Organ-function preservation in advanced oropharynx cancer: results with induction chemotherapy and radiation.

PURPOSE: To evaluate the feasibility and efficacy of a strategy using induction chemotherapy followed by radiation therapy (RT) as a means of organ-function preservation in patients with advanced oropharynx cancer. PATIENTS AND METHODS: From January 1983 to December 1990, 33 patients with advanced squamous cell oropharynx cancer whose appropriate surgical management would have required a tongue procedure and potential total laryngectomy were treated with one to three cycles of cisplatin (CDDP)-based induction chemotherapy. Patients with a complete response (CR) or partial response (PR) at the primary site then received definitive external-beam RT with or without interstitial implant with or without neck dissection with surgery to the primary tumor site reserved for disease persistence or relapse; patients with less than a PR after chemotherapy had appropriate surgery and postoperative RT recommended. RESULTS: With a median follow-up period of 6.2 years, actuarial overall and failure-free survival rates at 5 years are 41% and 42%, respectively. Chemotherapy toxicity contributed to the death of two patients and was possibly a factor in two others. Local control was achieved in 14 patients (42%) without any surgery to the larynx or tongue. Among 13 patients currently alive, all had a preserved larynx and only one required tongue surgery; 12 of 13 have speech subjectively described as always understandable; and nine of 13 have no significant restrictions in their diet. CONCLUSION: This treatment program is feasible and effective in patients with advanced oropharynx cancer and produces an excellent functional outcome in most long-term survivors. Modifications to optimize patient selection, minimize toxicity, and improve local control are indicated. The relative toxicity, efficacy, and functional outcome provided by this and other chemotherapy and RT programs versus either standard surgery and/or RT options can only be addressed in a randomized comparison of these therapies.

Prospective computer-assisted voice analysis for patients with early stage glottic cancer: a preliminary report of the functional result of laryngeal irradiation.

In January 1987 we began a prospective study aimed at evaluating objective parameters of vocal function for all patients treated with RT for early glottic cancer. All patients underwent vocal analysis using a voice analyzer interfaced with a computer. This allowed for the determination of percent voicing (%V) (normal = presence of phonation = 90-100%V). Other parameters such as breathiness (air turbulence or hoarseness) and strain (vocal cord tension) were also measured. Patients were recorded before RT, weekly during RT, and at set intervals after RT. There have been 25 patients studied. Eighteen (18) are evaluable at 9 months after treatment. All patients were male and ranged from 45-84 years old. Fourteen (14) and T1 lesions and received 66 GY/33 fractions to their larynx and 4 had T2 tumors and received 66-70 Gy/33-35 fractions. To date, all patients are locally controlled. Three distinct patterns of %V changes have been encountered. However, all patients demonstrated normal phonation pattern by 3 months after RT, and this is sustained at 9 months follow-up. In addition, 94% of patients have had significant decrease in breathiness after RT, which objectively documents diminished hoarseness. In 83%, breathiness is normal after RT. Most patients have had increased strain after RT, which documents increased vocal cord tension. However, strain remained within normal limits in 89%. Our preliminary analysis suggests that the majority of patients irradiated for early glottic cancer demonstrate a decrease in breathiness and an increase in strain after RT, and enjoy a resultant voice that has normal phonation maintained at 9 months after RT. Our data also demonstrate three distinct phonation patterns. Further follow-up will allow us to determine the prognostic significance, if any, of these patterns, and to continue to follow objective vocal parameters on larger numbers of patient.

Brachytherapy as part of the definitive management of squamous cancer of the base of tongue.

