Search

Background: The purpose of this study was to evaluate the efficacy of the modern diagnostic evaluation for squamous cell carcinoma metastatic to cervical lymph nodes from an unknown head and neck primary site., Methods: One hundred thirty patients were evaluated between June 1983 and June 1997. All underwent head and neck examinations, head and neck computed tomography (CT), and/or magnetic resonance imaging (MRI) scans, panendoscopies, and biopsies of head and neck mucosal sites. Twenty-four patients underwent 2-[fluorine-18]-2-deoxy-D-glucose (FDG) single photon emission computed tomography (SPECT); 34 patients underwent tonsillectomy., Results: The primary site was identified in 56 patients (43%); the likelihood was increased in patients with suggestive findings on physical examination and/or radiographic evaluation. Eighty-three percent of the lesions were located in the tonsillar fossa and base of tongue. Results of FDG-SPECT scans were positive in 20 patients (83%); the primary tumor was detected in 7 patients (35%). Twelve (35%) of 34 patients who underwent tonsillectomy had a primary tumor discovered in the tonsillar fossa. Multivariate analysis of successful primary site detection revealed that suggestive findings on physical examination (p= .0225) and suggestive findings on CT and/or MRI (p = .0013) were significantly related to this end point., Conclusion: The primary lesion will be detected in over 40% of patients with physical examination of the head and neck and CT and/or MRI followed by panendoscopy and biopsies. Limited data pertaining to FDG-SPECT suggest that this provides additional useful information in a small subset of patients. Tonsillectomy is useful for those with suggestive findings on physical examination and/or radiographic evaluation.

Background: Because only limited data are available pertaining to radiotherapy for chemodectomas of the carotid body and glomus vagale, we reviewed our experience., Methods: Fifteen patients with 23 chemodectomas of either the carotid body or glomus vagale were treated with radiotherapy at the University of Florida between 1981 and 1995. Eighteen lesions were previously untreated. One patient had received prior radiotherapy at another institution and four patients had received prior surgery., Results: The local control rate at 10 years, calculated by the Kaplan-Meier product-limit method, was 96% for the overall group of 23 lesions and 100% for the subset of 22 lesions without prior radiotherapy. The 10-year cause-specific survival rate was 89% for all 15 patients and 100% for the 14 patients who had received no prior radiotherapy. No patient experienced a significant complication secondary to irradiation., Conclusions: Irradiation offers a high probability of tumor control with relatively minimal risks for patients with chemodectomas of the carotid body and glomus vagale.

Background: We performed an analysis of treatment outcome at the University of Florida for 77 patients with previously untreated carcinoma of the skin metastatic to parotid area lymph nodes., Methods: Seventy-seven patients with 79 parotid node metastases were treated with curative intent between 1966 and 1994. All patients had a minimum 2-year follow-up. Treatment consisted of radiotherapy alone in 18 patients, preoperative radiotherapy in 6 patients (7 parotids), and postoperative radiotherapy in 53 patients (54 parotids)., Results: Overall rate of disease control in the parotid area at 5 years was 82%. Parotid area disease control according to treatment group revealed: surgery and postoperative radiotherapy, 43 of 48 (90%); preoperative radiotherapy and surgery, 4 of 5 (80%); and radiotherapy alone, 8 of 15 (53%). Multivariate analysis of parotid disease control revealed the following: treatment method (p = .0037), N stage (p = .0726), cranial nerve involvement (p = .1851), fixation (p = .1820), skin involvement (p = .1771), and sex (p = .5782). Overall 5-year absolute and cause-specific survival rates were 54% and 68%, respectively. The 5-year cause-specific survival rate was 72% after surgery and postoperative radiotherapy and 59% following radiotherapy alone. Multivariate analysis of cause-specific survival revealed the following: skin involvement (p = .0215), cranial nerve involvement (p = .1549), fixation (p = .0347), N stage (p = .4302), sex (p = .6806), and treatment method (p = .9599)., Conclusions: Our current policy is to treat patients with tumor that is initially resectable with surgery followed by radiotherapy, whereas those who have initially unresectable disease undergo preoperative radiotherapy followed by surgical resection if the tumor becomes resectable. If surgery cannot be performed for medical reasons, patients still have better than a 50% chance of being cured with radiotherapy alone.

