Pharmacokinetic evaluation of gemcitabine and 2',2'-difluorodeoxycytidine-5'-triphosphate after prolonged infusion in patients affected by different solid tumors.

BACKGROUND: The study determined pharmacokinetic parameters, toxicity profile and preliminary clinical activity of gemcitabine administered i.v. at different infusion rates in patients with a range of solid tumors. PATIENTS AND METHODS: Twenty patients were enrolled for both pharmacokinetic and clinical studies. Gemcitabine 300 mg/m(2) was administered during 1 h, 2 h or 3 h, and as a conventional dose of 1000 mg/m(2) during 30 min infusion. Administration was on days 1, 8 and 15 every 4 weeks. RESULTS: Patients were randomly assigned to one of the four arms. After 30 min infusion of 1000 mg/m(2) gemcitabine the plasma concentration remained above the saturation level of 10-20 microM, whereas after 1, 2 or 3 h infusion 300 mg/m(2) gemcitabine it remained below the saturation level for most of the time (being in the range 2.5-10 microM). Gemcitabine triphosphate was determined in the four arms in white blood cells; for infusion times from 0.5 to 3 h there was a progressive enhancement of gemcitabine triphosphate levels. In all evaluable patients the toxicity was mild, myelosuppression being the main toxicity. No grade 3 or 4 toxicities occurred. Clinical response was similar in patients receiving 300 mg/m(2) gemcitabine in 2 and 3 h and in the 1000 mg/m(2) arm. CONCLUSIONS: 300 mg/m(2) gemcitabine during 3 h infusion produced the highest accumulation of gemcitabine triphosphate. Thus, to achieve the highest possible gemcitabine triphosphate level, prolonged infusion time would appear to be more important than a high dose administered as a short infusion. However, there was no substantial difference in toxicity or antitumoral activity in the all different patient groups.

Ifosfamide in the treatment of head and neck cancer.

Ifosfamide (IFO) has demonstrated activity in recurrent/metastatic squamous cell head and neck carcinoma with an overall response rate of 24-26%. Better results are reported for chemotherapy-naive patients; in heavily pretreated cases results are poor and toxicity unacceptable. Cisplatin-IFO combination in stage III-IV is probably more active than IFO alone (ORR = 60-72 vs. 50%) but is indicated in patients who desire aggressive treatment and are physically able to tolerate the drugs. The carboplatin-IFO scheme is better tolerated than the cisplatin-IFO regimen with superimposable clinical results (ORR = 69%; CR = 15%). Carboplatin-taxol-IFO is one of the most active regimens in recurrent (ORR = 59%; CR = 17%) and in locally advanced (ORR = 81%; CR = 31%) head and neck cancer. Its role in the multidisciplinary treatment of advanced head and neck cancer is under investigation. In recurrent/metastatic undifferentiated nasopharygeal carcinoma, IFO combinations have proven to be effective as first- and second-line treatment.

Update and perspectives on non-surgical treatment of salivary gland malignancies.

Surgery is the treatment of choice for major and minor salivary gland malignancies. Herein, the role of radiation and medical treatment in the multidisciplinary management of salivary gland tumours is discussed. Neutron irradiation and hyperfractionated external beam mega voltage irradiation improve local control. Combination of three dimensional conformal radiotherapy and intensive-modulated radiation therapy provide better local tumour delineation, better field design to encompass the tumour allowing dose escalation to target while sparing the surrounding normal tissue. Cisplatin-based chemotherapy provides a response rate > or = 45%, in a palliative setting. Concomitant chemo-radiotherapy could improve local control. Recent studies evaluated the expression of molecular targets in salivary gland carcinomas (c-kit = 53-90%, EGFR = 25-85%, c-erb-B2 = 11-58%, p53 = 11-67%, H ras = 18%); these targets are very important since new targeted drugs are now available. Anti-androgen therapy might have a role in the management of patients with ductal carcinoma. These new targeted drugs could be integrated with chemotherapy and radiotherapy in the treatment of locally advanced/metastatic salivary gland malignancies.

Concomitant chemoradiotherapy followed by adjuvant chemotherapy in parotid gland undifferentiated carcinoma.

