Although stage I and II squamous cell carcinomas of the head and neck may be curable with either surgery or radiotherapy, recurrences affect 10 to 40 per cent of these patients. Patients with stage III and IV disease are usually treated by combined surgery and radiotherapy. The addition of chemotherapy to this approach has been tested in many trials over the past 10 years without conclusive evidence for benefit.

Most trials using surgery and radiotherapy in stage III and IV head and neck cancer have resulted in survival patterns similar to the United States National Cancer Institute (NCI) Contracts Program. Apparent differences in survival rates between trials are often due to the reporting of data without accounting for all the possible causes of death and treatment failure. For example, in the NCI study the overall 5-year survival was 35 to 45 per cent, while the local–regional control rate was 71 per cent. The difference in these numbers is explained by the causes of treatment failure and death in the study shown in Table 1 611.

Patients with stage III and IV glottic tumours in this study had a different survival pattern compared with those with stage III and IV tumours at other sites: 75 per cent of patients with glottic tumours were alive at 5 years compared to only 40 per cent of those with tumours at other sites. Nasopharyngeal tumours were not part of the NCI study, since surgery is rarely used in the treatment of this disease, and survival is usually much longer than that in patients with tumours of other head and neck primary sites.

The use of chemotherapy has been tested extensively in patients with stage III and IV resectable disease, mainly as neoadjuvant treatment, before surgery and radiotherapy. Although the results of early trials appeared encouraging, subsequent randomized trials have not shown any improvement in survival rates Table 2 612 and Table 3 613. Nearly all studies have reported survival rates of 50 to 60 per cent at 2 years and 35 to 45 per cent at 5 years, as in the NCI Contracts Program. Even with two or three cycles of the most active chemotherapy regimens (cisplatin and 5-fluorouracil) the results are the same, despite response rates of 80 to 90 per cent and complete response rates of 23 to 54 per cent. The reasons for the failure of neoadjuvant chemotherapy to improve survival are uncertain. Perhaps chemotherapy as used today may be capable of eliminating only microscopic disease.

Three trials (Table 4) 614 suggest that using chemotherapy after surgery and radiotherapy in the classic adjuvant method may prolong survival. Johnson used methotrexate-leucovorin and 5-fluorouracil for 6 months after standard therapy in very high-risk patients with extracapsular spread of cancer at the time of surgery. Two years later, survival was approximately twice that expected from historical control information. The NCI Contracts Program and a trial at the Dana-Farber Cancer Institute both studied patients randomized to further cisplatin-based chemotherapy after standard surgery and radiotherapy. Less than half of the patients would or could tolerate the planned adjuvant chemotherapy. However, survival was 10 to 20 per cent higher at 5 years for those patients randomized to receive the adjuvant chemotherapy, suggesting that further trials using this approach are warranted. The main disadvantages of adjuvant chemotherapy are time, cost, and toxic reactions. Over the short term there may be nausea and vomiting, mucositis, renal toxicity, and leucopenia. There may also be long-term effects such as tissue fibrosis due to the combination of chemotherapy and radiotherapy.

Neoadjuvant chemotherapy has been effective in predicting which patients will respond well to subsequent standard therapy. Several studies have shown that 60 to 90 per cent of patients who achieve complete response with chemotherapy are cured, while non-responders rarely live 5 years. This information can be useful in planning subsequent treatment. One commonly used approach to avoid laryngectomy or other handicapping surgery is to see if a patient achieves a complete response with the neoadjuvant chemotherapy. If a complete response occurs, then prognosis is very good, and the patient receives only radiotherapy as the definitive therapy. If the patient does not achieve a complete response, then the definitive therapy becomes combined surgery and radiotherapy.

The use of only radiotherapy or a combination of radiotherapy and chemotherapy in patients with advanced head and neck carcinoma has become more common. Approximately 15 to 28 per cent of patients remain alive without evidence of disease for 3 to 5 years after radiotherapy alone, given by standard techniques. A recent trial using hyperfractionation (daily dose divided into two sessions each day) has reported 5-year survival rates of 30 per cent for stage IV disease and 60 per cent for stage III disease, indicating a likely advance over the once-a-day method. Trials using chemotherapy in addition to radiotherapy also seem to produce higher disease-free survival rates than the once-a-day radiotherapy trials. However, benefits only seem to be seen in patients receiving chemotherapy during or after radiotherapy, not before radiotherapy. In one study 55 per cent of patients were alive 2 years after simultaneous cisplatin and radiotherapy while 5-year survival among patients treated with bleomycin and methotrexate after radiotherapy was 30 per cent.

