Skip to main content
Premium Trial:

Request an Annual Quote

Q&A: Stephen Kingsmore on Scaling Up Genome Testing at Rady Children's Hospital

Premium
Stephen Kingsmore

NEW YORK (GenomeWeb) – Last week, Rady Children's Hospital in San Diego announced that Stephen Kingsmore will be the first president and CEO of the Rady Pediatric Genomics and Systems Medicine Institute. He joins Rady after almost five years at Children's Mercy Hospital in Kansas City, where he was the founding director of the Center for Pediatric Genomic Medicine and pioneered rapid genome sequencing to diagnose critically ill infants with suspected genetic disease.

The institute was launched in the spring of last year and is funded by a $120 million gift from the Rady family and $40 million from Rady Children's Hospital, a nonprofit organization with 520 beds that serves children in the larger San Diego area.

Kingsmore started his new appointment this week and joins a number of recent hires by the nascent institute, which aims to bring together scientists and clinicians to accelerate the translation and implementation of research discoveries into pediatric medicine. In April, the institute announced its first faculty member, Tina Chambers, a professor of pediatrics at the University of California San Diego, who will lead a project to sequence the genomes of children in San Diego County who are born with birth defects of unknown cause. This summer, the institute hired Joe Gleeson, who has been spearheading exome and genome sequencing for diagnosing neurodevelopmental disorders, as chair of neuroscience. He will also be a professor of neurosciences and pediatrics at UCSD and a Howard Hughes investigator.

GenomeWeb caught up with Kingsmore on his last day at Children's Mercy to talk about his plans for the Rady Pediatric Genomics and Systems Medicine Institute. Below is an edited version of the conversation.

What is the mission of the new institute?

The mission of the institute is to deploy genomic and systems medicine in the hospital. The mission is not just translational research but it's actually implementation science. It's taking the types of things that we prototyped here at Children's Mercy, making them scale to enterprise level, and have them increasingly be adopted as the new standard of pediatric care.

One thing I don't think many people realize is that Rady Children's Hospital is intimately tied into the department of pediatrics at UCSD. When I looked at this research institute, it was not just a matter of having a great children's hospital, it was also about having [UCSD] right on our doorstep, and all of the systems biology and stem cell technologies and mathematical modeling. The dean of the medical school at UCSD is on the board of the new institute — that's a key connection, and was a key piece in the puzzle for me, that we would be able to get things going very rapidly because of the intellectual richness of UCSD. But it's not just UCSD; the Salk Institute, Scripps Clinic, and others are also collaborators with Rady. So, yes, we're announcing a new research institute, but it's going to be a highly collaborative, San Diego-focused effort.

Can you talk about some concrete projects you have planned?

The first thing we want to do is expand our NIH-funded NSIGHT [Newborn Sequencing in Genome Medicine and Public Health] program for genome sequencing in the neonatal intensive care unit. Rady Children's Hospital has the only level IV NICU in the San Diego region, which has a population of nine million individuals. Furthermore, Rady neonatologists provide support for all of the NICUs in the San Diego region. So there is an exceptional opportunity to take our 26-hour STAT-seq genome test and scale that up and put it in play in neonatal care.

We will be splitting the funds of our current NIH-funded program between Rady and Children's Mercy, so that the work can continue here [in Kansas City], but we will have a new, very large site at Rady. Our goal is to do a series of pivotal randomized clinical trials that will define which babies should get a rapid genome. And it will start to parameterize how a genome sequence translates into precision medicine for newborns. This is a brand new field of endeavor — there is no randomized prospective evidence base.

Over the next three years, we will be doing a number of studies to ask very practical  questions, like, 'For which indications should a genome test be ordered?' And should it be singletons or trios, should it be a very rapid protocol or a slightly slower protocol, say, on a HiSeq X Ten? And then how does that influence care plans?

Many of the babies with genetic diseases who are in a NICU are going to die. Our paper in Lancet Respiratory Medicine showed that 57 percent of the babies that we gave a diagnosis to by genome sequencing had died by day 100. So we have a really huge need to go from delivering a genome sequence test to delivering clinical practice guidelines related to that baby's condition. Some of the babies need changes in management, and for some, there is no treatment today. There, the emphasis shifts to discussion of end-of-life care with parents. So, we suddenly get involved in the practice of medicine, and we need ways to be able to scale genome testing that neonatologists can understand. That means use of clinical practice guidelines, artificial intelligence, and electronic support, so they can maximize the utility of genome information in the acute management of those babies.

How many patients are you planning to enroll in your studies over the next three years?

