Does natural infection with SARS-CoV-2 offer more protection than a vaccine against new variants?

As the current coronavirus disease 2019 (COVID-19) pandemic stretches on, vaccines have provided a gleam of hope. This is threatened by the emergence of multiple variants of the causative pathogen, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which can offer resistance to the humoral immunity elicited by both natural infection and vaccines.

A new study, released on the preprint medRxiv* server, demonstrates that vaccines may be less effective at countering the newer variants compared to natural infection.

Study: mRNA vaccination compared to infection elicits an IgG-predominant response with greater SARS-CoV-2 specificity and similar decrease in variant spike recognition. Image Credit: ktsdesign / Shutterstock

What are mRNA vaccines?

The twenty or so vaccines currently approved against the virus have been developed on several platforms. The earliest to receive emergency use authorization (EUA) both use the messenger ribonucleic acid (mRNA) platform – the BNT162b2 (BioNTech/Pfizer) and the mRNA-1273 (Moderna/NIAID) vaccines.

These are unlike conventional vaccines that present the antigen to the host, because they use mRNA to encode the desired antigen in the recipient. Their safety and efficacy were demonstrated in clinical trials, which preceded their authorization.  

The use of mRNA means the cell’s protein-synthesizing machinery is used to produce a properly folded viral antigen within the host, thus stimulating robust antibody and cellular responses. Comparative studies remain to be done, showing how these fare relative to lipid nanoparticles and other vaccine platforms or to natural infection.

In COVID-19, protection is a correlate of neutralizing activity. Most neutralizing antibodies are specific to the receptor-binding domain (RBD) of the viral spike, the protein that drives viral attachment to the host cell receptor, the angiotensin-converting enzyme 2 (ACE2) receptor.

Both the mRNA vaccines encode the prefusion stabilized spike antigen, which generated strong immune responses and neutralizing antibody titers in phase III clinical trials. However, the encoded spike belongs to the early Wuhan-Hu-1 spike antigen.

Variants of concern

Recently, three variants of concern (VOCs) have emerged, with grouped mutations in the spike and especially the RBD. These include the United Kingdom (B.1.1.7 lineage, UK variant), South Africa (B.1.351 lineage, SA variant), and Brazil (P.1 lineage). They threaten the control of the pandemic by reducing vaccine efficacy, probably by interfering with the binding of the RBD to the receptor.

While the RBD N501Y mutation is common to all of them, the Brazil and SA variants have two more alterations in the RBD. The E484K mutation is associated with the lack of efficacy of many neutralizing antibodies, by altering the epitopes or antibody recognition sites on the RBD.

Neutralizing antibodies that work by binding to the RBD and thus preventing RBD-ACE2 binding thus fail in efficacy when confronted with this unrecognizable RBD variant.

The current study follows antibody responses following vaccination with the BNT162b2 vaccine, and in COVID-19 patients, and describes the characteristics of the antibody response.

High IgG responses

Vaccination produced a potent and targeted response to the viral spike proteins and their domains.

It resulted in high titers of anti-SARS-CoV-2 IgGs, among those naïve to the infection as well as to the few with a prior history of documented infection. Anti-spike, anti-RBD and anti-NTD (N-terminal domain) IgG titers were initially negative in naïve subjects, began to increase by day 21 post-vaccination, and peaked at days 28 and 42.

The speed of increase was increased in almost all previously COVID-19-positive subjects, with three of four showing an increase in IgG specific to the spike and its domains by day 7. Titers on day 21 were higher than for the naïve group.

Anti-nucleocapsid antibody levels were negative at baseline and after vaccination, in infection-naïve patients. While positive at baseline in the previously infected group, no increase was observed after vaccination, as expected.

The IgG responses were strong and comparable between the two groups, except in two, one of whom received only one dose, and one an elderly individual with a history of oral cancer in the past. Younger recipients had higher antibody titers compared to those over 60 years.

Side effects included local tenderness, myalgia, headaches and fatigue, but were not predictive of the antibody response.

Two recipients on the immunosuppressive agent methotrexate showed comparable antibody responses. The immune response was dominated by IgG antibodies targeting the spike, RBD and NTD.

Variation in Ig isotype response

Severe COVID-19 is associated with higher titers of specific SARS-CoV-2 antibodies compared to asymptomatic or mild illness. The researchers, therefore, compared the antibody response in moderately and severely ill patients compared to that in vaccine recipients.

The IgG antibody titers to the NTD, RBD and the spike proteins in vaccine recipients were similar to those in severe COVID-19, being higher than those in milder illness, especially for antibodies targeting the NTD and RBD.