AIMS AND BACKGROUND: Undifferentiated carcinoma of the parotid gland is a poor-prognosis lesion. Results in unresectable lesions, treated with radiotherapy alone, are very disappointing. METHODS: Six patients with T3-4 N0-1 inoperable lesions were treated with conventional radiotherapy (64-70 Gy, 2 Gy per fraction 5 times a week) and concomitant cisplatin (100 mg/m2, days 1, 22 and 43). Four weeks after radiotherapy, adjuvant chemotherapy (cisplatin, 80 mg/m2, day 1, + VP16, 100 mg/m2, days 1, 3 and 5, q = 3 weeks, for 3 cycles) was given. RESULTS: A median dose of 66 Gy (range, 64-70 Gy) was delivered, and all patients received 3 courses of cisplatin during radiotherapy. Five of 6 patients received all three chemotherapeutic adjuvant courses. Two months after the end of treatment, 3 CR (50%), 2 PR (33%) and 1 NC (16%) was observed. Median CR and PR duration was 26+ and 10 months, respectively. Median overall survival was 18 months. No severe acute or late toxicity was observed. CONCLUSIONS: Concomitant chemoradiotherapy followed by adjuvant chemotherapy in advanced unresectable undifferentiated parotid carcinoma is feasibile and well tolerated. The high percentage of long-lasting CR is encouraging.

Feasibility and long-term results of autologous PBSC transplantation in recurrent undifferentiated nasopharyngeal carcinoma.

BACKGROUND: Recurrent undifferentiated nasopharyngeal carcinoma (UNPC) is a chemosensitive illness. Here we report long-term results of high-dose chemotherapy (HDC) as late intensification, with autologous peripheral blood stem cell (PBSC) support. METHODS: Six patients (5 men, 1 woman; median age 41years; median ECOG PS = 0) with recurrent UNPC (local, 2; local + nodal, 2; bone metastasis, 2) have been enrolled. All patients had been previously treated with neoadjuvant chemotherapy and radiotherapy; 3 of 4 local relapses had received a re-irradiation. Every patient received three courses of cisplatin + epirubicin and 1 cycle of epirubicin followed by PBSC collection. A median of 7.2 x 10(6)/kg (range, 4.5-18) CD34+ cells were reinfused. HDC was according ICE scheme: ifosfamide, 2.5 g/m(2)/d, + carboplatin, 300 mg/m(2)/d, + VP-16, 300 mg/m(2)/d days 1 through 4. RESULTS: After conventional chemotherapy, we had 1 CR (16%), 3 PR (50%), and 2 NC (34%). After HDC, we had 4 CR (66%),1 PR (17%), and 1 MR (17%). Toxicity was manageable. After a median follow-up of 30 months (range, 14-50), two patients are alive without disease (34%), one is alive with bone disease (16%), and three (50%) died of disease at 16, 18, and 24 months. CONCLUSIONS: HDC has an acceptable toxicity, can convert PR in CR, and seems effective, with long-lasting CRs.

Survival in patients with recurrent squamous cell head and neck carcinoma treated with bio-chemotherapy.

BACKGROUND: We have shown that rIL-2 administration in recurrent head and neck cancers induces a tumor-specific T-lymphocyte reactivity and tumor regression; in a pilot study we have shown a safe and effective administration of rIL2 after cisplatin + 5-fluorouracil. Long-term results are not known. METHODS: Thirty patients with recurrent-persistent head and neck cancer were treated with cisplatin (100 mg-m(2)) d.1,5-fluorouracil (1 gr-m(2)-d c.i. 96 h), and SQ rIL-2 (4.5 M IU day 8 to 12 and 15 to 19) every 3 weeks. RESULTS: The overall response rate was 53.3% (95% CI; 34.4-72.3%): 26.6% complete response (CR) (8 patients) and 26.6% partial response (PR) (8 patients); 6 patients had SD (20%), 8 had PD (26.6%). The median follow-up was 36 months (range, 28-44). The median CR duration is 16.2 months (8.5-39+); the median survival duration of this group has not been reached. The median PR duration was 7.2 months (3-10); the median survival was 13.3 months (10-26). The median overall survival was 14 months. CONCLUSIONS: The most impressive finding is the very long survival of CRs patients. This outcome has been reported in other cancer patients with a CR after IL-2 therapy.

Phase II randomized trial comparing vinorelbine versus vinorelbine plus cisplatin in patients with recurrent salivary gland malignancies.