Between 1981 and 1986, 17 patients were treated at the Department of Radiation Oncology at the Memorial Sloan-Kettering Cancer Center with squamous cancer of the base of the tongue whose definitive treatment included brachytherapy. The patient sample consisted of 14 men and 3 women with age range of 35 to 71 years (median = 58). There were four patients with T1 lesions, six with T2, six with T3, and one with T4. In general, treatment consisted of 5000-5400 cGy with external beam radiation and 2000-3000 cGy boost to the base of tongue via an Ir-192 implant using afterloading catheters. Necks were managed with elective radiation alone in the N0 group (n = 5) or with radiation plus neck dissection in the N+ group (n = 12). Five patients who would have required laryngectomy had they undergone primary surgery received neoadjuvant chemotherapy followed by external beam and implant as part of a larynx preservation study that was being done at our institution (4-T3, 1-T2). The range of follow-up is 8 to 59 months, with median follow-up of 24 months. No patients have been lost to follow-up. Crude local control by T-Stage are as follows: T1-4/4, T2-5/6, T3-5/6, T4-1/1. Actuarial local control at 24 months is 87%. There have been no neck failures. There have been five patients who had soft tissue ulceration (STU) and one patient who had osteoradionecrosis (ORN). All soft tissue ulceration patients have been successfully managed conservatively. The patient with osteoradionecrosis is currently being managed. In 4 of these 6 cases, the implant was the initial therapeutic intervention and the entire tumor bed was implanted. On the other hand, when external beam was the initial treatment, the boost was administered to the smaller volume of residual disease. Overall, 4 of 7 patients who had implant first developed either soft tissue ulceration or osteoradionecrosis, as opposed to 2 of 10 patients who had implant after external beam and/or chemotherapy. The numbers are too small to be statistically significant, but our current policy is to perform brachytherapy after the external beam. In addition, all those with either soft tissue ulceration or osteoradionecrosis were implanted with a non-looping technique. Overall, 6 of 12 patients treated with a non-looping technique developed an injury, whereas none of the five treated with a looping technique has developed one.(ABSTRACT TRUNCATED AT 400 WORDS)

Postoperative radiation therapy for squamous cell carcinomas of the oral cavity and oropharynx: impact of therapy on patients with positive surgical margins.

PURPOSE: The presence of a positive or close margin after resection of a squamous cancer of the head and neck is associated with a significant risk of local recurrence. To determine the efficacy of postoperative radiation therapy for patients with advanced oral cavity and oropharyngeal cancers with inadequate margins of resection, the present retrospective analysis was undertaken. METHODS AND MATERIALS: One hundred and two patients were treated with surgery and postoperative radiation therapy for advanced squamous cell carcinomas of the oral cavity and oropharynx. The anatomic subsites treated include oral tongue (n = 29), floor of mouth (n = 22), base of tongue (n = 31) and tonsillar fossa (n = 20). Twenty-five patients (25%) had positive margins, 41 patients (40%) had close margins (< or = 0.5 cm from the surgical margin) and 36 (35%) had negative margins. The median radiation dose was 6000 cGy. RESULTS: With a median follow-up of 7 years, the actuarial control rate for patients with positive, close and negative margins was 79%, 71%, and 79%, respectively. When postoperative doses of > or = 60 Gy were delivered to patients with positive/close margins (excluding patients with oral tongue lesions), the 7-year actuarial control was 92%. In similar patients receiving < 60 Gy, the actuarial control was 44% (p = 0.0007). Compared to other anatomic subsites, inferior control rates were obtained with oral tongue lesions. For this subsite, the control rates for positive, close, and negative margins were 50%, 62% and 69% respectively. CONCLUSION: We conclude that excellent local control can be achieved with postoperative radiation therapy, despite the presence of inadequate margins of resection, when doses of > or = 60 Gy are used. Future strategies must be directed at further improving these results in patients with oral tongue lesions.

Paraspinal tumors: techniques and results of brachytherapy.

Because of their proximity to nerve roots and the spinal cord, it is frequently difficult to achieve complete resection of paraspinal tumors. We have used brachytherapy in an attempt to prevent local recurrence and its associated neurological sequelae. This report analyzes our experience with 35 patients to determine the feasibility, optimal techniques, and efficacy of this approach. The tumor types were non small-cell lung cancer (18), sarcomas (9), and other tumor types (8). Temporary, single plane implants using Ir-192 (median minimum peripheral dose 3000 cGy) were used in 21 patients, and permanent I-125 implants were used in 14 cases (median matched peripheral dose 12,500 cGy). Local control was achieved in 51% (18/35). However, local control was poor when lung cancers were implanted and in cases where the dura was exposed. Radiation myelitis did not occur despite the combined effects of previous external beam radiotherapy (N = 21) and brachytherapy. Our experience demonstrates that combined surgery and paraspinal brachytherapy can be performed with acceptable toxicity and is reasonably effective in preventing local relapse and its neurologic sequelae, particularly for tumors other than lung cancers.

Concomitant chemotherapy-radiation therapy followed by hyperfractionated radiation therapy for advanced unresectable head and neck cancer.

