptides. However, in their supplement Braun et al. show absence of high similarity between the sequences of the predicted MHC-II epitopes of the S protein of SARS-CoV-2 versus the common coronaviruses https://www.medrxiv.org/content/medrxiv/suppl/2020/04/22/2020.04.17.20061440.DC1/2020.04.17.20061440-1.pdf . I don’t understand why Science doesn’t raise some questions to this matter.

Personally, I trust most of the data presented in these studies, but feel that the observation of CD4 T cell activity generated by adding SARS-CoV-2 S protein peptides to peripheral blood mononuclear cells (PBMC) of "naïve" blood donors needs more discussion.

Anyway, apparently, a different history of infections of common coronaviruses should be added to the list of possible causes creating differences between countries in the spread of COVID-19. I just doubt that those (largely ignored because quite harmless) viruses are systemically and continuously monitored in different countries, and guess that it will remain speculation.

In large part, the reports so far shed a good news because most of the infected individuals displayed activation of B cell system by making antibodies with proper titers (igM, IgG and IgA) and most importantly, activation of cytotoxic T cells and the memory T cells. How long those helpful immune activities could be sustained among us could predict whether we would have another pandemic in the future.

Shih-Wen Huang, MD

To get back to the possibility of surfactant concentrations being a non-immunology parameter that may determine whether the lungs can be infected. Maybe bats are a likely source of coronaviruses that can infect other species because in the mucosal layers of bats the viruses need to withstand a wide range of daily variation (see a reference with its abstract below). It may make them more robust for jumping to other species.

Article Periodic Fluctuations in the Pulmonary Surfactant System in ...

Codd JR1, Slocombe NC, Daniels CB, Wood PG, Orgeig S.

Abstract

Pulmonary surfactant is a mixture of phospholipids, neutral lipids, and proteins that controls the surface tension of the fluid lining the lung. Surfactant amounts and composition are influenced by such physiological parameters as metabolic rate, activity, body temperature, and ventilation. Microchiropteran bats experience fluctuations in these parameters throughout their natural daily cycle of activity and torpor. The activity cycle of the microchiropteran bat Chalinolobus gouldii was studied over a 24-h period. Bats were maintained in a room at constant ambient temperature (24 degrees C) on an 8L : 16D cycle. Diurnal changes in the amount and composition of surfactant were measured at 4-h intervals throughout a 24-h period. The C. gouldii were most active at 2 a.m. and were torpid at 2 p.m. Alveolar surfactant increased 1.5-fold immediately after arousal. The proportion of disaturated phospholipid remained constant, while surfactant cholesterol levels increased 1.5-fold during torpor. Alveolar cholesterol in C. gouldii was six times lower than in other mammals. Cholesterol appears to function in maintaining surfactant fluidity during torpor in this species of bat.

Could my speculation that Japan was protected from a more severe COVID-19 outbreak by the abundance of hay fever really be right (see the original question of this thread)? Martijn Hoogeveen in the Netherlands did not just speculate but investigated the correlation between influenza symptoms versus pollen density and hay fever symptoms, and found a clear inverse relation Article Pollen likely seasonal factor in inhibiting flu-like epidemi...

Likeable but unlikely.

Have a look at this map: https://news.google.com/covid19/map?hl=de&gl=DE&ceid=DE:de

It is clear that outside of Asia we find the major part of the COVID cases. Considering the fact that COVID = SARS, and that it may have been circulating in the Asian population since 2003, we might expect immunity in Asia. The rest of the world will be more affected because SARS is new there.

Article A SARS-like Coronavirus was Expected, but nothing was done t...

Dear Dr. Borger,

SARS is a rather deadly disease and therefore has not been circulating widely in the Asian population. As for the map, in this thread we are discussing that Type 2 immunity in the lungs induced by hay fever (Japan) or helminths (poor countries) might explain a lot of the global COVID-19 distribution. More in general, apart from Type 2 immunity polarization, immune stimulations of the lungs by pathogens, allergens, or irritants (smoking), can induce lung-associated immune tissue (iBALT), so that exposed individuals may have a faster and better reaction against a viral infection. Please see earlier posts in this thread for that.

