This is the third of my posts, in which I cover some of the hypotheses and research into why some cases of COVID-19 are more severe than others. Much of this information comes from clinical observations of COVID-19 patients that are correlated with poor disease outcomes, and scientists and physicians and hypothesize that these could be linked to disease progression. There is a strong push to initiate clinical trials, even small ones, so that observations can be generalized, but for now, most of the data is anecdotal.
Age, Comorbidities, and Being Male Increase COVID-19 Severity
A big question surrounding COVID-19 is why some people infected with SARS-CoV-2 get more severe symptoms than others. There is no preexisting immunity in humans against SARS-CoV-2 because no virus that is very similar to it has broadly infected the population. The SARS and MERS outbreaks infected fewer than 10,000 people each, so there was no herd immunity in the population, and the true number of people who have had recent colds caused by the mild coronaviruses (called HCoV-229E, -NL63, -OC43, and -HKU1) is unknown.
People who are older are at greater risk of severe COVID-19 and death due to preexisting chronic health conditions (comorbidities). SARS-CoV-2 primarily affects the respiratory tract and the ability of the blood to carry oxygen, and many of the common chronic conditions in developed countries, such as diabetes, hypertension, asthma, chronic obstructive pulmonary disease (COPD), obesity, and cardiovascular disease, all affect the cardiovascular and respiratory systems. These conditions don’t necessarily make it easier for SARS-CoV-2 to infect cells, but they increase the severity of COVID-19 and risk of death by making it more difficult for patients to recover.
Differences between people’s immune systems may be the reason for different disease outcomes, but this encompasses many factors. Immunosenescence – aging of the immune system – makes the immune system slow to respond to future infections. The adaptive immune system is more so affected, making it difficult for older adults to retain immunity to an infection after they recover (side note: this is something to consider when developing vaccines because they may not work as well in the elderly). As people age, fewer and less diverse T cells (T lymphocytes), which are responsible for many functions in adaptive immunity, are produced because of shrinking of the thymus (a small organ in the chest), where T cells mature 1, 2 .
Based on data of around 72,000 COVID-19 cases (including asymptomatic ones) in Wuhan, China through mid-February 2020, the case fatality rate (deaths per number of diseases) and the infection fatality rate (deaths per number of infections, including asymptomatic ones) are higher for men than for women 3, 4 . In Italy, Spain, and New York City, men are more likely to be in the intensive care unit (ICU) and die from COVID-19 5 , but there is a severe lack of data in the United States on demographics of COVID-19 infections and deaths. However, data from previous SARS and MERS outbreaks suggests that men are more likely to have severe disease and poor outcomes 6, 7 . The discrepancy between men and women may be due to immunological differences.
Female humans have two X chromosomes, while males have only one. The X chromosome contains the largest number of immune-related genes of any of the chromosomes 8 . In each cell of a female, one of the two X chromosomes is inactivated, and the inactivated one is different in each cell, 9 as shown below.
Women produce more antibodies in response to infection, vaccination, and trauma, while men exhibit more severe inflammation 10 . In addition, more than three quarters of the 20 million Americans with autoimmune diseases are women, suggesting that the immune systems of women tend to be more robust than those of men, but they can, in fact, be too strong 10 . A stronger immune system could be an evolutionary advantage for women during pregnancy to allow them to tolerate the developing fetus (half of the fetus’ antigens come from the father, so they are foreign to the mother’s immune system). This could make women better at clearing SARS-CoV-2 and other pathogens early before diseases worsen. There is also some evidence that approximately 5-10% of X chromosome genes may escape inactivation, possibly bolstering the immune system 11 .
An interesting point is that the gene for ACE2, the receptor used by SARS-CoV-2 to bind to host cells, is found on the X chromosome. Although this could be a potential cause for different case fatality rates between men and women, a study of common genetic polymorphisms, or forms, of ACE2 in SARS patients showed that even though men had a higher mortality rate 12 , the difference is not due to differences in ACE2 13 . Since SARS-CoV and SARS-CoV-2 both use ACE2 to attach to cells, it is reasonable to assume that ACE2 polymorphisms would not affect disease outcomes for COVID-19 either.
