Background: A major episode of drinking water contamination by PFAS (poly/perfluoroalkyl substances) affected a wide area of the Veneto Region in Italy. The contamination was officially detected in 2013, but evidence suggests it may have originated decades earlier. A voluntary biomonitoring campaign conducted in 2017 measured 12 PFAS compounds, including PFOA, in residents from 30 municipalities (around 135,000 people) served by the contaminated aquifer, falling between the three provinces of Padova, Verona, and Vicenza. PFAS exposure has been associated with several health outcomes, notably testicular and kidney cancers. Because the data are available only at an aggregated geographic level, ecological regression approaches are necessary.
Objectives: This study evaluates, at the aggregate level, the relationship between PFOA exposure and kidney cancer mortality in 21 municipalities within the PFAS-contaminated area of Veneto, applying an ecological regression.
Methods: PFOA quantitative exposure was measured via the geometric mean blood concentration (30ng/ml) from 2017 biomonitoring data. Death certificates for primary ICD-10 code for Renal Cell Carcinoma (RCC) C64, referring to malignant neoplasm of the kidney, excluding the renal pelvis, for both sexes for the years between 2015-2018 were extracted by the National Institute of Health from the National Mortality Database. Municipal deprivation index (DI) components, extracted from the 2011 census, were included as confounders. Expected cases and standardised mortality ratios were calculated using as reference the population of the three provinces of Padova, Verona, and Vicenza. Subsequently, an ecological regression was implemented, including the DI variables.   
Results: A total of 42 kidney cancer deaths were observed between 2015 and 2018 in the red area of Veneto, against 31.2 expected ones. The municipality of Cologna Veneta was the one registering the highest excess risk, with an SMR of 3.97 (9 observed cases vs 2 expected). The ecological regression resulted in an incidence rate ratio (IRR) of 1.69 (95% CI [1.126, 2.543]) associated with a one point increase in the geometric mean of PFOA blood concentration in 30ng/ml.
Conclusions: A clear positive dose-response relationship was highlighted at the aggregate level between PFOA exposure and kidney cancer mortality in the red area of Veneto.