This https://www.sciencemag.org/news/2020/05/why-do-some-covid-19-patients-infect-many-others-whereas-most-don-t-spread-virus-all# is an interesting article that explains that COVID-19 expands in clusters of infection. It suggests that Japan was protected by prohibiting these clusters: "Japan, which was hit early but has kept the epidemic under control, has built its COVID-19 strategy explicitly around avoiding clusters, advising citizens to avoid closed spaces and crowded conditions." Additionally important, interpreting this article, is that in Japan the people, compared to other nations, are not very vocal.

Just submitted a followup paper to this one Article Pollen likely seasonal factor in inhibiting flu-like epidemi...

In the first study we just studied correlations, but also checked for incubation time (the temporality criteria of causality). And hypothesized 2 different explanations: 1) immuno activation/allergic responses 2) pollen aerosol is antiviral (viruses have documented anti-viral properties).

In the followup study, with Erasmus MC and Jeroen Bosch MC, we also checked for Covid-19, and meteorological variables as co-inhibitors. The conclusion is that still pollen are the strongest predictor of flu-like seasonality, including covid-19, and that solar radiation is a (weaker) co-inhibitor. We can predict exactly the start and end of flu season, based on the passing of now exactly defined pollen thresholds (allergenic and non-allergenic ones), and the passing of a solar radiation (UV) threshold strengthens the effect. It's not just the allergenic ones that effect hay fever, also the so-called non-allergenic ones. Very likely there are no non-allergenic pollen, but only very low allergenic ones.

We added a third explanation: circadian/seasonal rhythms in the immune system (which is affected by melatonin levels), whereby also pollen can have a pre-programmed trigger function.

All to ideas to be further tested in labs, and in other countries. Next, I want to control for pollution, behavioral factors, and climate change shifts.

Dear Dr. Hoogeveen,

Thank you for your very important contribution. Explaining seasonality of influenza, and maybe COVID-19, would be huge from scientific point of view (for the readers here, see Article Sick time

Are also data available for Holland about the seasonality of the common coronaviruses (OC43, HKU1, 229E, and NL63) which are among the possible agents of the common cold?

Just now they start to differentiate better, although probably not on the level you would like to see it. In general, all flu-like viruses are effected in the same way as the influenza virus, by pollen-flu-seasonality (acc to stats). But, too little data to split it up.

https://www.rivm.nl/griep-griepprik/feiten-en-cijfers

Dear Dr. Hoogeveen,

Thank you.

In this thread we have been discussing that the SARS-CoV-2 virus is like a double-phenotype virus, with, surprisingly, what happens in the lungs and the severity of symptoms hardly relating to what happens in the upper respiratory tracts.

If I am correct, when looking at COVID-19, you mainly focus on the spread of the virus, thus monitoring the infection of the upper respiratory tract. It would be interesting to compare SARS-CoV-2 viral loads in the upper respiratory tract outside and inside the pollen season. Is there any group in Holland performing such analyses? It might provide definite evidence that there is a reason other than social distancing that currently limits the virus spread.

It would be good if at least one country would select a group of representative individuals of the entire population and follow them systematically over time in regard to health, behavior (roughly), and COVID-19-relevant features (at least PCR, and if positive also viral load and immune features). That would greatly help to make conclusions. Maybe 1000 people, and one year long, would be enough for that.

Probably good. Johannes, are you aware of a great pollen (open) data set for Japan?

Dear Martijn,

Funny, just at this moment I am listening to your voice on the podcast of the Telegraaf newspaper (in Dutch, starting from the last ten minutes) https://podcastluisteren.nl/ep/Corona-update-Tweede-golf-wordt-heftiger-corona-zal-hele-winter-huishouden .

To your question, there are many published reports on pollen related topics in Japan (see PubMed). There are also on-line information sites with continuous pollen updates. I took a look already but didn't figure out yet how to translate the unit that you use, pollen per cubic meter, to the per square cm unit used in most publications on Japanese pollen/hay fever (I am not a pollen specialist).

