COVID-19 Antibody Responses in Cystic Fibrosis

The symptoms of SARS-CoV-2 infection range from mild upper respiratory symptoms
in the majority of cases to pneumonia and acute respiratory distress syndrome
(ARDS) with overall case fatality rates of 1-5%.1 The diagnosis of SARS-CoV-2
infection is based on clinical symptoms and polymerase chain reaction (PCR)
testing. However, given the asymptomatic nature of infection in a large
proportion of cases, and the limitations of sampling the upper respiratory
tract for PCR analysis, serologic assays may play a significant role in helping
us to understand the epidemiology of SARS-CoV-2 infection. A seroprevalence
study in California from community drawn samples in a non-CF population
demonstrated a 4.06% prevalence of SARS-CoV-2 antibodies, with a
population-weighted prevalence of 4.65%, significantly higher than the reported
prevalence of SARS-CoV-2 infection in this region.2 Antibodies typically become
detectable between 7-14 days following SARS-CoV-2 infection,3,4 but the degree
and duration of immunity is not known. Antibody levels (titres) have been
demonstrated to vary significantly between patients. Although animal models
have suggested that antibody responses to SARS-CoV-2 can protect rhesus
macaques from re-infection,5 it is not known whether this is true in humans.
Re-infections with all seasonal coronaviruses are known to occur, often within
three years, although the period of protective immunity is typically much
shorter, lasting approximately 6-12 months.
Commercial serologic assays are in development, which test both IgG and IgA
antibodies to the spike protein (S1 and S2) of SARS-CoV-2.7 However, additional
research based serologic tests are also needed to detect antibody
isotypes/subclasses (eg. IgG1, 2, 3, 4, IgA and IgM) and the specific region of
the spike protein to which they bind (e.g. receptor binding domain) as these
tests may be more predictive of future protection. Antibodies to the receptor
binding domain (RBD) of SARS-CoV-2 may neutralize the virus more effectively
than antibodies to other parts of the virus, as has previously been shown with
SARS-CoV-1.8,9 The neutralizing capability of selected sera can be determined
using a SARS-CoV-2 pseudo-virus neutralizing assay 9 using 293T cells
engineered to express the SARS-CoV-2 receptor, ACE2 as target cells.
Many countries have focused on enhanced infection prevention recommendations to
shield those populations perceived to be most at risk. Consequently, there has
been limited testing in pwCF to date. Given this relative paucity of data of
SARS-CoV-2 epidemiology in CF, and the potential increased risk due to
underlying structural lung disease, serologic testing in combination with
clinical assessment will be key to understand the impact on this population.
This study will be the first (to our knowledge) to assess the seroprevalence of
SARS-CoV-2 in the CF population on a longitudinal basis and may guide the risk
assessment for pwCF during further peaks of outbreak. Furthermore, this study
will be the first to assess IgG antibody responses and correlates of clinical
disease in the CF population both according to severity and age. We hypothesize
that pwCF will have a lower seroprevalence to SARS-CoV-2 compared to the
general population due to their enhanced infection prevention practices;
seroprevalence will be greater than the reported number of confirmed cases
collected in the European CF Society (ECFS) Patient Registry.

Our overall aim is to assess the seroprevalence of SARS-CoV-2 antibodies in an
international population of pwCF and to examine associations between SARS-CoV-2
infection and clinical risk factors for infection, as well as subsequent
clinical outcomes for pwCF.
We anticipate the seroprevalence of SARS-CoV-2 will be lower in people with CF
than the general population. The benefit in the prospective longitudinal design
of this study is that it provides a measure of baseline seroprevalence and
subsequent change in prevalence as the pandemic progresses. We expect that the
seroprevalence in this study will be higher than the number of cases reported
to the national CF registries given that the registry data does not include all
cases and asymptomatic cases are likely to be missed. This study also presents
a unique opportunity to compare antibody responses in pwCF following natural
infection with those produced following vaccination should a vaccine
subsequently become available. Specifically this study enables a longitudinal
comparison of the development of anti-SARS-CoV-2 antibodies post vaccination
and post natural infection as well as comparison of the progression of
anti-SARS-CoV-2 antibodies over time following these events.
Given the paucity of data surrounding SARS-CoV-2 infection and antibody
responses in pwCF, we believe our international multi-centre study will provide
critical knowledge about the risks to this vulnerable population and
potentially inform clinical care and practice for subsequent SARS-CoV-2 surges
and future pandemics.

CAR-CF is a prospective, longitudinal cohort study in pwCF with repeated serial
sampling of participants. This study design was chosen to provide
comprehensive information on SARS-CoV-2 seroprevalence changes over time and
the subsequent clinical impact on people with Cystic Fibrosis.

This study is intended as a pragmatic study seeking to cause minimal additional
disruption to enrolled patients and accordingly only serum samples will be
taken alongside standard clinical samples without additional samples for
sputum, viral or other specimen collection. However, this study could generate
an increased workload for data and blood sample collection.
As discussed previously, there is increasing evidence that antibody levels
decrease rapidly post infection and therefore more frequent blood sampling may
be required to avoid missing potential seroconversion amongst patients.
However, the need for more frequent blood sampling needs to be balanced against
the potential burden for enrolled patients with regards to blood-work that is
additional to standard annual blood-work collection. This is particularly
relevant for the paediatric CF cohort and could led to difficulty in patient
recruitment and increased subject drop-out. Accordingly, at present we have
included more frequent bloodwork on an opportunistic basis to be undertaken at
any additional visits for routine blood-work outside of annual clinic visits.
We recognise this as a potential significant limitation for the study and will
aim to strongly encourage sites to include these additional samples,
particularly within adult cohorts.