Effect of a delay in definitive surgery in melanoma on overall survival: A NCDB analysis.

Abstract

e21586 Background: Time delay from diagnosis to definitive surgery has negative impact on the overall and disease-specific survival in breast cancer but the same has not been proven for colon cancer. Using National Cancer Data Base (NCDB), we report the impact of delay in surgery on overall survival of cutaneous melanoma. Methods: We identified patients with stage I to III cutaneous melanoma, age ≥ 18, using the NCDB (2004-2014). Days from diagnosis to definitive surgery was calculated in the increments of the 30 days (d) as 0-30d, 30-60d and > 60d. Overall survival (OS) was calculated using Kaplan Meier analysis with hazard ratio (HR) from cox regression model. Results: The NCDB cohort had 313,329 patients, diagnosed between 2004-2014.Of which patients who had surgical interval from date of diagnosis of < 30d, 31 to 60d and > 60d accounted for 59.05%, 31.89% and 9.05% respectively. Five-year OS and nine-year OS were 80% and 65.6% vs 79.2% and 63.3% vs 72.7% and 56.2% for 0-30d, 31-60d and > 60d time lag respectively. Hazard ratio (HR) with 95% confidence intervals (CI) for the respective groups is shown in table 1. More percent of stage I melanoma were likely to get surgery within first 30 d of diagnosis, whereas more percent of stage III melanoma had surgery after 60 d. Conclusions: OS was statistically significantly lower among patients with melanoma who were to receive surgery after 60d of diagnosis compare to within 60d. Table 1- HR estimates and 95% CI from Cox regression Groups ( in days) Adjusted HR 95% CI 0-30d vs 31-60d 1.025 0.999-1.052 0-30d vs > 60d 0.943 0.908-0.980 31-60d vs > 60d 0.920 0.884-0.957