Please specify exactly what kind of database you would like to access, so that I and maybe some Japanese colleagues can try to find such. I am pretty sure that what you are looking for exists, since they have so many datasets on pollen here. Hay fever is really a very serious problem in Japan.

Dear Martijn,

The short answer to your question “…are you aware of a great pollen (open) data set for Japan?”, probably should be http://kafun.taiki.go.jp/ (although in Japanese)

The longer answer:

I now found data in pollen per cubic meter for Japan. A monitoring on-line site http://kafun.taiki.go.jp/ (in Japanese; e-mail [email protected]) is maintained by the Ministry of the Environment http://www.env.go.jp/en/index.html (e-mail https//www.env.go.jp/en/moemail/ ), and appears to be directly used in scientific articles (e.g. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0131710 and see below). If you want access to rawer data, or have information about how they were gathered etc., it may be best to (i) contact those e-mail addresses and hope that you will be connected to a capable/friendly person, or (ii) contact an author of one of the articles on pollen in Japan that you deem of good scientific quality. If it doesn’t work, let me know and I may try something else. Alternatively, if someone reads this and can help, please let Martijn know.

I added a figure, showing that in the Kanto region where Tokyo is situated, in many years even the average of pollen around 30 micrometer in diameter (like from pine trees) in the period February-1 to May-31 is higher than the 70 pollen per cubic meter that you indicated as threshold for association with reduction of viral disease. The figures are from a paper by Ishibashi and Sakai, 2019, https://www.nature.com/articles/s41598-019-47870-6 (I hope that I understand the paper correctly). I guess you could try to contact them, as for them it should be easy to organize the data per time, period and place in whatever way you want. Hokkaido is a region which stands out a bit from most of Japan by having been hit a bit harder by COVID-19, and it is also a region of which the rest of Japan often says that there is no hay fever (although that is not perfectly true). I feel it is not responsible of me from statistical point of view to show a figure like this, but it may raise interest in the question.

Comparison of influenza and COVID-19 distribution in Japan may also be interesting. This newspaper article explains that this year the influenza cases were exceptionally low https://www.japantimes.co.jp/news/2020/02/21/national/influenza-wave-drastically-wanes-japan-amid-spread-coronavirus/#.Xsh1g8BS-Uk . Influenza seems to spread relatively independent from cluster spreading https://www.sciencemag.org/news/2020/05/why-do-some-covid-19-patients-infect-many-others-whereas-most-don-t-spread-virus-all , which means that either:

1. The Japanese people did more social distancing than required by their government, since the official policy was centered on the avoidance of cluster spreading (see a post above). That would also mean that the Japanese situation can't be used as an example that avoiding cluster spreading is enough.

2. This year the winter felt unusually short. It might have been a weather thing (pollen? temperature?). Hokkaido is the coldest part of Japan, which may explain their extra trouble with COVID-19. See this article https://english.kyodonews.net/news/2020/05/3d7b544c27f9-urgent-tokyo-area-hokkaido-to-remain-under-virus-emergency-beyond-thurs.html with a small movie within about the number of cases in Japanese prefectures; from that movie I took the March 16 image shown here, revealing that at the time Hokkaido had most cases.

Dear Drs. Johannes and Hoogeveen,

I quickly checked the units of flower pollen data stored in

http://kafun.taiki.go.jp/

The unit of the data listed on the page below is pollens/m3

http://kafun.taiki.go.jp/library.html?20190712

Here is a translated spreadsheet of one of the data set from the prefectures belonging to the great Tokyo area (Tokyo, Chiba, Tochigi, Saitama, Ibaraki, Kanagawa, Gumma) in 2017.

https://docs.google.com/spreadsheets/d/1diUjTFP_EII-1qpm80iZ19DljKTPnCqgw1ClPxIFdwQ/edit?usp=sharing

If you can specify which dataset you want, I can download them when I am free.