In January 1988, we initiated a prospective study evaluating a new treatment approach with chemotherapy and radiotherapy for unresectable head and neck cancer. Weeks 1-4 were the initial debulking phase. Radiotherapy was delivered using 1.8 Gy/day to large portals including gross disease and all areas at risk. Cisplatinum, 100mg/m2, was given concomitant with radiotherapy on days 1 and 22. Weeks 5 and 6 were the boosting phase. This involved twice-a-day irradiation. The AM fraction of 1.8 Gy was given to the entire area at risk, whereas the PM dose of 1.6 Gy was limited to the gross disease alone. Thus a total of 70 Gy/6 weeks was delivered. A total of 24 patients were treated, 22 of whom have completed this protocol and are evaluable. All patients had massive disease, with 15 having gross involvement of brain, orbit, skull base, or carotid artery. Follow-up ranged from 3-22 months (median 12 months). Major responses were achieved in 96%, with 64% complete responses and 32% partial responses. Two patients with PR were able to undergo complete surgical resection, making the overall rate of complete response, including surgery, 73%. At 1 year, actuarial survival was 69%, and local progression-free survival was 56%. Distant metastases developed in 5 (23%). This approach appears both safe and effective in producing excellent regression and local control for far advanced head and neck cancer. More time is needed to see if these results are sustained. Efforts to build upon this experience appear warranted.

Twice daily thoracic irradiation for limited small cell lung cancer.

Thoracic failure is a significant obstacle to the cure of limited stage small-cell lung cancer (LSCLC) patients treated with combined modality therapy. In 1985 we initiated a prospective trial to evaluate the impact of twice daily thoracic irradiation without concomitant chemotherapy on control of intrathoracic tumor in LSCLC. Twenty-nine patients treated in this fashion were compared with 36 patients treated from 1979-1982 with once daily thoracic irradiation and concomitant chemotherapy. Both groups received the same induction chemotherapy; cyclophosphamide, Adriamycin, and vincristine (CAV) alternating with cisplatin and etoposide. For consolidation, the twice daily patients received thoracic irradiation, 45 Gy in 1.5 Gy fractions given twice daily, and the once daily patients received thoracic irradiation, 45 Gy in 2.5 Gy fractions given once daily with concomitant cyclophosphamide and vincristine. After completion of radiotherapy both groups received maintenance chemotherapy. The complete response (CR) rate after thoracic irradiation was higher for twice daily patients (86% (25/29) compared to the once daily patients [61% (22/36), p = 0.02]. However, this advantage was offset by the shorter duration of thoracic control among CR patients treated with twice daily thoracic irradiation compared to once daily thoracic irradiation (32% vs 67% at 2 years, p less than 0.05). In view of the enhanced initial response of LSCLC to twice daily thoracic irradiation, this basic radiotherapeutic approach seems appropriate, but new strategies using higher doses of twice daily thoracic irradiation or concomitant chemotherapy appear to be necessary to enhance long-term thoracic control.

Improved dose distributions for 3D conformal boost treatments in carcinoma of the nasopharynx.

This study was designed to demonstrate the feasibility of 3-dimensional (3D) treatment planning in patients with carcinoma of the nasopharynx, and to explore its potential therapeutic advantage over the traditional 2-dimensional (2D) approach in this disease. Qualitative and quantitative comparisons between the two techniques were made for the boost portion of the treatment (19.8 Gy of a total 70.2 Gy treatment schedule) in 10 previously untreated patients and for the entire treatment in 5 patients with locally recurrent disease. The 2D and 3D plans were compared in each patient using dose-volume histograms (DVH's), tumor control probabilities (TCP's), normal tissue complication probabilities (NTCP's), and a new biologic figure of merit that describes the probability of uncomplicated control. Although there was no attempt to optimize the 3D treatment approach by using this method throughout the total treatment course (rather than for the boost only), it was still found that for each of the endpoints examined the 3D approach resulted in improved plans. An average of 22% of the target volume was underdosed at the 95% isodose level with the 2D plans compared to 7% with the 3D plans. The improved treatment planning by 3D increased the mean dose to the tumor volume by an average of 13% over 2D planning. The dose to normal structures such as the mandible and parotid glands was reduced with the 3D plans while the brain stem and spinal cord remained within tolerance limits. The probability of uncomplicated tumor control was increased by an average of 15% with 3D treatment planning compared to the 2D approach. Our findings demonstrate the potential of 3D planning for improving the treatment of carcinoma of the nasopharynx, but prospective studies are required to define the true clinical advantages of this methodology.

