Latest data on immune response to COVID-19 reinforces need for vaccination

17th June 2021
Vaccine syringe and calendar with clock
  • Immune response to COVID-19 following natural infection varies between individuals 6 months after natural infection, one of the world’s most in depth studies of immune memory has found. 

  • Findings from DHSC-funded PITCH study show that previous COVID-19 infection does not necessarily lead to long-term immune responses against COVID-19, leaving people potentially vulnerable to reinfection. 

  • COVID-19 vaccines generate higher immune responses than natural infection, underlining the need for everyone to get vaccinated for maximum protection against this disease. 

  • Study shows an individual’s immune response at one month post infection can predict if they will have a durable immune response at six months. 


This new study led by University of Oxford, in collaboration with the Universities of Liverpool, Sheffield, Newcastle and Birmingham with support from the UK Coronavirus Immunology Consortium, has found that previous infection, whether it was symptomatic or asymptomatic, does not necessarily protect you long-term from COVID-19, particularly against new Variants of Concern.

The Protective Immunity from T cells to COVID-19 in Health workers study (PITCH) examined how the immune system responds to COVID-19 in 78 healthcare workers who had experienced either symptomatic or asymptomatic disease (66 vs 12). An additional 8 patients who experienced severe disease were included for comparison. 

Blood samples were taken monthly from 1–6 months post infection to examine different elements of the immune response. This included different types of antibodies – such as Spike-specific and Nucleocapsid-specific antibodies which are produced to target different parts of the virus, alongside B cells, which manufacture antibodies and keep the body’s memory of the disease, and several types of T cell.

The preprint report posted on Research Square today details a highly complex and variable immune response following COVID-19 infection.  

The researchers used a new machine learning approach – nicknamed SIMON – to identify detailed patterns in the data and to see if initial disease severity and the early immune response could predict longer-term immunity.

They found an early immune signature, detectable one month post infection and linked to both cellular and antibody immunity, which predicted the strength of immune response measured at 6 months post infection. This is the first time that such a signature has been found and improves understanding of the development of lasting immunity. When serum samples (containing antibodies) obtained at 1 and 6 months post infection were tested, the majority of samples from people who produced a weak immune response signature failed to show any neutralising antibodies against the Alpha variant, with none mounting a neutralising antibody response against the Beta variant. This raises the possibility that the immune memory of these individuals does not provide sufficient protection to prevent reinfection by these variants. 

While the majority of people who had symptomatic disease did have measurable immune responses at six months post infection, a significant minority (17/66; 26%) did not. The vast majority of people who experienced asymptomatic disease (11/12; 92%) did not exhibit a measurable immune response at six months post infection. This implies that people who have previously been infected with COVID-19 should not assume they are automatically protected against reinfection and highlights the importance of everyone getting their COVID vaccination when they are offered it. 


Health Minister Lord Bethell said:

“This powerful study addresses the mysteries of immunity and the lessons are crystal clear. You need two jabs to protect yourself and the ones you love. I call on anyone invited to get vaccinated to step forward and finish the job so we can all get out of this.”


Key findings from the PITCH study:

  • Immune memory following COVID infection is measurable at 6 months but is highly variable between people

  • Previous infection does not necessarily protect you long term from SARS-CoV-2, particularly variants of concern Alpha and Beta. Individuals who show little or no evidence of immune memory to COVID at 6 months post infection are not able to neutralise the variants of concerns. 

  • We can use the immune response characteristics at one month post COVID infection to predict which people will have durable immune responses at six months.

  • People with COVID symptoms have variable immune responses that may decline over time and are not necessarily protected from SARS-CoV-2 variants.

  • People who experienced asymptomatic infection tend to have lower immune responses across the many immune parameters we have measured.


Understanding the strength and durability of the immune response to natural COVID infection remains highly relevant as it will help us reduce reinfections, better understand immune responses to vaccination and tackle new variants of concern.  Further research will continue to deepen our understanding of the immune responses over the longer term and what it means for protection against COVID-19 in the real world.

This study reinforces how important it is that everyone gets their COVID vaccination when offered. COVID-19 vaccines generate higher immune responses than natural infection, underlining the need for everyone to get vaccinated for maximum protection against this disease and in particular against Variants of Concern. 
Even if you think you’ve had COVID previously, you may not have protection against reinfection. 


Dr Christina Dold, study author from University of Oxford, said:

“Our study is one of the most comprehensive accounts of the immune response following COVID-19 in both symptomatic and asymptomatic individuals.  We found that individuals showed very different immune responses from each other following COVID-19, with some people from both the symptomatic and asymptomatic groups showing no evidence of immune memory six months after infection or even sooner. Our concern is that these people may be at risk of contracting COVID-19 for a second time, especially with new variants circulating. This means that it is very important that we all get the COVID vaccine when offered even if you think you may have previously had COVID-19.”


Dr Adriana Tomic, study author from University of Oxford said:

“Using machine learning is a new and exciting tool that allows us to dig down into large and complex datasets to draw out patterns that we would otherwise be unable to detect.  For this study, we wanted to try to understand if there are immune factors that can predict how likely it is that a person can maintain immunity against SARS-CoV-2 over time.  

“We found a signature present in the immune response one month after infection that can predict whether an individual will exhibit strong immune memory against SARS-CoV-2 six months later. This ‘immune signature’ comprises aspects of both T cell and antibodies responses, indicating that both immune components are important for strong and durable protection against SARS-CoV-2.”


Professor Eleanor Barnes, PITCH study lead from University of Oxford said:

“Gaining a detailed understanding of the immune response after COVID-19 infection is critical in allowing us to reduce reinfection, tackle variants and to helping us to design future vaccination strategies. 


Professor Susanna Dunachie, PITCH study lead from University of Oxford said:

"The PITCH study has allowed us to work collaboratively at scale and speed to better understand how our bodies respond to COVID-19 infection and the immune responses they develop. While the setting up a study of this type during a pandemic is challenging, the clear data found by PITCH shows how informative this approach can be.” 


Our grateful thanks go to all the healthcare workers who volunteered to take part in this study.


This press release reports on findings in the following preprint: Tomic et al. 2021. Divergent trajectories of antiviral memory after SARS-CoV-2 infection. Research Square, doi: 10.21203/


Sign up now to receive latest updates.

Sign up