Children produce weaker coronavirus antibodies, study finds

Children infected with the coronavirus produce weaker antibodies and fewer types of them than adults do, suggesting they clear their infection much faster, according to a new study published Thursday.

Other studies have suggested that an overly strong immune response may be to blame in people who get severely ill or die from COVID-19. A weaker immune response in children may paradoxically indicate that they vanquish the virus before it has had a chance to wreak havoc in the body, and may help explain why children are mostly spared severe symptoms of COVID, the disease caused by the coronavirus. It may also show why they are less likely to spread the virus to others.

“They may be infectious for a shorter time,” said Donna Farber, an immunologist at Columbia University in New York who led the study reported in the journal Nature Immunology.

Having weaker and fewer antibodies does not mean that children would be more at risk of re-infections, other experts said.

“You don’t really need a huge, overly robust immune response to maintain protections over some period of time,” said Deepta Bhattacharya, an immunologist at the University of Arizona in Tucson. “I don’t know that I would be especially worried that kids have a little bit lower antibody response.”

The study looked at children’s antibody levels at a single point in time, and was too small to provide insights into how the levels may vary with age. But it could pose questions for certain antibody tests that may be missing children who have been infected.

Farber and her colleagues analyzed antibodies to the coronavirus in four groups of patients: 19 adult convalescent plasma donors who had recovered from COVID without being hospitalized; 13 adults hospitalized with acute respiratory distress syndrome resulting from severe COVID; 16 children hospitalized with multi-system inflammatory syndrome, the rare condition affecting some infected children; and 31 infected children who did not have the syndrome. About half of this last group of children had no symptoms at all.

Individuals in each group had antibodies, consistent with other studies showing that the vast majority of people infected with the coronavirus mount a robust immune response.

“This further emphasizes that this viral infection in itself, and the immune response to this virus, is not that different from what we would expect” from any virus, said Petter Brodin, an immunologist at Karolinska Institutet in Stockholm.

But the range of antibodies differed between children and adults. The children made primarily one type of antibody, called IgG, that recognizes the spike protein on the surface of the virus. Adults, by contrast, made several types of antibodies to the spike and other viral proteins, and these antibodies were more powerful at neutralizing the virus.

Children had “less of a protective response, but they also had less of a breadth of an antibody response,” Farber said. “It’s because those kids are just not getting infected as severely.”

Neither group of children had antibodies to a viral protein called the nucleocapsid, or N, that is entangled with the genetic material of the virus. Because this protein is found within the virus and not on its surface, the immune system would only see it and make antibodies to it if the virus were widely disseminated in the body, she said.

“You don’t really see any of that in the children, and that suggests that there’s really a reduced infection course if these kids are getting infected,” she explained.

The finding could undermine the results from tests designed to pick up antibodies to the N protein of the virus. Many antibody tests, including those made by Abbott and Roche and offered by Quest Diagnostics and LabCorp, are specific to the N antibodies and so may miss children who have successfully cleared the virus. “That’s absolutely an interesting implication of that finding,” Brodin said.

Lower levels of virus in the body would also explain why children seem generally to transmit the virus less efficiently than adults do.

But experts urged some caution in interpreting the results because they represent samples taken from people at a single point in time.

Samples from the more severely affected children and adults were collected within 24 to 36 hours of being admitted or intubated for respiratory failure; those from children with mild or no symptoms were banked after medical procedures.

The type of antibodies produced by the body varies over the time course of an infection. This was a limitation of this study because the researchers may have been comparing people at different points in their infection, Brodin said. “You risk comparing apples and oranges.”

Other experts cautioned that the study was too small to draw conclusions about how the immune response may vary in children of different ages. The children in the study ranged in age from 3 to 18 years, with a median age of 11. But some studies have suggested that teenagers may be just as much at risk from the coronavirus as adults.

“It’s very important to understand what happens in children,” to understand the nature of their illness, but also how they contribute to spread of the virus in the community, said Dr. Maria L. Gennaro, an immunologist at Rutgers University. But “to try and stratify by age, it’s a little bit of a stretch in the analysis,” she said.

The researchers were also not able to explain why children have a more limited antibody response.

Having fewer types of antibodies may seem like a bad thing, but “having a ton of antibody isn’t necessarily a marker of a good thing,” said Bhattacharya. “It usually means that something went wrong early in the response.”

At least one other study has suggested that children have a powerful inborn immune system, intended to combat the many new pathogens they encounter, and that this first line of defense may clear the infection early without needing to rely on later antibodies.

Another possibility is that the children have some protection — in the form of immune cells called memory T cells — from previous encounters with common cold coronaviruses.

“Is it all innate? Or could there actually be some preexisting memory?” Bhattacharya said. “I think those are both possible.”