Background: Pre-clinical studies report a critical role of c-KIT mutations in mucosal, acral, and solar melanomas. Clinical trials of KIT-directed therapy with imatinib and sunitinib demonstrate activity in these subtypes. Unlike other TKIs, dasatinib inhibitory activity is seen in melanoma cell lines with the most common KIT mutation at exon 11L576P. E2607 tests the efficacy of dasatinib in treatment-naïve or previously treated patients (pts) with these unresectable subtypes. Methods: Pts receive dasatinib 70 mg PO twice daily for this two-stage phase II trial. The primary objective is response rate (RR: RECIST). Stage I was open to KIT mutated (KIT+) and wild-type (KIT-) tumors with these subtypes (n=56). Depending upon the interim analysis, stage II would pre-select for KIT+. If the overall response rate were < 5%, the study would be terminated. Secondary objectives include progression-free survival (PFS), overall survival (OS), safety, and KIT mutation status. Results: Between May 2009-Dec 2010, 57 pts have been accrued to stage I. Of 54 (2 no therapy, 1 ineligible), 26 have mucosal (48%), 13 acral (24%), and 15 solar melanomas (28%). As of Jan 2012, the median f/u is 15 months (mo, range: 4-24 mo). Best responses (n=52) are 1 complete (CR: 2%), 3 partial (PR: 6%), 13 stable disease (25%), 29 progression (56%), and 6 unevaluable (11%). 51/52 have progressed (median PFS: 1.9 mo, 95% CI: 1.5-2.8) and 40/54 died (median OS: 7.4 mo, 5.7-10.8). KIT status has been assessed in 42 pts: 39 KIT- (93%) and 3 KIT+ (7%). PR is seen in 1/39 KIT-. All KIT- have died: median PFS (1.8 mo, 95% CI: 1.4-2.9) and OS (6.8 mo, 95% CI: 5.2-10.8). Of the 3 KIT+ (1 acral, 2 mucosal), 2 have died (PFS 1.4-2.5 mo, OS: 1.4-10.5 mo), and 1 refused f/u (OS: 13.3+ mo). The pt with a CR is alive at 14.4+ mo (unknown KIT status). Toxicities include 1 grade IV (elevated lipase) and grade III dyspnea (n=7), nausea (n=6), and vomiting (n=3). Conclusions: Further evaluation of dasatinib should be pursued in pts selected for KIT+ and subtype, given the lack of promising results in Kit-(<5% RR). The current trial is revised to enroll only KIT+ acral, mucosal, and vulvovaginal melanomas and available via CTSU.