Nasopharyngeal brachytherapy with access via a transpalatal flap.

We used a posteriorly based transpalatal flap for interstitial nasopharyngeal brachytherapy in five patients with nasopharyngeal carcinoma. All of these implants were done with iodine 125 seeds. The disease in three of five patients has been controlled locally for 27, 30, and 40 months, respectively. Two patients experienced local failure and died. There have been no palatal healing problems despite high doses of previous radiation in all of our patients. No palatal incompetence has been seen. This technique appears safe and should be considered for discrete lesions of the high posterior or superior nasopharynx that cannot be approached by less invasive methods.

Malignant tumors of major salivary gland origin. A matched-pair analysis of the role of combined surgery and postoperative radiotherapy.

Between 1966 and 1982, 46 patients with previously untreated malignant tumors of major salivary gland origin received combined surgery and postoperative radiotherapy. They were compared with 46 patients treated with surgery only between 1939 and 1965, who were matched according to prognostic criteria. Radiation doses ranged from 4000 to 7740 cGy (median, 5664 cGy). The 5-year determinate survival rates for patients given combined therapy with stage I and II disease vs patients given surgery only was 81.9% vs 95.8%, while for stages III and IV it was 51.2% vs 9.5%, respectively. Local control for stage III and IV disease in patients given combined therapy vs patients given surgery only at 5 years was 51.3% vs 16.8%. For patients with nodal metastases, 5-year determinate survival for the combined-therapy group vs the surgery-only group was 48.9% vs 18.7%, and the corresponding local-regional control was 69.1% vs 40.2%. The results of this analysis suggest that postoperative radiotherapy significantly improves outcome for patients with stage III and IV disease and for patients with lymph node metastases.

Larynx preservation with combined chemotherapy and radiation therapy in advanced hypopharynx cancer.

Twenty-five untreated patients with advanced, resectable squamous cell carcinoma of the hypopharynx, for whom standard treatment would have required total laryngectomy, were treated with one to three cycles of cisplatin-based chemotherapy with larynx preservation as the goal. Patients with a major (complete or partial) response to chemotherapy at the primary site were treated with definitive radiation therapy, with total laryngectomy reserved for salvage; patients with less than a partial response to chemotherapy had total laryngectomy and postoperative radiation therapy recommended. Four patients had a poor response to chemotherapy and thus were not candidates for laryngectomy. Total laryngectomy was required for initial induction chemotherapy failure in five patients and for local recurrence in five others. Three additional patients had unresectable recurrence. Successful larynx preservation was achieved in 32% (8 of 25). With a median follow-up period of 41 months, the actuarial overall and failure-free 2-year survival rates were 44% and 32%, respectively. These preliminary data suggest larynx preservation is feasible in patients with advanced lesions of the hypopharynx. Improved local and regional control must be incorporated into the larynx preservation approach for hypopharyngeal lesions. A prospective, randomized study is necessary for a more valid comparison with conventional therapy, including comparative assessments of survival, morbidity, cost and functional results.

Radiation therapy for oral cavity cancer.

Radiation therapy, using both external beam and brachytherapy, is one of the mainstays of treatment for oral cavity cancers. For early lip, tongue, and floor of mouth lesions, radiation alone is highly effective and produces an excellent functional result. More advanced lesions are frequently treated with combined therapy or radiation alone with surgical salvage. Buccal mucosa and gingiva lesions are generally treated with surgery but can be managed by radiation therapy if they are early. More advanced lesions are treated by surgery plus postoperative radiation. Retromolar trigone lesions can be managed by radiation alone if they are small, but larger lesions are generally treated by combined therapy. Clearly, the goal of treatment of squamous cell cancer of the oral cavity is cure of the disease with optimal function results. These goals can be achieved only by cooperative interdisciplinary evaluation and management by the surgeons, radiation oncologists, and dentists involved in the care of these patients.

Limited small cell lung cancer: prognostic significance of a complete response to the induction phase of chemotherapy followed by thoracic irradiation.