It's really tough to give a number. What we've told NIH is 1,000 — I think that's an underestimate. We have tried to map out how many kids with genetic diseases live in the Rady catchment area [and how many of these families qualify for and might agree to whole-genome sequencing], and that number maps to about 24,000 genomes per year that are needed to implement genomic medicine for all kids. Obviously, most of these kids are not in a NICU setting, so the NICU number is smaller. But that gives you an idea of the overall vision over the next five years, as we go from the NICU to then also the pediatric intensive care, the cardiac intensive care unit, inpatients with genetic diseases, outpatients with genetic diseases, and then into perinatal care for mothers in the second or third trimester who are being seen by a fetal pediatrician in a maternal medicine clinic.

The INSIGHT grant from NIH is the starting point for a partnership between Children's Mercy and Rady Children's Hospital. But our goal is, rapidly over the next six months, to set up a consortium of like-minded children's hospitals that want to deploy this. The task is too big for any single children's hospital. And the pace of change is too rapid to think about a study that could take five years to do, so there is going to be a need for several children's hospitals to collaborate and to have enough numbers to be able to run clinical trials in a timely manner.  I think we've got the basis for that. There is a lot of exciting news that will be happening through the rest of the year with regard to that, how we fund that, and who is going to participate.

Are you planning to develop and deploy other genomic tests besides whole-genome sequencing?

We are strong believers in genome sequencing. STAT-seq is important for some children who have an emergency, whether it's a cancer emergency or an intensive care emergency. But for most children, [slower] whole-genome sequencing on something like a HiSeq X Ten is going to be a better option. A turnaround time of a month is acceptable for those kids, and it's more cost-effective for them to have this type of test.

But we want to not just be delivering diagnoses, we also want to get into the types of secondary findings that are important in children. That includes things like pharmacogenomic guidance for medications. Many drugs that are used in children have not been through formal testing with the FDA, but instead, the dosing in adults is scaled down for children. The randomized trials to show their effectiveness and to optimize the dosing in children have never been done for many of these drugs. There is a huge need for pharmacogenomic guidance, and it gets even more complex in newborns who have not yet developed all of the drug metabolism enzymes.

Physicians are not used to dealing with pharmacogenomic information, so they don't really need, let's say, a CYP2D6 genotype — what they need is dosing guidance. Again, we need to build clinical practice systems that sit on top of the medical record that can allow physicians to make use of that type of information in practical patient management.

And then there are many childhood genetic diseases that it's worth surveying patients for. Things like bone disease, metabolic disorders, and endocrine disorders, where that information is highly relevant to the care of the child. So, the child may get a genome sequence for one condition but we want to identify other, secondary findings that will be used in that child's care as they grow up.

How will you perform the genome sequencing tests — are you planning to build your own facility, or outsource the sequencing?

One of the most exciting things about moving to Rady is that San Diego is arguably the genomics capital of the United States. There are companies like Human Longevity, with X Ten instruments, that are focused on clinical markets. And of course Illumina is headquartered there, and there has been a strong relationship between Rady and Illumina before I came on the scene. I don't have a crisp answer, but I do know that there is an abundance of sequencing capacity in the San Diego area. It will be a matter of sitting down with the various organizations and figuring out what's in the best interest of San Diego's children. The kids of the employees of all of these organizations actually receive their healthcare from Rady, so there is a unique opportunity for the San Diego population to be the first in the country to really get the benefits of the technologies that are manufactured in San Diego.

Reimbursement for genome sequencing tests has been difficult. How is that going to change?

Reimbursement is tough. In the past, it was sufficient to show diagnostic utility of a test in order for it to be reimbursed. Today, in 2015, we need to show clinical utility and cost-effectiveness. That's why I'm so keen for us to do a series of randomized prospective clinical studies, using the endowment that Mr. Rady has given us, and to make those multi-center studies in order to get the evidence that shows payors that for certain subsets of children, this is associated with better outcomes and with cost-effectiveness.

The beautiful thing about our founding gift is that we can take a 10-year time perspective on our goals. We don't need to figure out how to get tests reimbursed this year. Instead, we can make sure that we're thinking about the needs of the children, and how we put in place solutions that real-world physicians can use day in and day out. And then, how we accrue evidence over the next decade so that we actually transform pediatric care.

What about efforts other than genomic testing?

I don't want to leave you with the idea that it's only genomics. All of us now realize that our understanding of the genome today is insufficient to rely on the genome alone. We need a metabolome, a proteome, a transcriptome, and a natural language-processed medical record to be able to make sense of the genome. For most of the variation in the genome, we don't understand the impact. So by adding in other 'omes', we will have an opportunity to integrate the information and actually understand much more about what the genome means.

Do you have concrete plans for that yet?

We have a pilot project with SomaLogic. They make aptamer-based protein assays and can measure almost 5,000 proteins in 70 microliters of plasma. That's a great example, where we can take a single blood sample from a baby and get 5,000 protein measurements simultaneously, along with the genome. By integrating that information, we will [learn to better understand] variants of uncertain significance, for example.