Intensified hyperfractionated accelerated radiotherapy limits the additional benefit of simultaneous chemotherapy—results of a multicentric randomized German trial in advanced head-and-neck cancer

Abstract

$\text{Purpose:}$ To demonstrate the efficacy of radiochemotherapy (RCT) as the first choice of treatment for advanced unresectable head-and-neck cancer. To prove an expected benefit of simultaneously given chemotherapy, a two-arm randomized study with hyperfractionated accelerated radiochemotherapy (HF-ACC-RCT) vs. hyperfractionated accelerated radiotherapy (HF-ACC-RT) was initiated. The primary endpoint was 1-year survival with local control (SLC).

$\text{Methods and Materials:}$ Patients with Stage III and IV (UICC) unresectable oro- and hypopharyngeal carcinomas were randomized for HF-ACC-RCT with 2 cycles of 5-FU (600 mg/m²/day)/carboplatinum (70 mg/m²) on days 1–5 and 29–33 (arm A) or HF-ACC-RT alone (arm B). In both arms, there was a second randomization for testing the effect of prophylactically given G-CSF (263 μg, days 15–19) on mucosal toxicity. Total RT dose in both arms was 69.9 Gy in 38 days, with a concomitant boost regimen (weeks 1–3: 1.8 Gy/day, weeks 4 and 5: b.i.d. RT with 1.8 Gy/1.5 Gy). Between July 1995 and May 1999, 263 patients were randomized (median age 56 years; 96% Stage IV tumors, 4% Stage III tumors).

$\text{Results:}$ This analysis is based on 240 patients: 113 patients with RCT and 127 patients with RT, qualified for protocol and starting treatment. There were 178 oropharyngeal and 62 hypopharyngeal carcinomas. Treatment was tolerable in both arms, with a higher mucosal toxicity after RCT. Restaging showed comparable nonsignificant different CR + PR rates of 92.4% after RCT and 87.9% after RT (p = 0.29). After a median observed time of 22.3 months, l- and 2-year local-regional control (LRC) rates were 69% and 51% after RCT and 58% and 45% after RT (p = 0.14). There was a significantly better 1-year SLC after RCT (58%) compared with RT (44%, p = 0.05). Patients with oropharyngeal carcinomas showed significantly better SLC after RCT (60%) vs. RT (40%, p = 0.01); the smaller group of hypopharyngeal carcinomas had no statistical benefit of RCT (p = 0.84). For both tumor locations, prophylactically given G-CSF was a poor prognostic factor (Cox regression), and resulted in reduced LRC (log-rank test: ± G-CSF, p = 0.0072).

$\text{Conclusion:}$ With accelerated radiotherapy, the efficiency of simultaneously given chemotherapy may be not as high as expected when compared to standard fractionated RT. Oropharyngeal carcinomas showed better LRC after HF-ACC-RCT vs. HF-ACC-RT; hypopharyngeal carcinomas did not. Prophylactic G-CSF resulted in an unexpected reduced local control and should be given in radiotherapy regimen only with strong hematologic indication.

Introduction

Patients with advanced carcinomas of the head and neck have a dismal prognosis. Locoregional control poses a major therapeutic challenge. For tumors considered to be unresectable, the treatment has traditionally been conventionally fractionated radiation therapy up to total doses of 60–75 Gy administered in 6–8 weeks and resulting in 2-year survival rates of less than 30% 1, 2. Attempts to improve this poor outcome have been made by combining conventional radiotherapy regimen with simultaneously given chemotherapy, as well as introducing different alterations of radiation fractionation using hyperfractionated and/or accelerated treatment schedules 3, 4, 5, 6, 7. There is, however, no evidence that neoadjuvant chemotherapy followed by radiation therapy (RT) is more efficient than radiotherapy alone (8). The concurrent administration of chemotherapy and radiation therapy is a proven highly effective approach resulting in improved tumor response and control but associated with increased acute toxicities, especially mucositis. Several cytotoxic drugs have been tested with concomitant radiotherapy. By using cisplatin or carboplatin alone, randomized trials of concomitant radiochemotherapy (RCT) have reported either survival improvements (9) or no benefit of the combined modality treatment (10). Browman et al. (11) suggested a potential benefit of the combination of 5-fluorouracil (5-FU) and RT. Other studies have used multidrug combinations with alternating chemo- and RT, and they reported better results compared to standard radiation alone 12, 13. Compared with sequential application of chemo- and radiotherapy, based on a large meta-analysis of the role of chemotherapy, the concomitant treatment with cytotoxic drugs and radiation appeared to be the most effective (14).

Recently, accelerated repopulation of tumor clonogens during the conventional fractionated radiotherapy regimen has been recognized as an important determinant of local control and possible cause of treatment failure in head-and-neck carcinoma, especially if the overall treatment time is prolonged. This hypothesis is supported by laboratory as well as clinical data, which showed median potential doubling times in many tumors of only approximately 4 days 15, 16, 17. These observations have led to the development of accelerated fractionation schemes, whereby radiotherapy is administered in a shorter overall time using multiple daily fractions and/or including the weekends (17). The reduction of overall treatment time has the potential for improving tumor control by minimizing tumor clonogen regeneration. Various accelerated regimens have been used, e.g., the Polish trial (CAIR) (18), the Danish trial (DAHANCA) (19), the Vancouver trial (20), and the concomitant boost concept (21). These different trials have shown improved tumor control rates but also increased treatment-related acute toxicities.

Based on promising results of a prospective Phase II trial with simultaneous RCT using the concomitant boost concept up to a total dose of 66 Gy in 5 weeks with concomitant carboplatin in weeks 1 and 5 (22), in 1995, our German study group of 5 centers (in each center, the Departments of Otorhinolaryngology/Head and Neck Surgery and Radiation Oncology) initiated a prospective randomized multicenter Phase III trial. The purpose of this study was to test the hypothesis that combined simultaneous radiochemotherapy with 5-FU/carboplatin and accelerated radiotherapy using the concomitant boost concept leads to a better locoregional progression-free survival than the same regimen of accelerated radiotherapy alone. Despite the fact that the combination of 5-FU/cisplatin might be one of the most widely accepted standard regimens of chemotherapy in head-and-neck cancer, there is no randomized trial revealing a statistical superiority of cisplatin vs. carboplatin in this tumor entity. Carboplatin was used because of its reduced renal, digestive, and neurologic toxic side effects compared to cisplatin. This study was reviewed by the Leading Commission for Therapeutic Clinical Trials of the German Cancer Society and received generous financial support by the “Deutsche Krebshilfe.”