Dose escalation using twice-daily radiotherapy for nasopharyngeal carcinoma: Does heavier dosing result in a happier ending?

Purpose: To present our experience using a twice-daily radiotherapy (RT) technique, including hyperfractionated and accelerated-hyperfractionated RT, on nasopharyngeal carcinoma (NPC) patients. The dose to the primary tumor was increased in the hope that local control could be increased without the cost of increased late complications. We analyzed acute and late complications and local control and compared the results with the results of NPC patients treated during the same period using conventional once-daily RT. Methods and Materials: Between October 1991 and July 1998, 222 histologically confirmed, Stage MO, previously unirradiated NPC patients completed RT at our hospital. Most patients had American Joint Committee on Cancer (AJCC) 1992 Stage III and IV disease. Among them, 88 received altered fractionated, twice-daily RT; 76 patients received hyperfractionated RT and 12 accelerated-hyperfractionated RT. The remaining 134 patients received a conventional once-daily regimen. Hyperfractionated RT was delivered using 120 cGy b.i.d. separated by 6-h intervals throughout the course. For the accelerated-hyperfractionated patients, 160 cGy b.i.d. was given, also at 6-h intervals. The median dose in the twice-daily group was 7810 cGy (range 6840-8200). In the once-daily regimen, RT was delivered using 180-200 cGy q.d. The median tumor dose to the primary tumor was 7000 cGy (range 6560-8100) given during about 8 weeks. The median follow-up time was 70.5 and 72 months for the twice-daily and once-daily groups, respectively. Results: The incidence of acute toxicities was higher in the twice-daily group with more severe mucositis and moist desquamation than in the once-daily group. Both groups had a similar incidence of late complications, except for 3 cases of temporal lobe necrosis in the twice-daily group, all in patients treated with 160 cGy. No difference was noted in recurrence-free local control between the two groups when the individual T stage was compared using AJCC 1992 or 1997 criteria (p = 0.51 and 0.59, respectively). The 5-year local control rate for T1-3 (AJCC 1997) was 93.2% for the twice-daily group and 86.4% for the once-daily group (p = 0.45). In Stage T4 (AJCC 1997) patients, the local control rate dropped drastically to 43.5% and 36.9% for the twice-daily and once-daily groups, respectively. The overall neck control rate at 5 years was 87.3% and 80.3% for the twice-daily and once-daily patients, respectively (P = 0.16). The overall locoregional control rate was 82.7% for the twice-daily group and 66.6% for the once-daily group. The difference was again not statistically significant, but showed a tendency in favor of the twice-daily regimen (P = 0.055). Locoregional failure occurred mainly in Stage T4 patients with central nervous invasion for v horn local control was particularly poor, with a failure rate of about 60%. Conclusion: The present data suggest that NPC patients can be safely treated using a 120-cGy twice-daily program with a 6-h interval up to 8000 cGy. The accelerated-hyperfractionated technique is not recommended. A large discrepancy in local control between patients with T1-3 and T4 disease was noted. For T1-3 disease, an excellent local control rate >90% was achieved using the twice-daily regimen. In contrast, failure in the T4 patients was as high as 55% in the twice-daily group and reached 65% in the once-daily group. More rigorous treatment is needed using either additional dose escalation or other strategies for T4 NPC patients. With a dose escalation of 1000 cGy using 120-cGy twice-daily RT, a trend toward better locoregional control and disease-specific survival was noted in the twice-daily group. Whether this difference was truly the result of an increased dose needs additional confirmation in studies with larger patient numbers. (C) 2002 Elsevier Science Inc.