BACKGROUND: Some previous studies have shown that vinorelbine (VNB) is active in recurrent salivary gland tumors. METHODS: Between April 1993 and April 1997, 36 patients in a Phase II randomized trial received either cisplatin, 80 mg/m(2), on Day 1 plus VNB, 25 mg/m(2), on Days 1 and 8 (every 3 weeks) (for a minimum of 3 cycles (Arm A [16 patients]), or VNB, 30 mg/m(2)/week, (for a minimum of 9 wks) (Arm B [20 patients]). There were 23 males and 13 females with a median age of 59 years (range, 20-74 years) and a median Eastern Cooperative Oncology Group performance status of 1 (range, 0-2). Four patients had been treated with prior surgery (S) or radiotherapy (RT), 27 patients had been treated with S plus RT, and 5 patients had been treated with S plus RT plus mitoxantrone. Eighteen patients had major salivary gland tumors, and 18 patients had minor salivary gland tumors; 9 patients had adenocarcinoma, 22 patients had adenoid cystic carcinoma, 1 patient had a malignant mixed carcinoma, 3 patients had undifferentiated carcinoma, and 1 patient had a mucoepidermoid carcinoma. The site of recurrence was local in 16 patients, local plus metastatic in 5 patients, and metastatic only in 15 patients. These characteristics were well balanced between the 2 arms. RESULTS: In Arms A and B a complete response (CR) was noted in 3 patients (19%) and no patients, respectively; a partial response (PR) was noted in 4 patients (25%) and 4 patients (20%), respectively; no change was noted in 6 patients (37.5%) and 9 patients (45%), respectively; and progressive disease was noted in 3 patients (19%) and 7 patients (35%), respectively. The median duration of the CR was 15+ months (range, 6-27+ months) and for PR the median duration was 7.5 months (range, 3-11+ months) and 6 months (range, 3-9 months) in Arms A and B, respectively. Number of patients surviving > 12 months was 6 versus 1 in Arms A and B, respectively (P < 0.05). Grade 2-3 nausea and emesis was statistically higher (P < 0.001) in Arm A; there was no significant difference with regard to other side-effects between the two treatment arms. CONCLUSIONS: VNB is a drug with moderate activity in salivary gland malignancies. The combination of cisplatin plus VNB was found to be more active than VNB alone, with a good number of CRs and long-term survivors reported in the current study.

Clinical and pharmacokinetic data of a docetaxel-epirubicin combination in metastatic breast cancer.

In order to explore activity and pharmacokinetic data of a docetaxel-epirubicin combination we analyzed a population of 60 metastatic breast cancer patients. All the patients had an ECOG performance status < 3; 41 patients (68%) had visceral metastases as dominant site of disease, including 33% with liver metastases. Three or more involved organs were present in 43% of patients; 35% had received prior hormonotherapy; 10% for metastatic disease. Twenty-five patients (42%) had received prior adjuvant chemotherapy; 15% a CAF regimen. Twenty per cent of patients had less than 12 months disease-free interval. Docetaxel and epirubicin were both given at a dose of 75 mg/m2 i.v. d. 1 every 3 weeks. After a median of six cycles we had 5 CR (8.3%), 40 PR (66.6%), 7 NC (11.6%), and 8 PD (13.3%). Response rates in patients with visceral and liver metastases were 78% and 55% respectively. Premenopausal status, < 1 year disease free survival and > 3 metastatic sites were associated with a lower response rate. After a median follow-up of 19 months (12-36), median disease-free survival is 11 months and median overall survival has not been reached. Grade 4 neutropenia was observed in 75% of courses but with febrile neutropenia in 6.2% of courses only. Non-hematologic toxicity wasn't clinically important. A NYHA class III reversible cardiac failure was observed in one patient (1.6%). The pharmacokinetic evaluation in 16 patients has shown that docetaxel transiently interfered with epirubicin plasma level when docetaxel was administered 1 h after epirubicin.

Paclitaxel and carboplatin for recurrent salivary gland malignancies.

BACKGROUND: The use of chemotherapy for recurrent salivary gland carcinomas is under investigation. PATIENTS AND METHODS: Fourteen patients (10 males, 4 females; median age 55 years, range 20-70) with recurrent carcinomas of major (9 patients) and minor (5 patients) salivary gland origin (histology: 1 adenocarcinoma, 10 adenoid cystic carcinoma, 2 undifferentiated carcinoma, 1 mucoepidermoid carcinoma) were treated with carboplatin AUC 5.5 + paclitaxel 175 mg/m2 (3-hour infusion) on day 1 (interval = 3 weeks). All patients had been previously treated with surgery + radiotherapy and 8 with a cisplatin combination. One patient had a local lesion, 7 locoregional recurrence and metastases and 6 patients had metastases only. RESULTS: Overall 65 courses were given (median 5; range 2-6). Responses were: PR in 2 patients (14%) lasting 5 and 12 months; 7 NC (50%) with a median duration of 8.5 months (5-12); and 5 PD (36%). The median survival time was 13.5 months for PR/NC patients, 6 months for non responders; median overall survival was 12.5 months (3-17+). CONCLUSION: This combination had a moderate activity; the treatment was well tolerated and toxicity was manageable.

