Robert H. Lustig, M.D., Professor of Clinical Pediatrics
University of California, San Francisco
Obesity has rapidly become the largest (both figuratively and literally) public health problem, both in the U.S. and in numerous Westernized countries. The causes of this epidemic are myriad, but they still must obey Newton's First Law of Thermodynamics, which states that energy can neither be created nor destroyed. In human terms, "If you eat it, you had better burn it, or you're going to store it." Most social critics point to increased energy intake due to high-fat foods and soft drinks, and decreased energy expenditure due to television and cutbacks in physical education programs, as being the basis of the increase in obesity. These "behavioral" changes are thought to account for the majority of affected individuals.
However, there are now genetic, biochemical, and hormonal syndromes that also manifest with obesity. Although rare, these syndromes give physicians and scientists a window on how and why the body stores energy, rather than burns it. These syndromes are also providing physicians with possible new targets for obesity treatment.
An example of such an "organic" form of obesity is seen in patients who suffer damage to an area at the base of the brain called the hypothalamus. This area, the size of a fingernail, whose primary function is to release hormones which control the function of the pituitary gland immediately below it, also controls energy balance. Damage to this area has long been known to promote excessive eating (hyperphagia) and weight gain, termed "hypothalamic obesity." This form of weight gain is not responsive to diet and exercise. Victims of this form of obesity continue to gain weight despite their best efforts, and suffer greatly from this unrelenting scourge.
Hypothalamic obesity is an unfortunate complication in some survivors of brain tumors, especially those diagnosed in childhood. For instance, it has been estimated that as many as one-third of all the survivors of craniopharyngioma's develop severe obesity after diagnosis and treatment. At St. Jude Children's Research Hospital in Memphis, Tennessee, our team recently reviewed the Body Mass Index (BMI) curves (weight adjusted for height) of 148 children who survived brain tumors for at least five years, to see if we could determine risk factors for the development of future obesity in this vulnerable population. We looked at the roles of tumor location, tumor histology (type), extent of surgery, use of glucocorticoids (which can also cause weight gain), radiation therapy, chemotherapy, and the development of hormone disturbances. We found relations between the development of obesity and: 1) tumor location (hypothalamus), 2) tumor histology (craniopharyngioma and hypothalamic astrocytoma), 3) having had surgery (either a biopsy and/or a gross total resection), 4) the amount of radiation directed at the hypothalamus (greater than 51 Gy), and 5) the presence of hypothalamic hormone disturbances. All of these risk factors inferred a degree of hypothalamic damage in their expression. Thus, we are led to believe that obesity in this setting is due directly to damage to the hypothalamus, whether it is from the tumor itself, or the surgery or radiation treatments thereof. This is important because it gives physicians an idea of who is at risk, so that potential preventative measures can be instituted early, and suggests methods for limiting damage to the hypothalamus during brain tumor treatment in the future.
A rat model of hypothalamic obesity has been available to scientists for approximately 50 years. Experimental damage to an area of the brain known as the ventromedial hypothalamus (VMH) triggers non-stop eating and weight gain in these rats. Even in the face of severe caloric restriction, animals with VMH lesions continue to gain weight, because their metabolism is geared toward energy storage instead of energy burning. Prior to the development of the severe obesity, it has been shown that VMH-damaged rats manifest increased insulin levels (which converts blood sugar into storage as fat). However, investigators have also found that cutting the nerve that leads to from the brain to the pancreas (called the vagus nerve) in rats prevents the hyperphagia, insulin rise, and weight gain. Thus, it has been proposed that the VMH may control energy balance by influencing insulin secretion through the vagus nerve. These data also suggest a potential target for therapy. While suppressing the function of the vagus nerve with drugs has numerous side effects, the use of drugs to suppress insulin production by the pancreas is easier, more specific, and less dangerous.
