Adjuvant Gm-Csf Therapy For Patients With Resected Stage Iii/Iv Melanoma: A Retrospective Review Of A Single-Center Experience.

8596 Background: Published phase II clinical data suggest a survival advantage of adjuvant GM-CSF (250 mcg/day x14 days every 28 days) therapy in patients with resected advanced stage III/IV melanoma (Spitler et al, JCO, 18:1614, 2000). Preliminary reports of a randomized phase III clinical trial (E4697) of adjuvant GM-CSF with/without a peptide vaccine in the same patient population do not suggest a statistically significant survival benefit of GM-CSF over placebo. Thus, we sought to retrospectively review our clinical experience with adjuvant GM-CSF therapy administered on a compassionate use protocol to patients with resected advanced stage III/IV melanoma. Methods: Following IRB approval we abstracted medical records of patients with resected advanced stage III/IV melanoma treated at our institution since 2000. Nearly 1000 patients with resected advanced stage III/IV melanoma were identified, some of whom were treated with adjuvant GM-CSF on a compassionate use protocol following complete surgical resection. Of those, we selected only patients with sufficient follow-up data to identify relapse-free and overall survival (OS). Herein are our preliminary findings; final case-control analysis will be presented at the Annual meeting. Results: Of the first 253 retrospectively reviewed patients, 219 met eligibility criteria. Only 70 patients in this cohort had resected stage IV melanoma and 149 patients had resected stage III disease. Analysis of median OS data by disease stage demonstrated that patients with resected stage III melanoma treated with GM-CSF (n = 56) vs observation (n = 93) exhibited median OS of 8.6 (95% CI = 5.1-NR) and 5.2 (95% CI = 3.7-NR) years respectively. Patients with resected stage IV melanoma treated with GM-CSF (n = 30) vs observation (n = 40) exhibited median OS of 6.6 (95% CI = 3.3-10.0) and 6.8 (95% CI = 2.4-NR) years respectively. Conclusions: Preliminary analysis of our data suggests further research is warranted to understand the impact of adjuvant GM-CSF in patients with resected stage III melanoma. Full case-control analysis of the entire data set will be presented at the Annual Meeting.