Background: This study was undertaken to determine whether pretreatment computed tomography (CT) findings can predict local control in pyriform sinus carcinoma treated with definitive radiotherapy (RT)., Methods: Twenty-three patients with pyriform sinus carcinoma (T1: n = 5; T2: n = 18) were treated with high-dose RT and followed for a minimum of 2 years. Tumor volumes and extent were determined on pretreatment CT studies. The specific CT parameters assessed were analyzed as predictors of local control., Results: There was a significant decrease in local control rate for tumors over 6.5 mL (1 of 4 [25%]) relative to tumors under 6.5 mL (17 of 19 [89%]; p = .021). Sensitivity and specificity for local control using this cutoff were 94% and 60%, respectively. Tumor score, as a measure of anatomic extent, was also found to be a significant predictor of local control (p = .033). The local control rate was not influenced significantly by the presence of "minimal" apex disease (< 10 mm in greatest dimensions as measured on CT) but decreased significantly when "bulk" apex disease (> OR = 10 mm) was present (p = .027). Laryngeal cartilage sclerosis was not a significant predictor of outcome., Conclusion: Computed tomography can stratify pyriform sinus carcinomas into groups more or less likely to be locally controlled with definitive RT.

Background: Postoperative radiotherapy is commonly used in treating head and neck cancer. A concern is the possibility of an unhealed surgical wound delaying or disrupting the treatment schedule. Our institution previously reported our experience on the outcome of radiotherapy on unhealed wounds from 1977 to 1984., Methods: In an effort to update and enlarge this series, we reviewed the charts of patients receiving postoperative radiotherapy from 1985 to 1990. Of 267 patients, 20 had unhealed wounds or fistulas at the beginning of treatment. The wounds healed spontaneously in 17 of these patients, 1 required surgical closure, and 2 never healed. Combining our current and previous series, 452 patients had onset of postoperative radiation therapy at our institution from 1977 through 1990, of whom 33 had unhealed wounds at the onset of irradiation., Results: In 22 of 33 (67%) patients, the wounds healed spontaneously (mean: 98 days, range: 36 to 304 days). Five additional patients achieved successful wound healing following surgical closure (mean: 281 days, range; 112 to 608 days). The remaining 6 patients died before healing was complete (4 patients) or were lost to follow-up (2 patients) at 16 and 27 months before wound healing occurred. Nine patients in this study are alive with no evidence of disease, 19 died of cancer, and 5 died of other causes., Conclusions: When there are indications to deliver irradiation in the postoperative setting, it is desirable to initiate treatment within 6 weeks of the data of surgery. Although it is preferable to have completely closed operative wounds prior to irradiation, some consideration must be given to initiating irradiation in the face of incompletely closed wounds when it is anticipated that healing time will be prolonged. In some patients, the wounds continue to heal during the course of irradiation, or will stabilize, and can be surgically managed after irradiation is completed or may spontaneously heal after treatment. Excessive delays in initiating appropriate cancer therapy may lead to recurrence prior to irradiation; such recurrences are rarely successfully salvaged.

Background: This study was undertaken to analyze the clinical course of patients who developed local (primary) recurrence after high-dose irradiation of T1 -T2 squamous cell carcinoma of the glottic larynx., Methods: Between May 1977 and December 1989, 247 patients with previously untreated T1 and T2 invasive squamous cell carcinoma of the glottic larynx were treated for curative intent with radiotherapy. Local recurrence occurred in 26 of 247 patients (11%). Successful surgical salvage was defined as no evidence of recurrent cancer for at least 2 years after salvage surgery and continuously thereafter., Results: Two patients refused to undergo salvage surgery and one patient had unresectable disease. Of 23 patients who underwent salvage surgery, 19 had a total laryngectomy and 4 had a voice-sparing procedure. One patient who was continuously free of disease, died less than 24 months after salvage surgery; this patient was excluded from all local control and survival analyses, leaving 22 patients available for analysis. Successful salvage was achieved initially in 13 (59%) of the 22 patients. Two patients with local failure after voice-sparing salvage surgery underwent completion laryngectomy; one was salvaged, so that 14 of 22 (64%) were ultimately salvaged. The rate of successful salvage did not correlate with preirradiation T stage, time to failure after irradiation, or time from clinical suspicion of recurrence to histologic proof. In a subgroup of patients (those with positive margins, tumor extension into the soft tissues of the neck, or two or more adverse histologic features), the risk of local-regional recurrence after salvage laryngectomy was 80%., Conclusion: There are few data in the literature regarding the clinical outcome of patients whose tumors are not controlled by initial radiotherapy. This series and previous publications from other institutions indicate that most patients who develop primary failure undergo salvage surgery, which is successful in approximately 50-80% of the patients. The rate of surgical complications is acceptable and is related to radiotherapy dose. At our institution, most patients who had recurrence after radiotherapy underwent total laryngectomy. More than 50% of the patients who experienced recurrence after irradiation originally had tumors of such extent that they would have required total laryngectomy if surgery had been recommended instead of radiotherapy for the initial treatment. It is reasonable to consider reirradiation in patients with a very high risk of local-regional recurrence after salvage laryngectomy.

