Plasma osteopontin, hypoxia, and response to radiotherapy in nasopharyngeal cancer.

PURPOSE Recent studies have suggested that osteopontin is induced by hypoxia in head and neck cancer cell lines and its plasma level may serve as a surrogate marker for tumor hypoxia and treatment outcome in head and neck cancer. We investigated the response of osteopontin to in vitro hypoxia in nasopharyngeal carcinoma cell lines, and determined plasma osteopontin levels in nasopharyngeal carcinoma patients, nonnasopharyngeal carcinoma head and neck cancer patients, and healthy controls. We explored the relationship of plasma osteopontin and response to radiotherapy in nasopharyngeal carcinoma. EXPERIMENTAL DESIGN Nasopharyngeal carcinoma cell lines HK1, HONE-1, C666-1, and CNE-2 were treated with 0 to 48 hours of hypoxia or normoxia, +/- reoxygenation. Osteopontin secretion in the supernatant was measured by ELISA assay. Cellular osteopontin protein and mRNA were detected by Western blotting and reverse transcription-PCR, respectively. Plasma osteopontin levels in patients (n=66; 44 nasopharyngeal carcinoma, 22 head and neck cancer) and controls (n=29) were measured by ELISA. RESULTS Hypoxia has no effect on osteopontin protein and mRNA level in nasopharyngeal carcinoma cells. Only CNE-2 secreted osteopontin, and there was no significant induction by hypoxia. Plasma osteopontin levels in patients of metastatic nasopharyngeal carcinoma and head and neck cancer, but not in locoregional nasopharyngeal carcinoma, were significantly higher than in controls. In patients with locoregional nasopharyngeal carcinoma receiving curative radiotherapy (n=31), a high (>median) pretreatment plasma osteopontin level was a significant predictor of poor response to radiotherapy (complete response rate, 40% versus 88%; P=0.009), which remained significant in multivariate analysis. CONCLUSION Our results suggested that the pretreatment plasma osteopontin level may be a useful biomarker of response to radiotherapy in nasopharyngeal carcinoma.