For 18 years, Wisconsin man Tim Friede has injected increasing amounts of venom into his body from the world's most deadly snakes.
His strange quest yielded results on Saturday, with a team of US researchers claiming to have created "the most broadly effective" snake antivenom to date.
After hundreds of injections and snake bites, Mr Friede's blood contained broad-spectrum immune proteins called antibodies, which were analysed and used to create an antibody "cocktail" that could potentially protect against 19 different species of snake.
The team's results have been published on Saturday in the journal Cell.
According to Jacob Glanville, one of the researchers working on the project, and CEO of start-up Centivax, Mr Friede's envenomed blood was the key to producing the snake antivenom.
"The first antibody [discovered] was the proof of principle that broadly neutralising antibodies existed in Tim Friede."
The new antivenom described in the study is very different to traditional antivenoms, according to Peter Kwong, a biochemist at Columbia University and one of the study's authors.
"The current technology hasn't really changed in over 100 years," he said.
"Whereas with [this new] approach … you can use all the tools of modern antibody therapy."
But while the team is extremely excited about their results, which has so far only been demonstrated to be effective in mice, other experts caution against too much hype.
How antivenom is made
If you were unlucky enough to get bitten by a snake tomorrow, the antivenom provided would be made in a similar way to what would have been provided decades ago.
Both then and now, antivenom is created using animal blood, like horses or other large mammals.
The animals are given small doses of the venom, and then over time given larger and larger doses. This helps the animal produce an increasing immune response without getting so sick they die.
"You're poisoning the animals over an extended period," Timothy Jackson, who is the co-head of the Australian Venom Research Unit at the University of Melbourne, said.
"They're probably feeling pretty crappy for a lot of the time because their immune system is fighting an active pathogen."
The immune cells — or antibodies — in the horse plasma can be stored at local hospitals and health centres ready for use when required.
Usually the antivenom will either be for a particular snake, or for a few snakes located in one region — called a polyvalent antivenom.
This technique works well in places like Australia where there are effective ways to produce and distribute the antivenom, but less so in India and other countries in South-East Asia.
"I really don't like the claim that Australia has the world's most dangerous snakes," Dr Jackson said.
"If [Australia's] snakes kill two people every year, and snakes in India kill 40,000 to 60,000 people every year, do we have the world's most dangerous snakes? Clearly not."
The World Health Organization (WHO) listed snakebite envenoming as a neglected tropical disease in June 2017 but while this resulted in a funding increase, millions of people are still being bitten by snakes in developing countries, many of those dying or becoming disabled through amputation.
Creating a broad-spectrum antivenom
The team found three antibodies, two of which were in Mr Friede's blood, that when used together produced a type of broad-spectrum antivenom.
This new antivenom was then tested in mice studies.
The mice were protected from the venom of most deadly snakes across one family called the elapids, including Australia's eastern brown snakes, inland taipans and tiger snakes.
"It got really exciting when we started seeing mice live," Professor Kwong said.
Out of the 19 species of venomous snakes tested, the researchers found the antivenom provided full protection in the mice against 13 snake species and partial protection for the remaining six.
Importantly, the process of creating the antivenom is also different. Mr Friede won't need to do regular donations to produce it.
Modern antibody therapy, like monoclonal antibody therapy used in cancer treatments, uses cells grown in labs to produce the product instead of antibodies regularly drawn from horses or other mammals.
The team hope they can produce a universal antivenom that protects against more species, no longer uses horses or other mammals to be created, and has a longer storage life than traditional antivenoms.
A lifelong project
The American team isn't the only one using this antibody technique.
Other broad-spectrum antibodies for snake venom, including one called 95Mat5, have already been created by other teams of researchers.
According to Dr Glanville, Mr Friede's antibodies have been particularly honed over the years by rotating hundreds of times between 16 species of venomous snakes.
"This was his lifelong project," Dr Glanville said.
Mr Friede began injecting himself with venom back in the 2000s to try and create an immunity to his venomous pet snakes, but because antibodies can deplete after just a few weeks or months, he continued to regularly inject himself to ensure continued immunity.
Dr Glanville believes that particularly in the body, antibodies can evolve to optimise their performance.
"There's this emerging appreciation that sometimes living systems … can give rise to superior therapeutic properties."
But Dr Jackson, who was not involved with the paper, said antivenoms could be created without using donors like Mr Friede.
"It's fun to put him in the loop, but it's not the only way of generating highly specific antibodies," he said.
It's also extremely dangerous, and Dr Glanville stresses that this is not something to be repeated.
"No-one else should do what Tim has done.
"He … produced something remarkable, but that means that there's no reason why anyone else needs to go try."
Just a first step towards helping humans
There's still a long way to go before the team can produce a universal antivenom for human use.
The first issue is that the new antivenom doesn't work on vipers — a large group of snakes found on most continents except Australia and Antarctica — so more toxins will need to be found to work on their particular toxins.
Vipers include include the saw-scaled vipers, which are responsible for the most snakebite cases and deaths in the world.
"Our goal is to produce a broad-spectrum cocktail for the vipers as well," Dr Glanville said.
"Then … we would have complete global protection from snake venom."
The team also need to test the current products on larger animals, and then humans.
Because snake venoms have so many types of toxins, there's potential that some others may have been missed by just the three antibodies used.
"[The study] looks good on paper," Dr Jackson said.
"But is it actually neutralising all of the clinically relevant toxins?"
Larger, animal studies will need to be done, and the team is now looking at testing the cocktail on dogs bitten by snakes at an unnamed veterinary clinic in Australia.
Finally, potentially the biggest hurdle is ensuring that the drugs are cheap and available in locations around the world where they might be needed.
"Having a good drug is the easy part when it comes to dealing with a problem like snakebite," Dr Jackson said.
"The WHO has referred to snakebite, aptly, as a multi factorial crisis — there's a lot of factors here."
"A drug is only as good as your capacity to get it into a patient."