Regards,

Ryuichi

Dear Dr. Nakajima,

Thank you for your important contribution. The spreadsheet information you added shows a lot of variation in pollen concentrations even between hours (dependent on when the trees drop their pollen and how far wind and moisture allow them to be carried). Given that the concentrations outdoors and indoors are also quite different, it may need consideration as to where the cut-off value of 70 pollen per cubic meter concluded in Dr. Hoogeveen's paper stands for. I feel that an immunological effect can easier explain an anti-virus effect of pollen than a direct anti-virus effect because of the heterogeneity of the pollen concentrations.

Dr. Dijkstra,

Pollen counts correlates with the number of patients who have hay fever. However, your interest is if hay fever may block Covid19 infection or its symptoms. Therefore, your focus should be on hay fever patients, not strictly on the pollen counts.

Shih-Wen Huang, MD

Dear Dr. Huang,

You are right that allergy seems to be the more likely mechanism by which pollen can affect COVID-19. I started this thread with that idea, since in Japan hay fever is so prevalent, and still believe in its likelihood. However, the advantage of a discussion thread is that we can also explore other, even opposite ideas. Dr. Hoogeveen did not only suggest allergy as a possible anti-virus mechanism by pollen, but also a possible direct anti-virus effect by pollen. I just like to check that idea for its merits. For the moment, I feel the allergy model is far more appealing, but I like to hear more arguments.

Vaccination may help reduce virus spreading. Good news is that newer studies in rhesus macaques found that immune memory induced by vaccination (Yu et al.Article DNA vaccine protection against SARS-CoV-2 in rhesus macaques

Article SARS-CoV-2 infection protects against rechallenge in rhesus macaques

Now foreign news media are interested in why Japan is in a good shape despite the incomplete government policy:

https://foreignpolicy.com/2020/05/14/japan-coronavirus-pandemic-lockdown-testing/

Japanese people need to know the real reason why they survived much better than the people in other countries. If it is really due to just the recent hay fever, Japan may experience quite a strong second impact in Autumn, because we do not have so much hay fever during summer. To be honest, it would be a good news for Japanese people if hay fever has nothing to do with Covid19 infection.

In terms of that, the question you raised is important. I hope we can figure this out.

Dear Dr. Nakajima,

Thank you for that very fine article! The author lives in Japan, and, like all of us who live here, is astonished why the problems are not much bigger. The rest of the world, however, hardly seems to be interested in the Japanese situation. Well, 50x lower death rates is something to consider I would say. Here is some information proving that the difference is real and that Japan is not undercounting its COVID-19 deaths: ( Sambridge and Jackson https://media.nature.com/original/magazine-assets/d41586-020-01565-5/d41586-020-01565-5.pdf in combination with http://jupiter.ethz.ch/~ajackson/benford.pdf ).

Are the Japanese relatively resistant against respiratory viruses? Let's compare with the 2009 H1N1 influenza epidemic. The figure, shown here, from Simonsen et al. Article Global Mortality Estimates for the 2009 Influenza Pandemic f...

So, there seems nothing special in general about Japan and protection from respiratory viruses. It also irritates me that Western media usually portray Japan as overcrowded by focusing on the subway in Tokyo, and now suddenly assume that the Japanese are protected from COVID-19 because they are so good in keeping distance. For the Japanese it is important to find out why they are better protected against COVID-19, since data suggest that if it behaves as influenza, then in October there will not be extra protection. Whether the hay-fever theory is right or wrong, it is time that Japan starts generating some COVID-19 patient data by which that can be estimated.

Note added on June 2: Integrity concerns have been raised about the Mehra et al. study https://www.sciencemag.org/news/2020/06/mysterious-company-s-coronavirus-papers-top-medical-journals-may-be-unraveling . Therefore, the data shown in this post may not be correct. My apologies for that.

An international study in The Lancet shows that only ~10% of hospitalized COVID-19 patients are current smokers, but that among those patients they are overrepresented in the non-survivor group (Mehra et al., https://www.thelancet.com/action/showPdf?pii=S0140-6736%2820%2931180-6 ). They did not compare percentages with the general population (which would need equilibration for age and sex), but 10% seems to be an underrepresentation; that would agree with smoking (also) having a protective effect as hypothesized by others and discussed earlier in this thread.

