When Ozempic fails: Study sheds light on why some patients lose half as much weight as others

Researchers have used artificial intelligence to find out why some people have a better response to GLP-1 agonists

Ozempic cuándo funciona
A patient injecting a dose of Ozempic.Europa Press News (Europa Press/ Getty Images)
Enrique Alpañés

Ozempic has not given Ana the expected results. She reads articles everywhere that talk about the so-called miracle drug, and it makes her angry, because in her case the results have been quite limited. “I started using it two years ago, by medical prescription,” she explains in an exchange of messages. “I had to stop due to stomach problems, when I had lost two or three kilos.” She gained them back.

With Saxenda, a previous medication from the same class of drugs, the results were somewhat better, as the side effects were not as strong. She lost five kilos in three months. But this weight loss is very far below the average. GLP-1 receptor agonists have led patients to lose up to 15% of their weight, but this magic number is an average; for some people, like Ana, the results are much more minimal. She has lost weight, but not as much as her doctor was expecting. Neither she nor her doctor clearly understand why this has happened.

A study presented this week at Digestive Disease Week could have the answer. According to the research, a genetic analysis can predict each patient’s response to treatment with this drug. This summary reinforces the work carried out over the last 12 years by the Mayo Clinic in Michigan.

The researchers used a machine learning process to identify patients with a phenotype, identified as “hungry gut,” associated with greater weight loss in response to semaglutide. After 12 months of treatment with semaglutide, patients with the hungry gut phenotype lost 8.8 kilos compared to 4.5 kilos in the other group. Almost double. “It is quite robust evidence,” says Andrés Acosta, a gastroenterologist and associate professor at the Mayo Clinic, in a videoconference.

GLP-1 agonists act in the intestine and send a signal to our brain to make us feel full. This is known as the gut-brain axis, the biochemical signaling between the gastrointestinal tract and the central nervous system. “We have taken all the genes that are in this axis,” explains Acosta. “And we have analyzed them with artificial intelligence to try to predict and create a model that tells us if this axis is normal or if it has many genetic variables.” And what they found, after analyzing 84 people, is that patients who do not have variants respond much better to treatment.

“They are pilot studies, which will have to be evaluated in more patients, but it is seen that in patients with intestinal hunger it is more effective,” explains Cristóbal Morales, endocrinologist at the Virgen de la Macarena hospital in Seville, Spain. “It is interesting, not as a maximum level of evidence, but it highlights the need to phenotype this very complex disease. This is going to help us treat it,” he says.

The World Health Organization defines obesity as a “multifactorial disease due to obesogenic environments, psycho-social factors and genetic variants.” The Mayo Clinic analysis focuses on the genetic causes of the disease. “We have focused on the genetics that explain the biology of the disease,” says Acosta. “This is one of the most important factors in understanding who are the best responders and non-responders.”

The analysis is in line with the current trend towards precision or personalized medicine. Javier Escalada San Martín, director of the Department of Endocrinology and Nutrition at the Clínica Universidad de Navarra in Spain, believes that it is an “interesting finding, which seeks efficiency, detecting those people who will respond better to certain drugs.” However, Escalada recalls that the proposal to divide patients into four genotypes is controversial. In addition to the hungry gut, the other phenotypes are brain hunger, emotional hunger and slow burn, according to the division of Acosta’s team. “But not the entire scientific community agrees because it is difficult to find pure phenotypes,” says Escalada.

Acosta defends his idea by comparing the way we understand obesity with that of cholesterol. “Everyone knows that there is good cholesterol, bad cholesterol, triglycerides… In reality, there are more than 35 types of cholesterol. But talking about three large groups helps us understand and better manage the disease,” he explains.

In a recent study in The Lancet, researchers analyzed how any intervention against obesity obtains heterogeneous results, with people only losing 5% of their initial weight and others losing up to 40%. Escalada points to this study to explain how no treatment works equally well for everyone. And in any case, he clarifies, the new generations of GLP-1 agonists are increasingly more efficient, “with retatrutide [currently in phase two] 100% of the treated people lost at least 5% of their initial weight; in other words, there were no non-responders.”

Albert Lecube, endocrinologist at the Arnau de Vilanova University Hospital in Lleida, highlights the importance of the Mayo Clinic study, especially for the future. “Knowing that there is a genetic basis that affects you now is important,” he says, “but there are not many pharmacological treatment alternatives either.”

All the experts who spoke to EL PAÍS agree that these studies endorse the idea, increasingly widespread, that we should not talk about obesity, but rather obesities. There is no single response to a treatment, and even less so with a multifactorial disease like obesity.

Sign up for our weekly newsletter to get more English-language news coverage from EL PAÍS USA Edition

More information

Archived In

Recomendaciones EL PAÍS
Recomendaciones EL PAÍS