Elevated Cytokines and Inflammatory Markers
In an update on This Week in Virology (TWiV) podcast, Daniel Griffin, chief medical officer for ProHealth in the New York metropolitan area, described that several inflammatory markers, such as the cytokine interleukin-6 (IL-6), are upregulated in patients with severe COVID-19 14 . Cytokines are signaling molecules involved in many immune functions, including promoting and inhibiting inflammation. IL-6 is released early in an immune response (the acute phase), is one of many contributors to fever, and has 3 primary functions 15 .
- Stimulate hepatocytes (liver cells) to release various acute phase proteins.
- Increase the number of neutrophils, a type of innate immune cell.
- Promote proliferation of B cells, adaptive immune cells that make antibodies.
High IL-6 levels have been detected in some patients with severe COVID-19 in China 16, 17, 18 . Elevated IL-6 has also been found in patients with severe COVID-19 in New York City, but this is more anecdotal and was not studied against patients with mild COVID-19 14 . IL-6 receptor inhibitors, such as tocilizumab (brand name Actemra), which is a monoclonal antibody, are used to treat rheumatoid arthritis because IL-6 is involved in several autoimmune and inflammatory diseases 19 . Physicians have been prescribing corticosteroids and IL-6 inhibitors to a few patients with the hope of reducing widespread inflammation. However, many of these drugs have only been tested in patients who are not actively fighting an infection, and they may suppress the immune system while it is trying to clear SARS-CoV-2, worsening the disease or making patients more susceptible to other bacterial or fungal diseases 20 .
Steroids may need to be administered on a schedule. Stanley Perlman, who studies coronaviruses at the University of Iowa, noted that administering interferons – intercellular signaling proteins – before an infection or in the beginning stages was protective in mice (likely because the interferons prevented viral infection) 21 . However, giving interferons later in a respiratory infection may worsen cytokine storm. The clinical recommendation from Daniel Griffin is to allow the immune system time to try to fight off SARS-CoV-2 when the disease is virus-driven (within a few days after the first mild symptoms appear), but give steroids later when the disease is immune system-driven 14 . Some people have worried that taking steroids for allergies or asthma could make it harder for your immune system to fight off SARS-CoV-2, but asthma attacks cause respiratory damage that will worsen COVID-19. The American Academy of Allergy, Asthma & Immunology recommends for asthma patients to keep taking prescription steroids 22 .
COVID-19 May Cause Blood Clotting
On TWiV, Daniel Griffin also described that there are growing concerns of abnormal clotting (thrombosis) in COVID-19 patients because high levels of a clotting marker called D-dimer have been associated with a greater risk of death 14, 23, 24 . Blood clotting is a normal response to local injury; when a tissue is damaged, blood flow is first increased to the site. Blood cells called platelets initiate clot formation to prevent uncontrolled bleeding and to allow the tissue to be repaired by other cells 25 .
Clot formation is promoted by high levels of cytokines, primarily IL-1β, IL-6, and IL-8 26, 27 . Clots may be caused by the innate immune response because in addition to plugging up a wound to stop bleeding, clots block and trap microbes that can enter through wounds. Thus, the cytokine storm seen in some COVID-19 patients could lead to formation of microclots. Once clots form, they are degraded, producing D-dimer. Besides having COVID-19 as a potential cause, clots can also form in hospitalized patients lying on beds for extended periods of time with little movement due to poor blood circulation. Sedation, administered to COVID-19 patients on ventilators, also reduces mobility. A large clot can dislodge and migrate elsewhere in the body, and if it lodges in a vessel leading to a vital organ, it may block blood flow and cause fatal damage. Physicians are increasingly prescribing anticoagulants (blood thinners) to hospitalized COVID-19 patients because even young patients with no underlying conditions could be at risk of cytokine-mediated clotting 28 .
