• 11/30/2004
  • Houston, TX
  • K. Kian Ang
  • The University of Texas M.D. Anderson Cancer Center, Houston, TX

Improving the outcome for patients with locally advanced head and neck carcinomas (HNC) by rational modification of radiation fractionation regimens or combinations of radiation with chemotherapy has been the subject of intensive clinical investigations for more than three decades. The two prototypes of biologically sound-altered radiation fractionation regimens are hyperfractionation and accelerated fractionation.1 Hyperfractionation was based on preferential sparing of late-responding tissues when the radiation dose per fraction is reduced. Accelerated fractionation regimens emerged through the recognition that tumor clonogen proliferation occurring during radiotherapy has a detrimental effect on outcome. Results of large randomized trials addressing the optimization of radiation fractionation collectively show that a number of biologically sound altered fractionation schedules improve the locoregional (LR) control rate on the order of 10% to 15%, but have only a modest impact on overall survival. Although several altered fractionation regimens consistently induce more severe acute mucositis than standard 7-week radiotherapy, the general consensus is that late toxicities are not appreciably increased.

Scores of clinical trials testing combined-modality therapy have also been published. Meta-analyses of studies completed before 1995 reveal that cytotoxic agents given before or after surgery or radiation do not significantly improve the therapeutic outcome over LR treatment alone. In contrast, chemotherapy given concurrently with radiation improves 2- and 5-year overall survival rates by 8%. Although a variety of cytotoxic agents have been studied, cisplatin is the most extensively investigated, and will be the emphasis of this commentary. The data from four recent trials using standard radiation fractionation, with or without high-dose single-agent cisplatin given every 3 weeks, in the treatment of locally advanced non-nasopharyngeal HNC have been reported in the last 2 years. All four trials showed that the combined regimen was superior to radiation therapy alone in LR control or organ preservation; two trials also showed improvement in overall survival. A common finding, however, is that 100 mg/m2 of cisplatin given every 3 weeks during the course of radiotherapy substantially increases the severity of mucositis, apparently more so than altered fractionation regimens without chemotherapy. In addition, cisplatin induces systemic toxicity requiring intensive premedication and supportive care. Unfortunately, recording and reporting of the late morbidity of combined radiation-chemotherapy have not been sufficiently consistent and systematic. A thorough report of the long-term results of a French cooperative group (French Head and Neck Oncology and Radiotherapy Group [GORTEC]) trial revealed that the late complication rate of the combination of radiation with concurrent carboplatin and fluorouracil was significantly higher than that of radiation alone. Due to a lack of adequate reporting, there is controversy as to whether the late toxicity of the combination of standard radiation with 100 mg/m2 of cisplatin every 3 weeks might also be higher than radiation alone. Hopefully, longer and more complete follow-up data on late morbidities will be reported in the future.

Although the collective data are strong in establishing the superiority of the combination of radiation with concurrent chemotherapy relative to standard radiation fractionation alone in the management of locally advanced HNC, there is variability of clinical trials in patient selection and regimens, leading to continuing debate as to which combined regimen should be considered standard. Furthermore, many questions remain to be answered, including whether the cisplatin dose can be altered to reduce acute and late toxicities without diminishing efficacy, whether cisplatin has benefit when added to altered fractionation, whether newer cytotoxic agents such as taxanes have higher efficacy, and whether neoadjuvant chemotherapy further improves the outcome of concurrent radiation and chemotherapy.

The randomized trial coordinated by the Swiss Group for Clinical Cancer Research, reported by Huguenin et al in this issue of the Journal of Clinical Oncology, was designed to address two of these questions. Although the sample size is relatively small, the study generated many interesting findings. With regard to chemotherapy, the total cisplatin dose was reduced (ie, from the usual three to two cycles), and for each cycle its administration was fractionated into five doses of 20 mg/m2 given over 5 consecutive days, rather than 100 mg/m2 given in a single administration. Compared with other cisplatin trials, this approach seems to reduce the severity of systemic toxicity and mucositis without diminishing the impact on LR control and, potentially, on occult metastasis. The incidences of grade 3 to 4 nausea or vomiting, hematologic toxicity, and mucositis were relatively low. This finding is consistent with the data of a phase III trial in nasopharyngeal carcinoma from the National Cancer Center of Singapore, reported by Wee et al at the 2004 Annual Meeting of the American Society of Clinical Oncology, in which each cycle of concurrent cisplatin was 25 mg/m2/d given on 4 consecutive days. It is also in agreement with the data from another multi-institutional trial using cisplatin at 6 mg/m2/d, 5 days per week, for 7 weeks. The data from these three trials, along with the observations from other studies showing that a substantial fraction of patients could not receive the third planned cisplatin dose of 100 mg/m2, suggest that a cumulative cisplatin dose of approximately 200 mg/m2, independent of the schedule, might be sufficient to yield a beneficial antitumor effect.

The choice of radiation fractionation regimen by the Swiss cooperative group is rather surprising, as 74.4 Gy given in 1.2 Gy per fraction twice daily has not been tested against the standard fractionation in a phase III trial setting. Hyperfractionation regimens shown to be superior to the conventional 70 Gy in 7 weeks by the European Organization for Research on Treatment of Cancer and the Radiation Therapy Oncology Group (RTOG) phase III trials, deliver 80.5 Gy and 81.6 Gy, respectively. Given the exponential nature of radiation cell killing, an 8% to 9% reduction in the total dose represents a considerable reduction in biologic effect. The data from Huegenin et al are consistent with other studies showing that regimens combining variants of altered fractionation plus concurrent chemotherapy yield better LR control than altered fractionation alone. However, the study does not answer the question as to whether altered fractionation plus concurrent chemotherapy is better than standard fractionation plus the same chemotherapy regimen. Two ongoing cooperative group trials have been designed to resolve this issue. The RTOG trial (H0129) compares the efficacy of an accelerated fractionation regimen (concomitant boost: 72 Gy in 6 weeks) plus two doses of cisplatin to standard fractionation (70 Gy in 7 weeks) plus cisplatin. The three-arm GORTEC study (99-02) compares accelerated fractionation (70 Gy in 6 weeks) plus carboplatin-fluorouracil to a more pronounced accelerated fractionation (64.8 Gy in 3.6 weeks) without chemotherapy and to standard fractionation plus carboplatin-fluorouracil. Both trials are approaching completion of accrual and the results will likely be reported in 2 to 3 years.

Finally, it should be noted that advances in the understanding of tumor biology has opened an exciting era for translational research. Preclinical and correlative biomarker studies conducted in a few centers, for example, have revealed epidermal growth factor receptor (EGFR) as a predictor of radiation response of HNC and have identified EGFR and its downstream signaling molecules as appealing targets for therapeutic intervention. The results of a recently completed international trial reported at the 2004 Annual Meeting of the American Society of Clinical Oncology meeting validated the latter notion by showing that adding an anti-EGFR antibody to radiation yielded improved LR tumor control and overall survival without increasing mucositis and dysphagia, as compared with radiation alone. These data have generated tremendous enthusiasm for exploring more selective strategies for improving therapy for HNC. The results of the Swiss trial contribute by delineating the baseline data, against which outcomes of novel combined therapies should be measured, and in defining a less toxic radiation-chemotherapy platform to which novel agents can be added in the treatment of locally advanced HNC.

OCF NOTE: The author Dr. Ang, is a valued member of the Oral Cancer Foundations Scientific Advisory board, and was the treating radiation oncologist for OCF’s founder, Brian Hill.