Limited-stage small cell lung cancer is frequently treated with induction combination chemotherapy (ICC), followed by consolidation with thoracic irradiation. It has been suggested that patients who do not have a complete response to ICC are unlikely to have control of occult distant metastasis and consequently have such a poor prognosis that thoracic irradiation is unlikely to be of benefit. To examine this hypothesis, 48 patients treated on prospective protocols who achieved a complete response to ICC or subsequently to thoracic irradiation were analyzed. Twenty-four patients had a complete response to ICC (CR-ICC), and 24 subsequently converted to complete-response status after thoracic irradiation (CR-TI). The two groups had similar prognostic factors and treatment. Comparing CR-ICC and CR-TI patients, survival was 40% versus 26% at 2 years and 35% versus 4% at 5 years, respectively (P less than .05). Freedom from distant metastasis was 41% at 5 years for the CR-ICC patients and 8% for the CR-TI patients (P less than .05). A modest number of CR-TI patients were long-term survivors, suggesting a value for thoracic irradiation as consolidation therapy.

Observations on the natural history and treatment of recurrent major salivary gland cancer.

In an attempt to define the natural history and the indications for postoperative radiotherapy (RT) in locally recurrent major salivary gland cancers, we reviewed 78 patients treated between 1965 and 1982. All patients underwent resection of the tumor. Group 1 (N = 38) had complete resection of tumors, with low or intermediate grade histology and without lymph node spread. Group 2 consisted of the remaining 40 patients who had high-risk features (HR) (high grade histology, lymph node metastases, and close or positive margins of resection). Half of the Group 2 patients received radiation therapy. Overall survival was 63% at 5 years, and 35% at 15 years. Survival of Group 1 was 83% at 5 years and 58% at 15 years. Local control for Group 1 was 69% at 5 years, 54% at 15 years, and was size-dependent. Group 2 survival was 40% at 5 years and 29% at 10 years. Local control at 5 and 10 years was 49% and 35%. Our data indicate that surgery alone yielded good local control in patients with small tumors (less than or equal to 3 cm) and no HR features, suggesting that postoperative RT may be unnecessary for these patients. Other patients have suboptimal local control, and although there is a rationale for the use of postoperative radiotherapy, we cannot assess its impact in this retrospective study.

Postoperative radiation therapy for major salivary gland malignancies.

Between 1966 and 1982 there have been 46 patients treated with surgery plus post-operative radiation therapy for malignant tumors of salivary gland origin. The indication(s) for radiotherapy included positive margins (42%), advanced local tumor (37%), positive nodes (33%), or high grade histology (48%). Overall actuarial local control at 5 years was 73%, being 100% for T1, 83% for T2, 80% for T3, and 43% for T4. Actuarial survival at 5 years was 80% for T1, 83% for T2, 60% for T3, and 48% for T4. Patients with positive nodes (N+) did worse than those with negative nodes (No), with locoregional control and survival at 5 years being 58% vs. 83%, (P = 0.025) and 38% vs. 80% (P = less than .01), respectively. We found no need for contralateral neck treatment even for those with positive nodes. Also, to date, none of eight patients with adenoid cystic histology has failed locally, as opposed to three of eight failures in patients treated with surgery alone. We believe that post-operative irradiation provides excellent locoregional control for appropriate patients with malignant tumors of major salivary glands.

Esthesioneuroblastoma: is there a role for elective neck treatment?

A retrospective review of the Memorial Sloan Kettering Cancer Center experience (MSKCC) with esthesioneuroblastoma was performed. From 1975 to 1985 14 cases were identified. Overall 5- and 10-year survival was 86% and 70%, respectively. Four local failures were observed and 4 neck failures were observed. A review of the literature since 1966 revealed an unexpectedly high incidence of neck failure, and of the 21 of 110 patients with neck failures that were identified through the literature, there were 6 subsequent deaths.

Postoperative radiotherapy for oral cavity cancers: impact of anatomic subsite on treatment outcome.

