Endoscopic ultrasonography in the quantitative assessment of response to chemotherapy in patients with adenocarcinoma of the esophagus and esophagogastric junction

Prediction of survival by tumor area on endosonography after definitive chemoradiotherapy for locally advanced squamous cell carcinoma of the esophagus

BACKGROUND

Definitive chemoradiotherapy (CRT) is a reasonable approach for patients with locally advanced esophageal cancer who are not surgical candidates. This study was performed to investigate whether endosonography (EUS) assessment of tumor area response is a useful prognostic marker in patients with squamous cell carcinoma (SCC) of the esophagus who receive definitive CRT.

METHODS

A total of 33 patients who received definitive CRT for locally advanced esophageal SCC were enrolled. The maximal transverse cross-sectional area of the tumor was measured before and after definitive therapy. EUS response was defined as a ≥50% reduction of the tumor area after definitive CRT.

RESULTS

Based on EUS evaluation, there were 20 nonresponders (60.6%) and 13 responders (39.4%). The median progression-free survival (PFS) was significantly longer in EUS responders than EUS nonresponders (p = 0.005). However, there was no statistical significance in overall survival according to EUS response (p = 0.120). During multivariate analysis, EUS response to definitive CRT was the only significant factor associated with PFS (p = 0.045), whereas EUS response to definitive CRT was not associated with overall survival (p = 0.221).

CONCLUSIONS

A reduction of the maximal cross-sectional tumor area measured by EUS correlates with a superior prognosis in patients with locally advanced SCC of the esophagus after definitive CRT.

Impact of fluorodeoxyglucose PET on the management of esophageal cancer

Esophageal cancer is the third most common malignancy of the alimentary tract. The incidence of esophageal cancer has steadily increased over the past three decades. Almost all therapeutic modalities for esophageal cancer are associated with a considerable mortality and morbidity. Consequently, there has been growing concern regarding effective management of esophageal cancer. 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) is playing an increasing role in the management of esophageal cancer, offering potential advantages in the accuracy of disease assessment at a number of decision points in the management pathway. This review evaluates the critical role of FDG-PET in (i) diagnosis, (ii) preoperative staging, (iii) monitoring of response to neoadjuvant therapy, (iv) assessment of recurrence and (v) prediction of prognosis of esophageal cancer. We have also compared diagnostic performance of FDG-PET and other current technologies such as computed tomography scan and endoscopic ultrasonography based on available evidence.

Using Q-RT-PCR to measure cyclin D1, TS, TP, DPD, and Her-2/neu as predictors for response, survival, and recurrence in patients with esophageal squamous cell carcinoma following radiochemotherapy

PURPOSE

The purpose of this study was to evaluate the use of thymidilate synthetase (TS), thymidilate phosphorylase (TP), dihydropyrimidin dehydrogenase (DPD), Her-2/neu, and cyclin D1 as predictors of therapy response, survival, and recurrence in patients with esophageal squamous cell carcinoma (ESCC) following radiochemotherapy.

MATERIALS AND METHODS

Twenty-six patients with histologically proven intrathoracic, locally advanced ESCC (cT3, cN0/+, cM0) underwent preoperative, combined simultaneous radiochemotherapy followed by R0-transthoracic esophagectomy. Because R0 resection is the strongest known independent prognostic factor in this tumor entity, only R0-resected patients were included in this study. Pre-therapeutically taken, formalin-fixed, and paraffin-embedded tumor biopsies were used for laser-assisted microdissection of tumor cells and RNA extraction and subjected to real-time (TaqMan) quantitative reverse transcriptase-polymerase chain reaction (Q-RT-PCR).

RESULTS

No significant correlation between clinical or histopathological parameters and the relative gene expression of TS, TP, DPD, or Her-2/neu was observed. However, patients with relative cyclin D1 levels below the median gene expression did not reach median survival compared to the 19.9 months seen in patients with relative cyclin D1 gene expression above the median (P = 0.02). Patients with low cyclin D1 levels experienced significantly less frequent recurrence of the tumor (20% versus 63%; P = 0.006), and there was a significant difference in the recurrence-free interval (P = 0.003).

