Researchers at 23andMe described what is believed to be the largest genomic study of rare diseases ever done in a groundbreaking new paper. The study, published in the pre-print server MedRxiv, offers up a potential alternative method for gaining insight into these challenging-to-study conditions.
Using data from more than 1.6 million research participants, including 19,000 who reported a rare disease diagnosis, 23andMe scientists were able to study 33 different rare diseases that included people of European, African, Latino, East Asian, South Asian, Middle Eastern, and other ancestries.
The researchers were then able to run genome-wide association studies (GWAS) for each of these different rare diseases. They replicated 29 previously identified associations across different rare diseases. For three of these rare conditions — Duane retraction syndrome, vestibular schwannoma, and spontaneous pneumothorax — the 23andMe study is the first GWAS ever done. The study identified new genetic associations for each.
The conventional method for studying rare diseases entails sequencing patients with the condition or their family members. Then to use that data to focus on potentially causal genes, but the challenge is to first simply finding enough patients for a study and then paying for the much higher cost of sequencing them.
A New Way
23andMe’s model is different in that the scale of the data allows researchers to quickly find sufficient numbers of cases, and those individuals, who have consented to participate in research, are already genotyped.
“This is exciting because it offers a viable alternative for studying rare diseases,” said Suyash Shringarpure, Ph.D., a 23andMe senior scientist, and statistical geneticist, who worked on the study. “This approach will enable us to discover new genetic associations across multiple populations much faster than could be done otherwise. And it’s complementary to a sequencing-based approach.”
It can often feel like being lost in the wilderness for patients living with a rare disease and their families. With so little information about the conditions, simply receiving a diagnosis can often take years. And once diagnosed, treatment options are often limited or even non-existent.
For the scientists trying to help, it can often feel like being asked to lead those patients out of the wilderness without a map.
Studying IPF and Systemic Sclerosis
The challenge of studying rare diseases is that they are rare. Simply finding enough people with a specific rare disease and then gathering enough data from those individuals to make meaningful insights is extremely difficult. In addition, there is the challenge of finding the money to pay for these studies. Rare diseases are defined as diseases that affect fewer than 200,000 people in the U.S. But there are more than 7,000 rare diseases that have been identified. Collectively they impact an estimated 25-30 million people in the United States alone.
23andMe is currently recruiting to study two rare diseases systemic sclerosis, a form of scleroderma, and idiopathic pulmonary fibrosis (IPF), a chronic progressive lung disease. There are no cures for either condition, but the hope is that by gathering enough participants our scientists will be able to gather insights that could lead to new and effective treatments.
Beyond working specifically on these two conditions, 23andMe’s unique research model allows scientists to more easily and quickly assemble data to study. This most recent paper illustrated how even looking broadly across all research participants, scientists were able to find enough individuals with certain rare conditions to garner insights. In some cases, these insights surprised researchers.
For example, many rare diseases are Mendelian diseases, meaning that they are usually inherited and are typically caused by a mutation in a single gene. From this study, the researchers found that common genetic variants may also play a role in some of these conditions. Given the variety in the genetic architecture of rare diseases, it surprised the scientists working on this study that they found associations in rare causal genes by conducting genome-wide associations studies on 23andMe’s data. Typically genome-wide association studies are used to study common diseases — those that are affected not just by many different genes often with small impacts. These conditions are also influenced by environmental factors as well as gene-by-gene or gene-by-environment interactions. So it surprised researchers that they were able to look at a dataset of unrelated individuals — albeit a very large one — and find genetic associations. But they were able to do that for conditions such as Huntington’s disease and phenylketonuria, for example.
“The ability of a genome-wide association study to find associations at rare causal genes for monogenic diseases in an outbred population is to some extent unexpected,” Suyash said.