Background: Currently, many patients with early vocal cord cancers are treated with 6 MV photons, but almost all the published radiotherapy data are based on patients treated with 60Co, 2-MV, or 4-MV X-rays. A theoretical risk of underdosage exists with higher energy beams due to lack of dose build-up. This dosimetric study compares 6-MV photons with 60Co., Methods: A tissue-equivalent phantom was constructed of a stack of 0.5-cm-thick acrylic plates. With a male subject in treatment position as the model, the external surfaces of the phantom were machined to match the contour of the neck. To precisely represent the internal contour of the airway, computed tomography (CT) was performed on the subject in treatment position, with images at 0.5-cm intervals, and the airway shown on the CT was cut out of each corresponding acrylic plate. Thermoluminescent dosimetry (TLD) rods were inserted into the phantom. For each measurement, a calculated tumor dose of 10 Gy was delivered to the volume specified as the entire right true vocal cord in the phantom, with either 60Co or 6-MV photons (15 measurements were made with each). In a second series of eight experiments with each modality, TLD minichips were used to measure the dose received by the immediate surface of the vocal cords with delivery of a calculated tumor dose of 0.5 Gy., Results: The doses received at the vocal cords, as well as a point 6 mm beneath the anterior skin surface, did not differ significantly for the two energies compared. The dose delivered to the skin and a point 3 mm beneath the anterior skin surface was significantly lower with the use of 6-MV photons., Conclusion: Although there is no difference in the dose received by the vocal cords, underdosage of the anterior tissues may occur with the use of 6-MV photons.

Background: The purpose of this article is to present the results of radiotherapy with or without neck dissection for squamous cell carcinoma of the supraglottic larynx at the University of Florida and to compare these data with those obtained after conservation surgery., Methods: Continuous-course radiotherapy alone or combined with a planned neck dissection was used to treat 209 patients with 211 supraglottic carcinomas between 1964 and 1992; all patients had follow-up for > or = 2 years., Results: The 5-year rates of local control after radiotherapy were as follows: T1, 100%; T2, 83%; T3, 68%; and T4, 56%. Tumor volume, as calculated on pretreatment computed tomography (CT) scan, and vocal cord mobility influenced the likelihood of local control., Conclusions: Based on our data and the literature, early or moderately advanced supraglottic carcinomas may be treated successfully with either supraglottic laryngectomy or radiotherapy. Supraglottic laryngectomy probably produces a higher initial local control rate but, based on anatomic and coexisting medical constraints, is suitable for a smaller subset of patients and has a higher risk of complications compared with radiotherapy.

Background: Little information about the incidence of retropharyngeal adenopathy and its impact on prognosis has been published., Methods: For 774 patients with squamous cell carcinoma of the nasopharynx, oropharynx, hypopharynx, or supraglottic larynx, pretreatment CT and, in selected cases, MRI scans were reviewed to determine the presence of retropharyngeal adenopathy. Results were analyzed in 619 patients treated with curative intent to determine the prognostic impact of retropharyngeal adenopathy., Results: The highest incidence of retropharyngeal adenopathy was seen in patients with nasopharyngeal (74%) and pharyngeal wall (19%) cancers. The number of cervical nodal groups involved was the most significant factor (p < .0001) relating to the incidence of retropharyngeal adenopathy. The rates of neck relapse (40% at 5 years) and distant metastasis were significantly higher in patients with retropharyngeal adenopathy, and the rates of 5-year relapse-free survival and absolute survival were significantly lower., Conclusions: Retropharyngeal adenopathy is a strong predictor of poor prognosis, particularly for patients with advanced neck disease.

