Three major factors contribute to a person’s weight gain, said Steven Heymsfield, professor and director of the Body Composition-Metabolism Laboratory, Pennington Biomedical Research Center, Louisiana State University system: genes, epigenetic modifications that affect the expression of genes, and the interactions of genes and their modifications with environmental factors. However, he added, these interactions create “literally thou-
sands of potential ways any one individual person can develop obesity.” As a result, he asserted, deciphering the underlying causes of obesity in any given individual is “a tremendous scientific challenge.”
Heymsfield reviewed innovative new approaches to controlling obesity in three areas: medications, devices, and lifestyle interventions. In a later presentation at the workshop, Gary Bennett, Bishop-MacDermott family professor of psychology and neuroscience, global health, and medicine, Duke University, looked more closely at technological innovations, and his remarks also are summarized in this chapter.
In some people, targeted treatments have very strong effects, Heymsfield noted. For example, he elaborated, some young children develop severe obesity because of a problem known as hyperphagia, in which they have no control over their food intake and often die early in life. He noted that 10 to 20 genetic abnormalities, often characterized by a single amino acid defect in a hormone, can cause this and related conditions leading to obesity. For instance, some people have zero levels of the hormone leptin, and when they are given small physiological doses of leptin, their weight becomes normal. Although such opportunities are rare, said Heymsfield, “it shows it is possible.”
Finding a medication that can counter such genetic defects is challenging, Heymsfield acknowledged, but the treatment has been successful in some cases. For example, mutations in the melanocortin 4 receptor and closely related mechanisms are found in 2–5 percent of cases of severe obesity. Heymsfield reported that a recently introduced drug, setmelanotide, which is given subcutaneously, has normalized the weight of people with these mutations (Kühnen et al., 2016). “This is a curative mechanism,” he observed. “It shows that we have the ability to have targeted therapies.” He noted that repositories of people who have such mutations are now being created so they can be offered pharmacotherapy.
Setmelanotide is a peptide, and other small proteins may be valuable in treating obesity, Heymsfield continued. He cited another prominent family of peptides, glucagon-like peptide 1 (GLP-1), which affects glucose metabolism (i.e., diabetes and obesity), and explained that saxenda (liraglutide) is a GLP-1 drug now used in treating obesity that can produce weight loss over time. He characterized this as the most potent of the monotherapies for obesity currently approved by the Food and Drug Administration (FDA). A new GLP-1 analog, semaglutide, also can produce weight loss over time, he noted, adding that unlike liraglutide, which is injected subcutaneously once or twice per day, it needs to be injected just once per week.
Like all peptides, Heymsfield continued, including insulin, these medi-
cations need to be injected to prevent them from being digested in the gastrointestinal tract. He then explained that drug developers have been working to create stabilization factors that wrap around peptides and prevent their digestion by acid in the stomach and facilitate their absorption by the gastrointestinal tract. He cited the example of an oral version of semaglutide coformulated with an absorption enhancer known as sodium N-[8-(2-hydroxybenzoyl) amino]caprylate (SNAC). SNAC causes localized pH increase, which leads in turn to higher solubility and protection from enzymatic degradation. Heymsfield reported that initial results show that treatment with this medication causes significant weight loss in a majority of people at 1 year. With more development, he suggested, the result could be a once-per-week oral pill.
Another approach, Heymsfield continued, is to use combinations of medications to create double, triple, or quadruple agents. For example, a GLP-1 agonist such as liraglutide can be combined with a coagonist to create a double peptide. Heymsfield said he had identified more than 23 GLP-1 peptides in development, both alone and in combination therapies. For example, he explained, activating the proglucagon peptide in the brain breaks up the different peptides, each of which has effects on glucose metabolism and body weight. Two such breakdown products are GLP-1, which acts on the GLP-1 receptor, and oxyntomodulin, which works mainly on the glucagon receptor but also on the GLP-1 receptor. Heymsfield noted that one such drug that combines the two is in development, and that early trials have shown putting the two agonists together produces double the amount of weight loss (Finan et al., 2015; Pocai et al., 2009).
