TUMOR RECURRENCE IN LONG-TERM SURVIVORS AFTER TREATMENT OF CARCINOMA OF THE ESOPHAGUS

To evaluate the status of tumor recurrence and the possible factors re levant to tumor recurrence among patients who survived more than 5 yea rs after subtotal esophagectomy for the treatment of squamous cell car cinoma of the esophagus, a total of 104 patients who received treatmen t between 1959 and 1986 were reviewed. In 18 of these 104 patients, lo cal or distant tumor recurrence developed, for a tumor recurrence rate of 17.3%. Eleven (61.1%) of these 18 patients eventually died of carc inomatosis despite further radiotherapy or chemotherapy, or both, and 4 patients with the disease are still alive. Three patients continue t o survive after aggressive therapy was instituted for control of the l ocally recurrent tumor. Sixty-nine of the 104 patients are alive witho ut tumor recurrence after the initial esophagectomy, and the remaining 17 patients died of miscellaneous causes. Tumor recurrence appears to be the most important factor affecting the prognosis in long-term sur vivors with resectable esophageal carcinoma. Among the 11 patients who died of tumor recurrence, 10 died within 5 to 9 years of their esopha gectomy. The incidence of various modes of tumor recurrence among thes e 18 patients was as follows: blood-borne metastasis, 61%; lymph node recurrence, 33%; and locoregional organ recurrence, 33%. Factors that may be pertinent to a higher tumor recurrence rate include male sex, m oderate to poor tumor differentiation, the presence of lymph node meta stasis, and late stage of disease (stage IIb or worse). However, we co uld not find any statistical significance among these possible factors . These results suggest that the character of tumor recurrence in long -term survivors may differ from that in patients suffering from early tumor recurrence, and this may warrant further investigation. As time elapses, metachronous tumor resulting from an esophageal remnant or a secondary primary neoplasm originating from another site should be dif ferentiated from primary tumor recurrence.