Note added on June 2: Integrity concerns have been raised about the Mehra et al. study https://www.sciencemag.org/news/2020/06/mysterious-company-s-coronavirus-papers-top-medical-journals-may-be-unraveling . Therefore, the data shown in this post may not be correct. My apologies for that.

The same study showed that Asians among hospitalized COVID-19 patients in North America have the smallest chance of dying (Mehra et al., https://www.thelancet.com/action/showPdf?pii=S0140-6736%2820%2931180-6 ). Apart from that Asians are a very diverse group, I don't know how to compare these data with ethnic (combined with age) distributions among the non-hospitalized population. But the figure might suggest that Asians have a physiological advantage to fence off COVID-19; at the country level, that might explain differences in virus spreading (R) rates.

It becomes more and more common to sequence the virus per patient https://www.nature.com/articles/d41586-020-01573-5

Besides that such can help to trace how the virus spreads, it could also allow the detection of virus mutants that are less pathogenic. That would be helpful in the discussion of vaccination/herd-immunity.

I have a background in animal virology. In my opinion, many of the current COVID-19 vaccines that are currently developed can be expected to be either too weak or too dangerous (the general dilemma in vaccination). If everyone would be at risk from COVID-19, we could accept some degree of danger from the vaccine. But I doubt that young people are willing to vaccinate themselves and their children with a vaccine that has not been properly tested only to induce herd-immunity that protect the elderly and a small subset of young people. Vaccination of the elderly themselves, with their weakened immune responses, may not induce the desired protection. In my opinion, natural mutation of the virus to a weaker form will be faster than the development and implementation of world-wide or even country-wide vaccination.

Other than strategies for reducing the viral challenge (hygiene, social distancing, air space control), the development of medication should be more promising than the development of vaccines.

A recent article by Hou et al., in Cell, showed by in vitro analysis and autopsies that SARS-CoV-2 most readily infects cells of the nose and cannot as easily infect cells of the lungs; they showed that this difference coincides with a gradient of the ACE2 receptor abundance [see picture (1)] https://www.sciencedirect.com/science/article/pii/S0092867420306759?via%3Dihub . However, they also concluded that the infection pattern can’t be explained by receptor+coreceptor distribution alone. It is a nice, scholarly written article, which discusses how the virus most probably spreads through the respiratory tract, and that the infection most likely starts in the nose.

Thus, in common with this discussion thread (see earlier posts), a physiological and not immunological reason is assumed for the difficulty of SARS-CoV-2 to infect the lungs. However, whereas this article discusses ACE2 expression patterns, it is unclear how such could explain disease patterns in the different population groups (children, elderly, etc.). Instinctively (for what it is worth), if ACE2 abundance would be the sole explanation, I would expect more gradual differences between population groups. I still favor a model in which lower surfactant concentrations in lungs of the elderly make it easier for the virus to replicate.

Like others before them (see earlier posts), Hou et al. also found that IL-13, a typical Type 2 immunity cytokine, downregulates ACE2 expression [see picture (2)]. As discussed before, hay fever enhances IL-13 expression (e.g. https://pubmed.ncbi.nlm.nih.gov/27677865/ ), so that hay fever may help fight COVID-19 by not only immune reactions but also by physiological changes. Hou et al. state, regarding IL-13: “… IL-13, a cytokine associated with Th2-high asthma, inhibited ACE2 expression.”

Blood clotting appears to be a major cause of death in severe COVID-19 (Matacic, https://www.sciencemag.org/news/2020/06/blood-vessel-attack-could-trigger-coronavirus-fatal-second-phase ). In a review by van der Poll et al. Article Cytokines as regulators of coagulation and fibrinolysis

Studies published in the Lancet and The New England Journal of Medicine, both by Mehra et al. (Harvard University) have serious integrity issues https://www.sciencemag.org/news/2020/06/mysterious-company-s-coronavirus-papers-top-medical-journals-may-be-unraveling . I now have added notes to previous posts in which I referred to the Lancet study.

The Lancet article may have been directed at president Trump to ridicule his praise of hydroxychloroquine. Or not, it could be the same old problem of xxxxxx (self-censorship by JMD) in science for selling hot stories.

