NEW YORK – California, the nation's most populous state, recently began covering rapid diagnostic whole-genome sequencing (rWGS) for sick infants under Medicaid. The decision, in theory, should make the test more widely available, leading to faster diagnoses, better medical decisions, and lower healthcare costs in many cases.
On Jan. 1, a provision of a state budget bill from 2021 kicked in, expanding Medi-Cal benefits to include rWGS for beneficiaries under 12 months old who are receiving inpatient hospital services in an intensive care unit.
California is one of at least five states that now proclaim to cover that form of diagnostic testing as part of their health programs for low-income residents, expanding access to tens — even hundreds — of thousands of babies. Michigan was the first to do so in September of last year, while Oregon and Maryland joined California in offering coverage beginning Jan. 1 and Minnesota started coverage April 4. The states are following private payor Blue Shield of California, which has covered rWGS for critically ill newborns since June 2020.
Support for this policy from state governments comes as studies are providing more evidence that rWGS can improve medical care — often abbreviating the so-called diagnostic odyssey — and can even save money. Data from SeqFirst, a study of rWGS with broad inclusion criteria, have echoed the diagnostic yield of around 40 percent seen in Project Baby Bear, a pilot study funded by the state of California. In addition, cost analyses of providing rWGS to children older than one year suggest there are savings to be found in those cases, too.
"Genomic sequencing should more often be the first DNA test that is sent [out], and rapid WGS should be the standard of care when a child is hospitalized," said Erica Sanford Kobayashi, a pediatrician at Cedars-Sinai in Los Angeles who worked on Project Baby Bear while at Rady Children's Hospital in San Diego. And the faster, the better: Results should ideally be delivered within three days, she said, with no cost savings associated with 14-day turnaround.
Whether children will actually receive this diagnostic testing remains to be seen. "Anecdotally, [NICU doctors] are having a hard time ordering these tests," said Paul Kruszka, chief medical officer at GeneDx, a genetic testing company recently acquired by Sema4 that offers rWGS services, and an investigator on the SeqFirst study. Many doctors without clinical genetics training are simply not comfortable doing so, he said, noting that identifying patients, managing consent, and interpreting and delivering results are key challenges.
Moreover, byzantine medical billing rules mean that hospitals in California — and potentially elsewhere — may not be able to access the funding theoretically available for rWGS in the NICU under Medicaid.
California led the way in studying the potential benefits of sequencing the genomes of inexplicably ill newborns when it launched Project Baby Bear in 2018, providing $2 million for a pilot study in collaboration with Rady Children's Hospital and several other sites. Other states, including Michigan and Florida with Projects Baby Deer and Manatee, respectively, have run their own pilot programs.
Baby Bear showed that sequencing the genomes of 178 parent-child trios resulted in 76 infants receiving diagnoses — 35 of them rare conditions that occur in less than one in a million births. Of those, 55 saw a change in treatment or management, resulting in a combined 513 fewer hospital days, 11 fewer major surgeries, and a reduced testing burden that alone saved an estimated $300,000. Other costs were reduced by approximately $2.5 million, mostly attributable to shorter hospital stays, according to Sanford Kobayashi.
In a study of cost-effectiveness of rWGS published in January in Frontiers in Pediatrics, she wrote that trio sequencing costs approximately $7,500, including interpretation. However, the figure is "an average of total costs from previous cases sequenced at Rady," the authors noted.
Rady provides rWGS testing for itself and about 80 other hospitals, charging around $8,000 to $9,000 for sequencing and interpretation. "We have a bespoke, extremely rapid test," said Stephen Kingsmore, director of the Rady Children's Institute of Genomic Medicine and a champion for rWGS in pediatrics. "It can be done for less, but it's the difference between a Rolls-Royce and a Ford Fiesta."
Other state "baby animal" programs saw comparable results. Baby Manatee, led by Nicklaus Children's Hospital in Miami in collaboration with Rady Children's, enrolled and sequenced the genomes of 50 patients, leading to 20 diagnoses (diagnostic yield of 40 percent) and a change in care for 19 patients, or 38 percent. The estimated savings were more than $3.8 million — a $2.9 million return on investment, according to the final report.
