Bottle of Insulin injection with a syringe on black table and stainless steel background.

People with type 1 diabetes can't produce the hormone insulin.Credit: Bernard Chantal/Alamy

In mid-April, Finn Gnadt, an 18-year-old student from Kiel, Germany, learnt that he had been infected with the SARS-CoV-2 coronavirus despite feeling well. Gnadt’s parents had fallen ill after a river cruise in Austria, so his family was tested for virus antibodies, which are produced in response to infection.

Gnadt thought he had endured the infection unscathed, but days later, he started to feel worn out and exceedingly thirsty. In early May, he was diagnosed with type 1 diabetes, and his physician, Tim Hollstein at the University Hospital Schleswig-Holstein in Kiel, suggested that the sudden onset might be linked to the viral infection.

In most people with type 1 diabetes, the body’s immune cells start destroying β-cells — which are responsible for producing the hormone insulin — in the pancreas, often suddenly. In Gnadt’s case, Hollstein suspected that the virus had destroyed his β-cells, because his blood didn’t contain the types of immune cells that typically damage the pancreatic islets where the β-cells live.

Diabetes is already known to be a key risk factor for developing severe COVID-191 and people with the condition are more likely to die2. “Diabetes is dynamite if you get COVID-19,” says Paul Zimmet, who studies the metabolic disease at Monash University in Melbourne, Australia.

Now Zimmet is among a growing number of researchers who think that diabetes doesn’t just make people more vulnerable to the coronavirus, but that the virus might also trigger diabetes in some3. “Diabetes itself is a pandemic just like the COVID-19 pandemic. The two pandemics could be clashing,” he says.

Growing evidence

Their hunch is based on a handful of people such as Gnadt, who have spontaneously developed diabetes4 after being infected with SARS-CoV-2, and on evidence from dozens more people with COVID-19 who have arrived in hospital with extremely high levels of blood sugar and ketones5, which are produced from fatty deposits in the liver. When the body doesn’t make enough insulin to break down sugar, it uses ketones as an alternative source of fuel. “In science, sometimes you have to start off with very small evidence to chase a hypothesis,” says Zimmet.

Researchers cite other evidence, too. Various viruses, including the one that causes severe acute respiratory syndrome (SARS), have been linked with autoimmune conditions such as type 1 diabetes6. And many organs involved in controlling blood sugar are rich in a protein called ACE2, which SARS-CoV-2 uses to infect cells7.

The latest clue comes from an experimental study in miniature lab-grown pancreases published last week8 suggests that the virus might trigger diabetes by damaging the cells that control blood sugar.

But other researchers are cautious about such suggestions. “We need to keep an eye on diabetes rates in those with prior COVID-19, and determine if rates go up over and above expected levels,” says Naveed Sattar, a metabolic-disease researcher at the University of Glasgow, UK.

To establish a link, researchers need more robust evidence, says Abd Tahrani, a clinician–scientist at the University of Birmingham, UK. “Well-constructed epidemiological cohort studies and mechanistic and experimental studies are needed,” he says.

Diabetes database

One initiative is now under way. Earlier this month, an international group of scientists, including Zimmet, established a global database3 to collect information on people with COVID-19 and high blood-sugar levels who do not have a history of diabetes or problems controlling their blood sugar.

Cases are beginning to trickle in, says Stefan Bornstein, a physician at the Technical University of Dresden, Germany, who also helped to establish the registry. The researchers hope to use the cases to understand whether SARS-CoV-2 can induce type 1 diabetes or a new form of the disease. And they want to investigate whether the sudden-onset diabetes becomes permanent in people who’ve had COVID-19. They also want to know whether the virus can tip people who were already on their way to developing type 2 diabetes into a diabetic state.

The study in pancreatic organoids shows how SARS-CoV-2 could be damaging the organ8. Shuibing Chen, a stem-cell biologist at Weill Cornell Medicine in New York City, and her colleagues showed that the virus can infect the organoid’s α- and β-cells, some of which then die. Whereas β-cells produce insulin to decrease blood-sugar levels, α-cells produce the hormone glucagon, which increases blood sugar. The virus can also induce the production of proteins known as chemokines and cytokines, which can trigger an immune response that might also kill the cells, according to the study8 published in Cell Stem Cell on 19 June.

Chen says the experiments suggest that the virus can disrupt the function of key cells involved in diabetes — either by directly killing them or by triggering an immune response that attacks them.

The virus also attacked pancreatic organoids that had been transplanted into mice, and cells in liver organoids. The liver is important for storing and releasing sugar into the blood stream when it senses insulin.

The organoid study8 adds strength to the argument that SARS-CoV-2 might cause or worsen diabetes, but the paper itself is not enough to prove the link, says Tahrani.

There could be more going on than some scientists suggest, says Shane Grey, an immunologist at the Garvan Institute of Medical Research in Sydney, Australia. The virus could trigger an extreme inflammatory state, which would impair the ability of the pancreas to sense glucose and release insulin, and dampen the ability of the liver and muscles to detect the hormone, he says. This could trigger diabetes.

Fatigue and muscle loss caused by severe infection can also push people at risk of the condition into a pre-diabetic state, says Sattar. Only long-term studies will reveal what’s really going on, he says.