NEW YORK (GenomeWeb) –NYCKidSeq, one of several projects funded under the National Institutes of Health's Clinical Sequencing Evidence-gathering Research (CSER) program, is testing two interactive tools: one designed to improve the communication of genomic test results to families and the other to revamp the interaction between clinical providers and testing labs.
At a meeting of the New York State Genetics Taskforce at New York University School of Medicine last week, Melissa Wasserstein, associate professor of pediatrics and genetics at the Albert Einstein College of Medicine, said the goal of the NYCKidSeq project is to "incorporate genomics into the clinical care of diverse New York City children in a way that is personalized, simplified, and comprehensive."
More specifically, the program is conducting genome sequencing in children from Harlem and the Bronx with undiagnosed likely genetic diseases, comparing results of targeted and whole-genome analyses, and studying how to best communicate the test results to families. The hope is that the findings will help shape best practices for establishing such testing as a standard of care, both overall and for understudied and underserved populations.
The effort is one of several being supported by the NIH as part of CSER2, which was announced in late 2016 with a $21.8 million budget.
Joining NYCKidSeq are several similar projects, including an effort at Baylor College of Medicine focused on pediatric cancer genome sequencing in diverse populations in Texas and a project called SouthSeq at the HudsonAlpha Institute for Biotechnology that is sequencing newborns across communities in the deep south.
Many of the consortium projects are focused not only on collecting data on the utility of genome sequencing but are piloting strategies to improve the communication between patients, providers, and labs. HudsonAlpha, for example, is researching whether a secure website will encourage people to provide more detailed family histories and save clinicians time compared to more traditional methods of collecting patient data.
Speaking at the NYGTF meeting, Wasserstein discussed two interactive web-based tools that the NYCKidSeq project has developed. One, called GUIA (genetic understanding information and awareness) is meant to aid in the return of results to parents and families. The other, GenomeDiver, is designed to create a more interactive relationship between healthcare providers who order a genomic analysis and laboratories that conduct the testing and report out variants.
The trial itself — which is shared between Mount Sinai, Albert Einstein College of Medicine, Montefiore Medical Center, and the New York Genome Center — is recruiting 1,130 kids from Harlem and the Bronx.
The study is not only comparing the utility of whole-genome sequencing and targeted sequencing, which each participant receives, but is also randomizing participants to either receiving test results through genetic counselling alone or through counselling plus the GUIA tool.
Kids up to 21 years of age are eligible, and parents must be either English or Spanish speaking, although Wasserstein said the group is working to also allow Bengali speakers. Children can have neurologic, immunologic, and cardiac disorders, and need to be undiagnosed, but with a phenotype that suggests a genetic cause, though not one that would be so obvious as to obviate a need for broad genomics.
"For example, a kid who looks like they have Marfan syndrome who fulfills criteria would be excluded based on the fact that it's kind of an obvious thing for which you wouldn't typically do a whole panel," Wasserstein said.
The study has a "really simple design," she explained. "Families come in for baseline surveys [asking] 'What do you know about genetics? What do you think you're going to learn from this test? What impact do you think it might have on your child?'"
Patients then receive targeted panel sequencing from Sema4 and whole-genome sequencing from the New York Genome Center. Three months later, they come back and sit with a genetic counselor for the return of results, half of them with and the other half without GUIA. Participants then fill out another set of surveys focused on what they did and didn't understand and what they believe their child's results mean. Six months later, there is one last check in to measure how much families have retained longer term.
According to Wasserstein, GUIA is designed to fill some of the gaps that she and her colleagues identified in the standard genetic counselling and results-return process. As part of getting NYKidSeq off the ground, the group interviewed families who had undergone genomic testing for clinical purposes, including those who had positive results, negative results, and families who had received uncertain results.
"We wanted to learn about their interaction when they had those results returned to them. What were they thinking? What did they understand? What can we do better?" Wasserstein said.
"When you are explaining things [and the patient is nodding], for many people, what's actually going through their head when they are nodding is, 'What the heck are you talking about?'," she added. "And then when we say: 'Do you have any questions?', how many people actually have questions in your interactions? Most don't. And the reason some of the families gave for why they don't have questions is not because they understand everything, but in fact they didn't really understand it enough to even formulate a question. So that's a problem. We need to fix this."
Part of what the team gleaned in their initial surveys is that parents want to have choices in how they navigate results and be able to return to them over time. Showing an early version of the GUIA being deployed in the study, Wasserstein demonstrated how the application is also child-specific, with the report for a particular family referring to their child by name throughout.
"It's an interactive modular navigation, which is available on the home page and on the side of every screen the family will be looking at. And that basically allows parents to control how they want that flow of information to be. If they want to go right through everything, they can do that, and if they want to say 'let me run back and see if I remember what that means', they can, so they can incorporate it in their own learning style," she explained.
The tool employs illustrations to help explain genetic concepts and also displays varying depths of information. "We decided that we didn't want to be paternalistic in return of results and decide if we should give them a lot of information or a little bit of information because everybody wants a different amount," Wasserstein said.
By the end of the trial, the group hopes to glean from its surveys whether GUIA improves the communication of results to parents.
At the same time, the NYCKidSeq researchers have tried to improve another area of communication in the genetic diagnostic process — that between the providers engaging with the patient and the lab performing the test. The main stumbling block the group hopes to refine here is one of translation.
In standard practice, "We fill out [lab requests] describing patients using medical terms in English … but the lab looks at [those terms] and the first thing that goes through their mind is 'What is the corresponding HPO [human phenotype ontology] term?'"
HPO is standardized, she said, "so if we say 'epilepsy,' there's a number. If we have specific types of epilepsy, there's further numbers and HPO terms are in fact the language that most genomic diagnostic labs use to assess the meaning of particular variants." Essentially, Wasserstein argued, the two groups are speaking different languages, something that can lead to inaccuracies in the results that a family gets.
The team's GenomeDiver is focused on this communication between the genetics providers and the lab. In brief, the tool takes a more static, binary process where a provider orders a test and a lab reports results back and transforms it into more of an interaction.
The program helps to make the translation from English to HPO more congruent, prompting medical professionals to explore whether the way they are describing a phenotype is going to mean the right thing to the lab receiving a test order when translated to an HPO, and also allowing labs to communicate to providers if a finding seems to prompt HPOs that don’t reflect what the provider initially described.
"If the lab says we found variants in X, Y, and Z genes, they can ask the clinician to respond back 'Does the patient actually have any of these particular phenotypes based on those HPO term?'" Wasserstein explained. "There's a dialogue … and hopefully through this communication, we can get much better reporting and much more accurate [diagnoses]."