NEW YORK – Based on findings from a small pilot study in a relatively well-off patient population in London, a UK team has suggested that a combination of whole-genome sequencing, panel sequencing, and comprehensive medical testing and imaging is feasible in a family practice setting and may benefit patients and their families.
Ros Eeles, a researcher with the Institute of Cancer Research and the Royal Marsden NHS Foundation Trust shared early findings from the study at the American Society of Clinical Oncology's annual meeting in Chicago on Friday. Together with Michael Sandberg, a general practitioner and medical director at the private healthcare center 90 Sloane Street, she co-led the 90S Study, named in a nod to that central London center.
"We had a group of specialists who wanted to look at what would happen if you brought whole-genome sequencing into primary care," Eeles said, noting that the 90 Sloane Street center has built in electrocardiogram, echocardiogram, abdominal and pelvic ultrasound, and routine blood test capabilities.
In the context of cancer, for example, germline and somatic mutations can impact everything from disease screening to treatment strategies, for example. Indeed, large-scale clinical and population sequencing studies looking at germline genetic variation and their ties to human traits or disease are increasingly common, as are efforts to characterize germline and somatic mutations in cancer patients.
But studies on the potential for sequencing individuals in a family medicine setting are much rarer, and it remains to be seen whether the approach is feasible, scalable, cost-effective, and/or beneficial for patients with or without an existing diagnosis.
"What we wanted to ask was, 'Should you use whole-genome screening in next-generation health screening in primary care?'" Eeles said.
For the pilot stage of the study, the team had thirtyfold average whole-genome germline sequencing done commercially on 104 seemingly healthy individuals ranging in age from 29 to 82 years old, who were recruited between 2020 and 2022. The participants tended to have high socioeconomic status and more than 94 percent had a family history of cancer or cardiac disease in a first- or second-degree relative.
All but one of the participants was symptom-free, and 13 had a history of cancer. European ancestry was reported for more than 95 percent of those enrolled.
Since researchers had their sights set on finding as many informative variants as possible, they rounded out short-read genome sequence data with targeted panel sequence data to try to avoid missing deletions. They profiled pharmacogenomics-related variants, polygenic risk contributors, and other genetic features using whole-genome sequence data, which provided insights on up to 566 genes, and used targeted sequence data to search for actionable changes in dozens of cancer- or cardiac-related genes.
The team also considered participants' personal and family health histories, as well as results from a full medical with the same general practitioner. For that medical, participants were profiled with a variety of clinical screens ranging from blood tests to organ imaging.
The team opted to report only variants that could prompt actionable changes to a patient's care, she added, but they established a system for returning to suspicious variants in the genomes as the definition of actionable mutations continues to expand or evolve in the future and new drug targets turn up.
All told, the newly generated genome sequences led investigators to relatively rare, clinically actionable germline mutations in 27 study participants. They flagged cancer-related genetic risk variants in genes in 10 of the participants, for example, while two more individuals carried actionable cardiac rhythm disorder-associated variants, and still other patients had variants in genes involved in clotting, lipid metabolism, or other processes.
Far more individuals — around 80 percent of individuals in the study — were identified as carriers of variants implicated in autosomal recessive conditions, the team reported, including conditions that are relevant to reproductive screening programs. A significant proportion of sequenced individuals had at least one variant linked to pharmacogenomic responses.
The kinds of risky germline variants detected sometimes tracked with the new diagnoses made or with an individual's personal or family health history, but not always.
The medicals alone led to new heart disease or prostate cancer diagnoses in several study participants. On the other hand, 70 percent of individuals who ended up having cancer susceptibility-linked variants in their germline would not have been found without the genetics program, Eeles said.
"Clearly we are picking up mutations in those without a family history, and that's where geneticists giving advice is going to be important," she explained.
"The majority of patients had a significant change in management for themselves or their families," Eeles and her co-authors wrote in an abstract accompanying the ASCO presentation. "Whole-genome sequencing as part of health screening in family practice is feasible and is likely to have significant beneficial health management implications."
The team tapped into the common variant data collected from participants for a research arm of the study, which looked at polygenic risk for conditions such as colorectal cancer or prostate cancer.
While the research was done within a private clinic setting, the investigators also looked at what aspects of the program might be retained, jettisoned, or tweaked if attempting to apply it through a public health system such as NHS.
Speaking remotely at the ASCO meeting, Fergus Couch, a laboratory medicine and pathology researcher at the Mayo Clinic in Rochester, Minnesota, who was not involved in the study, noted that the 90S Study is "a really nice model for evaluating how exome sequencing, genome sequencing, or panel sequencing might actually be useful in the clinical setting."
For example, he noted that the program picked up risky variants that would have been missed by following current family- or personal medical history-based genetic testing guidelines in the UK.
Still, Couch questioned whether the patient population for the 90S Study was representative of the broader population, since family histories of disease and rates of clinically actionable changes in cancer-related genes such as BRCA1 were higher than that described in the past.
He also noted that care is needed in general when incorporating sequencing into clinical care for symptom-free individuals, not only when it comes to defining actionability or implementing information on specific variants or risk scores but also when selecting the sequencing strategy used and setting up appropriate referrals for patients with actionable variants or recessive disease carrier status.
"Could this type of sequencing and the data that comes from it actually be handled by general medicine, family medicine, or community practices around the country? Is this really just too much?" he wondered, noting that "in the absence of a very strong referral network, this could be really quite challenging for the vast majority of non-tertiary medical centers."