AFFILIATIONS
1 Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States
2 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Feinberg School of Medicine, Chicago, Illinois, United States
3 John H. Stroger, Jr. Hospital of Cook County, Chicago, Illinois, United States
4 University of Chicago, Chicago, Illinois, United States
5 University of Illinois Chicago, Chicago, Illinois, United States
6 John H. Stroger, Jr. Hospital of Cook County, Chicago, Illinois, United States
7 Rush MD Anderson Cancer Center, Chicago, Illinois, United States
8 University of Illinois Chicago, Chicago, Illinois, United States
9 Rush MD Anderson Cancer Center, Chicago, Illinois, United States
10 University of Chicago, Chicago, Illinois, United States
11 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Feinberg School of Medicine, Chicago, Illinois, United States
12 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Feinberg School of Medicine, Chicago, Illinois, United States
13 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Feinberg School of Medicine, Chicago, Illinois, United States
14 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Feinberg School of Medicine, Chicago, Illinois, United States
ABSTRACT
Introduction: Cutaneous T-cell lymphoma (CTCL) is a rare form of non-Hodgkin lymphoma that has been shown to have geographic variation in incidence. It is hypothesized that this variation may be due to environmental exposures and socioeconomic factors, although their contribution to spatial disease clustering remains incompletely understood and has thus far only been studied at the county level. This study evaluates the geographic clustering of CTCL across Chicago at the neighborhood level in a racially and ethnically diverse population, and examines environmental, social, and healthcare-access factors associated with neighborhood-level incidence.
Methods: An IRB-approved spatial epidemiologic analysis was conducted by using CTCL cases diagnosed between 2004 and 2024 across 77 Chicago community areas including patients from 5 major hospitals in Chicago, IL. Cases were identified from hospital-based tumor registries and geocoded to patients' residential addresses at the time of diagnosis. Age- and sex-standardized incidence ratios (SIRs) were calculated using Chicago-wide rates as the reference. Spatial clustering was assessed using global Moran's I and local indicators of spatial association (LISA). Environmental exposures, including benzene, trichloroethylene, polychlorinated biphenyls, carbon monoxide, and PM2.5, were obtained from the EPA National Air Toxics Assessment. Social vulnerability measures were derived from CDC Social Vulnerability Index (SVI) domains. Healthcare access was assessed using CTCL-relevant specialist density and travel time to the closest academic medical center. Univariable and multivariable negative binomial regression models were used to identify factors independently associated with CTCL incidence.
Results: Among the 687 patients, 51% were female, 40% were Black, and 52% were White. Mycosis fungoides accounted for 69% of cases. CTCL incidence demonstrated significant geographic clustering across Chicago (Moran's I= 0.234, p<0.001), with high-incidence clusters localized to the Loop (SIR= 2.70), Lincoln Park (SIR= 1.38), and Near West Side (SIR= 1.18).
In univariable analyses, higher benzene exposure, lower PM2.5 exposure, lower social vulnerability scores, increased specialist density, and shorter travel time to an academic medical center were associated with higher CTCL incidence. In multivariable analysis, each IQR increase in benzene exposure (Q1–Q3 range) was associated with a 7% increase in CTCL incidence (IRR= 1.07 per IQR, 95% CI= 1.02-1.12, p= 0.007). PM2.5 exposure, however, was inversely associated with incidence (IRR= 0.81 per IQR, 95% CI= 0.66-0.99, p= 0.048). Higher SVI household composition scores were associated with increased incidence, with each IQR increase associated with a 36% increase in incidence (IRR= 1.36 per IQR, 95% CI= 1.04-1.78, p= 0.026). However, higher SVI minority status scores were associated with lower incidence (IRR= 0.61 per IQR, 95% CI= 0.51-0.73, p<0.001) possibly due to this population having lower healthcare access. Greater CTCL specialist density was independently associated with higher observed incidence (IRR= 1.02 per 100,000 population, 95% CI= 1.00-1.03, p= 0.006), likely from more cases being diagnosed in areas with greater specialty healthcare access. The multivariable model explained a substantial proportion of geographic variation (Nagelkerke R²= 0.48), and residual spatial autocorrelation was no longer significant after adjustment (Moran's I= 0.073, p=0.117), indicating that included covariates largely accounted for the observed clustering.
Conclusions: CTCL incidence is non-randomly distributed across Chicago and exhibits significant geographic clustering. Benzene exposure, neighborhood social vulnerability, and healthcare access independently contribute to spatial variation in CTCL incidence. The inverse association observed with PM2.5 was unexpected and warrants further validation. These factors collectively explain much of the geographic clustering observed across Chicago, suggesting that environmental determinants and disparities in access to specialty care may shape the geographic burden of CTCL and influence patterns of diagnosis. Further analysis is needed to explore these exposures and patterns of clustering at the neighborhood level in other parts of the country.
DISCLOSURES
Madeline Jenkin, MD: No relevant conflicts of interest to declare.