Residence Proximity to Benzene Release Sites Is Associated with Increased Incidence of Non-Hodgkin Lymphoma.

Increased risk of Non Hodgkin Lymphoma (NHL) has been observed in persons occupationally exposed to benzene, but the risk among persons living near benzene release sites has been less studied.

In order to investigate the spatial patterns of NHL incidence and the association between NHL incidence and distance to benzene release sites, we linked cancer incidence data for the period 1999–2008 from the Georgia Comprehensive Cancer Registry (a CDC-supported a statewide population-based cancer registry collecting all cancer cases diagnosed among Georgia residents since 1995) with population data from the U.S. Census, and with Environmental Protection Agency's Toxics Release Inventory data on the locations of benzene release sites in Georgia between 1988–1998. NHL cases were aggregated by census tract. Descriptive spatial analysis was performed using ArcGIS. Choropleth maps were created to depict the standardized incidence ratios (SIRs) by census tract for NHL and NHL subtypes, and locations of benzene release sites were overlayed upon SIR maps. Spatial Empirical Bayes smoothing was performed on the SIR values using GeoDa 1.01. To assess spatial correlation in SIRs, we conducted global, local, and focal spatial analyses. The global Moran's I and a local Moran's I (also termed Local Indicators of Spatial Autocorrelation [LISA]) were calculated for SIR patterns of NHL and NHL subtypes. The Lawson-Waller Score test was used to individually assess each of the 19 benzene release sites for focal clustering of NHL with Bonferroni correction for the 19 comparisons. We performed Poisson regression on NHL incidence rates, using the mean distance between the tracts centroids and release sites as a marker of exposure.

12,716 incident NHL cases occurred among adults residing in Georgia during this period. There was a positive spatial autocorrelation for cases of NHL, B-cell NHL, T-cell NHL and for diffuse large B-cell lymphoma indicating that cases NHL in Georgia were geographically clustered (all global Moran's p-values < 0.05). LISA cluster maps of SEB-smoothed SIRs (Figure) show the locations of “hot-spots” (high-high clusters) and “cold-spots” (low-low clusters). High SIRs were clustered in the metro-Atlanta area for NHL and NHL subtypes, while low SIRs were mostly in the southern region of the state. The Lawson-Waller test scored each census tract for the difference between the observed and expected NHL incidence, weighted by inverse distance to the sites. Results were statistically significant at the á= 0.0026 level (Bonferroni correction) for 15 of the 19 benzene release sites. All sites located within the metro-Atlanta area had significant focal clustering, with greater incidence rates observed in the tracts near these sites than expected. Mean distance from benzene release site had a protective effect with a 0.4% decrease in the incidence rate of NHL for every mile the mean distance increased.

Clusters of NHL were significantly spatially associated with benzene release sites located in the metropolitan Atlanta area, but not with release sites in other areas of the state. Populations living in a census tract ≥100 miles away from all benzene release sites in Georgia have approximately a 40% decrease in the expected incidence of NHL when compared to populations living in a census tract adjacent to a benzene release site. While the overall incidence of NHL in Georgia is low (17.4/100,000 individuals in the population/year), this represents a significant change in incidence pattern. Additional studies are needed to identify factors that modify this risk.

Figure.

View largeDownload slide

LISA cluster maps of spatial Empirical Bayes smoothed standardized incidence ratios.

Figure.

View largeDownload slide

LISA cluster maps of spatial Empirical Bayes smoothed standardized incidence ratios.

Close modal

Flowers:Genentech/Roche (unpaid): Consultancy; Millennium (unpaid): Consultancy; Celgene: Consultancy; Celgene: Research Funding; Spectrum: Research Funding; Millennium: Research Funding; Gilead: Research Funding; Janssen: Research Funding.