PURPOSE: Chemotherapy and concurrent irradiation, intended to cure, are presently standard treatments for non metastatic, unresectable oesophageal cancer. The results of the combined therapy are superior to those of radiotherapy alone, attaining 25-35% 2-year survival rates. However these results mainly refer to stage I and II tumours as most of the available literature has focussed on these groups. The aim of our report is to present our experience with Stage III and IV patients. MATERIAL AND METHODS: Sixty-four Stage III and IV oesophageal cancer patients were referred to our Departments from January 1, 1990 to December 31, 1996. Diagnosis was obtained through oesophagoscopy and biopsy, stage was assessed by physical examination, chest CT scan, bronchoscopy, barium X-ray examination, upper abdomen ultrasonography and bone nuclide scan. Thirty-four patients, with no signs of blood-born metastases and in satisfactory medical conditions (i.e. age not exceeding 70 years, weight loss not exceeding 10% of body weight, normal serum values of BUN and creatinine, no other severe disease), were submitted to concurrent chemo-radiotherapy. The case features were as follows: histology of squamous cell carcinoma in 32 cases, of adenocarcinoma in 2; tumour in the upper third of the oesophagus in 11 (32.5%), in the middle third in 18 (53%), in the lower third in 5 (14.5%); male/female ratio 29/5, age 48-68 years (mean 56), Karnofsky performance status of 60% or higher. On referral, 18 out of 34 (53%) had a weight loss more than 5% of body weight and 22 (64.5%) had dysphagia. Twenty-one had Stage III (61.75%) and 13 stage IV (38.25%) cancer, with metastasis limited to the supraclavicular or coeliac nodes, which could be included in the radiation volume. In all cases chemotherapy consisted of 5-Fluoruracil (administered in a continuous i.v. infusion, from day 1 to 5, with a 750-1.000 mg/n.sq daily dose) and Cisplatin (75-100 mg/n.sq on the first day, or 20 mg/n.sq for 5 consecutive daily doses, administered by i.v. bolus). Three to 5 cycles were administered, one every 21 days. Irradiation started with the first cycle of chemotherapy in 5 patients, with the second or third cycle in 29. At least two cycles of chemotherapy were administered during the course of radiation. Radiotherapy was performed with 4 to 18 MeV linear accelerator X-rays, or telecobalt, through opposite anterior and posterior treatment portals or more complex field arrangements. The doses were in the range of 44-66 Gy, with fractionation of 5x180-200 cGy weekly sessions. After treatment, periodic follow-up controls were carried out in all cases. Thorough restaging was performed only in selected cases, thus a systematic evaluation of objective responses was not possible. Data on improvement of swallowing were always available, however, and the early therapeutic results were analysed accordingly. Toxicity was recorded according to the WHO parameters. Two-year survival after conclusion of the treatment was calculated according to Kaplan and Maier. Survival was analysed (log-rank test) according to stage, Performance Status, oesophagectomy and body weight loss. RESULTS: After treatment, subjective symptomatic relief occurred in 17 of the 22 patients presenting dysphagia (77.5%). Acute toxicity (Grade III or IV WHO) of the treatment accounted for 47% of hematologic adverse effects, 40% of mucositis, 20.5% of vomiting or diarrhoea not responding to drug treatment. Treatment delays of more than one week, due to toxicity, occurred in 23.5%. Moreover, we observed 20.5% of mild cardiotoxicity and 6% of mild nephrotoxicity. No symptomatic lung fibrosis was observed. No death could be related to toxicity. Overall 2 year survival was 13%, with a median value of 10 months. Survival analysis, according to stage, showed 2 year values of 24% in Stage III and 0% in Stage IV (p=0.09). No significant difference was related to Performance Status and weight loss. Six patients showed a remarkable improvement in symptoms and general conditions after treatment, and were restaged with oesophagoscopy, thoracic CT scan and bronchoscopy, which evidenced resectable residual tumors, and they were then operated. Although histologic examination showed tumour in all the resected specimens, 2 patients survived more than two years (33.5% survival, median 14 months). Due to the small number of operated patients, no attempt was made to assess the significance of this result, in comparison with the other cases. DISCUSSION AND CONCLUSIONS: Many Stage III and IV patients, selected for an aggressive chemo-radiation approach on the grounds of satisfactory medical conditions, can obtain relief of dysphagia. Toxicity can be severe, but is rarely life-threatening. Some cases, without extrathoracic spread of the tumor can achieve long term survival (in our experience 24% 2-year survival in Stage III, in our experience which favourably compares with the results obtained by other authors). Whether surgery may improve the therapeutic results of chemo-radiotherapy in patients whose tumour has become resectable, is an issue that cannot be satisfactorily addressed on the basis of our experience, nor are the results from the available literature exhaustive to this regard.