Obesity has been one of those therapeutic areas that drug companies can't quite stay away from. A glance at the potential patient ranks, expanding in every sense of the word, is enough to explain why. But as I've detailed here in the past, finding an effective obesity drug is not easy, for many good reasons.
One response has been that such a drug wouldn't have to work spectacularly well to be effective, since calories add up so alarmingly. The "one cookie a day" model of weight gain is often referenced. But just how useful is it? According to this paper in JAMA (PDF), not so much (emphasis added):
How much weight would an individual gain by eating an extra chocolate chip cookie every day for life? One approach to answering this question, frequently used in textbooks and scientific articles, is based on the assumption that a pound (454 g) of fat tissue has about 3500 kilocalories (kcal). Thus, a daily 60-kcal cookie would be expected to produce 0.2 kg (0.5 lb) weight gain in a month, 2.7 kg (6 lb) in a year, 27 kg (60 lb) in a decade, and many hundreds of pounds in a lifetime. This of course does not happen. . .
So what does happen? If a person takes in a steady excess of calories, they most certainly will gain weight. But some of those extra calories then go into maintaining (and carrying around) that new weight. The result is a slow climb up to a new equilibrium weight. The 60-calorie cookie example above would be expected to lead to about 6 lbs of total weight gain over a period of years. Most of that will be put on early, with an asymptotic rise to the final value.
Now, if you're trying to avoid gaining weight, this is probably good news. The effects of having some extra food now and then aren't quite as catastrophic as the usual calculations would make you think (although there is that faster-weight-gain-at-first effect, so you have to separate the short-term and long-term consequences a bit). For someone that's significantly overweight and wants to lose it, the implications are mixed. One way to look at it is that the maintenance costs of extra body mass are substantial, meaning that a person doesn't have to suddenly go on a 1400-calorie-a-day diet to see results. Another thing this paper tells us that significantly overweight people have gotten to that point only by having significant calorie imbalances. A cookie a day is not going to do it. A lot of people are taking in a lot more excess calories than that:
These calculations suggest that small changes in lifestyle would have a minor effect on obesity prevention. Walking an extra mile a day expends, roughly, an additional 60 kcal compared with resting—equal to the energy in a small cookie. Physiological considerations suggest that the apparent energy imbalance for much of the US population is 5- to 10- fold greater, far beyond the ability of most individuals to address on a personal level. . .
The authors go on to call from greater regulation of the food supply, greater public health efforts, and so on, but I have to say that I'm skeptical of the ability of these to do the trick (or, given my political leanings, of the amount of coercion that might be needed to make them really work). It's clear that many people would like to be less overweight than they are, but (revealed preference time), it's also clear that they have enjoyed running up their caloric imbalances even more. So while the authors may be right that significant weight loss may be beyond most people's ability to address on a personal level, it's hard for me to see any other level that's really going to work.
So what would help? Would an anti-obesity drug? On the one hand, a good one might alter the physiological underpinnings of weight gain (and weight loss) in such a way to help people escape the squirrel-cage aspects of dieting and rebound weight gain. But that's asking a lot, and I remain uncertain about whether a really good weight-loss drug is even possible. The other difficulty is that any such treatment still has to be coupled with all those things that patients don't want to hear about: less food, more exercise. The time course needed is also a hard sell, particularly when you consider all the scam ads that bombard everyone. It seems clear that any sustainable weight loss probably should take about as much time as the initial weight gain did, which is not what many people are wanting to hear, particularly when someone else is promising umpteen pounds a week of magic weight loss.
Then there's the commercial consideration: that a new approved obesity drug, regardless of whether it works very well in the real world or not, would nonetheless sell like crazy. And if it doesn't work quite as well as the rigorously controlled clinical trials indicated it would, well, you can always blame the patients themselves (and for that matter, in many cases you'd be right). It's enough to make me think that the whole therapeutic area is a moral hazard.