October 2, 2015

expert reaction to conference presentation on height and cancer incidence in the Swedish population

An abstract presented at the European Society for Paediatric Endocrinology annual meeting has examined cancer incidence in the Swedish population and reports an association between height and risk of developing cancer. A Before the Headlines analysis accompanied this Roundup.

Prof. Mel Greaves, Director of the Centre of Evolution and Cancer, The Institute of Cancer Research, London, said:

“It is difficult to assess the strength of this new study because there is no paper available and nothing has yet been through peer review or published.  But there are reasons for taking seriously the reported findings.  This observation has been made before in previous studies, and we already knew that breast and prostate cancers can be associated with height.  What would be a biologically plausible reason for this association?  It could be to do with growth hormone, and the suggestion for this is two-fold.

“Firstly, a previous study has reported that people who have genetic dwarfism appear to have very little cancer (even in an age-matched study, so not because of reduced life expectancy).  People with genetic dwarfism have a mutation in their growth hormone receptor, and we know that growth hormone and growth hormone receptor are critical to tumour growth too.

“Secondly, we know from experiments in mice that if they are engineered to make excess or very little growth hormone then their cancer rates escalate or decrease correspondingly (along with their size).  We know that in humans growth hormone not only stimulates bone growth during our growing years, but stimulates cell growth in general and blocks cell death.  So the level of growth hormone someone has could affect cancer risk by pushing up cell numbers.

“Of course for cancers like breast and skin cancer there are factors that have a much greater effect on risk, like family history, reproductive patterns and obesity, so tall people shouldn’t worry that they are destined to get cancer.  I’ll be interested to see the results of the present study once it has been written up in more detail and been through the peer review process, and in order to say how confident we can be in its findings we’d need to know whether these other factors that also influence cancer incidence were measured and controlled for.”

Prof. David Coggon, Professor of Occupational and Environmental Medicine, University of Southampton, said:

“It is unclear from the information provided (the abstract) whether the associations with height were adjusted for other known causes of the cancers examined, such as reproductive history (breast cancer) and sun exposure (melanoma).   Even if we knew that the study was robust and had controlled for all relevant confounding factors, the finding is such a non-specific observation that I cannot see it contributing usefully towards prevention of cancer.”

Prof. Eve Roman, Director of the Epidemiology and Cancer Statistics Group, Department of Health Sciences, University of York, said:

“The association between height and cancer incidence is old news and has been shown in previous studies – in this context the title of the press release, ‘Large-scale Swedish study discovers link between height and cancer’, is inaccurate in my opinion.  Although it seems from the information available that the study is likely to be robust and has been done well, as far as I can see this study contributes little that is new.”

Prof. Jack Cuzick, Director of the Wolfson Institute of Preventive Medicine and Head of Centre for Cancer Prevention, QMUL, said:

“The association between height and cancer has been known for some time; many cancers are increased in incidence in tall people. This study should add important further confirmation in a population based setting. The mechanisms for this effect are not clear and are worth further study. They may relate to the fact that the growth hormones related to height also are in some way stimulating cancer cells, but details are lacking.”

Dr Jane Green, Clinical Epidemiologist, University of Oxford, said:

“It is hard to give specific comments without a full paper being available, but I would not say this study ‘discovers’ a link between adult height and cancer incidence – this is by now well-established. It may well be the largest single study including both men and women, although there are several other very large studies which probably have more detailed information than this on other risk factors. This study adds to the existing evidence and the risks they found are very similar in size to those found by other studies.

“The very fact that similar associations are seen in studies on different populations, and for many different types of cancer, is interesting as it supports the view that the link between height and cancer may reflect basic (common) cancer mechanisms.  There is also already good evidence that increasing height is also related to an increased risk of death from cancer, both overall and for many individual cancer sites.

“This new study appears to be a record linkage study using state registry data on height and cancer incidence. This study design has strengths – data in Sweden are generally very well-recorded, and cancer data are complete for the population. It is not clear without seeing details of this study whether information was available for some, all or none of the population on important known cancer risk factors such as smoking, which could potentially affect (confound) measures of the link between height and cancer risk. Such information is not always available in registry linkage studies.

“In general, I would caution against interpreting a link as causal – however for height and cancer there is considerable evidence that suggests that the link is not explained by other known factors.  Clearly, adult height is not itself a ‘cause’ of cancer, but is thought to be a marker for other factors related to childhood growth.

“To put risk associated with a non-modifiable factor like height in context, it is worth noting that taller people have lower risks for heart disease, and a lower risk of death overall.”

Prof. Dorothy Bennett, Head of the Molecular Cell Sciences Research Centre, St George’s, University of London, said:

“As the abstract mentions, the results are consistent with previous studies of cancer related to height or body surface area (estimated from height).  Among others, a different quite large Swedish study that followed over 585,000 people concluded that melanoma incidence increased 17% in women and 11% in men per 5 cm height.  Deaths from breast and other cancers were also related to height.  (Wiren S et al, Cancer Causes Control, 2014;25:151-9).

“Thus the title of the press release seems a little overstated – Swedish study “discovers” link between height and cancer.  “Supports” might be a better description.

“It sounds an odd relationship at first glance, but it is actually very plausible that the risk of cancer in a person should be related to the number of cells in their body, since that determines the number of cells “at risk”.  A cancer arises by mutations from a single normal cell.  Bigger people have more cells (not bigger cells).   So melanoma risk for example might be expected to increase with surface area (amount of skin), which is related to the square of height.”

Abstract title: ‘Positive association between height and cancer in the Swedish population’ by Emelie Benyi et al. presented at the European Society for Paediatric Endocrinology Annual Meeting. 

Declared interests

Prof. Mel Greaves, Prof. Jack Cuzick and Prof. Dorothy Bennett declare that they have no conflicts of interest.

None others received.