This week a UK study has been generating headlines that suggest those who get coronavirus may develop antibodies for only a few months.
But what do we really know about how we generate an immune response to SARS-CoV-2, the virus that causes COVID-19? And is such research a setback for vaccine development? Three experts we spoke to didn't think so.
Researchers from Kings College London announced findings this week of a study that looked at people who had recovered from COVID-19 infection.
The research found levels of antibodies that can fight the virus peaked in the blood three weeks after people got sick, and then declined after as little as two to three months, in some cases to nothing at all.
This is important because getting sick and creating a good level of antibodies to a particular virus means your immune system is better able to fight the virus the next time you're exposed to it.
And this is why most vaccines aim to produce antibodies against the virus they are designed to protect you from.
The UK researchis yet to be peer reviewed or published, but according to Australian immunologist Stuart Tangye of the Garvan Institute of Medical Research, the findings are in line with those from a number of other studies.
"There was a paper in Nature Medicine recently highlighting that the longevity of that antibody response isn't great," said Professor Tangye, who is part of a global consortium of researchers looking at how people's genetics affects their ability to fight off the virus.
He said other findings, from researchers at the coalface in China, also showed great variability in antibody response, and a decline in a large proportion of people after eight weeks.
So what does this mean for vaccines against SARS-CoV-2?
"Ninety to 95 per cent of all successful vaccines that are out there at the moment really rely on antibody production," Professor Tangye said.
But along with other researchers he is still optimistic about future options.
"It's not a roadblock, it's more of a pivot.
"We just need to be clever and use the information in going forward."
What about the rest of the immune system?
There is much we still don't know about our body's immune response to coronavirus, and what role other parts of the immune system might play in developing immunity.
"Antibody levels specific for COVID-19 wane over time but we don't know what the implications of this are," said immunologist Emily Edwards from Monash University.
She pointed to the role of B cells that produce antibodies and 'remember' viruses they encounter, therefore working much quicker next time around.
"Just because the levels of antibodies in the blood have decreased it doesn't mean that the memory B cells aren't there to produce more antibodies if they see the virus again," she said.
Dr Edwards also said there was some evidence that another group of immune cells, called killer T cells, could kill SARS-CoV-2 infected cells, but more research was needed on this.
She said studies of recovered patients were helping in the development of vaccines by giving us a picture of what the immune system looks like after the virus leaves the body.
"It's like an imprint of what type of response we want to generate when looking for a vaccine," Dr Edwards said.
Finding other pieces of the puzzle
Immunologist Jennifer Juno of the Doherty Institute is helping define this imprint, with newly-published results in Nature Medicine this week.
Dr Juno and colleagues also found some people recovering from COVID-19 produce more antibodies than others and so theoretically could have better immunity, at least in the short term.
They are currently following up to see if they find the same drop-off over time.
But significantly, this team found these people had higher levels of a particular group of T cells.
These T cells, which also remember viruses, help B cells to make antibodies.
Dr Juno thinks this group of T cells may be a good marker to use in testing whether vaccines are doing their job.
"I am still quite optimistic about the vaccine landscape," she said.
"Some of the early data from vaccine candidates in animal models shows you can induce antibody levels from a vaccine that are either as high as what we see with natural infection, or even higher."
Will immunity to the common cold help?
Dr Juno said there was a lot of interest in the possibility that existing immunity to common colds, some of which are caused by other coronaviruses, could protect us from the new virus.
But she said there is no evidence for this, and this immunity is often short-lived anyway.
Dr Edwards added that the fact so many people are getting severely sick from COVID-19 also argued against this possibility.
Like Dr Juno, Professor Tangye is optimistic about a vaccine.
He said the fact that a small proportion of people infected with SARS-CoV-2 do actually produce lots of effective antibodies means it may be possible to mix ingredients called adjuvants into a vaccine to better stimulate the immune system and encourage production of more antibodies.
Although, Professor Tangye said, just like we get a regular flu shot we might need to have a yearly coronavirus jab — although it wouldn't be just in winter like the flu.
Whatever the case, Professor Tangye said it was important not to rely on any one thing.
"We need a multipronged approach."
"We need to be hitting every part of the immune system which is going to give you some protection against infection."
And, he added, even if a successful vaccine eludes us, as it has so far with HIV, then it might be possible to manage COVID-19 with drugs.
"While we still don't have a vaccine for HIV it's now a liveable chronic disease and that's been achieved through antiviral drugs," Professor Tangye said.
Apart from high profile candidates such as remdesivir, he points to an approach taken by colleagues at the Garvan Institute that involves making artificial antibodies in the lab that could be injected into people, called convalescent plasma therapy. He says this could be used as a "bridge" to a vaccine.
Another potentially helpful tool Dr Edwards is involved in developing, is a rapid test to tell who has immunity to COVID-19, who remains infectious, and who is at risk of developing a severe form of the disease.