There have been several trials of chemotherapy used before and after radiotherapy in patients with nasopharyngeal cancer. The one neoadjuvant trial showed no benefit from the addition of chemotherapy, while four studies of adjuvant chemotherapy given after radiotherapy all showed benefit. These four trials suggest that at 5 years the survival rate of patients treated with chemotherapy may be prolonged from the expected 35 to 55 per cent to 50 to 71 per cent; however, none of these trials was randomized. A recent report of hyperfractionated radiotherapy in nasopharyngeal cancer (without chemotherapy) showed an improved survival over historical controls.

Few patients who experience recurrent disease after surgery and radiotherapy treatment can be cured. Early trials using chemotherapy resulted in median survivals of 4 to 6 months. The most important drugs are cisplatin, methotrexate, bleomycin, and 5-fluorouracil; used as single agents, they produce response rates in patients with recurrent disease of 15 to 33 per cent. Although single-agent methotrexate was considered the standard treatment to which other regimens should be compared, now many investigators have reported better responses (40–70 per cent) and median survivals (5–10 months) using combinations of cisplatin and 5-fluorouracil infusions with or without methotrexate and bleomycin. Future areas for investigation include new schedules for combining the standard best agents, use of high-dose leucovorin with 5-fluorouracil and use of new agents such as carboplatin, taxol, interferon, and interleukin-2.

Amrein PC, Weitzman SA. Treatment of squamous-cell carcinoma of the head and neck with cisplatin and 5-fluorouracil. J Clin Oncol 1985; 3: 1632–9.

Blitzer PH. Epidemiology of head and neck cancer. Semin Oncol 1988; 15: 2–9.

Browman GP, et al. Methotrexate/fluorouracil scheduling influences normal tissue toxicity but not antitumor effects in patients with squamous cell head and neck cancer: results from a randomized trial. J Clin Oncol 1988; 6: 963–8.

Ervin TJ, et al. An analysis of induction chemotherapy in the multidisciplinary treatment of squamous cell carcinoma of the head and neck. J Clin Oncol 1987; 5: 10–20.

Hong WK, Bromer R. Chemotherapy in head and neck cancer. N Engl J Med 1983; 308: 75–9.

Jacobs JR, et al. Combined modality therapy utilizing a cisplatin combination for effective chemotherapy in patients with previously untreated head and neck cancer. Am J Surg 1986; 152: 451–5.

Jacobs JR, et al. Induction chemotherapy in advanced head and neck cancer. A D study. Arch Otolaryngol 1987; 113: 193–7.

Jacobs C, Goffinet DR, Goffinet L, Kohler M, Fee WE. Chemotherapy as a substitute for surgery in the treatment of advanced resectable head and neck cancer. A report from the Northern California Oncology Group. Cancer 1987; 60: 1178–83.

Johnson JT, et al. Adjuvant chemotherapy for high-risk squamous cell carcinoma of the head and neck. J Clin Oncol 1987; 5: 456–8.

Marcial VA, et al. Radiation Therapy Oncology Group (D) studies in head and neck cancer. Semin Oncol 1988; 15: 39–60.

National Cancer Institute Head and Necks Contracts Writing Committee. Adjuvant chemotherapy for advanced head and neck carcinoma. Final report of the Head and Neck Contracts Program. Cancer 1987; 60: 301–11.

Parsons JT, Mendenhall WM, Cassisi NJ, Isaacs JH, Million RR. Hyperfractionation for head and neck cancer. Int J Radiat Oncol Biol Phys 1988; 14: 649–58.

Schuller DE, Metch B, Stein DW, Mattox D, McCracken JD. Preoperative chemotherapy in advanced resectable head and neck cancer: final report of the Southwest Oncology Group. Laryngoscope 1988; 98: 1205–11.

Silverberg E, Boring CC, Squires TS. Cancer Statistics, 1990. Ca 1990; 40: 9–26.

Toohill RJ, et al. Cisplatin and fluorouracil as neoadjuvant therapy in head and neck cancer. Arch Otolaryngol 1987; 113: 758–61.