The international study, led by Dr Richard McNally at the University of Newcastle upon Tyne, was funded by Cancer Research UK, the Dutch Cancer Society, and the Christie Hospital Research Endowment Fund.
The British and Dutch team* analysed a database of adult brain tumours diagnosed in patients from the North Brabant province of the Netherlands between 1983 and 2001. They found clusters of cases of glioma tumours, which make up about half of all brain tumours, at different time intervals in different geographical locations.
This 'space-time clustering' of cases is a pattern typical of diseases caused by infections, adding weight to the theory that outbreaks of viruses are a potential contributory cause of brain tumours. Diseases caused by more constant environmental factors, such as pollution, produce clusters of cases in one place over a much longer time period.
However, the research team stresses it is too early to say exactly which infections could be the cause, and say that more research is needed to pinpoint what they are.
The findings, published in the European Journal of Cancer, may potentially be a step towards developing better preventative measures for cancer and may also result in better treatment.
Dr Richard McNally, of Newcastle University's School of Clinical Medical Sciences (Child Health), said: "Very little is known about the cause of brain tumours and we think our research brings us closer to understanding more about this disease.
"We only found clustering of cases in the East of the province we investigated, and we think it could be something to do with the way infections spread in less densely populated areas."
However, the researchers stress that people cannot 'catch cancer'. Infections are only likely to trigger cancer in a very small number of individuals who are already genetically susceptible to the disease.
The team looked at data from the Eindhoven Cancer Registry. They analysed results for the East and West of the 5000 square-kilometre province separately. They found an irregular pattern where many cases occurred at the same time in men and women over 15-years old in the East but not in the West. Around seven per cent more cases of brain tumours were observed to occur in 'clusters' than would be expected by chance.
This is the first study to carry out this sophisticated form of statistical analysis known as 'clustering' in brain tumours in adults.
The results of this work are consistent with earlier research led by Dr McNally into childhood cancer in North West England. This discovered patterns in the diagnosis of two types of cancer - leukaemia and brain tumours - in that they tended to occur together at similar times and geographical locations.
Dr McNally, who also works with Newcastle University's School of Population and Health Sciences, added: "Future research should try to identify specific infections which could potentially be a trigger. If these are found, it could lead to future preventative measures."
Professor John Toy, Cancer Research UK's medical director, said: "Brain cancer is rare, accounting for less than two per cent of all new adult cancers diagnosed in the UK each year. These findings suggest a possible link between infection and this type of the disease but by no means provide proof."
*The team included UK and Dutch researchers. In the UK, they came from the University of Newcastle upon Tyne and The University of Manchester/Central Manchester and Manchester Children's University Hospitals NHS Trust. In Holland the researchers came from St Elisabeth Hospital, Tilburg, Erasmus MC, Rotterdam, and Comprehensive Cancer Centre South, Eindhoven.