Background and purpose: To define the dose-response relationship of nasopharyngeal carcinoma (NPC) above the conventional tumoricidal dose level of 66 Gy when the basic radiotherapy (RT) course was given by the 2D Ho's technique.
Patients and methods: Data from all five regional cancer centers in Hong Kong were pooled for this retrospective study. All patients (n = 2426) were treated with curative-intent RT with or without chemotherapy between 1996 and 2000 with the basic RT course using the Ho's technique. The primary endpoint was local control. The prognostic significance of dose-escalation ('boost') after 66 Gy, T-stage, N-stage, use of chemotherapy, sex and age (< or =40 years vs >40 years) was studied. Both univariate and multivariate analyses were performed.
Results: On multivariate analysis, T-stage (P < 0.01; hazard ratio [HR], 1.58) and optimal boost (P = 0.01; HR, 0.34) were the only significant factors affecting local failure for the whole study population, and for the population of patients treated by radiotherapy alone, but not for patients who also received chemotherapy. The following were independent determinants of local failure for patient groups with different T-stages treated by radiotherapy alone: use of a boost in T1/T2a disease (P = 0.01; HR, 0.33); use of a boost (P < 0.01; HR, 0.60) and age (P = 0.01; HR, 1.02) in T3/T4 tumors. Among patients with T2b tumors treated by radiotherapy alone and given a boost, the use of a 20 Gy-boost gave a lower local failure rate than a 10 Gy-boost. There was no apparent excess mortality attributed to RT complications.
Conclusions: Within the context of a multi-center retrospective study, dose-escalation above 66 Gy significantly improved local control for T1/T2a and T3/4 tumors when the primary RT course was based on the 2D Ho's technique without additional chemotherapy. 'Boosting' in NPC warrants further investigation. Caution should be taken when boosting is considered because of possible increase in radiation toxicity.