Background
Despite decades of epidemiological research, a substantial proportion of cancer aetiology remains unexplained, particularly for cancers where environmental determinants are likely to be ubiquitous and difficult to measure but may be spatially patterned. At the same time, the rising cost of new primary studies and overburdened clinical infrastructures increasingly constrain exploratory research. In contrast, many settings now maintain extensive population-based cancer registries, routinely collected health and administrative datasets, and geospatial exposure resources that remain underutilised for hypothesis generation in environmental cancer epidemiology.
Objectives
To argue that systematic linkage and re-use of existing registry, health, and geospatial data can provide a cost-effective platform for generating new hypotheses in environmental cancer aetiology, while explicitly acknowledging the methodological limitations inherent to such approaches and outlining opportunities for future enhancement.
Methods
Using examples from nationwide linkage studies, population-based cancer registry data were combined with a nationally representative cohort and multiple geospatial exposure surfaces. Individual residential locations at diagnosis or interview were linked to area-based exposure metrics using geographic information systems. These examples illustrate a generalisable framework for exploratory analyses based entirely on secondary data. Building on this framework, we highlight further opportunities for linkage with prescription databases, maternity and reproductive health records, occupational datasets, and longitudinal address histories, as well as the potential integration of registry-linked biobanks containing archived tumour tissue and selected biospecimens.
Results
These linkages consistently identified spatially structured associations between environmental exposures and cancer risk. However, effect estimates were typically modest and subject to expected limitations, including exposure misclassification, residential mobility, residual confounding, and partial ecological inference. Accordingly, such analyses are best viewed as tools for hypothesis generation and prioritisation rather than causal attribution. The availability of archived tumour material and biospecimens offers the possibility of follow-up molecular investigations to explore biological plausibility.
Conclusions / Implications for practice or policy
Maximising the scientific value of routinely collected data represents a pragmatic and scalable approach to advancing environmental cancer aetiology in an era of constrained research funding. Strategic investment in data linkage infrastructure, alongside modest enhancements to routine data collection - such as inclusion of residential and occupational histories within cancer registries or healthcare systems -could substantially improve exposure assessment and hypothesis prioritisation. Such efforts would strengthen the evidentiary pipeline from population-level signals to targeted mechanistic and preventive research, while maintaining transparency about uncertainty and limitations.