The results of Stephen Friend’s ambitious study was finally published online in Nature Biotechnology on 11 Apr. From sequence and genotype results of 589,306 healthy individuals gathered from various sources such as 23andMe, 1000 Genomes, the Children’s Hopsital of Philadelphia and more, researchers looked at mutations in 874 genes believed to cause 584 distinct severe Mendelian childhood disorders. The diseases were chosen based on three characteristics – high severity, early age of onset and complete penetrance (i.e. if the mutation is present, you definitely get the disease). This approach, performed on self-reported healthy individuals > 18 years old, allowed one to find the “Genetic Superheroes” or people that have overcome a mutation known to confer a very severe disease at a young age.
They found 15, 597 candidates initially, which was filtered down to 303 due to low confidence in sequencing reads, unexpectedly high allele frequencies, or the inability to access individual genomic data. Further filtering was performed by a panel of experts and excluded persons uncertain to be completely resilient to the disease, persons having heterozygous mutations (where one healthy allele prevents disease manifestation), and persons having mutations with insufficient evidence linked to disease. Finally, a group of 13 individuals were identified to carry heterozygous dominant or homozgous recessive mutations linked to 8 different severe Mendelian childhood disorders that would normally manifest before 18 years of age: cystic fibrosis, Smith-Lemli-Opitz syndrome, familial dysautonomia, epidermolysis bullosa simplex, Pfeiffer syndrome, autoimmune polyendocrinopathy syndrome, acampomelic campomelic dysplasia and atelosteogenesis.
Of these 13, there was enough DNA material to Sanger-sequence only 5 to reconfirm the findings. And most unfortunately, all 13 could not be recontacted due to lack of a recontact clause in the original consent forms. Something which will definitely be addressed in the future.
The inability to recontact any of the 13 was the main caveat of the study. Not only because of the missed opportunity to confirm that they were really “superheroes” absent of disease, but also because there was no chance of furthering the investigation to find out which genes conferred their resistance to disease. Furthermore, the individuals identified may also be genetic mosaics – where though the mutation may be in the cells collected (usually derived from cheek swabs), the mutation may not be expressed in the cells associated with the disease e.g. in lung cells involved in cystic fibrosis.
Because most of the data were genotypes and not whole genome sequences, it was difficult to identify “resilience genes” that conferred the protection from disease manifestation with the existing data . Carrying out such studies are also hindered by the small number of these identified “genetic superheroes”. The authors argue however that N of 1 studies (i.e. sample size of 1) can yield relevant information if the disease was highly penetrant and whole genome sequences were available.
Definitely an interesting concept with the potential to change the way drugs are made. Identifying “resilience genes” for example is a great way to find a gene candidate worth modulating with small molecules/biologics. It also allows researchers to understand the disease and narrow in on which pathways are of particular importance.
Already, there are ongoing efforts to utilize this approach to speed up drug development. NIH Director, Francis Collins, launched a $230 million Accelerating Medicines Partnership (AMP) among 10 biopharmaceutical companies and non-profit organisations focusing on four diseases – Alzheimer’s disease, type 2 diabetes, rheumatoid arthritis and lupus. The large databases containing information linking genetic and clinical information will be shared with the biomedical research community in an effort to speed up the identification of relevant gene targets or biomarkers.
With the dropping cost of genetic sequencing, these databases look set to grow. Proper database management and the ability to filter through all that information are current challenges. But at least it is a step in the right direction to revitalize the search for novel drug targets.