Short summary

Several viral infections have been suspected to be part of the Type-1 diabetes pathogenesis, with gradual loss of insulin production in the pancreas as a result. SARS-Cov-2 is the virus associated with the current Covid-19 pandemic. We will now investigate if previous infection with SARS-Cov-2 increases the risk of developing Type 1 diabetes, and if diabetes progresses more rapidly in these patients. The study is multi-center, ocurring in both Denmark and Portugal.

Rationale

Although recognized as an autoimmune disease, the etiology of type 1 diabetes mellitus (T1D) remains undetermined. One dominant hypothesis is that a recent virus infection elicits or enhances an autoimmune reaction, resulting in CD4+ and CD8+ T-cells recognizing pancreatic antigens and subsequently targeting the beta-cells of the pancreas. This results in a gradual but irreversible loss of beta-cell function. The hypothesis is most strongly supported by animal studies, where enterovirus infection in susceptible mice results in an exaggerated risk of T1D, and even more convincing that a vaccine against Coxsackie B virus protects against virus-induced diabetes in a mouse model of type 1 diabetes. In humans, this has been difficult to prove since immunity to earlier virus infections remains present and a time-correlation accordingly hard to substantiate. A seasonal variation of newly diagnosed T1D peaking from October to January with a through in June to August on the Northern hemisphere and a reverse pattern on the Southern hemisphere does support but does not prove a possible relation to virus infection. Interestingly, the seasonality in T1D cases does not seem to correlate with average temperature measurements or sunshine hours, but other environmental factors may be at play. As such, it must be considered if any viral link to T1D development is based on pancreatic damage itself, or whether infection could simply precipitate an already ongoing autoimmune process.

Several mechanisms may be involved in virus-induced T1D. Virus infection is a strong stimulator of the immune system, which may cause infiltration of the pancreas by natural killer cells. A virus infection may act directly on the beta-cells via surface-receptors, and even a limited cytolysis might unmask the beta-cells for CD-8+ T-cell recognition and further immune attacks. A more massive infection attacking the beta-cells might induce a manifest cell-lysis leading to beta-cells dysfunction and eventually result in insulinopenic diabetes. Various enterovirus species have for years been suspected as causative factors of T1D. However, many different viruses may carry the potential to induce T1D via one of the above-mentioned mechanisms, including respiratory infections.

SARS and SARS-Cov-2 (the latter being the virus that causes Covid-19 infection) are both corona viruses, which enter susceptible cells by binding to the ACE-2 receptors. The ACE-2 receptors are widely distributed throughout many tissues in the body. Abundant immunostaining for the ACE-2 receptor has been found also in the endocrine part of the pancreas, which is accordingly a potential target for the virus.

Most people are presently immunologically inert to SARS-Cov-2, but a large proportion of the world's population is expected to be infected over the next few years, with the global focus having now shifted to slowing the spread. The pandemic of COVID-19 therefore provides an opportunity to study a possible correlation between a novel type of virus infection and the development of T1D. If SARS-Cov-2 comprises a possible trigger (or precipitator) for the development of T1D, we would expect:

1. That SARS-Cov-2 antibodies are more frequently found in newly diagnosed patients with T1D than in the background population.

2. That the incidence of T1D increases following the pandemic of Covid-19.

3. That country differences in the incidence of T1D correlates to the country incidence of SARS-Cov-2 antibodies.

In addition, if the pancreatic beta-cell destruction is abundant we would expect that

4. Patients with newly diagnosed T1D and positive for SARS-Cov-2 antibodies have an even more reduced beta-cell function, compared to newly diagnosed T1D SARS-Cov2-antibody-negative patients.

Description of the cohort

The following study populations are established:

1. Study population 1: Newly diagnosed adult patients with T1D according to usual practice, stratified for the presence of SARS-Cov-2-antibodies (+/-), in both Denmark and Portugal.

2. Study population 2: A control population (currently undetermined) of the same age (+/- 3 years) and sex from Denmark and Portugal. This will provide the SARS-Cov-2 status of the background population. Each newly diagnosed T1D patient is matched with five control persons.

Data and biological material

In both Denmark and Portugal, all subjects from study population 1 (T1D patients) will undergo the following assessments at diagnosis of T1D:

1. SARS-Cov-2 antibody measurement.

2. GAD-antibodies (and if necessary islet cell antibodies (IA-2)).

3. HbA1c, fasting blood glucose and fasting C-peptide.

4. Standard diabetes biochemical analyses, including lipoproteins and triglycerides, kidney function, electrolytes (Na,K), urine-protein.

5. Questionnaire regarding symptoms prior to diabetes diagnosis, including any respiratory symptoms indicative of Covid-19, possible exposure to SARS-Cov-2, other medical conditions and occupation.

6. Mixed-meal tolerance test (MMTT). The test is done in the morning after an overnight fast, consisting of a standardized breakfast (Ensure Plus®) ingested over 10 minutes. Blood samples for C-peptide and glucose analyses are then collected at 30 minutes intervals over the next 150 minutes, providing a reliable measurement of residual endogenous insulin production (C-peptide area under curve (AUC)). If needed, the MMTT can be simplified with only C-peptide measurement at 90 minutes.

7. Follow-up one year after the diagnosis with HbA1c, fasting blood glucose, fasting C-peptide and MMTT.

The MMTT will not be performed in T1D patients having received bariatric surgery, due to the risk of severe dumping symptoms. These patients will, however, be included in the study and have the other biochemical analyses such as Hba1c, fasting glucose and C-peptide done at both enrolment and at 1 year.

All samples obtained (both in Denmark and Portugal) will be stored frozen and subsequently transported to the Hospital of Southwest Denmark for the biochemical analyses of C-peptide. This will therefore will be done centrally, and not at the local inclusions sites.

Collaborating researchers and departments

Steno Diabetes Center Odense

• Morten Bjerregaard-Andersen
• Kurt Højlund

Steno Diabetes Center Copenhagen

Steno Diabetes Center Aarhus

Steno Diabetes Center Northern Jutland

Steno Diabetes Center Zealand

Research Unit OPEN, Odense University Hospital