Treatment strategies for obesity
Professor JPH Wilding, Diabetes and Endocrinology Research Group, University Hospital Aintree, Clinical Sciences Centre, Longmoor Lane, Liverpool L9 7AL, UK. E-mail: email@example.com
Given the scale of the obesity epidemic and the profound implications for public health, it is essential that any strategies adopted for treatment are directed towards those at greatest risk of the medical complications associated with the condition. The aim of this review is to provide some insight into how such a strategy might be developed over the next four decades, also taking into consideration developments in therapy and delivery of care that might occur over the same timescale.
The likely lead times for new drug development and changes in clinical practice that result from ongoing studies are relatively long, so predicting developments over the next 10 years can be extrapolated to some extent from what is currently known about existing treatments, drugs currently in development and ongoing clinical studies that have the potential to change practice. The first part of this review therefore begins with a brief overview of the current situation and extrapolate current trends forwards to the next 10 years; the second section is more speculative and considers how treatment might change as scientific knowledge about obesity and its treatment develops in the subsequent 30 years.
Obesity treatment: current position
Important principles of obesity treatment
Treatment of obesity is a challenge to medical practitioners and other healthcare workers. Being obese is frequently seen as the patient’s own ‘fault’, and practitioners are rarely trained in obesity and its management. So treatment is often started in the absence of an appropriate support programme and is frequently unsuccessful. Such failure serves to reinforce the patient’s and practitioner’s view that treatment is not worthwhile. Effective programmes offer ongoing support, with realistic weight loss goals and monitoring of the effects of treatment. Long-term follow-up is essential if weight regain is to be avoided (1,2).
Weight loss goals
Achievement of normal or ideal body weight is not a necessary goal in the management of obesity, and is rarely reached in practice. There is evidence from epidemiological studies of intentional weight loss, including several long-term trials of four or more years’ duration, that modest weight loss, in the order of 5–10% from presentation weight, is associated with clinically worthwhile reductions in comorbidities, such as hypertension, dyslipidaemia and diabetes risk (3–6). Larger weight loss goals may be appropriate in extremely obese patients who are contemplating surgical interventions, but even with surgery, attainment of ideal weight is uncommon as weight tends to be regained over time (7). In some patients, particularly in those with severe comorbidity, prevention of weight gain may be a reasonable aim of treatment.
Aims of treatment
Although weight loss is clearly one aim of therapy, the main reasons for advocating weight loss are to help people improve their health, and so monitoring of the effects of treatment on risk factors and comorbidities is important. Because other treatments are often used to help manage comorbidity, the need for these should be reviewed on a regular basis as well – successful treatment should help reduce the need for other treatments. The aims of therapy should be tailored to the individual patient as much as possible.
A number of dietary approaches have been advocated for the treatment of obesity. Recent evidence-based reviews support the use of low-calorie diets, energy-deficit diets and diets that are low in fat as being most likely to be effective for modest weight loss (8). Very-low-calorie diets produce greater short-term weight loss, but weight regain is common.
Although modest physical activity has undoubted health benefits and can contribute to weight loss, it is not usually advocated as a sole treatment option. Many studies, however, do suggest that it can be helpful to improve weight loss maintenance, although activity levels equivalent to 45–60 min of brisk walking each day may be needed to achieve this (9).
Behavioural approaches aim to help subjects to implement and sustain changes to their eating and activity behaviour and require trained health professionals with good interpersonal skills to use the approach appropriately. There is strong evidence that combining a behavioural approach with more traditional dietary and activity advice leads to improved short-term weight loss and is currently the most effective lifestyle approach to managing weight (10). However, these studies are of relatively short duration, so the evidence base is limited to 1 year at present.
Combined approaches: lifestyle interventions
A number of large-scale clinical trials have demonstrated the effectiveness of lifestyle interventions in specific groups of patients, particularly those with impaired glucose tolerance, but there is also some evidence for benefits in those with hypertension and dyslipidaemia. In general, weight loss with these approaches is modest (about 4 kg or 4% of body weight on average).