Variations in Immune Cells and Antibody Types
Severe COVID-19 and death have also been correlated with a high neutrophil to lymphocyte ratio (NLR) 29 . Neutrophils are part of the innate immune system, and they rapidly appear at infection sites to engulf pathogens and promote inflammation 30, 31 . High NLR is associated with increased likelihood of death from severe inflammation (COVID-19 appears to manifest this way) and low activity of natural killer (NK) cells and T cells 32 . NK and T cells are lymphocytes that kill infected host cells, and they secrete IFN-γ, a cytokine that inhibits viral replication and increases further lymphocyte activation 33, 34 . A high NLR results in a weaker adaptive immune response and fewer memory B and T cells, which protect against reinfection by the same virus. Excessive inflammation caused by neutrophils can also inhibit memory B cell proliferation.
Among 61 COVID-19 patients in Beijing, China in January 2020, those with moderate/severe disease had a higher NLR and more bacterial infections than patients with mild symptoms 35 . Patients with severe COVID-19 symptoms in Wuhan in January and February 2020 had more cytokines and fewer T cells in the blood 36 , and T cells were the most severely reduced group of lymphocytes 37 . The diagram below shows how cytotoxic T cells (which have CD8 receptors) are necessary to clear viral infections by killing infected cells.
Immunoglobulin A (IgA) is another class of antibodies found primarily in mucous membranes. Mucous membranes, which contain specialized epithelial cells, protect against dehydration and pathogens, and IgAs are primarily found in respiratory and gastrointestinal secretions, such as tears, saliva, and mucus (only ~10% circulate in the blood). IgAs are important for neutralizing viruses and preventing them from infecting epithelial cells, but they also neutralize bacterial toxins. Many people with selective IgA deficiency are asymptomatic because their bodies still produce other types of protective antibodies, but some tend to have higher incidence of infectious diseases (primarily respiratory and gastrointestinal), allergies, asthma, and autoimmune diseases 38, 39, 40 . Robust IgA activity may contribute to mild COVID-19 cases because SARS-CoV-2 could be cleared early before it gets to epithelial cells in the lower respiratory tract. This is all speculation and would be an interesting topic to research.
Not All Antibodies are Created Equal
The goal of vaccines is to elicit neutralizing antibodies in individuals, but viruses often elicit production of non-neutralizing antibodies by the immune system. Neutralizing antibodies inhibit viral proliferation, while non-neutralizing antibodies bind to a virus, but do not prevent infection. The two types of antibodies tend to bind to different regions of a virus, and non-neutralizing antibodies can interfere with the activity of neutralizing antibodies, but there are no general rules for how and why non-neutralizing antibodies arise. This is a problem for antibody tests to estimate SARS-CoV-2 immunity because just because a person has SARS-CoV-2 antibodies DOES NOT mean they are immune to the virus because they may have non-neutralizing antibodies. There are currently no point-of-care tests that can rapidly identify neutralization. Most of the serological tests in development check for the presence of antibodies that bind to a piece of the virus. Until we have studied neutralizing and non-neutralizing antibodies in recovered patients, it is risky to assume that everyone who has recovered from COVID-19 is immune to SARS-CoV-2.
As I mentioned in post 2, there are concerns about antibody-dependent enhancement for SARS-CoV-2, in which antibodies could help the virus infect cells by simultaneously binding to Fc receptors and SARS-CoV-2 (called cross-linking), as shown below.
Therefore, some anti-spike antibodies may increase inflammation and acute lung injury 41, 42 . Other studies showed that antibodies against only the receptor-binding domain of the spike protein – the part that binds to ACE2, but not the entire spike – protected mice against future SARS-CoV infection 43. We therefore need ways to distinguish between effective antibodies and potentially harmful ones.
