RESUMO
AgNOR (Argyrophilic nucleolar organiser region) has been shown in recent times, to have value in knowing the prognosis of carcinoma oesophagus. We have evaluated the significance of AgNOR in oesophageal cancer with reference to prognosis following treatment. Fifty patients of histologically proven squamous cell carcinoma of the oesophagus were studied. Following oesophagectomy the specimens removed were evaluated for AgNOR number. Of the resected specimens, 25 (50%) had an AgNOR count < or = 3.0, 18 (36%) had an AgNOR count of more than 3 per nucleus and in the remaining 7 cases, AgNOR number was not quantifiable due to the total absence of tumour tissue in the postoperative specimen due to preoperative radiotherapy. When followed up for an average period of 25 months (3 to 47 months), it was seen that patients with AgNOR count of < or = 3.0 per nucleous had a similar mean survival (30.39 +/- 3.29 months) as those with counts > 3.0 per nucleus (27.80 + 3.33 months). The survival in the seven patients in whom no tumour was present following preoperative radiotherapy, was 30.30 +/- 2.42 months. An analysis was done for the presence of change in the AgNOR count before and after radiotherapy in twenty eight case of carcinoma oesophagus treated with preoperative radiotherapy. It was found that the counts on an average were lower in patients after radiotherapy (2.89 +/- 1.04 per nucleus), than before radiotherapy (3.17 +/- 9.69). This was found in 24 cases, while the remaining 4 cases showed no change in count after radiotherapy. This suggested that radiotherapy caused a reduction in AgNOR counts. Mean survival in those with decreased count after radiotherapy was 33.65 (+/- 3.35) months. Since the AgNOR counting is a simple method and can be applied to paraffin embedded section, estimation of the AgNOR number may help in determination of prognosis in patients with oesophageal carcinoma. Preoperative radiotherapy seems to decrease AgNOR count with improved survival. These observations need however, to be reproduced with a larger sample size.