Vinorelbine treatment of recurrent salivary gland carcinomas.

Twenty patients (13 males, 7 females, median age 61 years, range 27-74) with recurrent adenocarcinoma-like tumors of major (10 patients) and minor (10 patients) salivary gland origin (13 adenoid cystic carcinoma, 5 adenocarcinoma, 1 malignant mixed tumor, 1 undifferentiated carcinoma) were treated with vinorelbine at the dose of 30 mg/m2 i.v. weekly. Sixteen patients had been previously treated with surgery + radiation, 3 with surgery + radiotherapy + Novantrone and 1 with radiotherapy alone. Nine patients had local recurrence, 2 local relapse + metastasis and 9 metastasis alone. Site of metastases are: lung (7), bone (1), lung + bone (2), lung + bone + lymph-node + skin (1). Overall 174 courses were given (median 9, range 6-19). Responses were: PR in 4 patients (20%) with a median duration of 6 months (3-9), 9 NC (45%) with a median duration of 3.5 months and 7 PD (35%). The median survival time was 10 months for PR/NC patients, 4 months for non-responders. Median overall survival was 7 months. Vinorelbine has a moderate activity in these very advanced cases.

Serum thyroid hormone changes in head and neck cancer patients treated with microwave hyperthermia on lymph node metastasis.

Eleven patients with head and neck cancer (ten men, one woman; mean age, 65 years) (larynx, six; oropharynx, two; tongue, one; skin, one; thyroid, one) with regional lymph node enlargements either in contiguity or firmly adherent to the vascular structures of the neck have been treated by means of external microwaves (915, 434 MHz) applicators. All patients were treated by hyperthermia (42-45 degrees C) alone (ten sessions, twice a week, each lasting 30 minutes). Thyroid hormones (T4, T3, FT4, FT3, rT3, thyroid stimulating hormone less than obTSH]) were evaluated during sessions 1, 2, 3, and 8, just before the session (time [t] = 0) and at 10-minute intervals during the heating (t = 10, t = 20, t = 30). Blood was also taken 10 (t = 40) and 30 minutes (t = 60) after the end of each session. T4 showed a decreasing trend; T3 decreased significantly from t = 10 and reached the lowest values at t = 40 and t = 60; FT4 decreased at t = 40 and t = 60 at all sessions; FT3 and rT3 showed no change; TSH decreased at t = 10 until t = 40.

Combined chemotherapy for recurrent and metastatic nasopharyngeal carcinoma.

Thirty-two patients (24 males, 8 females; median age 54 yrs) with recurrent and/or metastatic undifferentiated carcinoma of the nasopharyngeal type were treated with chemotherapy. Remissions were observed in 17 of 32 (53.2%) with 5 complete (CR) (15.6%) and 12 partial responses (PR) (37.6%). A combination of cisplatin and 5-fluorouracil was the most effective regimen (CR + PR = 83.3%). Objective responses. (CR + PR) were 47% (CR = 11.7%) in schemes without cisplatin and 60% (CR = 20%) in cisplatin-based combinations. The median overall duration of response was 7.2 months. The median overall survival time was 10.3 months: 15.1 months for responders and 5.2 for non-responders. No important toxicity was observed.

Cisplatin, epirubicin and 5-fluorouracil combination chemotherapy for recurrent carcinoma of the salivary gland.

Nine patients (5 males, 4 females; median age, 62 years) with recurrent high-grade malignancies of major (7 cases) and minor (2 cases) salivary gland origin (4 adenoid cystic carcinomas, 2 adenocarcinomas, 2 poorly differentiated carcinomas, 1 mixed malignant tumor) were treated with cisplatin (60 mg/m2), epirubicin (50 mg/m2) and 5-fluorouracil (600 mg/m2) (CEF) by intravenous injections on the first day of a 21-day regimen. Previous therapy included surgery (1 case), radiotherapy (1 case), and surgery + radiotherapy (7 cases). There was 1 complete response (11.1%), 3 partial responses (33.3%), 2 unchanged lesions (22.2%) and 3 progressions (33.3%). Patients with local recurrence had a better response. Median remission duration was 7.5 months in CR + PR patients. Median overall survival was 8+ months; 14+ months for responders and 4 months for nonresponders. The major toxic effects were nausea/vomiting and alopecia; myelosuppression was less frequent and usually not severe.

