Pharmacological approaches to weight loss

Obesity is the most important modifiable risk factor for type 2 diabetes (Abdullah et al, 2010), and large-scale randomised controlled trials (RCTs) have demonstrated that lifestyle interventions incorporating weight loss and increased physical activity are effective at reducing the incidence of diabetes in individuals at high risk over the long term (Lindström et al, 2013; Diabetes Prevention Program Research Group, 2015).

However, in interventions undertaken under more pragmatic, “real-world” conditions, the extent of weight loss observed, while still clinically significant in general, has been only 30–50% of that achieved in the large-scale clinical efficacy trials (Dunkley et al, 2014). In this context, the potential of pharmacotherapy to enhance the effects of lifestyle intervention on weight loss and diabetes prevention warrants consideration. This has been tested in two recent trials.

In a phase IIa, double-blind, placebo-controlled RCT (summarised on page 44), Hollander and colleagues assessed the efficacy and safety of canagliflozin and phentermine, both as monotherapies and in combination, on top of a standardised lifestyle intervention (a 600-kcal/day energy-deficit diet and 150 minutes of exercise per week). Weight loss and safety outcomes were assessed at 26 weeks in 335 overweight or obese adults without T2D.

Participants were randomised 1:1:1:1 into placebo, canagliflozin, phentermine and canagliflozin plus phentermine intervention arms. The rationale was that the two drugs would potentially induce weight loss via complementary mechanisms: canagliflozin, a sodium–glucose cotransporter 2 inhibitor, would aid weight reduction via calorie loss through urinary glucose excretion, while phentermine, a sympathomimetic amine anorectic, would act by increasing satiety and thereby reduce energy intake.

Mean weight loss was 0.6% of body weight with placebo, 1.9% with canagliflozin, 4.1% with phentermine and 7.3% with the canagliflozin/phentermine combination. Furthermore, 66.7% of participants in the combination group achieved at least 5% weight loss, compared to 17.5% on placebo, 17.9% on canagliflozin and 41.7% on phentermine. The canagliflozin/phentermine combination, but not the individual drugs on their own, led to a significant reduction in systolic blood pressure compared with placebo (by 4.2 mmHg), but there was a significant increase in heart rate in the phentermine and canagliflozin/phentermine groups.

Intriguingly, the extent of weight loss (and blood pressure reduction) with coadministration of canagliflozin and phentermine was greater than the additive effects of canagliflozin and phentermine monotherapy, which the authors hypothesised may be due to the phentermine’s effects on appetite and satiety acting synergistically to offset potential adaptive increases in energy intake in response to increased urinary glucose excretion with canagliflozin.

Longer-term studies are needed to determine the effects on cardiovascular and diabetes outcomes, but these data suggest that coadministration of canagliflozin and phentermine may be a viable approach for long-term weight management.

In a longer-term study (also summarised), le Roux and colleagues undertook a 3-year, double-blind RCT in 2254 people with prediabetes. They were randomised 2:1 to liraglutide or placebo in addition to lifestyle intervention (a 500-kcal/day energy-deficit diet and 150 minutes per week of physical activity). The time to onset of type 2 diabetes and weight loss were assessed.

The probability of developing type 2 diabetes over the 160-week follow-up was 3% in the liraglutide group and 11% in the placebo group, with time to diabetes onset while on treatment being 2.7 times longer with liraglutide than placebo. Mean weight loss was greater in the liraglutide group (6.1% of body weight) than the placebo group (1.9%); thus, the relative contributions of liraglutide’s effect on glucose metabolism and its effect on body weight to the diabetes risk reduction are unclear and warrant further investigation. Interestingly, most individuals who developed type 2 diabetes lost less weight than the treatment group’s mean.

Thus, there is increasing evidence that pharmacotherapy in addition to lifestyle intervention can enhance weight loss and, in the case of liraglutide, reduce or delay the incidence of type 2 diabetes. Whether such therapies gain widespread use for weight loss and diabetes prevention will depend on cost-effectiveness, available resources and views on the extent to which prevention, as opposed to treatment, of type 2 diabetes should be medicalised.

To read the article summaries, please download the PDF