Three drugs are currently licensed in the UK for the treatment of obesity and available for use on the National Health Service. These are orlistat, an inhibitor of gastrointestinal lipases, and two centrally acting drugs, sibutramine and rimonabant. These agents have all been shown to produce greater weight loss than lifestyle interventions in clinical trials and, importantly, to help with weight maintenance once weight loss has finished. Success rates at achieving weight loss goals of 5% or 10% of baseline weight are generally two to three times greater than placebo (11–13). Their use has also been associated with improvement in risk factors, such as dyslipidaemia, hypertension and diabetes control, that in some instances may be partly independent of weight loss (14,15). Only orlistat has published clinical trial data supporting effectiveness beyond 2 years. The 4-year XENical in the prevention of Diabetes in Obese Subjects (XENDOS) study also showed that type 2 diabetes could be delayed or prevented in the subgroup of patients with impaired glucose tolerance (16). At present, it is not known whether any of these drug treatments is clinically or economically cost-effective as a strategy to support weight loss and long-term weight maintenance in terms of hard outcomes such as improvement in vascular disease or other obesity-related problems and there are few published data in conditions such as arthritis, fertility, respiratory diseases or quality of life.
Surgical treatments are currently only recommended for patients with severe obesity [body mass index (BMI) of >40 kg m−2 or >35 kg m−2 with comorbidity] who have not been able to lose weight after trying other available treatments. Most surgical treatment is now carried out laparoscopically. Three approaches are widely used.
Laparoscopic gastric banding
This is the simplest procedure and involves placing an adjustable silicone band around the upper part of the stomach, creating a low-volume pouch with a small outlet. This mainly works by restricting how much food patients can eat. The average weight loss is around 15–20% of body weight, although some weight regain occurs over time. Morbidity and mortality are relatively low (mortality <0.2%), but patients do need to return for band adjustments.
Gastric bypass (laparoscopic or open)
This involves creating a small-volume gastric pouch and producing a Roux-en-Y diversion so that food bypasses the duodenum and upper jejunum. This works by both restricting food intake and causing a modest degree of malabsorption. Weight loss is generally greater than with the band. Operative mortality is <0.2% for laparoscopic procedures and 0.5% for open procedures.
A variant of the older biliopancreatic diversion, this involves a partial (sleeve gastrectomy) and bypass of a long-loop of jejunum. Weight loss is greatest with this procedure, but malabsorption is more likely and patients need to pay very careful attention to their diet and require vitamin and mineral supplementation.
Most of the surgical literature is based on large case series, but one ongoing study (the Swedish Obese Subjects) study has provided 10-year outcome data for many complications compared with a contemporary non-operated control group. Diabetes incidence is substantially reduced and there are improvements in lipids and sleep apnoea. Although blood pressure improved initially, there was no difference at 10 years (6).
Genetic testing leading to patient-specific treatment
This is currently largely a research-based activity, and even the more common conditions are rarely encountered in routine clinical practice. There are a number of well-defined genetic abnormalities that directly lead to severe, early onset obesity, including MC-4 receptor mutations, pro-opiomelanocortin mutations, pro-hormone convertase-1 mutations, defects in leptin or its receptor and several syndromic conditions, such as Prader–Willi syndrome (17). Only one of these, leptin deficiency, is amenable to specific treatment, but with dramatic clinic effect (18).
Ameliorating the consequences of obesity
It should be recognized that significant advances have been made in the past three decades using medical treatments (drugs, devices and surgery) that ameliorate or treat the medical consequences of obesity. Examples include drugs used to treat vascular risk factors such as hypertension, dyslipidaemia and aspirin, treatments for type 2 diabetes, analgesics and joint replacements for arthritis, fertility treatments, drugs and surgical treatments for gastro-oesophageal reflux, continuous positive airways pressure ventilation for sleep apnoea. As these treatments are in general considered clinically and cost-effective, it will be necessary to show that management of body weight can effectively reduce the utilization of these treatments in a cost-effective way.
Obesity treatment in children and adolescents
Although obesity is being increasingly recognized as a problem in children, and a number of rare genetic causes have recently been characterized, current treatment approaches based on diet and increased physical activity are relatively ineffective. Emerging evidence on the use of drugs in adolescents is leading to some use in specialist centres. Very occasionally, obese adolescents are considered for surgery. It seems likely that more children will need treatment in the future, and that the evidence base for treatment will expand as more is known about the natural history of childhood obesity, and as more effective treatments are developed.
Obesity treatment: the next 10 years
As increasing emphasis is put on recording BMI and waist circumference in primary care, it seems inevitable that this will lead to a greater recognition of obesity and its comorbidities. The area of greatest concern is likely to be type 2 diabetes and pre-diabetic conditions, where there is now a good evidence base that lifestyle intervention is effective (19). This is likely to lead to increased need for treatment for patients and this will need to be managed in a sensitive and cost-effective manner.