Background: Merkel cell carcinoma is a relatively rare neuroendocrine carcinoma of the skin. It arises in the head and neck region in approximately 50% of cases. Its aggressive behavior predisposes patients to local-regional recurrence and distant metastases after surgical excision alone. In this article, we describe our experience with Merkel cell carcinoma of the head and neck., Methods: Of 18 patients with Merkel cell carcinoma treated in the Department of Radiation Oncology at the University of Florida, 12 patients who had primary tumors in the head and neck region are reported. Eight patients were treated at initial diagnosis (group A), and four were treated at the time of local-regional recurrence (group B)., Results: Local-regional control was achieved in seven of eight patients in group A and all four patients in group B. One patient in group A and all patients in group B developed distant metastases and eventually died of their disease. Bone exposure developed in one patient, requiring surgical debridement and hyperbaric oxygen treatment., Conclusion: Patients with Merkel cell carcinoma of the head and neck should be treated aggressively. Our data suggest that local-regional recurrence is a harbinger of distant metastases. We recommend that these patients receive treatment to both the primary site and draining lymphatics at initial presentation. The role of chemotherapy remains unclear.

Background: Oral tongue cancer may be treated primarily with radiotherapy or with surgery alone or combined with adjuvant radiotherapy; the choice between these two approaches is controversial., Methods: To evaluate the results of a shift in treatment policy in 1985 in favor of primary surgical treatment for carcinoma of the oral tongue, the results of radiotherapy (with or without neck dissection, 105 patients) were compared with those for surgery (with or without radiotherapy, 65 patients)., Results: Local control rates were improved for T3 (p = .03) and 14 (p = .08) patients treated surgically but were similar for T1-T2 patients. Local-regional control and survival rates were not significantly different. The rate of severe complications was significantly higher (p = .01) for T3 patients treated with surgery, particularly in the subset of patients who received postoperative radiotherapy., Conclusions: We generally recommend surgical treatment for T1-T2 patients with the addition of postoperative twice-a-day radiotherapy in selected cases. For selected T3-T4 patients we are investigating split-course twice-a-day preoperative radiotherapy in the hope that the extent of the surgical procedure, and hence the rate of severe complications, will be reduced.

Sixty-seven patients with 68 stage T4 carcinomas of the skin of the head and neck were treated with radical radiotherapy at the University of Florida between October 1964 and November 1989. Thirty-three lesions were previously untreated and 35 were recurrent. Twenty-nine lesions were squamous cell carcinomas, 37 were basal cell carcinomas, and 2 were basosquamous carcinomas. Minimum follow-up was 2 years. The 5-year local control, local control including surgical salvage, and cause-specific survival probabilities were 53%, 74%, and 75%, respectively. Local control rates with radiotherapy alone were poorer in patients with recurrent lesions (41% vs. 67%, p = .07) or bone involvement (40% vs. 62%, p = .08). Results were analyzed by multivariate methods using local control, local control with surgical salvage, and cause-specific survival as endpoints. The parameters analyzed were histology; size of primary lesion; previous treatment (previously untreated vs. recurrent); involvement of bone, nerve, or cartilage; and skeletal muscle invasion. Three important prognostic factors were identified, each predictive of poorer ultimate local control and cause-specific survival rates: (a) bone involvement (p < .01); (b) recurrent lesions (p < .01); and (c) nerve involvement (p < .02). Radiotherapy alone can control advanced carcinomas of the skin of the head and neck, although lesions that have recurred after prior treatment and those with involvement of bone or nerve are associated with a lower likelihood of cure.

Between October 1964 and June 1990, 243 patients with squamous cell carcinoma of the tonsillar region were treated with continuous-course radical radiotherapy at the University of Florida. All patients had a minimum 2-year follow-up. Patients were staged according to the AJCC system; stage IV was stratified into two subsets: favorable, IVA (T1-T3 or N2A-N3A); and unfavorable, IVB (T4 or N3B). The initial and ultimate local control rates (including surgical salvage) according to T stage were as follows: T1, 87% and 100%, T2, 79% and 92%; T3, 71% and 76%; and T4, 44% and 48%. Multivariate analysis demonstrated that anterior extension of the primary tumor (p = .0001) and T stage (p = .014) were the most significant parameters affecting local control after radiotherapy. For T4 lesions, twice-daily irradiation significantly improved local control (p = .04). The 5-year absolute and cause-specific survival rates as a function of modified AJCC stage were as follows: I, 37% and 100%; II, 55% and 90%; III, 55% and 85%; IVA, 35% and 60%; and IVB, 23% and 38%. The probability of a severe complication was 3% for the entire group of patients.

