Remarkable Age Distribution of OC43 vs. SARS-CoV-2 in China

In their preliminary analysis of the clinical parameters of COVID-19, from the first 425 patients, Li et al noted that there were no pediatric cases, and that the median age was 59 years old. While there are sure to be pediatric cases now that the total number of confirmed cases has topped 70,000, the impression has been that the pediatric age group has been very much spared a major impact from SARS-CoV-2.

Li et al surmised that perhaps the pediatric age group had some inherent resistance to SARS-CoV-2 infection via the non-specific immune response. While there is precedent for such clinical resistance in other viral infections, such as Hepatitis A and Epstein Barr virus, another explanation may lie in the annual exposure of this age group to other coronaviruses causing upper respiratory disease.

In 2018, Zhang et al published a five-year survey of upper respiratory disease in Guangzhou, to the south of Hubei province in Guangdong province, focusing on the principal upper respiratory coronavirus, OC43. Not only was OC43 circulating in four of the five years, but also throughout the year. Other coronaviruses HXU1, 229E and NL63 were less prominent, but there was an outbreak of upper respiratory coronavirus infection every year, likely continuing to this day. Furthermore, the age distribution of those viruses was markedly skewed to the very young pediatric age group.

Thus, infants in China are exposed to OC43 and other upper respiratory coronaviruses every year of their early life. It is likely that their surface immunity to these viral agents is regularly boosted.

The age distribution of OC43 is contrasted to that of SARS-Cov-2 from Li et al, adjusting the age stratification to reconcile the two studies, is shown here.

The contrast between the two age distributions could not be more stark, almost to the point of being mutually exclusive. It should be noted that both samples included hundreds of patient samples.

OC43 and SARS-CoV-2, while their sequences can be readily aligned, have little similarity in the spike S1 protein. However, they have several areas of high sequence similarity in S2, the fusion glycoprotein moiety (not shown here). One hypothesis for the marked age distribution of SARS-CoV-2 could be that the younger infant and pediatric age group is resistant to SARS-CoV-2 due to annual exposure to other coronaviruses sharing much S2 sequence, imparting secretory immunity that is not as readily reinforced in the much older age groups.

If true, such a finding would have powerful consequences for understanding the SARS-CoV-2 outbreak and how to suppress it. One corollary may be that primary and secondary schools, so often drivers of outbreaks of respiratory viruses, are not a significant factor here. I would note that two months into the outbreak, in a city of 11 million and province of 57 million, there are “only” 70,000 confirmed cases, a number concomitant with adult to adult transmission.

A second corollary could be that immunity to the more highly conserved S2, using OC43 or HKU1 as sources of immunogen, may afford a jump start on immunization of those who have not yet experienced the presence of SARS-CoV-2 in their communities.

Bill Gallaher

Zhang S-f, Tuo J-l, Huang X-b, Zhu X, Zhang D-m, Zhou K, et al. (2018) Epidemiology characteristics of human coronaviruses in patients with respiratory infection symptoms and phylogenetic analysis of HCoV-OC43 during 2010-2015 in Guangzhou. PLoS ONE 13(1): e0191789.

Li Q, Guan X, Wu P, Wang X, Zhou L, Tong Y, et al. 2020. Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia. N Engl J Med. Jan 29. doi: 10.1056/NEJMoa2001316.

Below is an alignment in the spike (S) protein of the common Coronaviruses HKU1A, HKU1B, OC43 and the Wuhan Coronavirus now renamed SARS-CoV-2. Identical amino acids are highlighted in green.

HK A B OC and Wuhan.pdf (87.5 KB)

The pattern of amino acid identity is most evident in only some regions of the spike protein. Notably, there is very little similarity in the receptor binding motif in S1, and large deletions in the receptor binding domain, consistent with SARS-CoV-2 utilizing a different receptor from the others. However, there is some relatedness in flanking regions of S1.

The furin site in SARS-CoV-2 is convincingly alignable with the more efficient versions in the upper respiratory coronaviruses, with flanking regions of high similarity and identity.

The highest identity is seen in regions of S2, the fusion subunit, especially in the area of the HR1a-Hinge-HR1b (helix-turn-helix) region.

A large percentage of the population is exposed to upper respiratory Coronaviruses annually, that cause 25-30% of the common cold. It can be presumed that most develop secretory immunity that is boosted by re-exposure in subsequent years. In 2019-2020 there has been unusually high activity of HKU1 in the US, peaking recently. Those who have recently been exposed to these upper respiratory coronaviruses HAVE been exposed to the amino acid sequences in SARS-CoV-2 highlighted in green.

The age distribution of morbidity, and the much higher mortality in the 70-79 yo group (8%), and 80+ yo group (14.8%) are reminiscent of morbidity of shingles, and suggestive that immunological amnesia may play a role in both high morbidity and mortality in the elderly with SARS-CoV-2.

Bill Gallaher
(William R. Gallaher Ph.D., Professor of Microbiology, Immunology and Parasitology, Emeritus. LSU Health New Orleans. 1973-Present)

When one normalizes country data to the onset of the outbreak in each country, it becomes apparent that there are slight, but potentially significant differences in the epidemic curves.

Progression 022920

Countries and cultures differ substantially. For instance, China has had a one-child policy for a very long time, and fewer children per household, or, for that matter, entire communities, than other countries. Countries with more children, and more grandparents caring for children, may increase the circulation of upper respiratory coronaviruses in the very young and boost immune responses in older adults. Natural herd immunity to conserved peptides in coronaviruses could significantly vary, affecting both morbidity and mortality going forward.

While Japan has closed its schools, I have seen nothing that suggests this was data-driven. The pediatric impact of COVID-19 has been notably slight. They may not even be harboring the virus hardly at all. If this is all adult to adult transmission, school closings are both unnecessary and highly disruptive, especially in two-earner families.

Data-driven, people. Data-driven. Not knee jerk or hysterical. Let public health professionals run the response and not politicians.

The wrong answers are not just counterproductive but can be harmful. There is the famous medieval example of eliminating cats in response to plague in England – letting the rats thrive – when what was needed was simply removing the Broad Street water pump. Data-driven. Data-driven.

Bill Gallaher

Sorry for conflating the plague and cholera there. Plague was rats. Cholera was the Broad Street pump, and it was just necessary to remove the pump handle. Politicians driving to pilot the pandemic ship drive me nuts.

Point remains, without data stupid mistakes are made.

Bill Gallaher


Recent news is that immunologists are discovering SARS-CoV-2 reactive T cells in patients never exposed to the virus, while noting that nearly everyone is exposed to upper respiratory coronaviruses in childhood.

On Feb 17 in this thread, I said the following:

[quote=“profbillg1901, post:1, topic:399”]
One hypothesis for the marked age distribution of SARS-CoV-2 could be that the younger infant and pediatric age group is resistant to SARS-CoV-2 due to annual exposure to other coronaviruses sharing much S2 sequence…

Public blogpost, five and one half months ago. In terms of the pandemic, an eternity ago.

Bill Gallaher