Tumor Hypoxia Has Independent Predictor Impact Only in Patients With Node-Negative Cervix Cancer

Abstract

PURPOSE: This prospective clinical study was begun in 1994 to validate the independent prognostic impact of tumor hypoxia in patients with cervix cancer treated with definitive radiation therapy. PATIENTS AND METHODS: Between May 1994 and January 1999, 106 eligible patients with epithelial cervix cancer had tumor oxygen pressure (PO 2 ) measured using the Eppendorf probe. Oxygenation data are presented as the hypoxic proportion, defined as the percentage of PO 2 readings less than 5 mm/Hg (abbreviated as HP 5 ) and the median PO 2 . RESULTS: The median HP 5 in individual patients was 48%, and the median PO 2 was HP 5 . Progression-free survival (PFS) for patients with hypoxic tumors (HP 5 > 50%) was 37% at 3 years versus 67% in those patients with better oxygenated tumors (P = .004). In multivariate analysis, only tumor size (risk ratio [RR], 1.33; P = .0003) and evidence of pelvic nodal metastases on imaging studies (RR, 2.52; P = .0065) were predictive of PFS. However, an interaction between nodal status and oxygenation was observed (P = .006), and further analysis indicated that HP 5 was an independent predictor of outcome in patients with negative nodes on imaging (P = .007). There was a significant increase in the 3-year cumulative incidence of distant metastases in the hypoxic group (41% v 15% in those with HP 5 < 50%; P = .0023), but not in pelvic relapse (37% v 27%; P = .12). CONCLUSION: Tumor hypoxia is an independent predictor of poor PFS only in patients with node-negative cervix cancer, in addition to tumor size. Its impact appears to be related to an increased risk of distant metastases rather than to an effect on pelvic control.