Development of risk assessment tools
Given the already high (and increasing) prevalence of obesity in the population, it will not be possible or cost-effective to provide intensive lifestyle intervention, with or without pharmacotherapy, to all. There is an urgent need to develop risk assessment tools that can be used to help clinicians decide who is likely to benefit from interventions to control body weight. This will need to be of greater scope and more flexible than the cardiovascular risk assessment tools already available, as with obesity it is necessary to consider a broader view of interventions. Priority groups with cardiac/metabolic conditions are likely to include those with pre-diabetic conditions, diabetes and/or pre-existing vascular disease. Other conditions that may benefit from intensive intervention include those with osteoarthritis, gastro-oesophageal reflux, sleep apnoea and obesity-related infertility. Severely obese patients may also require priority for assessment and treatment. Tools to assess the severity of these conditions could be used to monitor the response to treatment.
It seems unlikely that lifestyle approaches will undergo substantial changes in the next 10 years, although the underlying evidence base may improve. One area of active development is in the area of nutraceuticals. Such developments may eventually make adherence to an energy-controlled diet easier, and could also help ameliorate some obesity-related comorbidities.
Ongoing studies that may change practice
This is an ongoing trial of intensive lifestyle intervention aimed at weight loss in new onset type 2 diabetes, which is being conducted in the USA and funded by the National Institutes of Health. To date, 5144 participants have been recruited and the study has been running for nearly 2 years. The main outcomes are those related to vascular disease after 11.5 years’ treatment. If this study shows some benefit, this will have very clear implications for the management of newly diagnosed patients with type 2 diabetes.
Sibutramine cardiovascular outcomes trial
This study has recruited from 16 countries over 9000 patients at high risk of vascular events, including many with type 2 diabetes and pre-existing vascular disease. It is a trial of lifestyle treatment vs. sibutramine plus lifestyle treatment over 4–5 years. Results are expected in 2008. If this shows a long-term benefit of a weight loss strategy that includes sibutramine in these high-risk patients, it will have major implications for how weight loss treatment is seen in the context of the management of high-risk patients.
This study is funded by pharmaceutical company Sanofi-Aventis and will look at the effects of rimonabant on vascular events in high-risk patients over 5 years. It is likely to report in 2010 or 2011.
This study, also funded by Sanofi-Aventis, is a diabetes prevention study comparing rimonabant with lifestyle in patients with impaired glucose tolerance. It is likely to report in 2009.
Swedish Obese Subjects study
This is a case–control observational study comparing surgery to routine care in morbid obesity. Preliminary data have already been reported. Final outcomes, with effects on morbidity and mortality after 18 years of follow-up, were reported in abstract form in 2006, and showed significant reductions in overall and cardiovascular mortality. Once these results are widely disseminated, this is likely to lead to increased use of this treatment.
Other ongoing trials
A search of the US clinical trials website http://clinicaltrials.gov/ct/gui/action/GetStudy (accessed October 2006) reveals 214 ongoing trials and studies in obesity, so the evidence base for treatment is likely to increase substantially in the next 5–10 years. These include studies comparing different diet and exercise regimes, novel methods of delivery (for example, telemedicine and use of the internet), new drugs, and comparisons between different surgical procedures.
New drug treatments
There are four broad areas that provide valid drug targets for the treatment of obesity:
- • drugs that inhibit absorption of nutrients;
- • drugs that mimic or enhance peripheral satiety or adiposity signals;
- • drugs that alter metabolic rate or substrate utilization;
- • drugs acting at central nervous system (CNS) targets that result in altered energy intake or energy expenditure.
Some agents may have multiple modes of action that include more than one of these broad mechanisms. Potential targets in each of these groups are discussed below.
Inhibition of nutrient absorption
A new intestinal lipase inhibitor, cetilistat is currently in development. Preliminary results from phase 2 studies suggest that it has equivalent efficacy to orlistat, but with a better adverse-effect profile (20). This could become widely used if results from phase 3 studies support these preliminary data. Other potential mechanisms might include agents that interfere with lipid or other nutrient absorption across the intestinal brush-border membrane.
Enhancement of peripheral satiety or adiposity signals
There is no licensed drug in this group at present, although it seems a promising area for future development. Potential targets include gut hormones such as GLP-1, cholecystokinin (CCK), peptide YY (3-36), obestatin and oxyntomodulin (see contribution from Bloom). Leptin itself has not lived up to its early promise except in those rare individuals with leptin deficiency, but enhancement of leptin action in the CNS remains a possibility.
