Trends and variations in the use of adjuvant immunotherapy for stage III melanoma in the U.S. population.
9077 Background: High dose Interferon alfa-2b (IFN), an adjuvant immunotherapy for patients (pts) with stage III melanoma, was the only approved treatment option in the US from 1995-2011. There is limited information on how high dose IFN has been disseminated to eligible pts in general clinical practice, and whether variations exist in its adoption according to non-clinical factors. Methods: We obtained data on 34,208 pts diagnosed between 1998-2010 with stage III melanoma from the National Cancer Data Base (NCDB). IFN treatment was abstracted as immunotherapy. We investigated the use of immunotherapy according to pt demographic, socioeconomic, and clinical variables. We conducted multiple logistic regression analysis to examine the effect of these variables on the receipt of immunotherapy. Results: 62% of pts in our study population were male, 88% were Caucasian and 31% were over age 65. Overall, 27% of the pts received immunotherapy. There was no significant trend in its adoption between year 1998 and 2010. After adjustment for clinical variables, age at diagnosis, facility type, and geographic region are predictors strongly associated with use of immunotherapy. Only 16% of pts aged 65-74 and 3% over 75 received immunotherapy compared to 42% of those ages less than 45 (adjusted ORs 0.44 [0.32-0.59], 0.05 [0.04-0.14), respectively). Also 24% of pts treated at a comprehensive community cancer program received immunotherapy compared to 30% of those treated at an academic/research program (OR, 0.71 [0.51-0.99]). The frequency of immunotherapy was 25% in the Atlantic region and 17% in the Western region compared to 33% in Northeast (ORs 0.42 [0.18-0.99], 0.31 [0.13-0.74], respectively). Median household income, insurance type and comorbidity were not associated with adoption of immunotherapy after adjustment for all other variables. Conclusions: Less than one-third of all eligible patients received adjuvant immunotherapy in US general practice over the past decade. There is significant variation in its adoption according to non-clinical factors. Further exploration of the reasons for these variations and whether they are linked to important patient outcomes is needed.