CONCLUSIONS

Despite the small number of investigated patients, our data seem to show that high levels of cyclin D1 measured by real-time Q-RT-PCR before neoadjuvant radiochemotherapy correlate significantly with patient survival, tumor recurrence, and recurrence-free-interval. Cyclin D1 might be useful in identifying patients at high risk of poor prognosis and suffering from recurrence after neoadjuvant radiochemotherapy treatment and R0 resection. Further investigations with a larger cohort are warranted.

Response monitoring of neoadjuvant therapy using CT, EUS, and FDG-PET

Neoadjuvant or adjuvant multimodality therapy in oesophageal cancer is introduced in an effort to improve prognosis. However, in a substantial fraction of patients there is no response to this non-surgical therapy. Non-invasive imaging modalities such as computed tomography (CT), endoscopic ultrasound (EUS) and 18F-2-fluoro-2-deoxy-d-glucose positron emission tomography (FDG-PET) have been evaluated for assessing patient response to therapy, and these are described in this review. Currently, FDG-PET seems to be the best available tool for neoadjuvant therapy response assessment in oesophageal cancer.

Esophageal cancer: CT, endoscopic US, and FDG PET for assessment of response to neoadjuvant therapy--systematic review

PURPOSE

To compare diagnostic accuracy of computed tomography (CT), endoscopic ultrasonography (US), and fluorine 18 fluorodeoxyglucose (FDG) positron emission tomography (PET) for assessment of response to neoadjuvant therapy in patients with esophageal cancer by using a systematic review of the literature.

MATERIALS AND METHODS

MEDLINE and EMBASE databases and Cochrane Database of Systematic Reviews were searched for relevant studies. Two reviewers independently assessed the methodological quality of each study. Summary receiver operating characteristic (ROC) analysis was used to summarize and compare the diagnostic accuracy of the three modalities.

RESULTS

Four studies with CT, 13 with endoscopic US, and seven with FDG PET met inclusion criteria. Percentages of the maximum score in regard to methodological quality ranged from 15% to 100%. Summary ROC analysis could be performed for three studies with CT, four with endoscopic US, and four with FDG PET. The maximum joint values for sensitivity and specificity were 54% for CT, 86% for endoscopic US, and 85% for FDG PET. Accuracy of CT was significantly lower than that of FDG PET (P < .006) and of endoscopic US (P < .003). Accuracy of FDG PET and that of endoscopic US were similar (P = .839). In all patients, CT was always feasible, whereas endoscopic US was not feasible in 6% of the patients, and FDG PET was not feasible in less than 1%.

CONCLUSION

CT has poor accuracy for assessment of response to neoadjuvant therapy in patients with esophageal cancer. Endoscopic US and FDG PET have equivalent good accuracy, but endoscopic US is not always feasible after chemotherapy and radiation therapy. FDG PET seems to be a promising noninvasive tool for assessment of neoadjuvant therapy in patients with esophageal cancer.

Responders benefit from neoadjuvant radiochemotherapy in esophageal squamous cell carcinoma: results of a prospective phase-II trial

BACKGROUND

We present the results of a prospective phase-II-study of neoadjuvant combined radiochemotherapy followed by surgical resection in patients with histological proven locally advanced squamous cell carcinoma of the esophagus located at or above the level of the tracheal bifurcation.

METHODOLOGY

Between February 1995 and March 2000 a total of 76 patients with esophageal squamous cell carcinoma (uT3/4N0/+-categories) received simultaneous combined neoadjuvant radiochemotherapy consisting of a continuous intravenous infusion of 5-fluorouracil (300 mg/m2/day) 7 day per week concurrently with conventional fractioned external beam radiation therapy (2 Gy/day), five fractions per week up to a total dose of 30 Gy.

