A meta-analysis of hyperfractionated and accelerated radiotherapy and combined chemotherapy and radiotherapy regimens in unresected locally advanced squamous cell carcinoma of the head and neck

Abstract

Background: Former meta-analyses have shown a survival benefit for the addition of chemotherapy (CHX) to radiotherapy (RT) and to some extent also for the use of hyperfractionated radiation therapy (HFRT) and accelerated radiation therapy (AFRT) in locally advanced squamous cell carcinoma (SCC) of the head and neck. However, the publication of new studies and the fact that many older studies that were included in these former meta-analyses used obsolete radiation doses, CHX schedules or study designs prompted us to carry out a new analysis using strict inclusion criteria.

Methods: Randomised trials testing curatively intended RT (> or =60 Gy in >4 weeks/>50 Gy in <4 weeks) on SCC of the oral cavity, oropharynx, hypopharynx, and larynx published as full paper or in abstract form between 1975 and 2003 were eligible. Trials comparing RT alone with concurrent or alternating chemoradiation (5-fluorouracil (5-FU), cisplatin, carboplatin, mitomycin C) were analyzed according to the employed radiation schedule and the used CHX regimen. Studies comparing conventionally fractionated radiotherapy (CFRT) with either HFRT or AFRT without CHX were separately examined. End point of the meta-analysis was overall survival.

Results: Thirty-two trials with a total of 10 225 patients were included into the meta-analysis. An overall survival benefit of 12.0 months was observed for the addition of simultaneous CHX to either CFRT or HFRT/AFRT (p < 0.001). Separate analyses by cytostatic drug indicate a prolongation of survival of 24.0 months, 16.8 months, 6.7 months, and 4.0 months, respectively, for the simultaneous administration of 5-FU, cisplatin-based, carboplatin-based, and mitomycin C-based CHX to RT (each p < 0.01). Whereas no significant gain in overall survival was observed for AFRT in comparison to CFRT, a substantial prolongation of median survival (14.2 months, p < 0.001) was seen for HFRT compared to CFRT (both without CHX).

Conclusion: RT combined with simultaneous 5-FU, cisplatin, carboplatin, and mitomycin C as single drug or combinations of 5-FU with one of the other drugs results in a large survival advantage irrespective the employed radiation schedule. If radiation therapy is used as single modality, hyperfractionation leads to a significant improvement of overall survival. Accelerated radiation therapy alone, especially when given as split course radiation schedule or extremely accelerated treatments with decreased total dose, does not increase overall survival.

Figure 1

Difference of survival comparing conventionally fractionated radiotherapy with conventionally fractionated radiotherapy plus simultaneous chemotherapy using cisplatin, carboplatin, mitomycin C, and 5-fluorouracil as single drug or combinations of 5-FU with a platin-derivate (Group 1a). OS = Overall survival; * = 95% confidence limits of 2 year overall survival; delta months = differences of median overall survival; LCL = lower confidence limit of median survival; UCL = upper confidence limit of median survival; N = number of patients.

Figure 2

Difference of survival comparing hyperfractionated or accelerated radiotherapy with hyperfractionated or accelerated radiotherapy plus simultaneous chemotherapy using cisplatin, carboplatin, mitomycin C, and 5-FU as single drug or combinations of 5-FU with one of the other drugs (Group 1b). OS = Overall survival; * = 95% confidence limits of 2 year overall survival; delta months = differences of median overall survival; LCL = lower confidence limit of median survival; UCL = upper confidence limit of median survival; N = number of patients.

Figure 3

Difference of survival comparing altered fractionated radiotherapy with altered fractionated radiotherapy plus chemotherapy using combinations of 5-FU with cisplatin or mitomycin C (Group 1c). This group of trials showed a relevant difference/prolongation in overall treatment time (delta 7 – 28 days) between standard and experimental treatment arm, caused by split course treatment or rapidly alternating radio-chemotherapy. OS = Overall survival; * = 95% confidence limits of 2 year overall survival; delta months = differences of median overall survival; LCL = lower confidence limit of median survival; UCL = upper confidence limit of median survival; N = number of patients.

Figure 4

Difference of median survival with different drugs used for concomitant radiochemotherapy compared to radiotherapy alone, χ² = > p-value = 0.04. 5-fluorouracil: studies with simultaneus 5-fluorouracil chemotherapy [18,24,25,38]. Cisplatin: studies with simultaneus cisplatin based chemotherapy [26,27,29,33,34,35,39,40]. Carboplatin: studies with simultaneus carboplatin based chemotherapy [26,30,31,32,36]. Mitomycin C: studies with mitomycin C based chemotherapy [12,30,37]. LCL = 95% lower confidence limit, UCL = 95% upper confidence limit; N = number of patients.

Figure 5

Comparison of accelerated radiotherapy with conventionally fractionated radiotherapy (Group 2). OS = Overall survival; * = 95% confidence limits of 2 year overall survival; delta months = differences of median overall survival; LCL = lower confidence limit of median survival; UCL = upper confidence limit of median survival; N = number of patients.

Figure 6

Comparison of hyperfractionated radiotherapy with conventionally fractionated radiotherapy (Group 3). OS = Overall survival; * = 95% confidence limits of 2 year overall survival; delta months = differences of median overall survival; LCL = lower confidence limit of median survival; UCL = upper confidence limit of median survival; N = number of patients.