A major difference was observed with regard to the IgM and IgA anti-spike, anti-RBD and anti-NTD antibodies, which were elicited at minimal levels in vaccine recipients compared to IgGs. This was not the case after natural infection, where individuals with severe illness showed markedly higher IgA antibodies against the spike.

Lower vaccine-elicited responses to endemic coronaviruses

There is significant variation in the SARS-CoV-2 proteins compared to those of other human coronaviruses (hCoV), but at the epitope level, there is high sequence conservation. The result is cross-reactivity, which is difficult to distinguish by routine serology.

That is, antibodies elicited by SARS-CoV-2 induced reactivation of memory B cells that were specifically recruited by a pre-existing hCoV may be indistinguishable from newer cross-reactive antibodies elicited by SARS-CoV-2 infection or vaccination.

Natural infection resulted in higher IgG antibody levels to HCoV-OC43 and HCoV-HKU1 than observed after vaccination, but without affecting the overall SARS-CoV-2-specific antibody response, which was similar in both cohorts.

IgA and IgM levels against the endemic hCoVs were higher in severe COVID-19 than after vaccination. The former was especially high with alphacoronavirus HCoV-NL63 S but not in vaccinees.

SARS-CoV-2 variants evade antibodies

Many monoclonal antibodies (mAbs) that were potent against the Wuhan variant have shown reduced efficacy against the South African (SA) variant, as well as antibody-containing convalescent and patient serum samples.

A comparison of the three variants was therefore performed in plasma samples from the vaccine recipients and in COVID-19 patients. The researchers explored RBD binding and blocking of binding between ACE2 and RBD or ACE2 and spike protein.

They found that plasma from both vaccine recipients and infected subjects showed reduced antibody binding to spike and RBD of the variants compared to the wildtype virus. The greatest reduction was with the South African and Brazil variants.

The high blocking activity seen by day 28 in vaccinees, at seven days from the boost dose, was directed against the wildtype virus, with a progressive reduction in binding with the UK > Brazil > SA variant spike. A similar reduction was observed for spike-ACE2 blocking antibody efficacy.

The very similar decline in blocking for the SA and Brazil variant indicates that the K417N and K417T mutations in the RBD are synonymous as far as their effect on ACE2 blocking antibodies goes.

These data indicate that the effects of viral variants are remarkably consistent with an escape from polyclonal antibody responses elicited either by infection or BNT162b2 vaccination.”

What are the implications?

The study shows that severe COVID-19 and BNT162b2 vaccination both produce high IgG responses directed against the spike, RBD and NTD antigens, but the former has a higher breadth, with a short-lived but strong IgA and IgM response in addition to a durable IgG response.

With vaccination, IgG production is significantly dominant, at all ages of adulthood, though somewhat lower in people over 60 years. These antibodies are also significantly less cross-reactive with other endemic hCoVs.

These vaccines contain lipids that may act as powerful drives of IgG class switching early in the course of the immune response. This could be because of the Th1-biased CD4+ T cell responses, with robust germinal center formulation, following exposure to the vaccine, especially the lipid components.

The risk of reinfection after infection is about 83% lower as reported in the SIREN (SARS-CoV-2 Immunity & Reinfection Evaluation) study of health care workers. The BNT162b2 vaccine was reported to have 95% efficacy against primary infection, which may be due to the protective efficacy of the higher IgG levels.

The narrower breadth of the immune response in vaccine recipients could be due to the difference in the body locations where the immune system encounters the virus. In natural infection, the mucosal surface is the site of antigen-immune cell meeting, compared to the systemic circulation in vaccination.

However, cross-reactive antibodies do not seem to protect against SARS-CoV-2 infection.

The comparable proportions of polyclonal antibodies directed against the epitopes of the new variants are very similar in vaccine recipients and infected individuals. Thus, vaccinated individuals are just as likely to be reinfected by the SA and Brazil variants, especially as antibody titers fall over time.

Further studies will be important in measuring the unique features of vaccine-induced antibodies relative to those elicited by infection, especially to understand how these are involved in “long COVID.” The efficacy demonstrated by the mRNA vaccines against the vaccine antigens indicates that this platform will be useful in creating other viral vaccines as well.

*Important Notice

medRxiv 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.

Written by

Dr. Liji Thomas

Dr. Liji Thomas is an OB-GYN, who graduated from the Government Medical College, University of Calicut, Kerala, in 2001. Liji practiced as a full-time consultant in obstetrics/gynecology in a private hospital for a few years following her graduation. She has counseled hundreds of patients facing issues from pregnancy-related problems and infertility, and has been in charge of over 2,000 deliveries, striving always to achieve a normal delivery rather than operative.