Similarly, Heymsfield suggested, peptides that block the SGLT2 receptor, which is involved in reabsorbing sugar in the kidney, could be an effective treatment for diabetes. He observed that the body tends to compensate for the loss of sugar by making a person want to eat more, and a similar counter-regulatory mechanism is involved with exercise. Combining a SGLT2 inhibitor with a compound such as phentermine could block this counter-regulatory effect, he said. He reported that in another set of initial results, this combination produced weight loss of 5–10 percent at 6 months.
Heymsfield added that a compound that blocks the μ opioid receptor in the brain’s arcuate nucleus, which is related to hedonic mechanisms such as liking fatty food, was successful at blocking hedonic mechanisms, although it did not cause weight loss in general. However, he noted, people who had particular mutations in their μ opioid receptor pathways did experience weight loss through a combination of pharmacogenetics (Ziauddeen et al., 2013). This is “very early work,” he said, “but promising.”
Finally, Heymsfield described pills that include hygroscopic compounds so they expand like a sponge in the stomach. He reported that trials have
shown that such pills affect appetite ratings and produce modest weight loss over time.
Heymsfield also reported that researchers are currently gathering genomic, metabolomic, proteomic, and other kinds of “omic” data to search for modifiable pathways that could influence energy balance and weight. The eventual result, he predicted, will be personalized medicine reflecting the unique factors that regulate weight in each individual.
Heymsfield briefly touched on several devices for inhibiting food intake. For example, balloons can be swallowed or placed in a person’s stomach endoscopically and then blown up to cause weight loss. He explained that they generally are used in people who have a body mass index (BMI) of 30 or more, are kept in the stomach for 6 months or less, and produce modest weight losses of 5–7 percent over 6 months compared with placebo.
Heymsfield described another approach that entails stimulating the vagus nerve in the gastric region through a surgically inserted device, which tends to produce moderate weight losses. He noted that a recently approved device introduces a sham fistula into the stomach with which food contents can be removed, producing a daily calorie deficit. Another device shunts food from the stomach to the small intestine and has been shown to produce significant weight loss and reduction in diabetes.
“Creativity is booming now with the need to curtail the obesity epidemic,” Heymsfield observed.
Finally, Heymsfield touched on innovative approaches to lifestyle modifications. He cited standing desks, websites, fitness trackers, cellphone applications, and many other technologies that offer new ways to change behavior. “There is a tremendous wave of interest in behavioral therapies,” he said. “We know very little about how effective they are. We are beginning to learn.”
Heymsfield went on to note that some behaviors that do not appear obviously related to obesity could have an effect on weight. As an example, he cited research showing that too little sleep leads to obesity, so sleep treatment programs are being pursued as a way to control weight. Financial incentives such as taxes on sugar-sweetened beverages can change behaviors, he suggested, as can innovative lifestyle management plans. It may even be possible, he observed, to take brain scans of people to determine whether they will respond to particular treatments.
Heymsfield emphasized the importance of supervised medical care for
all treatment plans. “It is very important,” he asserted, “that we have trained physicians and other health care workers . . . who can manage obesity with these complex mechanisms we are developing.”
During his remarks later in the workshop, Bennett stated that “the advice that ‘you should ask your doctor’ works very well for most clinical conditions—with the exception of obesity.” Primary care providers often do not deliver counseling and comprehensive obesity care in the primary care setting, he explained. Behavioral weight loss treatment can work well, he suggested, but it works less well when translated into the primary care setting and with higher-risk populations. The lack of attention to obesity in the primary care setting “undercuts our efforts, in my view, to try to help patients realize the clinical significance of the condition,” he said. Primary care providers need the tools to screen, to stage, and to hold patients accountable, he argued. Elaborating, he suggested that such tools could, for example, equip providers to take BMI measurements quickly and easily, present a range of options to patients, help patients engage in a shared decision-making process, and hold patients accountable for the decisions that are made.
Bennett focused on the promise of digital technologies to improve obesity treatment. “The widespread availability of digital devices has utterly changed our lives—that probably goes without saying,” he observed. Even more important than the devices are the data they collect, which, he suggested, hold great potential for “change in a wide range of clinical conditions, but particularly in obesity.”
