Broad Institute scientists deploy RNA sequencing to nail down disease–causing gene mutations in patients for whom genetic analysis failed to return a diagnosis. The study demonstrates the power of RNA sequencing to augment standard diagnostic tools in the clinic.

Recent advances in large–scale clinical DNA sequencing have led to genetic diagnoses for many rare disease patients, but the diagnosis rate based on these approaches is still far from perfect. On average, clinicians are unable to provide a genetic diagnosis for over half of patients in the clinic. The lack of a clear genetic diagnosis can lead to profound uncertainty about patients’ long–term prognoses, treatment options, and family planning decisions.

In a new Science Translational Medicine journal study, a team led by researchers from the Broad Institute of MIT and Harvard and the National Institute of Neurological Disorders and Stroke adds RNA sequencing to the diagnostic toolkit to identify disease–causing mutations buried inside the genome.

The researchers sequenced the RNA from muscle samples of 50 patients with undiagnosed genetic muscle disorders – who had undergone extensive genetic testing – and, in conjunction with DNA sequence information and a reference database, successfully located pathogenic mutations that had previously gone undetected in one–third of the patients. The study firmly positions RNA sequencing as a tool that adds additional power to the existing set of technologies deployed to solve genetic disease mysteries.

“For some patients, we know that there is variation in the human genome, with an effect on the transcript, that we just haven’t been capturing with our traditional genetic sequencing methods,” says senior author Daniel MacArthur, co–director of the Medical and Population Genetics Program at the Broad Institute and group leader at Massachusetts General Hospital. “With RNA sequencing, we were able to take a set of patients who had gone through diagnostic odysseys — often lasting many years, where many methods had been used to try to detect the cause of their disease without success — and find the biological answers that previous technologies had missed.”

The study demonstrates that RNA sequencing, or RNA–seq, applied to relevant tissue samples and coupled with genetic analysis, can detect pathogenic mutations hidden in the noncoding sections of a gene, highlight relevant mutations missed in the noise of whole–genome analysis, and rule out other genetic variants suspected to cause disease. Previously, the technology was rarely applied in a clinical setting, and then only for single patients when specific mutations were already suspected — but the research team saw the potential for RNA–seq to augment other clinical tools earlier in diagnostics.

The researchers first tested and refined their approach on a group of 13 patients with genetic muscular disorders that had been previously diagnosed, and then applied it to a cohort of 50 patients for whom traditional genetic testing had failed to provide a diagnosis. The team sequenced RNA from samples of muscle tissue from the patients, looked for aberrations in the transcripts when compared with a reference dataset from the Broad’s Genotype Tissue Expression Consortium project, and then verified the culpable mutations in the patients’ DNA to successfully diagnose 17 participants.

In one example, four unrelated patients were diagnosed with a new genetic subtype of severe collagen VI–related dystrophy, caused by a mutation in a gene called COL6A1 that was detected through RNA–seq. Then, the research team went back to their remaining undiagnosed patients and reached out to external collaborators who had undiagnosed patients with similar symptoms, and successfully diagnosed another 27 patients with exactly the same mutation.