Prognostic factors and long-term survivorship in patients with recurrent or metastatic carcinoma of the head and neck - An analysis of two Eastern Cooperative Oncology Group randomized trials

BACKGROUND. The current study was conducted to identify prognostic factors and report the characteristics of long-term survivors in patients with recurrent or metastatic carcinoma of the head and neck who were treated with cisplatin-based combination chemotherapy in two randomized, Phase III trials conducted by the Eastern Oncology Cooperative Group (ECOG) (E1393 and E1395). METHODS. The authors analyzed prognostic factors for response and survival by combining data from the E1393 trial, which compared cisplatin plus paclitaxel at two dose levels, with data from the E1395 trial, which compared cisplatin plus paclitaxel with cisplatin plus 5-fluorouracil (5-FU), using logistic regression and Cox regression models. RESULTS. A total of 399 eligible patients were included. The median follow-up was 4.7 years. The 1-year overall survival (OS) rate for all patients was 32%, the median OS was 7.8 months, and the objective response rate was 32%. On multivariate analysis, the following were found to be independent unfavorable predictors of objective response: weight loss of > 5%, an ECOG performance status of I (vs. 0), residual disease at the primary tumor site, a primary tumor site other than the oropharynx, prior radiation therapy (RT) (P = 0.056), and well/moderate tumor cell differentiation (P = 0.067). Independent unfavorable prognostic factors for OS were weight loss, an ECOG performance status of I (vs. 0), well/moderate tumor cell differentiation, a primary tumor in the oral cavity or hypopharynx, and prior RT. The following were found to be independent unfavorable prognostic facotrs for time to disease progression: well/ moderate tumor cell differentiation, a oral cavity or hypopharyngeal primary tumor, and prior RT. Patients with greater than or equal to 2 adverse prognostic factors were reported to have a median OS of I year, whereas patients with 3-5 adverse prognostic factors were found to have a median OS of 0.5 years (P < 0.0001). Forty-nine patients (12%) survived for greater than or equal to 2 years and 6 patients were alive at 5 years. Two-year survivors were more likely to have achieved an objective response to chemotherapy, have poor tumor cell differentiation, be white, have an ECOG performance status of 0, and have received no prior RT. CONCLUSIONS. Clinical parameters and tumor cell differentiation appear to be strong pretreatment predictors of outcome in patients with carcinoma of the head and neck and should be considered in the design of future randomized trials. A small percentage of patients with recurrent head and neck carcinoma can achieve long-term survival. Cancer 2004;101:2222-9. (C) 2004 American Cancer Society.