RESULTS

Radiochemotherapy related acute severe toxicity rate (CTC-grade-III) occurred in 34 patients, two patients died. Sixty-four patients underwent surgery with a complete resection in 48 patients. Three patients died during a 90-day post-operative course. The histopathological workup revealed no viable residual tumour cells in eight patients (ypCR) and according to the modified criteria of Mandard in 26 patients a histopathological response. Twenty-two of these patients underwent a R0-resection. The median follow-up time was 5.4 years with an overall median survival time of 20.6 months. The median survival in the 26 responders was 32.3 months versus 19.5 months in 38 non-responders (p=0.03).

CONCLUSIONS

Patients with locally advanced squamous cell carcinoma of the esophagus, who respond to preoperative neoadjuvant combined radiochemotherapy, seem to have more benefit from subsequent resection than non-responding patients.

Surgery after induction chemoradiotherapy for oesophageal cancer

AIM

We discuss the role of surgery after induction chemoradiotherapy, when a complete response is apparent.

METHODS

We enrolled 247 consecutive patients who underwent chemoradiotherapy and oesophagectomy for cancer. Patients were classified in two groups as to whether they had chemoradiotherapy (n = 60) or not (n = 187) before surgery.

RESULTS

The 5-year survival rate of patients with complete response was 54% and was significantly higher than that of the other group (P = 0.018). Of the 60 patients responding, 34 (56.7%) were found to have a complete pathological tumour response (pT0). The overall sensitivity and specificity of the post-chemoradiotherapy restaging were 60.7% and 86.4%.

CONCLUSIONS

Complete response after induction chemoradiotherapy is not a reliable tool in the management of oesophageal cancer. We need to improve postchemoradiotherapy restaging to be sure that chemoradiotherapy will provide the same long-term survival rates as surgery.

Gastrointestinal endoscopic ultrasonography

Although conventional endoscopy provides excellent visualization of gastrointestinal mucosa, it provides little information about intramural or nearby extramural lesions. The imaging of intraabdominal structures by conventional transabdominal ultrasound is degraded by ultrasound energy attenuation with distance. The provision of an ultrasound probe on a flexible gastrointestinal endoscope, to form an echoendoscope, provides excellent imaging of the gastrointestinal wall and of adjacent extramural structures. During the last two decades, endoscopic ultrasound, using an echoendoscope, has revolutionized the diagnosis and treatment of gastrointestinal diseases that affect the submucosa, deep bowel wall, and adjacent extramural structures. This article reviews the role of endoscopic ultrasound in the diagnosis and treatment of gastrointestinal disease, including standard and promising new applications, as well as standard and emerging new technology.

The clinical efficacy of neoadjuvant chemotherapy in squamous esophageal cancer: a prospective nonrandomized study of pulse and continuous-infusion regimens with Cisplatin and 5-Fluorouracil

BACKGROUND

We evaluated cisplatin and 5-fluorouracil as preoperative adjuvant chemotherapy for patients with locally advanced squamous esophageal cancer and compared two different infusion regimens. The outcomes were also compared with those of our historical control patients treated by surgery alone.

METHODS

From 1991 to 1997, 83 consecutive esophageal cancer patients underwent surgical exploration after completion of two cycles of cisplatin and 5-fluorouracil chemotherapy regimens, either in pulse or in continuous infusion cycles. Outcomes were compared with those of 76 historical control patients. Both groups were comparable in demographic characteristics and tumor stages. The resection rates, operative morbidity, mortality, and survival rates were compared.

RESULTS

Partial response was achieved in 50% of patients who received chemotherapy. There was no chemotherapy-related mortality. The resection, morbidity, and mortality rates and median survival between the surgery-alone group and the chemotherapy group were 71.1% vs. 82%, 51% vs. 55%, and 4% vs. 10.8%, 12.0 vs. 13.5 months, respectively (P >.05). There was also no statistically significant difference between the two regimens.

