“The lung cancer and smoking connection is very clear, so why in other cancers that have a strong link with tobacco are we seeing such surprising differences?” asks Pablo Fernández-Navarro, the lead co-ordinator of the atlas from the Spanish side.

“This is what is so fantastic. If whole countries had uniform levels of mortality, the maps would be in plain colours. Given that it is not the case, now we have to investigate and explain these differences, eliminating one factor after another,” Fernández-Navarro told The Telegraph.

In the case of larynx cancer, the Spanish epidemiologist says the map confirms that smoking is by no means the only risk factor, and that other elements must also be at work, from alcohol intake to levels of pollutants such as asbestos or petrochemicals in the environment.

“It would be interesting to see the results with the impact of tobacco filtered out. There are clearly so many factors at play,” Fernández-Navarro adds.

In 2016, teams of researchers from Spain’s National Institute of Health Carlos III and its Portuguese counterpart, the Dr Ricardo Jorge institute, started working on the idea of producing the Iberian Atlas of Cancer Mortality, the first known international atlas of cancer based on data that reaches down to the municipal level, as opposed to a flat national comparison.

Their raw material looked at 840,000 deaths from 10 types of cancers in Iberia between 2003 and 2012, but the challenge was to harmonise the data across council areas that are different in size and factor out the impact of age.

“After a few years of work we began to understand why no one had done this before,” jokes Fernández-Navarro of the difficulties of blending the data.

“But we are confident the use of this atlas will have an impact on reducing deaths from cancer,” adds Carlos Matius Dias, the chief co-ordinator from Portugal.

“The objective is to develop a better understanding of risk factors, which stem from individuals’ life choices, environmental factors and accessibility to healthcare and treatment,” Dias says. Individuals do have genetic disposition to certain types of cancer but the scope of the study means the impacts from the environment and regional variations in lifestyles will shine through, say the scientists behind the Iberian Atlas of Cancer Mortality.

For example, Dias points to Portugal’s considerably higher mortality risk from prostate and stomach cancers as key areas of concern for his country’s health authorities. Excluding age, not much is known about risk factors for prostate cancer, and it may turn out that the marked cross-border difference in this case is to do with different approaches to prevention in the Spanish and Portuguese health systems.

“With prostate cancer, we know that screening is key. A person’s survival depends largely on early detection,” explains Fernández-Navarro. Likewise, experts believe that access to treatment may play a role in explaining the higher rate of mortality in breast cancer seen on both sides of the border in poorer southern areas of Spain and Portugal. 

“Geographical patterns help generate new hypotheses, both in terms of the causes of a cancer and about its clinical treatment,” says Professor Marina Pollán, director of Spain’s National Centre for Epidemiology. Finding out why patients in southern Portugal and south western Spain suffer higher mortality from breast cancer than their compatriots could provide vital clues to preventing death from the most common cancer in women worldwide.