Purpose: Here we report a phase II clinical trial, which was designed to test a novel hypothesis that treatment with gemcitabine (GEM)/doxorubicin (DOX) would be efficacious via reconstitution of C 18-ceramide signaling in head and neck squamous cell carcinoma (HNSCC) patients for whom first-line platinum-based therapy failed. Experimental Design: Patients received GEM (1,000 mg/m 2) and DOX (25 mg/m 2) on days 1 and 8, every 21 days, until disease progression. After completion of 2 treatment cycles, patients were assessed radiographically, and serum samples were taken for sphingolipid measurements. Results: We enrolled 18 patients in the trial, who were evaluable for toxicity, and 17 for response. The most common toxicity was neutropenia, observed in 9 of 18 patients, and there were no major nonhematologic toxicities. Of the 17 patients, 5 patients had progressive disease (PD), 1 had complete response (CR), 3 exhibited partial response (PR), and 8 had stable disease (SD). The median progressionfree survival was 1.6 months (95% CI: 1.4-4.2) with a median survival of 5.6 months (95% CI: 3.8-18.2). Remarkably, serum sphingolipid analysis revealed significant differences in patterns of C 18- ceramide elevation in patients with CR/PR/SD in comparison with patients with PD, indicating the reconstitution of tumor suppressor ceramide generation by GEM/DOX treatment. Conclusions: Our data suggest that the GEM/DOX combination could represent an effective treatment for some patients with recurrent or metastatic HNSCC, and that serum C 18-ceramide elevation might be a novel serum biomarker of chemotherapy response.