New research from an international consortium led by Professor Terry Dwyer of the George Institute for Global Health, Oxford Martin School, confirms that childhood cancer occurrences rise with increasing birthweight, but shows that this is unlikely to be primarily due to a mother’s weight gain during pregnancy – an important contributor to size of the baby. The finding is believed to be an important step toward eventually understanding how to prevent childhood cancer.

The study of data from from the International Childhood Cancer Cohort Consortium (I4C) was recently published in the journal Paediatric and Perinatal Epidemiology, and was the first designed to specifically collect data before the children developed cancer. It includes pooled data on infant and parental characteristics and cancer incidence, and covered 380,000 live births. It was collected from studies going back more than 50 years from six geographically diverse regions, including the UK, USA, Denmark, Israel, Norway and Australia. The total analysis included 377 children with cancer, including 115 with leukaemia and 262 with non-leukaemia-type cancers.

“We have been able to compile a significant amount of evidence that effectively shows that childhood cancer incidence rises with increasing birthweight,” said Professor Terry Dwyer, the Executive Director for The George Institute for Global Health. “While we observed a correlation between increased birthweight and higher risk of cancer incidence in children, there were no significant interactions with maternal pre-pregnancy overweight or pregnancy weight gain.”

The study showed that after adjusting for gestational age and sex an increased risk of 26 percent for every kilogram increase in birthweight was observed for all cancers. In older children (diagnosed at or after three years of age), cancers other than leukaemia are particularly related to high birthweight.

“At The George Institute we are focused on the prevention, treatment and care of noncommunicable diseases, like cancer,” said Professor Dwyer. “I am more optimistic that we’ll find ways of preventing childhood cancer than I was when we started this study 10 years ago. We know that there are no easy answers, but we are assembling more clues, like this piece of evidence, which will help us fill in the puzzle. Additional research into childhood cancer is needed so that we can provide actionable solutions to improve outcomes for future generations.”

I4C researchers plan to follow this study up with several lines of inquiry, including: looking at data in new cohorts such as in Japan and China to see whether this association exists there as well; investigating other factors that have some relationship to birth weight, such as birth order, to see whether the association is similar in infants of different birth order; and making cord blood measurements from infants in the cohorts to determine whether growth hormones of various types explain what we have found.

Dwyer, a Professor of Epidemiology at the Oxford Martin School and the Nuffield Dept. of Population Health, also leads the I4C, an international alliance of longitudinal studies of children. The I4C aims to make major contributions in advancing understanding of the role of early-life exposures in childhood cancers.

While previous studies have posited on the connection between childhood cancers and high birthweights, the collaboration of the six cohorts has allowed the examination of a wider range of potential factors measured prospectively including maternal age, marital status, education level, smoking propensity, previous live births, diabetes, pre-pregnancy BMI and total pregnancy weight gain.

Researchers used state-of-the-art statistical techniques to analyse the complex data set. The final dataset excluded multiple births due to higher rates of low birthweight and children with Down's syndrome due to previously observed higher risks of childhood leukaemia.

• Video: Professor Terry Dwyer discusses the implications of the research
• Blog: New research provides clues for unlocking the mystery of childhood cancer

Further information: Research FAQs

Q: What did this study find?

A: The main finding of this study was that the larger the baby, the higher the risk of cancer during childhood. The pooled analysis demonstrates that a one kilogram birthweight increase correlates to a 26 percent increase in the risk of all cancers.

Q: Is that just childhood cancer or all cancer incidences?

A: The finding only correlates to childhood cancers, not cancers that an individual may develop later on in life.

Q: Who did you study?

A: The pooled data was collected from studies going back more than 50 years from six geographically diverse regions, including the UK, USA, Denmark, Israel, Norway and Australia. The study was the first to collect data before the children developed cancer. It covered 380,000 live births and included 377 children with cancer, including 115 with leukaemia and 262 with non-leukaemia-type cancers.

Q: Why is this study important?

A: While the occurrence is rare, there is indeed a correlation, which gives us something to further investigate. Currently, very little is known about the causes of childhood cancer and this study provides an important clue for finding ways to prevent it. This is a prospective study, which is designed to determine the relationship between a specific condition (childhood cancer in this instance) and certain factors. This is the first study that was designed to collect specific data before the child developed childhood cancer and gives us data that is of better quality and higher validity.

Q: What should mothers and families do with this information?

A: While this data provides important direction for scientific research, it does not provide guidance for preventing childhood cancer at this point. It is, however, cause for optimism, that science is one step closer to being able to prevent childhood cancer.

Q: Should mothers worry about their own weight or how much weight they gain during pregnancy?

A: Overall, childhood cancers are very rare – only five in 2,000 children will develop cancer – and the risk increase is small. A big increase in the size of the baby – one kilogram – would only account for one extra case for every 2,000 babies born. While the BMI of the mother, for example, is related to the birth weight of the infant, it is generally a much smaller effect than to increase birthweight by 1 kilogram. For every one unit change in the mother’s body mass index (BMI), birthweight increases only by 25 grams. So, the possible influence of risk of childhood cancer from things the mother might do is really quite small.

We looked to see whether there might be a special interaction between the mother’s BMI and the birthweight of the baby – which might have caused a bigger increase in risk than that described above – but could find no evidence of that.

Q: How will this study advance efforts to prevent childhood cancers?

A: Now that we know there is a correlation between birthweight and childhood cancer, we can begin to investigate other factors that help us explain why there is a correlation. It is an important piece of evidence that begins to help fill in the puzzle about the causes of childhood cancer.

Q: What is the next phase of research?

A: This finding is an important clue that we anticipate will lead to other findings in the coming years. Based on these findings researchers plan to follow this study up with several lines of inquiry, including: looking at data in new cohorts such as in Japan and China to see whether this association exists there as well; investigating other factors that have some relationship to birth weight, such as birth order, to see whether the association is similar in infants of different birth order; and making cord blood measurements from infants in the cohorts to determine whether growth hormones of various types explain what we have found.

Q: Who conducted the study?

A: Researchers from the International Childhood Cancer Cohort Consortium (I4C), an international alliance of longitudinal studies of children. The study authors include: Ora Paltiel, Gabriella Tikellis, Martha Linet, Jean Golding, Stanley Lemeshow, Gary Phillips, Karen Lamb, Camilla Stoltenberg, Siri E. Håberg, Marin Strøm, Charlotta Granstrøm, Kate Northstone, Mark Klebanoff, Anne-Louise Ponsonby, Elizabeth Milne, Marie Pedersen, Manolis Kogevinas, Eunhee Ha and Terence Dwyer. The study was recently published in the journal Paediatric and Perinatal Epidemiology.

Q: What is The George Institute and why are you highlighting this study?

A: Professor Terry Dwyer, the Executive Director for The George Institute for Global Health, is a Professor of Epidemiology at the Oxford Martin School and the Nuffield Dept. of Population Health at Oxford University who also leads the I4C.

The George Institute for Global Health is improving the lives of millions of people worldwide through innovative health research. Working across a broad health landscape, the Institute conducts clinical, population and health system research aimed at changing health practice and policy worldwide. The Institute has a global network of medical and health experts working together to address the leading causes of death and disability worldwide. Established in Australia and affiliated with The University of Sydney, the Institute today also has offices in China, India and the United Kingdom, and is also affiliated with Peking University Health Science Centre, the University of Hyderabad and the University of Oxford. The Institute has been ranked among the top 10 global institutes for impact for the last several years, and its research has resulted in changes to medical guidelines and ways of thinking about some of the most common medical treatments around the world.