This is an analysis of 16 patients with adenoid cystic carcinoma of the palate who were treated with curative intent by radiotherapy alone (9) or surgery and radiotherapy (7). No patient received adjuvant chemotherapy. Follow-up ranged from 3 years to 19 years with 69% of the patients having follow-up of more than 5 years. Two patients in the radiotherapy alone group developed local recurrences, and both have had long-term surgical salvage. No local failures were seen in the combined therapy group. Twelve patients (75%) are alive and disease-free. Two patients died with distant disease, one within each treatment group. One patient in the combined therapy group is alive at 8 years with distant metastases, while one in the radiotherapy alone group died of intercurrent disease at 12 years. No significant complications were seen in either treatment group. The role of radiotherapy in the treatment of adenoid cystic carcinoma of the palate is reviewed.

Between March 1978 and March 1989, 419 patients with 439 moderately advanced or advanced primary squamous cell carcinomas of the head and neck received treatment with curative intent with twice-a-day radiotherapy (120 cGy per fraction with a 4- to 6-hour interfraction interval). Seventy-nine percent of the patients had AJCC stage III or stage IV cancers. The primary site in most patients was the oropharynx, hypopharynx, or larynx. All patients received radiotherapy alone to the primary site, with or without planned neck dissection after radiotherapy. Surgery at the primary site was reserved for salvage of radiotherapy failures. No patient received chemotherapy as part of the initial treatment plan. Total doses were 7440 cGy to 7920 cGy in the majority of patients. In 54 patients with oropharyngeal lesions, an interstitial implant was used to add a 1,000- to 1,500-cGy boost after the basic dose. Local control results were compared on a site-by-site and stage-by-stage basis with the results of continuous-course, once-a-day radiotherapy at the University of Florida; the control results were the same as or better than after once-a-day radiotherapy for each site and stage (T2 to T4). Severe complications of radiotherapy occurred in 4% of patients and correlated with tumor dose, T stage, and disease site (oropharynx or larynx/hypopharynx). The addition of a neck dissection 4 to 6 weeks after radiotherapy of patients with advanced-stage neck disease was accomplished with an acceptable rate of morbidity and significantly improved the rate of control of neck metastases compared with radiotherapy alone.

Between 1964 and 1987, 194 patients with previously untreated squamous cell carcinoma of the floor of mouth were managed at the University of Florida. A retrospective analysis was undertaken in order to evaluate the treatment results and associated complication rates. Surgery or irradiation alone was found to result in similar local control rates for stage I and II lesions, whereas more advanced tumors had better local control rates with a combination of surgery and irradiation. Radiotherapy had a higher incidence of minor and moderate complications, whereas a greater number of severe complications occurred after surgery. We recommend surgery for early lesions due to the lower overall incidence of associated complications. Despite a higher risk of severe complications, combination therapy is recommended for more advanced lesions due to improved local control as compared to single modality therapy.

Chemodectomas of the carotid body and of the glomus vagale are uncommon tumors of the head and neck. Between 1981 and 1986, 6 patients with advanced chemodectomas of either the carotid body (5 patients) or the glomus vagale (1 patient) were treated with radiotherapy at the University of Florida. Tumors were bilateral in 2 patients. Five patients have had no evidence of disease progression for 5, 5.5, 7, 7, and 8.5 years following irradiation. One of the 5 patients experienced complete regression, 2 noted partial regression, and 2 remained stable (without progression). One patient, who had received 2 prior operations and a course of irradiation at another institution, died of locally recurrent disease 5 years after re-irradiation at the University of Florida. No patient experienced a significant complication secondary to irradiation.

An analysis of 508 patients (660 heminecks) with head and neck squamous cell carcinoma and clinically positive neck nodes who were treated with radiotherapy alone to the primary lesion (with or without a neck dissection) was conducted to determine if open neck-node biopsy before definitive treatment adversely affected the probability of control of neck disease, the risk of distant metastasis, or the cause-specific survival rate. The prognostic factors analyzed included biopsy status of the neck, N stage, neck treatment, node mobility, node location, T stage, primary site, and control of disease above the clavicles. Sixty-six patients who had undergone an open neck-node biopsy before definitive radiotherapy were compared with a control group of 442 patients who did not undergo a neck-node biopsy; no detrimental effect of the biopsy on neck control, distant metastasis, or cause-specific survival was demonstrated. We conclude that the potential adverse effect of violating the neck before definitive treatment cannot be demonstrated if radiotherapy is the next step in the patient's management.