CONCLUSIONS

Preoperative adjuvant chemotherapy with cisplatin and 5-fluorouracil infusion, in pulse or continuous regimens, followed by surgery for squamous esophageal cancer patients had no added benefit in the overall survival.

Correlation of EUS measurement with pathologic assessment of neoadjuvant therapy response in esophageal carcinoma

BACKGROUND

EUS-measured reduction in tumor size after neoadjuvant therapy has previously been correlated with downstaging and improved survival in patients with esophageal cancer. The aim of this study was to determine whether tumor changes measured by EUS correspond to pathologically assessed chemoradiotherapy-induced tumor regression.

METHODS

Forty-one patients with esophageal cancer treated with combined modality treatment were studied. After initial EUS, patients completed a cisplatin/carboplatinum, 5-fluorouracil, and radiotherapy regimen and underwent repeat EUS before resection. A positive response on EUS was defined as a 50% reduction in maximal tumor cross-sectional area. Chemoradiotherapy-induced tumor regression was assessed in resection specimens by using a previously defined pathologic scoring system based on the extent of tumor proliferation into adjacent fibrosis.

RESULTS

Pathologic tumor regression was present in 23, indeterminate in 5, and minimal or absent in 13 patients. EUS measured a positive response in 20 of 23 (87%) patients with CRT-induced tumor regression and a negative response in 10 of 13 (77%) patients with absent tumor regression (p < 0.001). EUS had a positive predictive value of 80% for pathologic tumor regression.

CONCLUSIONS

Measurement of tumor size by EUS is a reliable clinical method for assessing pathologic tumor regression before surgery.

Measurement of outcome

The outcomes of treatment of oesophageal cancer include traditional biological and physical measures, such as mortality and morbidity data, disease free and overall survival, clinical and pathological response rates and symptom control. Such factors are essential and should be recorded prospectively for clinical audit. Using this type of information alone to evaluate effectiveness of treatment is inadequate, however, because the diagnosis and treatment of oesophageal cancer has a major impact on functional well-being (including psycho-social function), general health perceptions and overall quality of life (QL)/satisfaction with health and health care. These aspects of patients' well-being need to be considered, in addition to standard outcomes in the evaluation of treatment of oesophageal cancer. Recent needs to judge the economic efficiency of health care by comparing health outcomes with costs may also be part of treatment appraisal. This article reviews surgical, oncological, patient-based and economic outcomes in oesophageal cancer.

Preoperative (neoadjuvant) chemoradiotherapy in oesophageal cancer

BACKGROUND

Oesophageal cancer carries a poor prognosis. The 5-year survival rate following resection ranges from 10 to 35 per cent. Recent evidence suggests that the addition of non-surgical treatments to surgery may improve resection rates, reduce the risk of recurrence and improve survival. This review examines the role of preoperative chemoradiotherapy (CRT) in oesophageal cancer.

METHODS

A Medline-based literature review (1980-2000) was performed using the key words 'neoadjuvant or preoperative' and 'chemoradiotherapy or radiochemotherapy'. Additional literature was obtained from original papers and published meeting abstracts.

RESULTS

Forty-six non-randomized and six randomized trials of preoperative CRT were found. Resection rates, pathological complete response (pCR), treatment-related mortality rates and relapse patterns are documented. Improved 5-year survival rates approaching 60 per cent may be achieved following pCR. Three of the six randomized trials show a benefit in either overall survival or disease-free survival compared with surgery alone. Treatment-related toxicity can be significant.

CONCLUSION

Preoperative CRT may improve survival. Emerging evidence suggests that CRT alone can achieve similar survival rates to surgery alone. New imaging modalities may help to select which patients require surgery. Larger randomized trials of preoperative CRT or chemotherapy are needed to define optimal regimens and produce higher pCR rates with acceptable toxicity.

