Supplementary Material for: Trends in the Incidence of Rare Eyelid Malignancies: A SEER-Based Study (2000–2021)

Purpose: Analyzing temporal trends in the incidence of eyelid cancers, including malignant melanoma (MM), Merkel cell carcinoma (MCC), and sebaceous gland carcinoma (SGC), and predicting future patterns. Methods: A retrospective study was conducted on the Surveillance, Epidemiology, and End Results (SEER) database spanning 2000–2021. Time-series models were applied to estimate incidence rates (IRs) per 100,000. The best models were selected based on root mean square error (RMSE) and Akaike Information Criterion (AIC). Poisson regression was used to assess the impact of gender and age ≥60 on IRs. Results: The ARIMA (0,1,0) model (RMSE = 0.0144, AIC = -100.23) showed an increasing overall trend in eyelid cancers IRs from 2000 to 2015, declining slightly until 2020 (IR = 0.1028), then rising again in 2021. The 10-year forecast predicts an increase to 0.13 (95% PI: 0.11–0.15). For MM; ETS (M, N, N) model indicated a gradual IR rise until 2010, followed by a decline, and subsequent rise to 0.068 in 2020; predicted to reach 0.081 (95% PI: 0.062-0.100) by 2022. For MCC; NNAR (2,2) model showed IR stability (0.003-0.008), with a slight drop in 2021 (IR = 0.006), and predicted to decrease to 0.01 (95% PI: 0-0.01). For SGC; ARIMA (0,0,0) model demonstrated a decline in IR in 2011 (IR = 0.0192), increase until 2019, then decrease in 2021 (IR = 0.0246), with a predicted 10-year decrease to 0.026 (95% PI: 0.019–0.032). Poisson regression revealed that Age ≥60 significantly increased IRR for both MM and SGC (17.0 and 25.4, respectively). Gender was non-significant for all three cancers. Conclusion: Incidence of eyelid cancers revealed variable temporal trends, with a slight overall increase projected. Age ≥60 is a strong risk factor, particularly for MM and SGC.