Bennett continued by observing that new analytic strategies for parsing the data collected by digital technologies have begun revealing clinical meaningful insights. For example, he reported that the past 7 years have seen almost 300 million downloads of weight loss apps, all of which can collect data about the people using them. These data can be used, for instance, to identify whether people are adhering to the Dietary Approaches to Stop Hypertension (DASH) diet. “For patients who have obesity and hypertension,” Bennett said, “we are able to make corrective recommendations on a real-time basis using the data that they would be collecting already.”
As another example, Bennett cited the use of scales in people’s homes to generate text messages from registered dieticians if the user gains weight, followed by telephone calls if the weight gain continues. “We have known about these kinds of stepped care programs for a long time,” he said. “What is new is the ability to have a low-cost digital device in homes broadcasting through cellular networks in real time, and our ability to put
these data on platforms where registered dieticians in clinics can leverage them and deliver care inexpensively and efficiently.”
Bennett suggested that the emergence of artificial intelligence has even greater potential to change how data are leveraged, saying, “The key challenge in obesity treatment, I would argue, but particularly in digital health treatment, is enhancing engagement, keeping patients on the hook, over the extended time horizon that it takes to produce weight loss.” He observed that artificial intelligence is being used to deliver feedback to individuals when they need it and in the language that they want. To illustrate the point, he said, “If a patient is a woman who works outside the home and has kids and likes to exercise outside and it is going to rain, then she gets a recommendation that she find something to do with her kids inside over the weekend.” He believes these technologies can maximize the fit between various treatments and the needs of individuals. They also could be used, he observed, to develop community-level interventions. For example, he said, spatial data could reveal pockets of risk or food deserts, and these conditions then could be ameliorated, or patients could be given tools to overcome them.
Bennett and his colleagues are using self-weighing in the context of a multicomponent weight loss intervention in primary care. He explained that self-weighing produces substantial weight loss on average, but its effectiveness declines over the course of treatment. He observed that artificial intelligence can identify when people are likely to become disengaged, deliver appropriate prompts, and get them back on track. At the same time, he noted, the use of different technologies can allow people to receive information in a way they prefer. He expects to see increasingly more digital therapeutics as a result of the promising outcomes demonstrated in academic and commercial trials.
Bennett continued by observing that many groups have become interested in the potential of digital technologies to help bend the health care cost curve. For example, he said, successful treatment models related to diabetes are providing a window for other conditions and are making companies interested in developing digital treatments. “This is a good sign of where things are likely to go in the next 5 or so years,” he suggested.
Treatments tend to be better when humans are involved, Bennett acknowledged, but new technologies provide many different ways of getting humans involved in obesity treatment, whether through in-person care, telephone contacts, text messages, or other ways. “Adding humans to technological approaches tends to produce better outcomes,” he said. “Technology serves, in my view, to improve the efficiency and the quality of human-delivered care. It doesn’t supplant it.”
Bennett asserted that primary care physicians and other providers will continue to play an important role in triggering the cascade of different
care options, coordinating care, and holding the patient accountable, but that digital technologies can help providers do their jobs. For example, he described a technology that tracked patients who were using a new app and created short counseling recommendations that were inserted into a patient’s electronic health record for use by a provider. The recommendation might be to continue participation in a study, work on portion control, or avoid sugary drinks—simple, straightforward, basic advice that could be read to a patient. When providers counseled a patient on diet or exercise in a generic way, Bennett noted, the patients tended not to lose weight. However, if the providers counseled patients using the recommendations generated by the app, the patients lost a substantial amount of weight on average. “We need to armor providers to do what they do best,” Bennett argued.
Bennett observed that such approaches can especially benefit patients at highest risk. For example, he said, many people on the lower end of the socioeconomic spectrum and many racial and ethnic minority groups are likely to own and use smartphones. He has done much of his work with disadvantaged and medically vulnerable patient populations, often in community health centers in rural settings in North Carolina, and he noted that these patients engage at a high level and benefit in ways similar to what is seen in other populations. “It is an ideal approach for reaching into historically disconnected populations,” he added, “and allowing us to ensure that we are delivering comprehensive care from providers in the way that suits them best.”
This page intentionally left blank.