Abstract
The purpose of our study was to test the predictive value of cisplatin-DNA adduct levels in head and neck squamous cell carcinoma (HNSCC) patients treated with cisplatin-radiation. Patients with advanced-stage HNSCC were treated within a randomized trial, investigating the optimal route of cisplatin administration, concurrently with radiation. Cisplatin was administered intra-arterially (IA, 150 mg/m2, with systemic rescue by sodium thiosulfate) or intravenously (IV, 100 mg/m2). In a subgroup, adducts were quantified in normal tissue and tumor. 32P-postlabeling was used to quantify intrastrand guanosine-guanosine adducts (GG-adducts) and adenosine-guanosine adducts (AG-adducts). Adduct levels were correlated with treatment outcome. Thirty-five patients were included (21 IV and 14 IA). At median follow-up of 27 months, locoregional (LR) control was 75% at 1 and 70% at 2 years. Adduct levels in tumor were 4-5-fold higher than in white blood cells (WBC) for both IA and IV treatment (p = 0.01). Adduct formation in WBC and buccal cells was higher in IV treated patients compared with IA infusion (p = 0.049 and 0.005 for GG-adducts in WBC and buccal cells, respectively). Adducts in tumors after IA infusion were not statistically different from those after IV. A strong correlation was observed between GG- and AG-adduct formation (r = 0.86, p < 0.001). Patients with higher GG adduct levels (>median) in primary tumor had significantly better disease free survival (DFS) than patients with lower (< or = median) adduct levels (p = 0.02). For overall survival (OS), a nonsignificant trend was observed, again in favor of patients with higher adduct levels (p = 0.06). In conclusion, cisplatin-DNA adduct formation in primary tumor appears to be predictive for DFS in HNSCC. No differences were observed in intratumoral adduct levels between IA and IV treatments, despite selective infusion of high-dose cisplatin with the IA procedure. However, systemic adduct levels (WBC and buccal cells) from IV patients were higher than in IA patients, consistent with less systemic exposure after IA administration.