It is no news that obesity is bad for your health. But it is only in the past ten years or so that it has been recognized as a major contributor to cancer risk – According to Craig Thompson (Sloane-Kettering Institute), speaking on the first day of the inaugural BioMed Central conference on Metabolism, diet and disease, held at the end of May in Washington DC, it has now overtaken smoking as the primary preventable cause of cancer, accounting for an estimated 25% of cases.

The cancer-obesity connection was the topic of the panel discussion at the conclusion of the conference – now accessible as an online video. Here is an idea of what the panel of experts in oncology, nutrition, endocrinology, epidemiology and policy, had to say.

Gregory Petsko (Brandeis University), chairing, started the discussion by asking whether there is anything we can we do about obesity-related cancer short of tackling obesity itself.

Insulin and carbohydrates
Acknowledging that eliminating obesity is not an answer in the short term, the panel turned to the role of  insulin and the potential of specific anti-insulin drugs (Michael Pollak, McGill University), with Richard Bergman (Cedars-Sinai Medical Center, Los Angeles), urging that clinicians should be measuring insulin and encouraging patients to take more exercise to lower it, and perhaps prescribing drugs to increase insulin sensitivity – to which Gary Taubes (Berkeley) objected that it would surely be better to lower dietary carbohydrates to achieve the same end – and thus opened further questions.

Do carbohydrates contribute any more to metabolic disease than any other food category (yes, probably, said Stephen Hursting, University of Texas, and Jeff Volek, University of Connecticut) – and if insulin causes cancer cells to grow, should we be using insulin-sensitizing drugs for cancer? – a question that Michael Pollak raised in order give the answer that it all depends how you are achieving the sensitization. This prompted Barbara Kahn (Harvard Medical School) to point out the need to define what we mean by increasing insulin sensitivity, which is not the same as changing insulin signalling – and that promoting insulin signalling can have negative or positive effects on cancer, depending on which insulin signalling pathway is activated. Clearly, though not surprisingly, the issues aren’t simple.

Enter the elephant
About halfway through the discussion, the elephant lumbered out of the shadows, and when Leonard Guarente (Massachusetts Institute of Technology) pushed the point that we cannot ignore the need to tackle the obesity that is the root of the problem, a questioner from the floor asked why something isn’t done about the sale and promotion of high-carbohydrate, cheap foods and ubiquitous high-sugar drinks.

This led Petsko to confront the perverse economic incentives that are encouraging us to eat too much of the wrong things and Pollak to raise the counter-incentive of unsustainable health-care costs; and Linda Nebeling (National Cancer Institute, but representing her own views and not those of the Institute) to point out that although a recurrent theme of the meeting was the difficulty of changing people’s behaviour, the history of smoking shows that it can be done, even in the face of perverse economic incentives.

What do we mean by obesity? – And other crucial questions
Meanwhile it is still essential to understand the biology underlying the problems, not only because we cannot eliminate, or even reduce obesity in the short term, but because it may be a much larger issue than we think. When Hursting made the point that it is not the obesity itself that causes problems, but the metabolic perturbations that arise from it, it became clear that we do not have an adequate definition of obesity. Pollak observed that it is possible to have the metabolic features of obesity without being obese on the criterion of body mass index (BMI), and Rudolf Kaaks (German Cancer Research Center, Heidelberg) added that BMI doesn’t reliably measure excess adiposity, there is no distinct and defined obese population but a continuum, and the health burden of excess adiposity may be grossly underestimated – so we need better measures. And a better understanding of how adiposity is related to metabolic disruption – Metabolism can settle down in response to dietary and surgical interventions well before any detectable weight loss (Volek, Hursting).

The discussion concluded with the vexed issue (once again) of carbohydrates versus calories – All agreed that significant calorie restriction is hard to achieve and restricting a single food class may be easier, and there was general consensus that carbohydrates probably have greater adverse metabolic effects than other food classes, and full consensus that this needs to be properly researched.

Clearly, many fundamental questions remain open, both about obesity and its role in cancer, and about the contribution of diet to each. The conference on Metabolism, diet and disease was hosted by BMC Biology and BMC Medicine, which welcome submissions to their article series that may help to answer these questions; and the panel discussion was sponsored by a new BioMed Central journal, Cancer and Metabolism, now accepting submissions which will be welcomed by its Editors-in-Chief Chi Van Dang and Michael Pollak.

Miranda Robertson                             Sabina Alam
Editor, BMC Biology                            Editor, BMC Medicine