Endoscopic Ultrasound Does Not Accurately Assess Pathologic Stage of Esophageal Cancer after Neoadjuvant Chemoradiotherapy

The accuracy of endoscopic ultrasound (EUS) for initial staging of esophageal cancer is widely accepted. There is, however, considerable variability in the reported accuracy of EUS for restaging of esophageal neoplasms after neoadjuvant therapy. From June 1995 through December 1999, we prospectively studied a series of 26 patients who underwent neoadjuvant treatment for esophageal cancer and were subsequently restaged by EUS before resection. Twenty-four patients had adenocarcinoma (92%), and two patients had squamous cell carcinoma (8%). EUS correctly predicted tumor stage in seven of 26 patients for an overall accuracy of 27 per cent. EUS overestimated the depth of tumor penetration in 18 patients (69%) and underestimated depth of penetration in one patient (4%). Lymph nodes were correctly staged in 15 of 26 patients for an overall accuracy of 58 per cent. Levels of sensitivity for detecting N0 and N1 disease were 44 per cent and 80 per cent respectively. Patients with a complete pathologic response were staged as T4N1 (one patient), T3N1 (three patients), T3N1 (one patient), and T2N1 (two patients). EUS cannot distinguish tumor involvement of the esophageal wall and lymph nodes from the postinflammatory changes that characterize effective neoadjuvant treatment. EUS is of limited utility in guiding clinical decision making after neoadjuvant therapy.

Assessing chemotherapy response of squamous cell oesophageal carcinoma with spiral CT

45 patients with squamous cell carcinoma of the oesophagus were examined prior to, and following, pre-operative chemotherapy by spiral CT. Oesophageal CT was performed following gaseous distention of the oesophagus. TNM stage and perceived resectability on CT before and after chemotherapy were compared and related to surgical resectability and pathological staging. T-stage changed in 26% and N-stage changed in 9% of tumours after chemotherapy. Post-chemotherapy CT predicted pathological T-stage with an accuracy of 88% and N-stage with an accuracy of 84%. Six of 14 tumours considered irresectable on CT pre-chemotherapy were considered resectable on post-chemotherapy CT. Five of these six tumours were resectable at surgery. Post-chemotherapy CT predicted surgical resectability with an accuracy of 88%, the main pitfall being underestimation and overestimation of tracheobronchial invasion. CT prediction of chemotherapy response as judged by change in tumour volume was compared with a quantitative pathological assessment of chemotherapy response. 93% of oesophageal tumours changed volume after chemotherapy with 51% having a volume reduction of > or = 50%. However, no correlation was found between tumour volume reduction on serial CT examinations and either a quantitative pathological assessment of tumour response or patient survival.

Endoscopic ultrasonography in the staging of esophageal carcinoma after preoperative radiotherapy and chemotherapy

BACKGROUND

In past years multimodal neoadjuvant treatment for carcinoma of the esophagus has been used with increased frequency. Staging of the neoplasm still remains fundamental in evaluating the response to therapy and in planning operation. The aim of the present study was to assess the accuracy of endoscopic ultrasonography (EUS) in a group of patients with squamous cell carcinoma of the thoracic esophagus after undergoing radiotherapy and chemotherapy.

METHODS

Among a group of 111 patients with squamous cell carcinoma of the thoracic esophagus and treated with preoperative radiotherapy and chemotherapy, 87 were operated. In these patients it was possible to compare the results of EUS, with regard to depth of invasion of esophageal wall (T) and lymph node involvement (N), with the results of operation and histopathologic study.

RESULTS

Feasibility of EUS before and after neoadjuvant treatment was 71.2% and 83.9%, respectively. The overall accuracy of EUS regarding the wall invasion was 47.9%. The more frequent error was overstaging, especially in patients with complete response and in patients with minimal residual disease. In the assessment of lymph node involvement, EUS showed an overall accuracy of 71.2% with a moderate kappa value. Sensitivity for N1 and NO was 73.7% and 68.6%, respectively.

