By Deborah Borfitz
March 6, 2020 | Human Longevity Inc. (HLI) launched six years ago with the goal of building a million-genome database replete with phenotypic and clinical data but is now laser focused on helping its clients live to be 100, according to President and Chief Innovation Officer David Karow. The means is Health Nucleus—both a decision health platform driven by artificial intelligence (AI) and a bricks-and-mortar precision health wellness space in San Diego, California, that offers multimodal, data-driven assessments of clients’ current condition and future risk of chronic, age-related diseases.
HLI co-founder J. Craig Venter continues to “put his imprint” on the scientific mission of the company, Karow is quick to add.
Health Nucleus currently has 5,000 self-referred clients, with an average age of 54, 1,500 of whom return each year, says Karow. But it could increase in size with the planned opening of additional new locations in and outside the U.S.
Karow says HLI makes no apologies for focusing on early adopters who can afford the Core $7,500 price tag ($19,000 for the Platinum-level package that includes more tests). About 20% of clients opt for follow-on testing for another $3,750 for Core (or $10,000 for Platinum). Many precision health companies have gone bankrupt because they didn’t have a sustainable business model, he notes.
Five years from now, Health Nucleus could be a “more democratized offering” as the cost of genome sequencing and MRI scanners plummet, predicts Karow. Pricing has already come down considerably from an initial $25,000 for the full battery of tests, he says.
Close to 150 gigabytes of data is being collected on each Health Nucleus client from whole genome sequencing (WGS), MRI and CT scans, echocardiograms, EKGs, calcium scoring, blood and metabolomic analysis, stool DNA tests, gut microbiome testing, among potential others, plus an individual’s clinical and family history, Karow says. The typical visit takes four hours between all the assessments and visits with a physician trained in longevity and performance medicine as well as a radiologist.
The depth and diversity of the data collected, combined with AI and machine learning, get turned into meaningful risk prediction algorithms so diseases may be detected in their earliest, more treatable stages, and potentially prevented, continues Karow.
Thanks to multiple pilot programs with academic and big pharma partners in its early years, HLI has amassed one of the world’s largest WGS databases with matched phenotypic and clinical data, he says. These data contribute to the company’s ability to generate meaningful risk prediction algorithms.
HLI’s oncology division was sold to NeoGenomics earlier this year, says Karow. The deal provided HLI with $37 million in non-dilutive capital to help grow Heath Nucleus.
Sick To Well Care
The U.S. healthcare system today consumes $3.7 trillion annually—17% of the country’s gross domestic product, much more than overall military spending and far more than any other nation on earth—and yet life expectancy has been declining for three consecutive years after a century of uninterrupted increases, Karow says. The focus is predominantly on “sick care” rather than identifying and mitigating risks for preventable diseases.
Doctors today might find bladder cancer, but not until blood is detected in the urine when it is often too late to treat, says Karow. Health Nucleus specializes in pre-symptomatic diagnosis, but it is even more interested in taking actions to stop common diseases such as diabetes, stroke, and cancer from striking in the first place.
“Our vision is to change the healthcare system from sick care to well care, one patient at a time,” Karow says, and that takes more than a sophisticated annual exam (aka “executive physical”). “We want to be your partner in longevity and performance. Our goal is to get you to 100 healthily. If we find a tumor, it’s at an early stage and can sometimes be treated on an outpatient basis.”
A Health Nucleus assessment is equivalent to “presidential care,” and has three pillars—data turned into a deep-dive report on an individual’s current condition and future risks; physicians who understand the data and can help people manage their risks and well care; and referrals to world-class specialists if an early-stage complex disease or condition is found, says Karow.
Through its recent acquisition of DoctorsForMe.ai, Karow says, HLI can now send clients with life-altering, serious, and potentially complicated diseases to some of the best specialists in the world, a list that currently includes 1,400 physicians at Massachusetts General Hospital. A handful of Health Nucleus patients have already been (or are about to be) referred there. The number of partner hospitals will grow, he adds.
The model has an impressive track record. In one out of 50 clients, high-grade tumors and aneurisms have been caught at an early stage and, in one out of six clients, a rare pathogenic mutation has been found, Karow says.
A newly published study in Proceedings of the National Academy of Sciences (PNAS) points to the value of the multimodal assessment in identifying people with genetic disorders and age-related chronic health conditions that are leading causes of premature mortality in adults and have interventions, says Chief Medical Officer and study co-author C. Thomas Caskey, M.D.
The approach led to “highly actionable” findings among the self-referred cohort, most of which were not previously known, says Caskey. These include:
- Insulin resistance and/or impaired glucose tolerance (34.2%)
- Elevated liver fat (29.2%)
- Cardiac structure or function abnormalities such as valvular disorders (16.2%)
- Significant calcified coronary artery plaque (11.4%)
- Elevated liver iron (9.3%)
- Cardiac arrhythmias such as atrial fibrillation (6.1%)
- Cardiac conduction disorders (4.8%)
- Early stage tumors, most malignant (1.7%)
Next steps for patients may include a variety of medical interventions as well as dietary changes and behavior modification. “The whole point here is to put in place an intervention strategy so that the patient doesn’t experience a stroke, myocardial infarct, renal failure, neurodegenerative disease and all the [other] things associated with type 2 diabetes,” says Caskey.
Impact On Clinical Practice
Another important finding of the PNAS study is that abnormal biomarkers have a genetic basis and vice versa, says Caskey. Approximately one in six individuals (17.3%) had at least one pathogenic genetic variant, and when integrated with deep phenotyping one in nine (11.9%) were found to have genotype and phenotype associations supporting the clinical diagnosis of a genetic disorder.
Among HLI’s earlier publication credits are a 2017 WGS study (doi: 10.1038/ng.3809) in Nature Genetics that discovered many rare genetics variants were associated with abnormal blood metabolic phenotypes, and a 2015 precision diagnostics study in PNAS (doi: 10.1073/pnas.1508425112) where Caskey and his colleagues demonstrated that metabolomics could be used to assess gene mutations and uncover other potentially damaging and previously unappreciated mutations in healthy adults.
A 2018 paper published in Cell Metabolism (doi: 10.1016/j.cmet.2018.09.022) conducted in collaboration with a twin study group successfully demonstrated that biomarkers of metabolism hold up over time and are “just as reliable as DNA sequencing” at identifying clinically meaningful heterogeneity in obesity and therefore could help select patients for clinical trials, says Caskey.
The latest PNAS study expands on those findings and represents the largest such group (1,190 Health Nucleus clients) ever evaluated by such a broad a mix of technologies, he adds. It shows multiple biomarkers can be easily measured and remain consistent over time.
“This is the way one practices medicine… with an n-of-1,” he continues. “The client walks in the doctor’s office and advanced technologies make predictions of risk without studying other individuals.”
The diagnostic tools being used by Health Nucleus are all highly validated, Caskey says. “I can foresee the medical community adopting a number of things we’re doing right now, and over time [this approach] will become as transformational for standard practice as when automated analytics came into play—and have more impact.”
Caskey says the greatest strides are being made in the areas of cardiovascular disease and diabetes, two of the leading causes of death. Moving forward, he expects more focus will be put on improving risk assessments for low-frequency, high-burden neurodegenerative diseases to prevent those diagnoses.