We have retrospectively reviewed the treatment results of postoperative radiotherapy (RT) for advanced oral cavity cancers. The purpose of this study was to determine the impact of anatomic subsite on the results of treatment. Between 1975 and 1985, 51 patients with squamous cell carcinoma of the oral tongue (OT = 29 patients) and floor of mouth (FOM = 22 patients) were treated with combined surgery plus RT. All had an indication(s) for RT including advanced primary disease (T3 or T4) (29 patients), close or positive margins (34 patients), and multiple positive neck nodes and/or extracapsular extension (41 patients). With a median follow-up of 6 years, the 5-year actuarial local control rate was 74% and the rate of distant metastasis (DM) was 34%. Despite the similar T stage, margin status and median RT dose, the 5-year actuarial local failure rate was 38% for OT vs. 11% for FOM (p = 0.03). Furthermore, the median survival after recurrence was 9 months for OT and 40 months for FOM (p = 0.02). At 5 years the determinate survival for both sites was (55%), and the likelihood of developing a second malignancy was 31%. The likelihood of developing DM was 50% for FOM (N0-N1 = 3 of 12, N2-N3 = 8 of 10) and 21% for OT (N0-N1 = 4 of 21, N2-N3 = 1 of 8). This study highlights significant differences between FOM and OT cancers in response to combined surgery and RT. Future strategies should be directed at the enhancement of local control for OT and better systemic therapy for those with advanced N-stage FOM.(ABSTRACT TRUNCATED AT 250 WORDS)

Base-of-tongue cancer treated with external beam irradiation plus brachytherapy: oncologic and functional outcome.

Between January 1981 and June 1990, 36 previously untreated patients with squamous cancer of the base of the tongue were treated with radiation therapy. This therapy consisted of a 5,000-5,400-cGy external beam plus a 2,000-3,000-cGy boost to the base of the tongue with an iridium-192 implant. Necks with negative nodes were only irradiated, while necks with positive nodes were treated with irradiation plus neck dissection. Actuarial local control and survival at 2 years were 87.5%. Neck control was achieved in 35 of 36 patients. When implantation was performed with the nonlooping technique, the injury rate was statistically higher than in patients who underwent implantation with the looping technique (P = .02). Thirty patients participated in a function assessment involving a Performance Status Scale for patients with head and neck cancer. The authors concluded that management of base-of-tongue cancer with external beam irradiation plus brachytherapy is both oncologically and functionally successful.

Impact of the time interval between surgery and postoperative radiation therapy on locoregional control in advanced head and neck cancer.

Between January 1975 and December 1980, 111 patients with AJCC stages III and IV squamous cell carcinoma of the head and neck were treated with surgery followed by planned postoperative radiation therapy. A previous analysis of a subgroup of these patients showed that, when radiation was delayed more than 6 weeks from surgery, a higher incidence of regional failure occurred compared with the incidence observed when therapy began within a 6 week period. We have looked back at this group of patients plus others in an attempt to determine whether other factors played a role in the results obtained. In the current study, 50 patients had a delay of 6 weeks or more and, of these, 11 (22%) suffered a locoregional recurrence. However, 8 of these 11 patients received suboptimal radiation doses (less than 56 Gy) for permanent control of the disease. In fact, of 17 patients who received at least 60 Gy and had more than a 6 week delay, only 2 (12%) had locoregional failure. This was similar to the incidence of failure in the patients who received at least 60 Gy and who started radiation within the first 6 weeks from surgery (3/20 [15%]). The effect of delay was apparent only in those who received less than 60 Gy (27% vs. 7%, P less than 0.05). Therefore, we cannot validate the previous conclusion that a greater than 6 week delay in the delivery of postoperative radiation therapy in advanced head and neck cancers produces poorer results. The current analysis suggests that a prolonged delay in postoperative radiation therapy in itself does not have a negative impact on locoregional control as long as appropriate tumorcidal doses of more than 60 Gy are employed.

The indications for elective treatment of the neck in cancer of the major salivary glands.

To define the indications for elective neck treatment, the cases of 474 previously untreated patients were reviewed who had locally confined major salivary gland cancers treated between 1939 and 1982. Clinically positive nodes were present in 14% (67 of 474). Overall, clinically occult, pathologically positive nodes occurred in 12% (47 of 407). By univariate analysis, several factors appeared to predict the risk of occult metastases; however, multivariate analysis revealed that only size and grade were significant risk factors. Tumors 4 cm or more in size had a 20% (32 of 164) risk of occult metastases compared with a 4% (nine of 220) risk for smaller tumors (P less than 0.00001). High-grade tumors (regardless of histologic type) had a 49% (29 of 59) risk of occult metastases compared with a 7% (15 of 221) risk for intermediate-grade or low-grade tumors (P less than 0.00001). In view of the low frequency of occult metastases in the entire group, routine elective treatment of the neck is not recommended. High-grade tumors and larger tumors have a high rate of occult neck metastases, and treatment should be considered in this group.