DISCUSSION:
We recruited a cohort of SARS CoV-2 naturally-infected children under 16
years across the infection spectrum (asymptomatic to severe disease)
spanning the wild-type (Wu Hu-1/D614G) and Delta (B.1.617.2) infection
waves in Vellore and examined the relationship of viral load,
anti-nucleoprotein (anti-N) and spike (anti-S) antibodies in them. The
sampling design allowed a comparison of dynamics of natural infection
between variants (wild-type and delta). We showed that variants (delta)
with increased replicative fitness demonstrated increased viral loads in
mild-to-moderate disease and contributed to a robust antibody response
with a skew towards anti-S. An impaired anti-S immune response was seen
in severe disease.
Our study showed evidence of viral load driving antibody response. For
both anti-N and anti-S, higher viral load (Q1), resulted in higher
antibody levels compared to lower (Q3) loads. In the natural history of
SARS CoV-2 infection, virus levels reduce over time due to development
of virus-specific immunity14 that enhances virus
clearance. Lower Ct values are associated with increased
peak of IgG levels. However, children mount a lower antibody responses
compared to adults,15,19 possibly due to continual
innate immune stimulation and lower viral replication from pre-existing
coronavirus immunity.20 Higher viral loads (e.g. Q1),
may result in inefficient innate immune clearance , thus stimulating a
stronger adaptive response, manifesting as higher antibody levels. Low
viral levels (Q3), on the other hand, could result in efficient virus
clearance by the innate immunity resulting in lower activation of
adaptive responses. We show that delta exhibiting increased replication
efficiency as a driver of immune response.2,21,22
We observed a difference in anti-S levels between wild-type and delta
among quartile 3. This indicates differing dynamics between
variants.22.The viral
loads [mean (SD) Ct] in respiratory samples (Table
1) was 33.6 (3.4) and 26.3(6.7) for wild-type and delta, respectively,
representing a ~100-fold difference (Ctdifference of 3.3 ~log10(1) viral load
difference). Sampling earlier in the illness course leads to lower
observed Ct values.23 We excluded
sampling date as a cause of low Ct as children in our cohort were
sampled median 2 days post-symptom-onset for both wild and delta
infections. While Ct values are reported to be similar
between asymptomatic and symptomatic infections in
children,9 we found viral loads (Ct value) in delta
infections (Fig 3) to be ~50X (Ctdifference 4.8) higher in mild-to-moderate disease compared to
asymptomatic infections. We found virus loads for delta were
~1000X higher (Ct difference 9.5) than
wild-type in mild-to-moderate disease similar to previous
reports.24 This asynchrony in Ct seen
in delta is noteworthy as clinical severity for wild and delta variants
in children were similar, as has been shown
previously.25 When Ct value is used as
a metric of replication kinetics,24,26 a higher load
observed in delta compared to wild-type suggests improved replication
fitness in the former.
In adults with severe disease, antibodies (including nAbs) produced are
abundant and have increased reactivity to all viral proteins (breadth)
compared to mild disease.22 In children however,
antibody levels are similar between mild and severe
disease27 and importantly skewed towards anti-spike
antibodies.15 Higher anti-S antibody compared to
anti-N, a characteristic seen among children15 is
recapitulated in our cohort. Varying half-life dynamics between anti-S
IgG (145 days) versus anti-N IgG (86 days)28 could
partly explain differential levels observed in both wild-type and delta
sera. The absence of skewing seen among severe disease could be due to a
lowered anti-S or elevated anti-N levels. A higher anti-N levels are
associated with poorer outcome.29 This was also seen
in MIS-C compared to minimal or severe disease.30Elevated anti-N levels have been seen in antibody-dependent enhancement
(ADE), which causes robust inflammatory response and increased
viremia29 resulting in severe disease . Anti-S is
considered protective13 and poor antibody response is
associated with severe COVID-19 in children.31 The
lowered levels suggest an impairment of response towards the spike,
indicating a dysregulated immune response, as is also seen in
MIS-C.6 This is in contrast to adults who mount robust
responses after severe disease.14,15,32
Through the pandemic, majority of the under-12 children globally have
acquired SARS CoV-2 immunity by natural infection. Understanding the
dynamics of virus replication, clinical disease and antibody response in
natural infection in children has implications for spread of future
variants and potentially informing targeted vaccination when widely
approved for use in children (under 12 years of age). Infection by
future variants are likely to result in a robust immune response but
higher replication fitness may lead to increased transmissibility. Our
finding of an impaired anti-spike response among children with severe
disease is a cause for concern. Hybrid immunity resulting from
vaccination would likely boost protection, especially against these
future variants.33 Omicron (including subvariants
BA.4/5) which are more immune-evasive, have increased ACE2-binding
affinity and replication fitness than Delta3. Repeated
exposure to spike protein will provide an improved breadth of protection
against newer emerging variants.34
We note a few limitations of the study. Inclusion of adults would have
allowed direct comparison of immune dynamics with that in children.
Immune response to SARS-CoV-2 initiates 7-10 days after symptom onset
and reaches a maximum 2-4 weeks after illness. No sampling was conducted
in the early stages of infection to assess early immune dynamics.
Neutralizing antibodies and cellular immunity were not assessed due to
cost constraints.
CONCLUSION: We compared the dynamics of antibody response at
4-6 weeks after PCR confirmation across SARS CoV-2 variants (wild-type,
delta), virus load and disease spectrum. Our study reveals the differing
dynamics of SARS CoV-2 infection in children with respiratory virus
loads driving both anti-N and anti-S antibody levels and skewed,
protein-specific antibody response across severity strata, with impaired
anti-spike immunity in severe disease .
ACKNOWLEDGEMENTS: AJP acknowledges CMC Research office for
intramural research funds. MM acknowledges Department of Biotechnology,
India, for financial support through grant, BT/PR40390/COT/142/1/2020.
CONFLICTS OF INTEREST: The authors declare no conflict of
interest.
DATA AVAILABILITY STATEMENT: Original data can be accessed upon
request.
ETHICAL APPROVAL: The Institutional Review Board of CMC,
Vellore (IRB No. 13700)