Children's Mercy Hospital is poised to update its rapid whole-genome sequencing protocol, known as STAT-seq, in an effort to bring the time needed for genome sequencing and analysis down to around 24 hours.
STAT-seq was developed to help diagnose serious genetic conditions in the neonatal intensive care unit at Children's Mercy Hospital in Kansas City. At the American College of Medical Genetics and Genomics annual meeting in Nashville, Tenn., last week, Carol Saunders, director of the molecular genetic laboratory at Children's Mercy Hospital and clinical laboratory director of the hospital's Center for Pediatric Genomic Medicine, presented findings from 25 cases tested with the current rapid sequencing pipeline.
She noted that the group is working on further speeding up the STAT-seq protocol — which currently takes around 50 hours — as it gets ready to kick off a five-year study of the rapid genome sequencing protocol as a diagnostic tool in the NICU.
To achieve that boost in speed, the team plans to tweak its Illumina HiSeq 2500 instruments to run on an 18-hour sequencing protocol, rather than the 24-hour sequencing protocol used at the moment.
The updated sequencing and analysis pipeline is designed to have enhanced speed on the library preparation and informatics sides as well, Saunders told Clinical Sequencing News. "The most significant thing is testing a different [bioinformatics] pipeline that takes around two-and-a-half hours versus our current pipeline, which is 24 hours," she explained.
The Children's Mercy researchers are currently relying on "rapid understanding of nucleotide variant effect," or RUNES, software to annotate and sort through variants detected in sequence alignments made from the Illumina reads generated for each sample tested through STAT-seq.
They are also using analytical software called "symptom and sign-assisted genome analysis," or SSAGA, to interpret findings related to an infant's condition while weeding out secondary findings. That software takes into account clinical features as well as the available sequence data.
In 2012, Saunders and her colleagues published a proof-of-principle Science Translational Medicine study of the STAT-seq protocol in the NICU setting. And last year, Children's Mercy Hospital's Center for Pediatric Genomic Medicine was one of four centers that received funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development and the National Human Genome Research Institute to investigate the clinical utility of newborn sequencing for the "Newborn Sequencing in Genomic Medicine and Public Health" (NSIGHT) program.
The team, led by Center for Pediatric Genomic Medicine Director Stephen Kingsmore, received $5 million to study STAT-seq. While some centers funded through the NSIGHT program have more of a focus on newborn screening, the STAT-seq protocol is firmly focused on diagnosing genetic conditions.
Over the five years of the study, the researchers plan to enroll 1,000 infants from the NICU that present with a range of puzzling conditions — from multiple congenital abnormalities to seizures, indeterminate syndromes, or a general failure to thrive.
Half of the infants in the study will be directed to STAT-seq for diagnostic testing using the streamlined protocol that will be added in the next few weeks. The other half will get typical treatment and testing.
"Once we start the actual protocol, it will be a blinded, randomized study where we either put patients into the rapid genome sequencing group or get the standard of care — standard newborn screening and all of the things they would normally get clinically," Saunders said.
Still, she noted that there may be an opportunity for crossovers if one arm of the study offers an obvious advantage or if an infant clearly requires rapid testing.
Promising candidate mutations will continue to be confirmed by Sanger sequencing, since genome sequencing for the study is being done through a research protocol. Nevertheless, the nature of candidate mutations found initially is verbally conveyed to physicians in case those changes provide information that could improve infant care.
The NSIGHT-funded look at STAT-seq is intended to not only test whether there really is a benefit to implementing the rapid sequencing protocol for infants, but also to consider the psychosocial and ethical implications of such testing.
"We think there's a benefit. I can't help but think there's a benefit," Saunders said. "But it hasn't been studied."
She noted that some NICU cases may continue to be assessed by STAT-seq outside of the grant-funded study, including cases from other centers. In addition to the NSIGHT grant, which specifically covers the randomized study, STAT-seq has been supported by private foundation funding and funding from the hospital.
Based on its experiences implementing the rapid sequencing program so far, the team has developed a system for prioritizing the most dire cases so that they can be moved through the rapid sequencing protocol as quickly as possible.
For cases not classified as emergencies, genome sequencing is still swift but has sometimes taken a little longer — perhaps on the order of a week. "It would still be a rapid run," Saunders noted, but not "around-the-clock work on it."
During her presentation at the ACMG conference, Saunders offered several examples of infants who were diagnosed or received specialized treatment due to genetic alterations detected by genome sequencing through the current STAT-seq pipeline. For the first 25 cases assessed using the STAT-seq approach, whole-genome sequence information on an affected infant — and, in some cases, one or both parents — altered treatment and/or provided a diagnosis in seven cases.
While sequencing does not always produce clues for treating a newly diagnosed child's condition, Saunders noted that it often spares infants further invasive testing to reach a diagnosis.
She estimated that the cost of sequencing a single sample using the existing rapid run protocol is around $10,000. That cost dips to roughly $6,000 per sample for genome sequencing with the center's non-rapid protocol, she said.
The center has four HiSeq 2500 instruments in-house, including three that can run in rapid mode. All of the analysis and interpretation is done on site at the Center for Pediatric Genomic Medicine, and results are generally returned to families by referring physicians, though Children's Mercy Hospital also has several genetic counselors on staff.
As it stands, the STAT-seq protocol is designed to return only diagnostic results to families. Investigators at the Center for Pediatric Genomic Medicine are not looking for unrelated alterations in other risk genes, Saunders said, noting that the STAT-seq protocol was developed prior to the introduction of the ACMG's recommendations on incidental findings.
Because the protocol is also designed to deliver rapid diagnoses, there are concerns about slowing the interpretation process during the initial sequence analysis.
"This is something that was discussed with our clinical genetics group," Saunders said. "At that time, we really just wanted to deal with the diagnostic findings and that's what we're still doing at this point."
"I can imagine that if you have a baby in the NICU, the last thing you're worried about is whether that baby has some variant that predisposes to adult-onset cancer," she noted. "They may want to, at some time, come back and learn about those things, but not in that acute setting."
Yesterday, the ACMG updated its guidelines in response to feedback from its members to allow patients or parents to opt out of receiving incidental findings related to the set of disease-related genes defined in the group's original recommendations.
Along with the upcoming changes to speed up STAT-seq, Saunders said researchers at Children's Mercy Hospital are also working on developing a pipeline to detect deletions in sequenced genomes, which they hope to add to their genome sequence analysis in the future.