ST. PAUL, Minn., Aug. 26 (UPI) — New artificial intelligence models can yield much more nuanced and detailed assessments of genetic risks for 10 inherited diseases, researchers reported Thursday.
This kind of machine learning has the potential to be a powerful new tool for helping clinical geneticists more accurately screen for inherited diseases and can greatly improve on test results that are often murky or uncertain, according to a study of the AI models published in the journal Science.
Tapping more than 1.3 million electronic health records generated by routine lab tests, researchers at the Icahn School of Medicine at Mount Sinai in New York used their models to focus on 1,648 rare variants in 31 genes corresponding to 10 “autosomal dominant” diseases, meaning diseases in which risk can be inherited with only one copy of a mutated gene from one parent.
A machine learning, or ML, model was constructed for each of 10 diseases: arrhythmogenic right ventricular cardiomyopathy, familial breast cancer, familial hypercholesterolemia, hypertrophic cardiomyopathy, adult hypophosphatasia, long QT syndrome, Lynch syndrome, monogenic diabetes, polycystic kidney disease and von Willebrand disease.
The authors reported the models succeeded in generating scores for the “penetrance” of each of the hundreds of genetic variants — or how likely a variance is to ultimately result in a disease. The “ML penetrance” scores range from 0 to 1, with a higher score closer to 1 suggesting a variant may be more likely to contribute to disease, while a lower score indicates minimal or no risk.
Senior study author Ron Do, the Charles Bronfman Professor in Personalized Medicine at the Icahn School, said such AI-generated penetrance scores represent a vast improvement over existing testing which can yield only simple “yes/no” answers for diseases such as high blood pressure, diabetes and cancer, whose genetic risks don’t actually fit into such neat, binary categories.
“Our study shows that machine learning-based penetrance is valuable not only for classic hereditary conditions such as familial breast cancer, familial hypercholesterolemia, or long QT syndrome, but also for diseases with murkier boundaries like monogenic diabetes, cardiomyopathies and kidney disease,” he told UPI in emailed statements.
“These conditions exist on a spectrum, and our approach quantifies risk in a way that reflects that spectrum. By combining genetic information with real-world health data such as lab values and vital signs, we can provide more nuanced and clinically relevant risk estimates,” he said.
One of the problems with current genetic testing methods is that for many patients, “receiving a genetic test result that is labeled ‘uncertain’ can be frustrating and anxiety-provoking, because it leaves them and their families without clear guidance,” Do said. “Our work shows that ML-based penetrance has the potential to help reduce that uncertainty.”
The likelihood of an inherited disease risk manifesting itself can be refined by drawing on millions of routine health records, the authors state. For example, it showed that patients with high ML penetrance scores had measurable differences in cholesterol, heart rhythms or kidney function.
“For patients, this could mean more personalized risk assessments and earlier interventions if warranted,” Do added. “While this approach does not replace conventional penetrance metrics, it adds an additional layer of evidence that can help patients and their clinicians make more informed decisions.”
The AI-generated data produced some surprises, as well: some genetic variants which had been considered “uncertain” showed clear signs of producing disease, while others previously thought to be likely culprits manifested few real-world effects.
The next step is to expand the model to include more diseases and to address a wider range of genetic changes and more diverse populations, the authors say.
Generally, the need for improved and expanded screening for inherited diseases, especially among children, is acute, and AI advances seem especially poised to impact the field of clinical genetics, according to a National Institutes of Health study published earlier this year.
One reason for that is that there aren’t enough clinical geneticists and other experts available while there are thousands of known genetic conditions, many of which are rare. Even if there were enough expert geneticists, no single clinician can be knowledgeable about all — or even most — genetic disorders, the study found.
Meanwhile, these conditions — which are often severe and very impactful on a patient’s life — require early detection to be fully treated, but patients now often experience years-long diagnostic delays and lack of access to state-of-the-art testing, especially in historically disadvantaged populations and less wealthy geographic regions.
Thus, many see AI as a way to help fill the gaps in efficiently and accurately diagnosing, investigating, managing and communicating with patients and families.
Should AI-based risk assessment tools such those in the current study ultimately be proven accurate and reliable enough for widespread use, the impact on patients’ lives could be profound, said Sue Friedman, executive director and founder of FORCE, a patient advocacy group dedicated to helping individuals and families facing hereditary breast, ovarian, pancreatic, prostate, colorectal and endometrial cancers.
“In our community, the difference between having a 30% risk for breast cancer and a 50% lifetime risk for some people may be the difference between having risk-reducing surgeries and mastectomies or not, or removing their ovaries at a young age or not,” she told UPI.
“The utility of this, once it’s become part of clinical practice and is included in the medical guidelines, would be huge. I’ve been an advocate in this community for 27 years, and I’m a 29-year breast cancer survivor with a ‘BRAC’ mutation. These are decisions I had to make, and these are decisions our community struggles with.
“Having these wide risk ranges and inconclusive test results are very frustrating, so this would be a very powerful tool. It’s just that we still have to put it through its paces, and eventually it could become part of the national guidelines,” Friedman said.
AI-enhanced genetic risk assessment tools do indeed have the potential to trigger big changes, agreed Dr. Marc Succi, director of the MESH Incubator, an in-house innovation and entrepreneurship center at Mass General Brigham and Harvard Medical School.
Should these tools become commonplace, “genetic testing could potentially shift from binary labels (yes/no) to patient-specific risk estimates,” he told UPI. “This would include information like a variant’s predicted likelihood of causing disease, meaning individualized decision making such as earlier screening vs. no screening.”
Higher-risk carriers might start surveillance or therapy sooner, whereas lower-risk carriers could avoid unnecessary tests, he said.
“At a population level, health systems could proactively flag the right patients for prevention clinics and genotype-guided trials,” Succi added. “Essentially, these advances could increase the fidelity of precision medicine further than where it is today.”
There are still barriers to implementation, he cautioned, such as the need to access extensive amounts of quality patient data and establishing oversight guardrails.