Sixty-nine patients with metastatic squamous cell carcinoma in neck nodes with an unknown primary lesion were treated with curative intent between October 1964 and December 1986. Sixty-five patients received radiotherapy to the neck and at least part of the mucosa of the head and neck, whereas 4 patients received treatment to the neck only. Mucosal doses were approximately 5,000 cGy-6,000 cGy at 170 cGy-180 cGy per fraction. Eight patients (12%) subsequently developed mucosal site failures, a figure that did not differ significantly from the incidence of a second metachronous head and neck cancer following definitive irradiation in a series of 393 patients with lesions of the supraglottic larynx, pharyngeal wall, pyriform sinus, or tonsillar area. This suggests that radiotherapy was highly effective in preventing the appearance of cancer at the unknown primary site from which the neck metastasis arose. Absolute and cause-specific 5-year survival rates were 48% and 66%, respectively.

An analysis of 129 patients with 131 squamous cell carcinomas of the supraglottic larynx treated between October 1964 and April 1987 with radiotherapy alone or radiotherapy followed by neck dissection is presented. All patients had a minimum 2-year follow-up. Patients were excluded from analysis of disease control at the primary site and/or neck if they died within 2 years of treatment with the site(s) continuously disease-free. Local control rates with radiotherapy and ultimate local control rates, including patients successfully salvaged after a local recurrence, were as follows: T1, 13 of 13 and 13 of 13; T2, 34 of 42 (81%) and 37 of 42 (88%); T3, 25 of 41 (61%) and 34 of 41 (83%); and T4, 3 of 9 and 6 of 9. There was no significant difference in local control rates when comparing patients who were anatomically suitable for a supraglottic laryngectomy with those who would have required a total laryngectomy. Local control rates were slightly diminished in patients with T2-T3 lesions who had impaired or absent vocal cord mobility. The overall rates of ultimate local control with voice preservation for the entire series of 129 patients were as follows: T1, 100%; T2, 87%; T3, 69%; and T4, 57%. Cause-specific survival rates at 5 years by stage were I, 2 of 2; II, 10 of 12 (83%); III, 9 of 13 (69%); IVA, 4 of 6; and IVB, 7 of 22 (32%). The incidence of severe complications was 4 of 115 (3%) for T1-T3 lesions and 4 of 14 (29%) for T4 lesions.

Between March 1978 and April 1986, 21 previously untreated patients with T3 squamous cell carcinoma of the true vocal cord received radical courses of twice-a-day radiotherapy (120 cGy twice daily) to total doses of 7,440-7,680 cGy. The rate of local control at the primary site was 67% after radiotherapy. The ultimate control rate after salvage surgery was 83%. Five-year absolute and cause-specific survival rates were 59% and 81%, respectively. It is concluded that radiotherapy is a rational approach for many patients with T3 glottic cancer.

This is an analysis of 16 patients with excisionally biopsied (TXN0) squamous cell carcinoma of the oral tongue (nine patients) and floor of mouth (seven patients) treated with radiotherapy. All patients had a minimum 2-year follow-up and 81% had at least 5 years of follow-up. One patient died of intercurrent disease 16 months after treatment and was excluded from analysis of local control. All patients were evaluable for analysis of survival and complications. Local control was achieved in 14 of 15 patients; one patient with local recurrence underwent a surgical salvage procedure, which was unsuccessful. No patients developed recurrent disease in the neck or in distant sites. Five-year absolute and cause-specific survival rates were 10 of 13 and 10 of 11, respectively.

This is an analysis of 25 patients with carcinoma of the skin of the head and neck with clinical evidence of perineural invasion, who were managed with curative intent by radiotherapy alone (16) or surgery and radiotherapy (9). All patients had a minimum of 5 years of follow-up. Local control following treatment was observed in 20% of patients treated with radiotherapy alone and 38% of those treated with surgery and radiotherapy. Local control was diminished for recurrent lesions and for tumors that also invaded bone and/or cartilage. The 5-year absolute survival rates were 31% for patients treated with radiotherapy alone and 33% for those in the combined-treatment group. The incidence of significant complications was similar: 25% with radiotherapy alone and 33% with surgery and adjuvant radiotherapy.

Between March 1978 and April 1986, 56 patients underwent planned unilateral, and five patients underwent planned bilateral, radical neck dissections following high-dose twice-a-day radiotherapy, usually with 60Co. Neck dissections were generally performed 6 weeks after radiotherapy. The overall rate of control of disease in the neck in all 61 patients was 81% at 5 years. Patients who underwent neck dissections following radiotherapy had significantly higher rates of neck-disease control than patients treated by radiotherapy alone during the same time period for stages N2A-N3A and N2B-N3B, but not for N1 disease. No carotid ruptures or fatal complications occurred. The rate of wound complications was judged to be acceptable.