For the 89 children in the Baby Deer study, 35 received a diagnosis (39 percent yield) with 24 (27 percent) receiving a change in management. At least 95 hospital days were avoided, contributing to total savings of $252,938.
In general, across 12 studies looking at the impact of rWGS on care of critically ill newborns, the diagnostic rate was 35 percent, with up to 77 percent of those patients receiving a change in management, Sanford Kobayashi said at a presentation in February at the Molecular Tri-Con meeting in San Diego.
The specific Medicaid benefits available for rapid diagnostic genome sequencing in infants and the ordering criteria vary by state. For example, in Michigan, rWGS requires prior authorization and isn't covered when the patient has an infection, trauma, or a confirmed pre- or postnatal genetic diagnosis, among other reasons. The state covers $6,278 in costs for rWGS testing specifically from Rady Children's, or $4,165 for the more general procedure of WGS billed under current procedural terminology (CPT) code 81425, "Genome (e.g., unexplained constitutional or heritable disorder or syndrome); sequence analysis" and $2,243 for CPT 81426, the code for comparator genomes, such as the parents. The state also covers genetic counseling. For fiscal year 2019, Medicaid covered approximately 46 percent of the 109,000 total births in Michigan.
Maryland also requires prior authorization and reimburses $3,999.80 for CPT 81425 and $2,154.40 for CPT 81426, according to data provided by Illumina. In 2020, the state had 68,554 newborns, of which nearly 40 percent were covered by Medicaid.
Minnesota requires an evaluation by a medical geneticist "or other physician subspecialist with expertise in the conditions or genetic disorder for which the testing is being considered," a spokesperson for the state's Department of Human Services said in an email.
Oregon, which already covered whole-exome sequencing under Medicaid for children under 18, may also cover rWGS for children over the age of 1 "if, based on individual consideration by the plan, the test will benefit the patient in terms of growth, development, or ability to participate in school," Philip Schmidt, a spokesperson for the Oregon Health Authority, said in an email. Data provided by the state showed that the Oregon Health Plan covered 53 WES tests during 2020. No data were available yet on rWGS coverage.
While California budget acts are only law for one year, the coverage expansion was included in a budget trailer bill, according to Nannette Miranda, director of communications for Assemblymember Phil Ting, D-San Francisco, chair of the budget committee. "Trailer bills become permanent law, just like regular bills," she said. Rady Children's hired lobbyist John Valencia to help get rWGS covered, she noted. The state's 2021-2022 budget provided $6 million to cover such testing, although it is not clear how much, if any, has been used. The state estimates that Medi-Cal covers around four in 10 children, and in 2020, the state had approximately 420,000 births.
In both Oregon and Maryland, San Diego-based Illumina helped precipitate the coverage decisions. Minutes from a Nov. 18, 2021, virtual meeting of Oregon's Health Evidence Review Commission, which makes coverage policy decisions for new treatments or procedures for the Oregon Health Plan, noted that "recently, Illumina contacted HERC staff to request a review of coverage of WGS." HERC had last considered the test in 2014. Illumina Senior Medical Director for the Americas John Fox, a veteran of Project Baby Deer, also testified at the meeting.
"Both Illumina and many supportive providers wrote the state [of Maryland] asking them to expand coverage for this important innovation," Illumina Senior Director of Communications Adi Raval said in an email. "We are grateful to the state for its action in a year of significant evolution, with COVID and other issues demanding their limited time."
Illumina's market access teams are pushing to get rWGS covered in other states, including Iowa and Ohio, as well as in other countries. "What we found is, if Illumina does not take the lead, it's very hard to accelerate adoption in the marketplace," said Ammar Qadan, Illumina VP of global market access.
Aside from the state-funded studies, other results are supporting the expanded use of sequencing in newborn care. At the American College of Medical Genetics and Genomics annual meeting in March, researchers from the SeqFirst study — which received donated sequencing reagents from Illumina — presented data suggesting that more than half of the 125 study participants received an abnormal result that led to an explanatory, or at least partially explanatory, diagnosis.