GLP-1 is a gut hormone that is secreted by intestinal L-cells. It is a potent stimulus to insulin secretion, and its main physiological role is thought to be as an incretin hormone (21). GLP-1 has also been shown to reduce food intake in preclinical studies, an effect that appears to be mediated via receptors in the brain stem and hypothalamus, and is due to enhancement of satiety rather than induction of anorexia or by producing non-specific illness (22). GLP-1 and its analogues also slow gastric emptying, an effect that may also confound measurements of hunger and satiety (23). Several human studies (reviewed in Verdich et al. (24)) have shown that infusion of GLP-1 into healthy volunteers, obese subjects and people with type 2 diabetes may acutely reduce food intake by up to 35%. Exenatide (synthetic exendin-4) is a long-acting GLP-1 analogue used for the treatment of type 2 diabetes (25–27). The glucose-lowering effects of exenatide mainly occur because of its incretin mimetic effects – stimulation of insulin secretion and suppression of glucagon secretion. Unlike most other drugs used to treat type 2 diabetes, exenatide also results in weight loss that may contribute to its antihyperglycaemic effect. Despite this evidence, the use of GLP-1 analogues has not been explored as a specific treatment for obesity.
Oxyntomodulin is a further product produced from the L-cells from the preproglucagon precursor. It has been shown to decrease food intake in rodents and humans, and appears to act at the same receptor site as GLP-1 (28,29). One short-term study that involved thrice-daily subcutaneous administration in obese people has suggested that these effects may result in weight loss (30). The results of longer-term trials are awaited.
Cholecystokinin may be considered the archetypal satiety signal. It is released after meals and signals via the vagus nerve to receptors in the brainstem. Administration of CCK to rodents and humans decreases food intake, and blockade of CCK-A receptors peripherally and CCK-B receptors in the CNS increases energy intake (31,32). CCK receptor agonists have been developed, but have not been demonstrated to be effective in clinical trials to date.
Two stomach-derived hormones, ghrelin and obestatin, are also worthy of mention. These are produced by differential splicing from the preproghrelin gene. Ghrelin acts via the growth hormone secretagogue receptor in the hypothalamus and is a potent stimulus to food intake; plasma concentrations rise in anticipation of meals and fall after food ingestion (33). However, circulating concentrations are low in obese individuals (34). It remains to be seen whether ghrelin antagonism will reduce food intake in obese subjects. Obestatin is more recently discovered. It does reduce food intake in rodents, but its relevance to humans is currently unknown (35).
Peptide YY (3-36) is released from K-cells in the large intestine (36). Its concentrations rise rapidly after ingestion of food, and it is thought to act on the neuropeptide Y Y2 receptors in the hypothalamus, to suppress NPY secretion and thus reduce food intake. PYY concentrations are low in obesity, and PYY infusion decreases food intake in lean and obese subjects (37). Further investigation of its potential to treat obesity is underway.
Drugs that alter metabolism/substrate utilization
This area has remained one of active interest, although there is no licensed drug that works exclusively via a peripheral metabolic mechanism. Sibutramine probably has some thermogenic activity that contributes to its efficacy. Drugs acting purely as thermogenic agents include β3 adrenoceptor agonists, and enhancers of uncoupling protein action in mitochondria (38). Other potential targets in this area include activators of AMP-kinase, drugs that alter fatty acid or triglyceride synthesis and modification of peripheral steroid metabolism (39). Some drugs may also have peripheral metabolic effects that are independent of weight loss; examples include the effects of sibutramine and rimonabant on lipid metabolism.
Drugs acting on central nervous system targets that affect energy balance
The CNS is a rich source of potential targets for the treatment of obesity. Over 50 peptide and classical neurotransmitters have been identified that influence energy balance and most of them have been considered or tested in preclinical models at some point – nevertheless, it has proved difficult to translate this into drugs that can proceed into testing in humans. These can be broadly divided into transmitters that decrease energy intake and maybe also increase energy expenditure and those that increase energy intake and/or decrease energy expenditure – the aim is of course to develop agonists of the former and antagonists of the latter. The most promising target in this group at present is the cannabinoid receptor system – the CB1 receptor antagonist rimonabant has completed phase 3 trials, and is now available for clinical use in Europe. Many other possible targets are under active investigation. These include neuropeptide Y Y5 receptor antagonists (40–42), galanin antagonists (43), ghrelin receptor antagonists or inverse agonists (this is a peripheral hormone that increases food intake via a CNS receptor) (44,45) and melanin-concentrating hormone antagonists (46–48). Agonism of receptors that are involved in satiety or reduction of food intake is also possible. Classical targets include the serotonin and nor-adrenaline systems – this is the target for the licensed reuptake inhibitor sibutramine. Peptide targets include melanocortin 4 receptors, CART, neurotensin, GLP-1 and many others (22,49–51).
