A phenotype-guided strategy for systematically matching weight-loss patients to their potentially ideal weight-loss drug roughly doubled treatment efficacy, compared with usual practice, in a single-center, randomized study with 268 patients.
Dr. Andres J. Acosta
In contrast, in 200 patients who received weight loss–drug therapy selected by routine means, 35% achieved greater than 10% loss compared with their starting weight, Andres J. Acosta, MD, said at the virtual ObesityWeek® Interactive 2020 meeting.
The phenotype-guided strategy also led to an average 16% weight loss from baseline after 12 months, compared with a 9% average loss among the usual-care controls, reported Dr. Acosta, a gastroenterologist at the Mayo Clinic in Rochester, Minn.
A “one-size-fits-all approach to weight loss treatment is not working,” he declared. “Our long-term goal is to develop a personalized approach to obesity management.”
Personalized weight loss treatment isn’t new
“The better we can match treatment to a patient’s needs, the more likely it will succeed. That’s not a brand new idea. They are trying to standardize the way that we classify the disorders that play a role in why a person gains weight or has trouble losing weight,” commented John D. Clark III, MD, an internal medicine physician and weight-management specialist at UT Southwestern Medical Center in Dallas.
The increased weight loss levels that Dr. Acosta reported in patients who underwent the study’s phenotyping protocol and received tailored treatment “are similar to the numbers we see when a patient’s treatment is the right fit for them. You see weight loss in these ranges,” Dr. Clark said in an interview.
The study run by Dr. Acosta and his associates consisted of two phases. First, they established normal and abnormal ranges for four different obesity phenotypes by studying 100 patients with obesity. The patients underwent an extensive and uniform workup designed to classify their obesity phenotype.
Four obesity phenotypes
The researchers categorized patients into one of four types:
- Disordered initial eating satiation, called ‘hungry brain,” and assessed by measuring food intake at a buffet, ad libidum meal.
- Disordered maintenance of satiety, called “hungry gut,” assessed by both a gastric-emptying study as well as patient self-assessment for postprandial fullness.
- “Emotional hunger,” assessed with two questionnaires.
- Disordered energy expenditure, called “slow burn,” assessed by measuring basal metabolic rate, and self-reports of both exercise and nonexercise activity.
Dr. Acosta estimated that the complete workup to assess all four potential phenotypes costs about $1,200.
The researchers then applied the 75th percentile value from each of these assessments to 450 patients with obesity in their clinic to see the prevalence of the four phenotypes. They identified a single phenotype in 58% of these patients, including 18% with hungry gut, 16% with hungry brain, 12% with emotional hunger, and 12% with slow burn. An additional 27% of the patients were positive for two or more phenotypes (including 9% who were positive for all four phenotypes), and 15% did not test positive for any of the four phenotypes.
Phenotype-guided treatments
They then applied their findings in a prospective randomized study that matched a drug intervention to each of the four phenotypes during a year-long, comprehensive weight-loss program at the Mayo Clinic’s Weight Management Clinic. The study randomized 100 patients to the phenotype-driven arm, with 68 of these patients receiving their assigned drug, and 200 patients served as controls. Patients averaged about 47 years old, and their average body mass index was about 41 kg/m2.
The investigational arm included 30 patients classified as having a hungry brain, with 20 of these patients treated with phentermine plus topiramate and 10 treated with lorcaserin (before it was withdrawn by the Food and Drug Administration); 12 with hungry gut and treated with liraglutide (Saxenda); 19 with emotional hunger who received naltrexone SR/bupropion SR (Contrave); and seven with slow burn who received phentermine.
The control arm included 200 patients seeking weight loss treatment at Mayo who did not undergo phenotyping and received their drug treatment based on their personal preference in consultation with their Mayo physician. In this group, drug treatment broke down as 106 patients (53%) on phentermine plus topiramate, 41 (21%) on liraglutide, 34 (17%) on phentermine alone, 14 (7%) on naltrexone SR/bupropion SR, and 5 patients (3%) on locaserin (percentages total 101% because of rounding).
Overall, phenotyping led to more patients treated with naltrexone SR/buproprion SR and lorcaserin and fewer treated with phentermine or phentermine and topiramate ER. All patients were eligible to also receive behavioral interventions as needed.
“We do a lot of testing to identify the phenotype,” in addition to gathering additional clues from a detailed history, said Dr. Acosta. Patients identified with more than one phenotype in routine practice at Mayo are often begun on more than one drug. When phenotyping fails to classify a patient, Dr. Acosta puts the patient on a low-calorie diet and then does a follow-up assessment “to see if the phenotype pops up as a metabolic adaptation.”
“This is something we’re all working toward” in the obesity management field. “How can we better identify the underlying causes in a way that can fit into the work flow. How can we move from research to things we can use daily in the clinic,” observed Dr. Clark. “We need a lot more investigation to determine how well this works in the real world. Are there other tools we can use that are not as expensive” as what Dr. Acosta used for this study?
“For this proof of concept study, it made sense to be very rigorous, but that probably is not realistic for every patient. What are other ways to get this information, or perhaps only use an extensive workup when initial weight loss attempts are unsuccessful,” Dr. Clark suggested.