"Two-thirds were suspected of having a genetic disorder, but one-third were not," said Mike Bamshad, chief of genetic medicine in the pediatrics department at the University of Washington and Seattle Children's Hospital, and principal investigator on SeqFirst. Of that one-third, nearly all (93 percent) saw a change in management.
The premise of the SeqFirst study is to improve access to precision genetic diagnostics. There are only a few reasons not to do sequencing, GeneDx's Kruszka suggested. For the study, children are only excluded if their illness can be explained by infection, trauma, or prematurity.
The researchers had hypothesized that by expanding enrollment, the diagnostic yield would drop, potentially by as much as 50 percent. "So we were impressed that our explanatory rate remained up around 50 percent," Bamshad said.
"These are results that scream to me 'we need to look at a much larger population of kids,'" he said. "Clearly, this strategy does exactly what it's intended to do."
SeqFirst, through GeneDx, was able to deliver verbal results in an average of five days, Kruszka said. But according to Sanford Kobayashi, results need to come back more quickly to yield the full benefit of testing.
"Project Baby Bear showed that savings are influenced by test turnaround time," she said. The best outcomes come when results are back in three days or less, she added. "If you want to have the biggest impact, it does have to be that fast."
While ultra-rapid sequencing in the NICU may yield the most medical and economic benefits, "diagnosis is beneficial at any stage in life," Sanford Kobayashi said. Her new research is focused on using rWGS in the pediatric intensive care unit, which admits older children.
"All the kids born 10 years ago with maybe long QT syndrome or cardiomyopathy, they're still showing up to the PICU," she said. "Until we catch up and start sequencing everybody earlier, there are still going to be years where it would be useful to apply it to older children."
In the cost-effectiveness study published in January, Sanford Kobayashi's team assessed the cost of sequencing in the pediatric, rather than neonatal, ICU. Of the 38 participants, 17 received a diagnosis from WGS and seven had a change in management, allowing the team to model costs for them. While such cost modeling for rare disease is hard, because there are not always good data to compare to, the team used a so-called "Delphi consensus method" that compared the children to a counterfactual trajectory — a hypothetical scenario in which rWGS was not ordered. Each trajectory was sent to 10 different pediatric institutions for review. "In doing that, we found that we saved about $185,000 in hospital costs," she said, again mostly due to shorter stays. Sequencing the trios cost about $240,000, resulting in net spending of about $55,000. Moreover, the study found that WGS added a total of about 12 quality-adjusted life years (QALY), an advanced metric that seeks to quantify the value of medical procedures beyond the simpler calculation of extended survival. "We were spending about $4,500 per QALY, which is super reasonable," she said, adding that spending $50,000 to get just one QALY is considered cost effective.
Still, most of the benefits of rWGS in older children are concentrated in a small number of cases. One of the children in the study was a 9-month-old who had been admitted to the PICU "without a good reason." Imaging showed a brain bleed and WGS detected a factor 13 deficiency, a hemophilia-like rare disease that is very unlikely to be tested for. Later, the child fell off a couch and was given extra factor 13 as part of the treatment, preventing another brain bleed.
"The moral of the story is that there are going to be times we don’t make a diagnosis and it doesn't save any money," Sanford Kobayashi said. "But often enough, you're going to make a big enough impact that it makes up for those other times."
When asked why Oregon's HERC decided not to cover rWGS for children over 1, even though it covers WES for them, a spokesperson said, "the Commission decided that this group of patients (severely ill hospitalized infants) would have the clearest benefit from this service, because it would be most likely to affect treatment planning for this group."
"If the evidence for WGS for older, less severely ill children develops in the future, the HERC will consider expanding coverage beyond the group that is currently covered," she said.
While states are saying they'll cover this testing, there's still no guarantee that the children who need it will have it ordered for them. "Eventually, most big children's hospitals will do their own WGS," Sanford Kobayashi said. "But it's a big lift. The equipment is expensive, and people who can do the analysis are expensive."
