CTCL Clustering in Iowa, Illinois, and Kentucky and Its Association with Environmental Factors

Introduction: Cutaneous T-Cell Lymphoma (CTCL) is a rare form of non-Hodgkin’s lymphoma that primarily affects the skin. While the exact causes of CTCL remain unclear, CTCL clustering was found in Georgia that correlates to increased levels of benzene and trichloroethylene (TCE) (Clough et. al, 2020). This suggests that environmental toxins may contribute to the clustering of CTCL cases.

Methods: 1,224 CTCL incident cases from 2010-2020 were obtained for counties in Illinois, Iowa, and Kentucky from the Surveillance, Epidemiology, and End Results (SEER-22) Program. County-level Standardized Incidence Ratios (SIRs) were calculated by dividing observed cases by expected cases, using state-specific subgroup rates to determine expected cases. Environmental exposure data for Benzene, TCE, and Polychlorinated Biphenyls (PCBs) was gathered from the Environmental Protection Agency, averaging data from 2002-2005. Index of Concentration at the Extremes for Race and Income (ICE) scores were computed using 5-year estimates from the American Community Survey spanning from 2010-2020. The associations between CTCL incidence, environmental exposures, and ICE scores were examined using multivariable linear and spatial lag regression models. CTCL incidence was log-transformed to satisfy regression assumptions. Effect estimates for Benzene, TCE, PCBs, and ICE scores were calculated based on 999 Monte Carlo simulations, accounting for both within-county (direct) and cross-county (indirect or spillover) effects in the spatial model. Spatial patterns were assessed using Global and Local Moran's I statistics.

Results: Multivariable linear regression modeling demonstrated a significant association between Benzene and PCB concentrations and CTCL (β = 0.24, 95% CI: 1.05-1.55, p=0.016; β = 3.7, 95% CI: 1.43-1096.63, p=0.03). Furthermore, a 1 µg/m³ increase in benzene exposure is associated with a 27% increase in CTCL incidence. The effect of PCBs appears to be particularly strong, with 1 ng/m³ associated with a 4,044% increase in CTCL incidence. However, the wide confidence interval suggests possible uncertainty in this estimate. TCE concentration and ICE scores do not show a significant association with CTCL (p=0.7, p=0.5). The spatial lag regression model demonstrated a better fit with the data compared to the linear model and revealed positive clustering of CTCL incidence (rho= 0.24, 95% CI: 0.08-0.39, p=0.003). Monte Carlo simulations within the spatial model showed a total increase of approximately 39% in CTCL incidence per 1 µg/m³ increase in benzene exposure (β=0.33, 95% CI: 1.06-1.84, p=0.02), considering both the immediate impact within the county and the ripple effects across neighboring counties. Unlike the linear regression model, PCB exposure shows a marginally significant positive association with CTCL incidence in the spatial lag model (p = 0.07). TCE exposure and ICE scores remain non-significant predictors of CTCL incidence (p>0.9, p>0.9). Moran’s I statistic indicates significant spatial clustering for all variables examined across the 321 counties, with moderate spatial clustering for CTCL incidence (Moran’s I= 0.11, p=0.002).

Conclusions: This analysis suggests that exposure to Benzene and PCBs are significantly associated with incidence and clustering of CTCL. To our knowledge, no prior study has demonstrated an association between PCB exposure and CTCL incidence. This information enhances our understanding of CTCL etiology and can inform targeted prevention strategies and interventions to mitigate environmental risks and reduce health disparities in affected communities. Further prospective analyses are needed to better understand the intersection of social determinants of health, environmental contaminants, and the incidence of CTCL.