Combining drug treatment
At present, there is very little evidence available about the possibilities of combining different drug treatments with different modes of action or using pharmacotherapy to prevent weight regain after bariatric surgery. Such approaches are now routine in the management of other chronic conditions (e.g. diabetes, hypertension, dyslipidaemia, asthma, rheumatoid disease). As new drugs become available, it is likely that combination therapy will be tried, and if successful, this could result in much greater weight loss and larger reductions in obesity-related comorbidity.
Genetic testing and individualized treatment
This is likely to become more widely available, especially for severe, early onset childhood obesity. However, in practice, this is unlikely to lead to changes until specific therapy to treat some of the more common defects, such as melanocortin-4 receptor mutations, becomes available. This seems relatively unlikely on a 10-year timescale. Use of pharmacogenetic testing outside of the research arena may begin to have an impact, but this will depend on whether such studies reveal clinically useful information that can predict responses to specific therapies.
Ameliorating the consequences of obesity
As cardiovascular risk factor management becomes more sophisticated (with use of high doses of statins, ACE inhibitors, better control of blood pressure in hypertension and better treatment of hyperglycaemia in diabetes), the potential additional benefits from weight loss in patients at high risk of cardiovascular disease may diminish. It is also possible that drugs, such as PPAR gamma agonists, GLP-1 analogues and DPP-IV inhibitors, that have potential effects to preserve β-cell function will be shown to alter the natural history of type 2 diabetes. This will again serve to reduce the additional benefits seen with weight loss.
Obesity treatment (10–40 years ahead)
It is, of course, hoped that public health measures will have started to have an effect on the epidemic in 10 years’ time, but even if the year-on-year increase in obesity is slowed or reversed, there is still likely to be a major burden of obesity and its complications on society.
Diet and lifestyle
As awareness of the obesity problem and potential solutions increase, and recording and collection of data becomes routine, it should be easier to track the progress of individuals and offer lifestyle counselling and other treatments early in the course of weight gain, so as to prevent people reaching the more severe degrees of obesity.
It is to be expected that more targets will be identified, and the development of pharmacogenetics may allow much more individualized targeting of treatment that is likely to include more than one drug to achieve optimal effects. The various pathways and potential targets have already been discussed, but at present it is not possible to predict which of these are most likely to enter clinical practice, or which combinations are most likely to be effective.
Although surgical treatment is likely to remain an important part of treatment of morbid obesity for some time to come, it should be recognized that procedures, such as gastric bypass, work at least in part by modulating endogenous signals from the GI tract, which may be tractable via pharmacotherapy. It is possible that combinations of drugs and lifestyle, particularly if used early in those with rapid weight gain trajectories, may eventually make surgical therapy obsolete.
Most of the major single-gene disorders will be known and should be identifiable either early in childhood when it is clear that there is an abnormal rate of weight gain, or even at birth in some cases if the technology develops to the point that such multiple testing is feasible and cost-effective.
Ameliorating the consequences of obesity
As medical therapy for conditions such as diabetes improve, the diagnosis may become less of a burden. Much will depend here on whether some treatments that are available now (such as thiazolidinediones, incretin mimetics and DPP-IV inhibitors for diabetes) live up to their early promise, and are able to alter the natural history of the condition. Other areas to watch from this perspective are agents that may prolong life expectancy by mimicking the life-prolonging effects of calorie restriction; examples here include sirtuins (activators of SIRT-1 receptors), insulin sensitisers and inhibitors of glycolysis (52). Ameliorating the mechanical consequences of severe obesity may turn out to be a more difficult challenge, although it is reassuring that for some conditions at least, the presence of less severe obesity has little effect on outcomes (53).
Conflict of Interest Statement
The author has received honoraria, consultancy fees and research support from several pharmaceutical companies involved in the marketing and development of drugs for obesity treatment.