SARS-CoV-2 in Younger Individuals
Young people make up a smaller percentage of COVID-19 deaths, and this has been the case in all countries hit hard by SARS-CoV-2. Children have lower mortality rates, less severe disease, and require less hospitalization 44. A large reason for this is that young people have fewer comorbidities, and their innate immune systems tend to respond faster to infections than those of older people.
There is some speculation that children may have different ACE2 expression than adults, but this is unlikely to be the cause of more severe outcomes in older adults because SARS had a 0% mortality rate in people under the age of 24 and a mortality rate higher than 50% for people aged 65 and older 45. SARS-CoV and SARS-CoV-2 both use ACE2 to enter cells, so if ACE2 expression was the reason for different COVID-19 outcomes, we would not already have seen COVID-19 deaths in children, adolescents, and younger adults.
Another hypothesis is that children get more colds, on average, and some of these colds are caused by mild coronaviruses. This may provide them with some protection if their immune systems react to sequences that are similar across many coronaviruses, including SARS-CoV-2. Children may get milder forms of COVID-19 if they have cross-reactive antibodies, which react to similar but not identical epitopes on multiple coronaviruses.
A very interesting possibility is that some childhood vaccines may induce innate immunity to other viruses, including SARS-CoV-2. Live-attenuated vaccines may especially stimulate the innate immune system because they simulate virus infection and replication in cells, but the virus is too weak to cause disease in humans. These vaccines could temporarily boost the immune system, even against viruses that are completely different from the virus in the vaccine. Children may have more residual non-specific immunity because they recently received vaccines like measles, mumps, and rubella (MMR) and varicella (chickenpox). Some countries with high tuberculosis incidence still vaccinate against it; this live-attenuated vaccine is called Bacille Calmette-Guérin (BCG), and it may similarly stimulate the immune system, even though tuberculosis is caused by a bacterium.
Some scientists are calling for clinical trials to study the efficacy of such vaccines to provide short-term protection against SARS-CoV-2. They are especially interested in Albert Sabin’s oral polio vaccine (OPV) because there was evidence that it protected people in the Soviet Union against influenza in the late 20th century 46, 47, 48. These vaccines will NOT provide long-term adaptive immunity against SARS-CoV-2, only a short-term boost to the immune system for broad protection. But repurposed live-attenuated vaccines could be a way to slow the number of new COVID-19 infections and deaths until we can make a SARS-CoV-2 vaccine.
Possible Cross-Reactivity Between Coronaviruses
Some studies indicated that people infected with SARS-CoV had some reactivity against the cold-causing coronaviruses 49, and serum taken from healthy people before the current COVID-19 pandemic appeared to react to the cold-causing coronaviruses, but not to the severe coronaviruses (SARS-CoV, MERS-CoV, and SARS-CoV-2) 50. There are conflicting reports about whether there could be antibody cross-reactivity (and cross-neutralization, which is more important) between SARS-CoV-2 and SARS-CoV 51, 52, so we have to assume that the entire human population is immunologically naive to SARS-CoV-2.
The diagram below is a phylogenetic tree that shows the evolutionary relationships between many (but certainly not all) coronaviruses. It includes the 7 identified human coronaviruses and some coronaviruses found in bats, such as SARS related (SARSr) viruses. Viruses that are closer together and connected by fewer branching lines are more related to each other. Related viruses are more likely, but not guaranteed, to have cross-reactivity and cross-neutralization. 53
The differences we’ve seen in how COVID-19 patients fare is most likely due to different immune responses to SARS-CoV-2. Some trends, like old vs. young, are broadly applicable because weaker immune systems and more comorbidities make it more difficult for people to recover from many viral and bacterial infections. But COVID-19 can be deadly even in young, healthy people because it causes severe pneumonia and life-threatening damage to the lungs and other organs. As we better understand how the inflammatory response progresses, we may be able to develop treatments, such as steroids or antibodies, to prevent patients from progressing to fatal acute respiratory distress syndrome (ARDS).