Anyway, for a moment I felt that, with its importance and the whole world watching, the COVID-19 research was a purer type of research than what we were used to. But it seems that, as usual, again we have to learn to distinguish a minority of correct papers from all the crap. My apologies for the dark thoughts.

Not just I got fooled by these articles https://www.theguardian.com/world/2020/jun/03/covid-19-surgisphere-who-world-health-organization-hydroxychloroquine .

The company Surgisphere which provided the "data" was founded by its chief executive Dr. Sapan Desai (for an impression see https://www.youtube.com/watch?v=KUxbJbPv0_4 ), who is co-author of the accused papers. The first and corresponding author was Dr. Mandeep Mehra of Harvard University (for an impression see https://www.youtube.com/watch?v=3W5rd_3Wjks ), who seems to be a pretty good scientist (though this was xxxxxxxx). Good for the peace in the world is that the misinformation appears to be homegrown in the US.

Interesting - https://www.webmd.com/lung/covid-allergies.

Allergies - a kind of Antigen - Antibody reaction ( Hypersensitivity Reaction ) - Increases the body defence mechanism .

The "LancetGate" issue depressed me, as the kind of literature investigation that I am doing in this discussion thread seems so useless if those papers can't be trusted. But I'll continue. Just a last post here on the LancetGate topic to get it out of my system. Dr. Elisabeth Bik, a famous sleuth having a magic eye for uncovering picture manipulations in science, found that the author of the Lancet paper that generated the "data" for that paper, Dr. Sapan S. Desai, in his younger years also was first author of a paper with the "Melba toast" pictures shown here. Please read her report on it https://scienceintegritydigest.com/2020/06/06/the-surgisphere-founder-and-the-melba-toast-figure/ ; apart from that it is very sad, the "Melba toast" part also makes it funny.

Dr. Martijn Hoogeveen now has his second manuscript out on pollen and flu-like diseases titled Pollen Explains Flu-Like and COVID-19 Seasonality https://www.medrxiv.org/content/10.1101/2020.06.05.20123133v1.full.pdf .

The strength of the study is that it shows a strong reverse correlation between flu-like infections and pollen-density/hay-fever in the Netherlands. However, the interesting statement "Meteorological factors alone do not predict seasonality, given substantial climate differences between countries that are subject to flu-like epidemics or COVID-19." would benefit from pollen density analyses for countries other than only the Netherlands. Nevertheless, the article is a good read and the concepts are interesting, and I advise readers here to read it. One of the interesting hypotheses in the paper is that many animals, not only people or only hay fever sufferers, may use pollen as a season signal which they use as cue for adapting their bodies (including immune their immune systems). A very important assumption in the study is that:

"On the basis of current data, we can conclude that also the covid-19 pandemic is seasonal and as a consequence multicycle, and will thus likely return from week 33 on, like all other flu-like viruses, when pollen season is over in the Northern Hemisphere." Thus, COVID-19 may be expected to gradually come back after August (see figure). If you live in the Northern hemisphere, my advice is to enjoy the coming months (in a safe manner), because afterwards COVID-19 may hit much harder than what we have experienced so far (see also the explanation in the Hoogeveen study that hitherto we only experienced the tail of a full COVID-19 season).

Johannes, it is not “global” but a prediction for the Northern Hemisphere and limited to the moderate climate zones. For the Southern Hemisphere/mcz the seasonal relation is 6 months flipped. Same principle. For tropical countries, flu season = rainy season (less pollen).

Martijn, thank you for recommending my post about your work. At first, I didn't see that recommendation and interpreted your last post as criticism. "Global" is not what I intended to suggest (see the text of my post), and your paper did not clearly divide the Northern Hemisphere into multiple zones. If you like, to this thread, you could add a world map with the zones where you think the 33 week rule applies.

Thx, how to interpret my respons was maybe not that clear as I struggled with the rather limited mobile app of RG. :-D

On itself the division NH/SH is discussed in the paper, but indeed is more implicit in the abstract as it focuses on NL. I am now going to work on a conf lecture. Maybe a geo map is a good idea as it's for a geographical society :-D