Cisplatin and 5-fluorouracil in recurrent head and neck cancer: results of an outpatient schedule.

Thirty patients with recurrent squamous cell carcinoma of the head and neck were treated with an outpatient schedule: cisplatin (100 mg/m2) day 1 and an 8-hour infusion of 5-fluorouracil (1000 mg/m2) on days 1-4 every 28 days. Twenty-eight patients were evaluable for response and toxicity: there were 5 complete responses (17.8%), 12 partial responses (42.8%), 6 stable disease (21.6%) and 5 progressions (17.8%). Patients with good performance status had a better response; patients who received prior chemotherapy had less positive responses. Median remission duration was 30+ weeks in patients who had a complete response, 25+ weeks in patients with a partial response. Median overall survival was 28+ weeks: 36+ weeks for responders and 14 weeks for non-responders. The major toxic effect was nausea/vomiting, while myelosuppression and stomatitis were less frequent and never severe.

Comparison of methotrexate and sequential methotrexate-5-fluorouracil for patients with recurrent squamous cell carcinoma of the oral cavity.

Forty-eight fully assessable previously treated patients with biopsy-proven recurrent squamous cell carcinoma of the oral cavity were randomized to receive either methotrexate (MTX), 40 mg/m2 iv push weekly, or sequential MTX and 5-fluorouracil (5-FU) (MTX 150 mg/m2 iv for 1 h; 1 h after the end of MTX, 5-FU 600 mg/m2 iv for 2 h; 24 h later, leucovorin rescue 10 mg/m2 iv and the same dose was given orally every 6 h 4 times; the treatment was repeated every 10 days). There were 1 complete response (CR) and 5 partial responses (PR) in the MTX group; median remission duration = 84 days. There were 3 CR and 11 PR in the MTX-5-FU group (overall response 14/24, 58.3%--p less than 0.05); median remission duration = 125 days. Median survival was 6.2 months in the MTX group and 8.1 months in the MTX-5-FU group. There was no difference in mucositis between the two groups, and a prevalence of leukopenia and moderate gastro-intestinal toxicity in the MTX-5-FU group.

Gastrointestinal cancer-associated antigen (GICA) in oral carcinoma.

Gastrointestinal cancer-associated antigen (GICA) is detected by means of a monoclonal antibody in the serum and pathologic tissues of patients with gastrointestinal tumors. This article compares serum and salivary GICA and carcinoembryonic antigen (CEA) levels in 19 healthy control subjects, 17 patients with benign oral cavity lesions, and 11 patients with squamous cell carcinoma of the oral cavity. Serum CEA levels were similar in all three groups, whereas salivary CEA levels were higher in patients with squamous cell carcinoma than in the control subjects (p less than 0.001) and the patients with benign lesions (p less than 0.025). Serum GICA levels gave the opposite result and were significantly lower in squamous cell cancer when compared with control subjects (p less than 0.0001) and patients with benign lesions (p less than 0.02). Values of GICA in saliva of patients with oral cancer were also lower than in the control subjects (p less than 0.02). The possible significance of this difference between the two antigens is discussed.

Combined cryosurgical, chemotherapeutic, and radiotherapeutic management of T1-4N0M0 oral cavity cancers.

