The link between overweight and obesity, and surviving head and neck cancers
Nicholas Timpson's research is a pilot study assessing direction and causality in the association between BMI and head and neck cancer survival
Topic: Head-Neck cancer
Institution: University of Bristol
Country: United Kingdom
Cancers of the head and neck (HNC) include different cancers in the mouth, throat and nose. When HNC patients are compared, those who are defined as overweight or obese appear to have better outcomes than those with normal or low weight. By being overweight at diagnosis, patients may be at an advantage because they have extra energy resources during treatment. We call this a ‘causal’ relationship.
Secondly, we know cancer causes a person to lose weight. Often, the earlier a cancer is detected the less weight the patient has already lost and so weight can be considered a marker of how early a cancer was diagnosed. If a patient goes on to survive better this may be because their cancer was picked up early, and not because they were heavier. This is referred to as ‘reverse causality’.
Finally, we may see this relationship because the relationship is biased. This could be because other factors that often go along with being very underweight (such as tobacco use) make patients do worse after treatment. It is important then that we identify which of the theories is correct. Head and Neck 5000 (HN5000) is a study which has identified and collected data on over 5,500 HNC patients, from their diagnosis to the present day. It is the largest study looking at cancers of this type and provides an excellent opportunity to investigate these problems.
Aims and objectives
We aim to identify if being overweight or obese improves or worsens HNC patients’ survival after treatment. Our objectives are to:
Use the HN5000 study to describe the difference between survival of those cancer patients who are underweight versus those who are overweight, and compare with other studies.
See if the relationship between weight and cancer survival is different across different cancers of the head and neck, such as mouth and voice-box.
Develop an existing study of HNC patients by measuring the participants’ genes.
Look across patients’ genes to identify ones that may be involved in cancer advancement.
Identify which type of relationship (causal, reverse causal, biased) is responsible for the apparent protective effect of being overweight if you have HNC.
Use this study to design a larger study to better address this question.
How it will be done
We will look at those cancer patients in the HN5000 study who are overweight and underweight, and compare if and how long they survive, and if their cancer returns. The genes of participants in HN5000 will be measured, and we will then look across all the participants to see if any small changes in their genes make them survive better or worse on average. To identify if this relationship is causal we need to use a method that can help to rule out the other options. By measuring Body Mass Index (BMI) before they have cancer, we can rule out the possibility of the cancer causing the weight loss (reverse causality). One way to achieve this is to look at whether a person has particular types of genes that will mean they are, on average, overweight or underweight. Genes are unlikely to be related to other factors which bias the relationship, so by using them we also rule out the biased theory. This stage will involve separating the participants into groups that will (according to their genes) be on average overweight or on average underweight. We can then compare the length of survival (or whether the cancer returns) between these groups.
In these studies, if we see a relationship between being overweight and better survival, we will have found evidence to support the causal association between being overweight and improved survival from HNC. This would allow us to be more confident in what the effects of treatments that focus on increasing weight during and after cancer treatment may be. In addition, we can look more deeply at what it is about being overweight that may improve cancer survival. This may provide clues for new types of treatment in the future. Measuring the genes of the participants in HN5000 will mean this study is the only one of its type and size with genetic information looking at how well HNC patients do after treatment.