Also, most doctors are not experts in genetic medicine. "We're not training a whole lot of medical geneticists," Kruszka said. "There are fewer than 60 residency slots per year and only half are being filled."
Even in the best-case scenario, where a patient is in the NICU, often the ordering physician will not have expertise in genetics. Those doctors continue to express discomfort ordering a test that requires consent and the results of which need to be explained and put into context, Kruszka said. And that doesn't even begin to address ordering the test outside of ICUs.
The good news is that when doctors see the power of rWGS up close, they're easily sold. "When neonatologists start using the technology, they love it," Kruszka said.
"Intensivists are eager to see us scale up," Bamshad added. "In the end, we're making their workflows much easier. And the changes in [patient] management are sometimes profound. They're now concerned they’ve been missing things all along."
The SeqFirst researchers suggested that relying on clinical geneticists will only hinder access and are now considering ways to help nonexperts, possibly with handheld devices that help navigate the explanation and consenting process.
"Interpreting the result is key," Bamshad said. "That is probably done most effectively face to face with a genetics provider. … To be honest, we probably still don't have the workforce to do that." At Seattle Children's, wait times to be admitted to the genetics clinic can be a year, even two; the hospital also has nearly two dozen genetic counselors to help interpret results. "That's a luxury that doesn't exist in most places," he said. "Eventually, we'll have to get to a point where return of this information is facilitated by technology and interpretation is done by nonspecialists."
Expanding the practice area for genetic counselors may be part of the answer. Michigan is in the process of licensing genetic counselors as independent clinicians, which could happen this year, Bob Wheaton, the spokesperson for the Michigan Department of Health and Human Services, said in an email. "Once this occurs, Medicaid will begin enrolling these practitioners and will allow direct reimbursement of genetic counseling services."
At the federal level, there are two avenues that could lead to increased coverage for rWGS in newborns under Medicaid. Last year, US Sens. Susan Collins, R-Maine, Mark Kelly, D-Ariz., and Bob Menendez, D-N.J., introduced the "Ending the Diagnostic Odyssey Act," which would have the federal government pay 75 percent of the cost of rWGS, with states picking up the rest.
The 21st Century Cures 2.0 bill, introduced last November by US Rep. Diana DeGette, D-Colo., also could lead to funding of "demonstration projects" in up to 15 states.
But unless the mechanisms by which Medicaid pays for inpatient care are addressed, hospitals are unlikely to facilitate ordering of rWGS in the NICU.
States reimburse for hospital stays with diagnostics-related group (DRG) payments, essentially a lump sum for a particular type of hospital stay. "As a genetics physician, that puts us in a tough spot," said Caleb Bupp, a physician at Helen DeVos Children's Hospital in Grand Rapids, Michigan, and a leader of Project Baby Deer. "The cost of genetic testing eats up how much the hospital gets reimbursed," he said, potentially resulting in a net loss. "My perspective is, [rWGS] only works if payment happens separately. That's the elephant in the room."
In Michigan, the state created a DRG "carve-out," a pipeline to directly fund rWGS testing in the NICU, which takes the onus for the cost of testing off hospitals. Getting that carve-out, however, was only possible due to trust that the various parties, including the state health department, had built up after years of dialogue, Bupp said. "You have to call it out as something you problem-solve for," he said.
California does not have a DRG carve-out, Kingsmore said, and Minnesota has not made one, either. "The performing lab outside the hospital would submit for the test," the DHS spokesperson said. Whether Oregon and Maryland have carve-outs is not clear — state health departments did not immediately respond to follow-up questions about payment mechanisms.
States that have not addressed payment for rWGS "are stealing from themselves by disincentivizing doctors from ordering this test," Kingsmore said. "It's really bizarre; they're saying 'no' to something that would save them money."
Rady has scheduled meetings with the California Department of Health Care Services to address payment mechanisms, but any celebration about establishing rWGS coverage would be premature.
"We have additional work to do," Kingsmore said. "Without such payment, uptake will be essentially zero."