Background: The prognosis of upper thoracic esophageal cancer is poor when compared with middle and lower thoracic esophageal cancer because the tumor easily infiltrates the respiratory tract and surgical en-bloc resection is difficult. Recently, preoperative chemoradiation therapy has been shown to lead to down-staging of the disease and improve prognosis. But the benefit of this therapy for tumors infiltrating the respiratory tract remains unknown.
Study design: Fifty-six patients with thoracic esophageal cancer infiltrating neighboring organs, but with no hematogeneous metastasis, were given preoperative concurrent chemotherapy (5-fluorouracil and cisplatin) and radiation (40 Gy) therapy. When a clinical response was observed, making a curative resection potentially possible, patients were scheduled for esophagectomy with extended lymphadenectomy. Patient prognosis with respect to the organs infiltrated by the tumors was estimated by calculating survival curves using the Kaplan-Meier method and comparing the curves by the log-rank test.
Results: The prognosis was significantly poorer for patients with tumors infiltrating the respiratory tract (T) or aorta plus respiratory tract (A + T) than for patients with tumors infiltrating the aorta alone (A) or other organs (Oth) (p < 0.05 for Oth versus T; p < 0.05 for Oth versus A + T; p < 0.0001 for A versus T; p < 0.0001 for A versus A + T by log-rank test). Patients positive for respiratory tract invasion (T, T + A), compared with those negative for respiratory tract invasion (A, Oth), showed a poorer clinical response to chemoradiation (3.0%, 45.5%, 39.4%, and 9.1% versus 4.3%, 82.6%, 4.3%, and 8.7% in complete response (CR), partial response (PR), nonresponse (NC) and progressive disease (PD), respectively, p = 0.0156) and surgical resectability (36.4% vs. 87.0%, p = 0.0003). Histologic effectiveness (8.3%, 50.0%, and 41.7% versus 25.0%, 70.0%, and 5.0% in grade 3, grade 2, and grade 1, respectively, for patients with respiratory tract invasion versus those without it, p = 0.0189) and histologic stages (8.3%, 8.3%, 8.3%, 8.3%, 25.0%, and 41.7% versus 20.0%, 0%, 15.0%, 25.0%, 40.0%, and 0% in pathologic CR, stage I, stage IIA, stage IIB, stage III, and stage IV, respectively, for patients with respiratory tract invasion versus those without it, p = 0.0496) were significantly better in patients negative for respiratory tract invasion; the percentages of patients with lymph node metastasis did not differ significantly between the two groups. Comparison of the recurrence patterns showed that local failure was most common in patients with respiratory tract invasion, and distant failure was the leading cause of recurrence in patients without it.
Conclusions: Because the prognosis of patients with thoracic esophageal cancer infiltrating the respiratory tract is extremely poor, partially because of the low local effectiveness of preoperative concurrent chemotherapy and radiation therapy, caution is needed when deciding on salvage surgery.