This story was originally published Nov. 3.
NEW YORK (GenomeWeb) – An early adopter of clinical sequencing, the Inova Translational Medicine Institute has now sequenced more than 7,000 genomes for a variety of translational research projects, and is looking to implement whole-genome sequencing as a way to augment newborn screening.
In addition, the company wants to launch targeted sequencing diagnostic panels, as well as an exome and/or a whole-genome sequencing test, according to Benjamin Solomon, chief of the division of medical genomics at Inova.
Speaking late last month in a presentation at the American Society of Human Genetics meeting in San Diego, Solomon said that a main goal of the Inova Translational Medicine Institute is to figure out how to integrate genomic data into the practice of medicine. As such, the research institute, which was launched in 2011 and is part of a six-hospital system that performs around 20,000 deliveries per year, has been conducting trio-based whole-genome sequencing on "everyone that wants to take part."
In a follow-up interview, Solomon told Clinical Sequencing News that by the end of 2015, if Inova keeps enrolling families at its current rate of between 60 to 100 per week, it will have more than 20,000 genomes with comprehensive clinical data attached to them.
Inova is currently working on three major research projects involving whole-genome sequencing: a preterm birth study involving around 1,100 patients, a longitudinal study of 5,000 patients to study how genomic information may be able to predict future health, and a congenital anomalies study.
Earlier this year, Inova said it had partnered with Personalis to provide data analysis and interpretation and said that it planned to launch whole-genome sequencing as a diagnostic test for newborns with unexplained syndromes by the end of the year.
During his presentation, Solomon provided an update of a study evaluating the potential of whole-genome sequencing to augment standard newborn screening. He discussed data from 702 newborns on whom both standard newborn screening and whole-genome sequencing had been done.
For the 702 newborns, 966 newborn screens were performed, and 117 infants, or 17 percent, received more than one abnormal or invalid result. Infants that were born prematurely were eight times more likely to have an abnormal or invalid result.
The standard newborn screening was compared to whole-genome sequencing with a selected analysis of the 127 genes corresponding to the disorders currently being evaluated by blood-based newborn screening.
The standard newborn screen yielded four false positive results for cystic fibrosis, while whole-genome sequencing identified 56 carriers of heterozygous variants in the CFTR gene, but no affected individuals
Whole-genome sequencing was concordant with newborn screening for hemoglobinopathies, with both techniques identifying two babies with sickle cell anemia. However, whole-genome sequencing was also able to identify nine other Hb types that were not detected by newborn screening.
For example, in one case an infant with a normal result on the standard newborn screen suffered from hepatosplenomegaly — liver and spleen enlargement. Whole-genome sequencing identified bi-allelic mutations to the gene MPC1, which causes Niemann-Pick disease, a metabolic disorder that can manifest as hepatosplenomegaly.
While whole-genome sequencing demonstrated it could be as good, if not better, than newborn screening, there are still a number of challenges that will need to be addressed, Solomon said. For instance, the initial analysis of the sequencing results for the 127 genes revealed that 546 infants, or 78 percent, had at least one predicted pathogenic variant before doing manual annotation. In addition, the researchers used both HGMD and ClinVar for initial prediction, and found that "there was not much overlap between the two databases," Solomon said, underscoring the need for a well-annotated and curated database.
Nonetheless, Solomon said the initial study of incorporating sequencing is promising. "What we're trying to demonstrate is how looking at the genomic data would increase the accuracy of [standard newborn screening]," he said. Standard newborn screening was designed to have very few false negatives, but as a consequence, there are a lot of false positives, particularly for babies born prematurely. The high false positive rate "makes it easier for things to fall through the cracks."
Inova would like to determine whether babies receiving a standard newborn screen and a gene panel at the same time would help reduce false positives and pick up true positives, Solomon noted.
Moving forward, Solomon said that Inova is developing a variety of sequencing-based diagnostic tests. The institute currently has Illumina's HiSeq and MiSeq instruments as well as Thermo Fisher's Ion Torrent PGM and Sanger sequencing capabilities.
The question of whether to do panels or whole genomes is always a tricky one, he said. "We don't want to take away from the whole-genome sequencing we're doing, which is enabling tons of research" on a variety of conditions like obesity and asthma. At the same time, "we want to focus in on the newborn screening genes" for clinical implementation and are "ramping up separate protocols for panels of those genes." The goal will be to design a rapid-turnaround sequencing-based test with results available in conjunction with or even before the standard newborn screen. "That's definitely very doable," he said.
In addition, he said, the group is working with specialists like pediatric cardiologists to figure out what types of sequencing-based tests would be the most helpful, whether small or large panels, or even whole exomes.