CONCLUSIONS

Endoscopic ultrasonography was feasible in most patients after preoperative radiotherapy and chemotherapy, but our study documented a worsening of accuracy of EUS in the evaluation of T attributable to the confounding presence of radiation fibrosis and soft tissue reaction after radiotherapy and chemotherapy.

Endoscopic ultrasound cannot determine suitability for esophagectomy after aggressive chemoradiotherapy for esophageal cancer

OBJECTIVE

Endoscopic ultrasound (EUS) provides important information in the initial staging of patients with esophageal cancer. With recent modifications in chemoradiotherapy protocols, a significant number of patients have no residual tumor at esophagectomy. The high surgical morbidity and mortality might be avoided if complete response to chemoradiotherapy could be predicted. Previously published clinical trials, with relatively small patient numbers, have suggested that EUS may accurately stage esophageal cancer after chemoradiotherapy. The aim of this study was to verify the accuracy of EUS in staging esophageal cancer after effective chemoradiotherapy.

METHODS

EUS staging was performed before and after concurrent cisplatin, 5-fluorouracil, and hyperfractionated radiotherapy in 59 patients with newly diagnosed esophageal cancer. All patients underwent subsequent esophagectomy and pathological staging. The accuracy of preoperative, postchemoradiotherapy EUS was evaluated in a retrospective fashion by comparison to pathological staging.

RESULTS

After chemoradiotherapy, 18 patients (31%) had no residual disease at pathological staging (T0N0). However, EUS correctly predicted complete response to chemoradiotherapy (T0N0) in only three patients (17%). The accuracy of postchemoradiotherapy EUS for pathological T stage was only 37%, and its sensitivity for N1 disease was only 38%. EUS was unable to distinguish postradiation fibrosis and inflammation from residual tumor.

CONCLUSION

When aggressive preoperative chemoradiotherapy is provided to patients with esophageal cancer, the predictive value of postchemoradiotherapy EUS is inadequate for use in clinical decision making.

Endoscopic ultrasound in restaging of esophageal cancer after neoadjuvant chemoradiation

BACKGROUND

Endoscopic ultrasound (EUS) is established as the most accurate method currently available for determining the depth of primary cancer invasion (T stage). Standard EUS criteria may not be accurate in assessing depth of cancer invasion and nodal status after patients have received chemotherapy or radiotherapy.

METHODS

We conducted a prospective study to determine whether EUS estimation of tumor size could be used to assess response to preoperative chemoradiation. Using EUS, TNM stage was assessed in 31 patients (22 men, 9 women; mean age 62 years) with cancer of esophagus or cardia (19 adenocarcinoma, 12 squamous cell cancer) before initiation of combined radiation and 5-fluorouracil/cisplatin (and/or carboplatinum) chemotherapy. The cross-sectional area of the tumor in the transverse plane at the location where the tumor had maximal thickness was calculated to estimate tumor size. EUS staging and measurement of maximal cross-sectional area were repeated at completion of chemoradiation just before surgery. Response to preoperative chemoradiation was defined as 50% reduction in maximal cross-sectional area. Surgical staging was compared between responders and nonresponders.

RESULTS

Eight patients who did not undergo surgery were excluded from analysis. EUST stage in the remaining 23 patients before therapy was as follows: 3 T2, 16 T3, and 4 T4. After chemoradiation, EUS T staging was changed in 6 patients (3 T4 downstaged to T3, 2 T3 downstaged to T2, and 1 T3 downstaged to T1). At surgical pathological examination, 3 patients had no residual tumor in the esophagus (T0), 5 had T1, 3 had T2, 10 had T3, and 2 had T4 tumors. EUS T staging accuracy after adjuvant therapy was only 43%. Maximal cross-sectional area decreased from a mean of 5.5 +/- 2.4 to 1.6 +/- 0.9 cm2 in responders, whereas maximal cross-sectional area went from 7.0 +/- 3.0 to 5.4 +/- 2.2 cm2 in nonresponders (p = 0.009). Ten of thirteen patients with at least a 50% reduction in maximal cross-sectional area (responders) had T0, T1, or T2 tumors at surgery, whereas 9 of 10 nonresponders had T3 or T4 tumors at surgery (p = 0.001).

