Abstract
Purpose :
Geospatial survival disparities remain an unknown frontier within the state of Texas. This retrospective cohort analysis assesses the influence of geographical locations within survival in cases of choroidal melanoma (CM).
Methods :
The Texas Cancer Registry (TCR) provides an authentic and comprehensive population-based directory inclusive of survival endpoints and tumor staging at time of diagnosis. 1966 cases of CM diagnosed between January 1, 1995 and December 31, 2018 were extracted. This represents 23-year time interval. All patient with unknown cause of death were excluded from analysis. Geospatial variables of interest include Metropolitan Statistical Area (MSA), Border Regions (15), and Border Regions (32). Analysis utilizing Cox proportional hazards modeling was computed to determine hazards ratios (HR) with 95% confidence intervals (CI). Statistical significance was achieved at p<0.05. All statistical analyses were performed with IBM SPSS Version 27.
Results :
Unadjusted survival analysis reveals a statistically significant influence of MSA on cause-specific survival in patients diagnosed with choroidal melanoma. More specifically, a 26.4% decrease in cause-specific hazards was observed in patients living in MSA, as compared to patients living in non-MSAs (HR=0.736, p=0.003). Adjusted survival analysis with MSA as the primary independent survival predictor sustained overall significance (p<0.001), with MSA providing 29.3% increased odds of survival, as compared to patients living in non-MSA areas (HR-0.707, p=0.015). For patients living in non-metropolitan areas, cases receiving multi-agent chemotherapy experienced significantly increased hazards of death as compared to patients who did not receive chemotherapy as planned first course of therapy (HR=28.230, p<0.001).
Conclusions :
This state-level geospatial analysis demonstrates a poor prognostication of survival for cases of CM diagnosed in non-MSAs. Ocular oncologists may be encouraged to increase ophthalmic care delivery in less accessible regions within the state of Texas.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.