Eighty-four previously untreated patients (69 males, 15 females) with squamous carcinoma of the tongue (30 patients), floor of the mouth (30), cheek (16), and retromolar region (8) were treated using a protocol comprising cryosurgery + chemotherapy, followed by external 60Co radiotherapy. The follow-up period was at least 6 months (median, 50 months). Cryosurgery (1-2 sessions in 49 T1-2 cases; 2-4 in 35 T3-4 cases) was accompanied by a CMF (cyclophosphamide, methotrexate, 5-fluorouracil) schedule (T1-2, two courses; T3-4, three courses). Radiotherapy was given 15 to 20 days after combined cryochemotherapy (T1, 50 Gy on tumor and lymph nodes; T2-3-4, same with an extra dose of 10 to 15 Gy on the primary lesion). Complete remission was reached 4 months after treatment in 76 of 84 patients (90.5%). Survival with no evidence of disease (NED) in the 57 patients (27 T1-2, 30 T3-4) with a follow-up of more than 3 years was 59.6% for the series as a whole, 70.3% for T1-2, and 50.0% for T3-4; 78.2% for the tongue, 52.6% for the floor, 66.6% for the cheek, and 0% for the retromolar region. The picture was much the same after 5 years. Actuarial survival at 6 years was 66% in the series as a whole, 75.5% in T1-2, and 57.5% in T3-4 (tongue 86.9%, floor 56.1%, cheek 68.4%, and retromolar region 0%). It is believed that the results obtained in tumors of the tongue, floor and cheek, coupled with the conservative aspects of the protocol, make it a suitable subject for a controlled trial.

Carcinoma of the tongue and floor of the mouth. Preliminary results of a multidisciplinary approach.

UNLABELLED: Eighty-five patients (74 males, 11 females) with untreated squamous-cell carcinoma of the tongue (40 patients) and the floor of the mouth (45 patients) underwent a multidisciplinary treatment. FOLLOW-UP: 6-72 months (median 52 months). T1-4N0M0 lesions (tongue; 32 patients; floor of mouth, 32 patients) were treated with cryosurgery (T1-2: 1-2 sessions, T3-4: 2-4 sessions) and contemporaneously with CMF (cyclophosphamide, methotrexate, fluorouracil) (T1-2: 2 courses; T3-4: 3 courses). 15-20 days after the end of cryo-chemotherapy the patients underwent TCT (T1: 50 Gy to the tumour and lymph nodes; T2-3-4: same with an extra dose of 10-15 Gy to the primary lesion). T1-4N1-3M0 patients (tongue: 8, floor of mouth: 13) received the same cryotherapy and chemotherapy, followed by surgery (13 extended suprahyoid dissections, 8 conservative laterocervical dissections, 1 RND). The actuarial survival rate of patients with tongue tumours after 6 years was 81.4% (N0 87.1%; N+ 60.0%). Four months after treatment, 35 patients had reached complete remission (CR). The probability of remaining in CR for 6 years was 53.6% (N0 56.1%; N+ 50.0%). For tumours of the floor of the mouth the actuarial survival rate was 55.2% (N0 56.5%; N+ 48.6%). Four months after treatment, 38 patients had reached CR. The probability of remaining in CR for 6 years was 59.9% (N0 61.5%; N+ 59.3%). A controlled study is recommended in the light of these results and the conservative nature of the protocol.

Combined chemotherapy, cryosurgery, and radiotherapy/surgery for oral cancer.

UNLABELLED: One hundred and twelve patients (95 males, 17 females) with untreated squamous-cell carcinoma of the oral cavity underwent a multidisciplinary treatment. FOLLOW-UP: 6-72 months (median 51 months). T1-4N0M0 lesions (tongue, 31 patients; floor, 31 patients; cheek, 17 patients; retromolar, 9 patients) were treated with cryosurgery (T1-2: 1-2 sessions, T3-4: 2-4 sessions) and contemporaneously with (cyclophosphamide, methotrexate, fluorouracil)(CMF) (T1-2:2 courses; T3-4: 3 courses). The patients, 15-20 days after the end of cryo-chemotherapy, underwent TCT (T1: 50 Gy on tumour and lymph nodes; T2-3-4: same with an extra dose of 10-15 Gy on the primary lesion). T1-4N1-3M0 patients (tongue: 8, floor: 13, cheek: 2, retromolar: 1) received the same cryotherapy and chemotherapy, followed by surgery (16 widened suprahyoid dissections, 8 conservative laterocervical dissections, 1 radical neck dissection). Overall actuarial survival at six years was 61.9%: T1-4N0M0 (88 patients) 66.6% (T1-2 76.0%; T3-4 56.9%); T1-4N1-3M0 (24 patients) 44.4% (T1-2 66.6%; T3-4 33.3%). Six-year actuarial survival by site was: tongue (39 patients) 79.2% (N0 85.9%); floor (44 patients) 55.2% (N0 57.1%); cheek (19 patients) 74.2% (N0 68.9%); retromolar (10 patients) 0% (N0 0%). Complete remission was reached four months after treatment by 97 patients.(ABSTRACT TRUNCATED AT 250 WORDS)