Glucagon-like peptide-1 agonist drugs approved for weight loss like Wegovy and Saxenda—and even GLP-1s not explicitly approved in that indication, such as Ozempic—have recently skyrocketed in popularity for their often fast-working approach to weight loss. But treatments for obesity aren’t necessarily one-size-fits-all, and GLP-1s may not work effectively for every patient.
Phenomix Sciences claims to have cracked the biological code on matching patients to the most promising weight-loss treatments, based on a combination of genetic, environmental and behavioral factors. Those factors can be used to classify individuals into one of four obesity phenotypes, according to Phenomix, which announced Thursday the launch of a saliva-based test aimed at the first of those phenotypes.
The test identifies patients who fall into what Phenomix has termed the “Hungry Gut” category, defined by a lack of feeling full between meals, since the body moves food out of the stomach more quickly than usual. People within the Hungry Gut group are more likely than others to respond well to GLP-1 medications, the company said, as part of a broader treatment plan that could also include a specialized diet and the insertion of an intragastric balloon in the stomach.
In an interview with Fierce Medtech, Andres Acosta, M.D., Ph.D., co-founder of Phenomix and a gastroenterologist at the Mayo Clinic, described the potential benefits of the test as twofold. For one, matching patients to the correct treatments could save non-responders from the often prohibitively high costs of GLP-1 drugs, for which insurance coverage is typically “very weak,” per Acosta.
For another, he said, “When a patient is ready to lose weight, they’re ready to engage, they want to start right away—so they should not be trying something, waiting three to six months to see if it works and then trying something else. This should get them the right intervention from the beginning.”
The MyPhenome Hungry Gut test will first be made available to a small group of obesity specialists who have been educated about the categorization system, before a full rollout begins later this year, according to Phenomix.
In the meantime, Acosta said, the company is already working on filling out its slate of obesity-classifying tests. The other three phenotypes they’ve identified are known as Hungry Brain, in which a patient's brain can't decipher when to stop eating; Emotional Hunger, describing patients who eat to manage their emotions, whether positive or negative; and Slow Burn, comprising those whose caloric intake outweighs their activity levels and metabolism.
“We are developing a Hungry Brain phenotype biomarker that will be launched in mid-summer, and together with that, the Emotional Hunger phenotype,” Acosta said. “So we’ll have the first three phenotypes launched by the middle of the year.”
As for why Phenomix started with the Hungry Gut test? As Acosta explained, “the whole world’s moving toward GLP-1.”
The debut of the phenotype-specific tests comes about two years after Phenomix unveiled its original MyPhenome test, which aimed to distinguish between all four phenotypes at once. At the time, the startup claimed that its four categories encompassed more than 90% of obese patients.
Phenomix ultimately decided not to roll out that test, Acosta said, after determining that it “was not robust enough to launch to guide medical decisions.”
With the new tests, however, the company feels more “comfortable” in their specificity and sensitivity, he said, and therefore more confident in rolling them out to doctors working to treat obesity.
Phenomix’s tests are based on about 12 years’ worth of work, according to Acosta, who co-founded the company in 2017 alongside his Mayo Clinic colleague Michael Camilleri, M.D., D.Sc. During that period, they built out a framework detailing the four obesity phenotypes and put it to the test, concluding in a series of trials that using the phenotypes to guide treatment improved outcomes.
“With that said, the question and the problem we had ahead of us was that doing the obesity phenotypes was very complicated, expensive, invasive—as we require radioactive materials—and only limited to a few centers such as the Mayo Clinic,” Acosta said. “So, for that reason, we started building biomarkers against the obesity phenotypes.”
The resulting tests use a machine learning artificial intelligence model that analyzes a patient’s DNA and other environmental and behavioral data to decipher whether they match up with the phenotype in question.