CONCLUSIONS

(1) Standard EUS staging criteria are not accurate after neoadjuvant chemoradiation, (2) reduction in maximal cross-sectional area of tumor appears to be a more useful measure for assessing response of esophageal cancer to preoperative chemoradiation, and (3) responders have an increased likelihood of downstaging at surgery than nonresponders.

Pre-operative staging of gastro-oesophageal junction carcinoma: comparison of endoscopic ultrasound and computed tomography

Fifteen patients with carcinoma of the gastro-oesophageal junction were pre-operatively staged with endoscopic ultrasound (EUS) and computed tomography (CT). The accuracy of tumour and nodal staging using both modalities was compared to the final histological staging of the resected specimens. In staging depth of tumour growth, EUS was significantly more accurate (87% of lesions correctly staged) than CT (40% correctly staged). In staging nodal involvement, EUS was again significantly more accurate (73% correctly staged) than CT (33%). Two-thirds of the lesions were traversable with the endoscopic probe, but most of the nontraversed lesions were correctly staged on EUS. In this study, CT has performed poorly as a staging modality for carcinoma at the gastro-oesophageal junction. Other studies have shown CT to be less accurate at this location than at other oesophageal sites. The orientation of the gastro-oesophageal junction, lack of surrounding fat planes, proximity of adjacent organs and patient motion contribute to the poor staging performance of CT at this location. In contrast, EUS has been an accurate staging modality at the gastro-oesophageal junction in this study and compares well with other studies evaluating EUS in the more proximal oesophagus. Endoscopic ultrasound is therefore a necessary modality if accurate pre-operative staging of gastro-oesophageal junction carcinoma is to be achieved.

Feasibility of five courses of pre-operative chemotherapy in patients with resectable adenocarcinoma of the oesophagus or gastrooesophageal junction

The purpose of this study was to examine the feasibility of administering all chemotherapy pre-operatively to patients with resectable adenocarcinoma of the oesophagus or gastrooesophageal junction. 32 patients with potentially resectable adenocarcinoma of the oesophagus or gastrooesophageal junction were studied in a stepwise fashion in which combination chemotherapy with cisplatin, high-dose arabinoside and 5-fluorouracil was administered. In the first part, 15 patients were to receive three chemotherapy courses pre-operatively and two chemotherapy courses postoperatively. In the second part, the next 15 patients were to receive all five chemotherapy courses pre-operatively, provided there was an objective response after three courses. Endoscopic ultrasonography was also performed, when feasible, prior to chemotherapy and surgery, and in some patients sequentially between chemotherapy courses. All of the 14 assessable patients in the first group tolerated all three courses of pre-operative chemotherapy, and 86% of patients in this group completed all protocol chemotherapy. In the second group, 9 of 18 (50%) assessable patients tolerated all five courses of preoperative chemotherapy, and 100% of patients in this group received all protocol chemotherapy. The median number of chemotherapy courses for the entire group (32 patients) was five (range one to five). Forty-one per cent (13/32) of patients had a major response to chemotherapy. Sixty-nine per cent (or 76% of 29 patients taken to surgery) had a curative resection. One patient had a pathological complete response. The median survival time of 32 patients was 17 months (range 2-36+ months). 14 patients (37%) remain alive at a median follow-up time of 26+ months. There was a correlation between endoscopic ultrasonographic tumour and nodal stage and pathological tumour and nodal stages in 16 patients. The tumour stage correlation was higher (75%) than the nodal stage correlation (62%). Our data suggest that it is feasible to administer five courses of cisplatin-based chemotherapy to patients with potentially resectable adenocarcinoma of the oesophagus or gastrooesophageal junction. More effective chemotherapy regimens that might result in higher pathological complete response rates and acceptable toxic effects are needed.