Examining prognostic factors and patterns of failure in nasopharyngeal carcinoma following concomitant radiotherapy and chemotherapy: impact on future clinical trials.

PURPOSE: Concomitant chemotherapy and radiotherapy (CCRT), followed by adjuvant chemotherapy, has improved the outcome of nasopharyngeal carcinoma (NPC). However, the prognosis and patterns of failure after this combined-modality treatment are not yet clear. In this report, the prognostic factors and failure patterns we observed with CCRT may shed new light in the design of future trials. METHODS AND PATIENTS: One hundred forty-nine (149) patients with newly diagnosed and histologically proven NPC were prospectively treated with CCRT followed by adjuvant chemotherapy between April 1990 and December 1997. One hundred and thirty-three (89.3%) patients had MRI of head and neck for primary evaluation before treatment. Radiotherapy was delivered either at 2 Gy per fraction per day up to 70 Gy or 1.2 Gy per fraction, 2 fractions per day, up to 74.4 Gy. Chemotherapy consisted of cisplatin and 5-fluorouracil. According to the AJCC 1997 staging system, 32 patients were in Stage II, 53 in Stage III, and 64 in Stage IV (M0). RESULTS: Univariate analysis revealed that WHO (World Health Organization) Type II histology, T4 classification, and parapharyngeal extension were poor prognostic factors for locoregional control. Multivariate analysis revealed that T4 disease was the most important adverse factor that affects locoregional control, the risk ratio being 5.965 (p = 0.02). Univariate analysis for distant metastasis revealed that T4 and N3 classifications, serum LDH level > 410 U/L (normal range, 180-460), parapharyngeal extension, and infiltration of the clivus were significantly associated with poor prognosis. Multivariate analysis, however, revealed that T4 classification and N3 category were the only two factors that predicted distant metastasis; the risk ratios were 3.994 (p = 0.02) and 3.390 (p = 0.01), respectively. Therefore, based on the risk factor analysis, we were able to identify low-, intermediate-, and high-risk patients. Low-risk patients were those without the risk factors mentioned above. They consisted of Stage II patients with T2aN0, T1N1, and T2aN1 categories and of Stage III patients with T1N2 and T2aN2 categories. Their risk of recurrence is low (4%). Intermediate-risk patients were those with at least one univariate risk factor. They are Stage II patients with T2bN0 and T2bN1 categories and Stage III patients with T2bN2 and T3N0-2 categories. The risk of recurrence is modest (18%). High-risk patients have risk factors by multivariate analysis. They are stage T4 or N3 patients. Their risk of recurrence is high (36%). CONCLUSION: Low-risk patients have an excellent outcome. Future trials should focus on reducing treatment-associated toxicities and complications and reevaluate the benefit of sequential adjuvant chemotherapy. The recurrence in treatment of intermediate-risk patients is modest; CCRT and adjuvant chemotherapy may be the best standard for them. Patients with T4 and N3 disease have poorer prognosis. Hyperfractionated radiotherapy may be considered for the T4 patients. Future study in these high-risk patients should also address the problem of distant spread of the disease.

Duke Scholars

Author Andrew T. Huang Medicine, Hematologic Malignancies and Cellular Therapy