The currently raging pandemic of coronavirus disease 2019 (COVID-19) is taking more lives today than over the first five months after its onset. The difficulty in gaining control of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) stems partly from the fact that most infections are asymptomatic or very mild. Moreover, the frequent emergence of escape variants that overcome immune inhibition elicited by earlier variants has also challenged attempts to arrest the virus’s spread.
A new study, released as a preprint on the bioRxiv* server, discusses variations in viral pathogenicity and in transmission, which are responsible for asymptomatic infection and for superspreading events. The ability to understand this would open the way to understand how this virus behaves the way it does, in terms of its virulence, its spread, and its emergent variants of concern.
Why control theory?
The surging number of infections is mainly because of the steep differences in viral virulence and spread between individuals. The two main determinants of such variation are the low levels of the host receptor, and the suppression of interferon, the innate antiviral host response.
To understand how this variation operates, the researchers chose to use control theory, which is a mathematical framework for the analysis of complex feedback systems. Feedback is thus viewed as a control tool to change the behavior of unstable systems.
As scientists in an earlier paper have observed:
The COVID-19 pandemic isn’t an obvious or typical engineering problem. But in its basic behavior it is an unstable, open-loop system. Left alone, it grows exponentially: Like many such systems, it can be stabilized effectively and efficiently by applying the principles of control theory, most notably the use of feedback.”
Control theory could help, they say, to offer an exit strategy that uses medical and socioeconomic feedback to both restrict viral transmission and manage the burden of cases over the longer term, when restrictions on mobility and social interaction are relaxed.
The advantage of using this approach in the current study is that its design overcomes the lack of knowledge about the specifics of the immune response and the epidemiology of COVID-19.
In addition, it allows the identification of mechanisms by which the virus can show differences in virulence and in its spread even in the best-case situations.
The aim was to find the highest possible immune response by the host, by aggregating all the changes in the immune functions as a single control function. The highest or best-case immune function would be that which reduces virulence to the lowest possible, while also limiting viral spread. The various constraints used are the viral load, immune activity, and the entry of new virus.
The problem has both open-loop and closed-loop components. The first is intracellular and has to do with viral replication: cell entry, intracellular replication, and virion release. The outcome will be the number of productively infected cells arising from the originally infected cell.
The second involves immune control due to innate immune factors acting outside the cell.
Symptoms provide warning
The proportion of respiratory epithelial cells vulnerable to SARS-CoV-2 infection is small, relative to rhinoviruses, respiratory syncytial virus, and influenza virus. The number of cells that show productive infection following the infection of one cell can vary widely from sample to sample. This is borne out by earlier single-cell-based reports.
When the number of secondarily infected cells is higher, the viral load can be brought down only by a stronger immune response. This, in turn, will produce symptoms, which cause recognition of the illness.
As a result, the patient takes steps to avoid spreading the infection to others, such as self-isolation and quarantine. The researchers factored in a simplified version of such behavioral changes, with the warning signal being a scaled form of the immune response causing the patient to avoid social interactions with others.
The immune response also reduces viral shedding.
The researchers also implemented a long presymptomatic period with subsequent severe symptoms in all cases. If only contacts can be warned about the possibility of infection and quarantined, this model would make containment very efficient. This is because all infected individuals would be aware of the infection and take appropriate steps.
This chain fails when cases are completely asymptomatic, even if they are not as infectious as presymptomatic cases – and, in fact, they need not make up a major share of all infections for this to happen.
Extracellular immune responses vary more with the level of productive infection in neighboring cells than interferon-based immune responses, which are in turn lower than interferon-suppressed responses.
The early use of interferon could help reduce the final symptom severity in patients who seem to have progressive disease.
However, interferon-mediated suppression of immune responses may prolong viral replication and shedding, as well as precluding behavioral changes prompted by warning symptoms.
The researchers also applied this to other viruses, such as the seasonal endemic coronaviruses (CoV) like HCoV-NL63, as well as the earlier SARS-CoV, are also ACE2-tropic viruses. While the first causes mild cold-like symptoms or none at all, the latter is associated with severe respiratory symptoms in most cases.
These infections may vary markedly in severity, causing extremely mild or extremely severe cases to be left out of the clinical case count, as they escape detection altogether or cause rapid death before any clinical suspicion of the true diagnosis arises. This is true of the spread of SARS-CoV, indicating that such variability in the transmission is probable.
However, HCoV-NL63 has low virulence, perhaps because the spike glycoprotein has low infectivity. Thus, few cells show secondary productive infection from a single infected cell.
Interferon a double-edged weapon
Interferon administration has been shown to modulate the severity of infection with CoVs, if given early enough. However, the greatest effect is if administered before symptoms begin.
On the other hand, clinical trials have shown that the presence of immunosuppression early in SARS-CoV-2 infection increases the severity of the illness.
The suppression of antiviral interferon production or release within infected cells is a common mechanism used by respiratory viruses. Thus, the use of such interventions may be better guided by control-theory-based estimates of efficacy.
If the host benefits most from immune control, viral replication is rapidly blocked, and both transmission and symptomatic disease are prevented. With interferon suppression, viral spread is maximum when the virulence is lowest.
Host variation plays a role here, such that even under such conditions, increased spread may occur as the virulence increases.
What are the conclusions?
The researchers found that according to this model, virulence and transmission are closely linked. The virus is mostly transmitted by presymptomatic cases that will eventually develop severe infection.
These patients have a highly productive infection of cells adjacent to the infected cell. The effect is most pronounced when groups of such individuals interact since one group is shedding virus at a high rate, while the high proportion of susceptible cells makes the other group very vulnerable.
This can be reduced by an advance warning which induces behavioral changes to limit transmission. Moreover, control theory may well guide the use of interferon in future novel pathogens, due to the extremely common observation of interferon suppression by respiratory viruses.
bioRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.