Our team has been working for the past five years to test this idea. Insulin release from the pancreas is controlled by several factors, including the vagus nerve. When the vagus nerve is activated, the pancreas releases more insulin into the bloodstream than it should. We sought to curtail this insulin "hypersecretion" in patients with hypothalamic obesity by administering a hormone called octreotide. Octreotide is a longacting derivative of the hormone somatostatin, which is a naturally occurring "brake" on insulin release within the pancreas. Octreotide attaches to the cells that make insulin, and limits the amount of insulin released during a meal. We postulated that, if we could reduce insulin secretion with octreotide, patients would eat less and lose weight. In 1977, we treated our first patient with octreotide for six months. During that time, she lost 28 lbs. She ate significantly less each day, according to her mother, and she became more physically active. Decreases in dietary intake and increases in physical activity preceded the weight loss. We tested her insulin production before and after octreotide therapy. Prior to treatment, her pancreas released approximately three times the normal amount of insulin during a glucose tolerance test. While receiving octreotide, her insulin response to glucose became normal. We treated a total of eight patients over six months. Three lost over 20 lbs., two lost approximately 10 lbs, and 3 maintained their weight, and, thus, did not gain more during treatment. All subjects reported decreases in appetite and increases in physical activity and stamina. Two of the patients remained on octreotide for one year and lost 48 and 36 lbs, respectively.
Since then, we conducted a double-blind, placebo-controlled trial of octreotide in pediatric hypothalamic obesity. In this study, some of the patients received octreotide, while others received a placebo, and neither the patients nor the treating doctors knew who was getting drug and who was getting placebo. Eighteen patients participated, and had, on average, a starting weight of 216Ibs., along with an average weight gain of 35 lbs/year. The nine subjects who received octreotide for six months stabilized their weight, while the other nine receiving placebo continued to have weight gain, adding another 20 lbs. Again, insulin responses decreased in the drug-treated group, but not in the placebo-treated group. Lastly, an objective quality-of-Iife questionnaire demonstrated improvements in physical, psychological, and social well-being in patients receiving octreotide, which related to their degree of insulin suppression. Thus, there is now suggestive evidence that insulin overproduction causes the weight gain in patients with hypothalamic obesity, and that treatment with octreotide initiates weight loss or prevents further weight gain in these individuals.
This mechanism for obesity can potentially be generalized to other patients, even those without hypothalamic damage, who might also respond to octreotide with weight loss. Some data suggest that the main factor in weight gain in adults is the initial insulin production that occurs immediately after eating. Therefore, we initiated a pilot study of octreotide therapy for adults with obesity, using the new, long-lasting preparation of octreotide, known as octreotide-LAR, given once a month, for a six-month period.
Forty-four severely obese, but otherwise healthy adults were enrolled. Patients were not directed to change their lifestyle and were given no instructions on beginning diet or exercise regimens. Of the 44 adult subjects treated with octreotide, eight had substantial weight loss of more than 10% of their total body weight. Those who responded decreased their carbohydrate intake by approximately two-thirds while on the medication, and made alternative food choices. Measurement of insulin levels revealed a clear relationship between insulin production and weight loss. Prior to treatment, the insulin levels of the responders rose quickly and then declined just as rapidly. This pattern disappeared with octreotide treatment. Thus, insulin overproduction was a good predictor of response to octreotide; that is, the higher the pretreatment insulin rise, the more weight that was lost. Furthermore, we showed a strong relationship between insulin overproduction and increased vagus nerve activity, suggesting that the vagus nerve was the cause of the insulin hypersecretion and the weight gain in these subjects.
Our research shows a cause, a mechanism, a diagnosis, and a treatment for one type of obesity. The VMH integrates (brain) and peripheral (pancreas) signals to control energy balance. VMH damage from tumors, surgery, or radiation leads to insulin overproduction, probably mediated through the vagus nerve. Children with hypothalamic obesity overproduce insulin, a process that results in sustained and sometimes unrelenting weight gain. Some obese adults (without brain tumors) also overproduce insulin" as a result of increased activity of the vagus nerve. Octreotide limits insulin release and thereby promotes weight loss, but only in some patients. Thus, all obesity is not the same. As our understanding of the cause of obesity improves, we will be able to improve our diagnostic efficiency and therapeutic success in patients who are victims of this unfortunate epidemic.