Esophageal cancer surgery - factors influencing survival

Sammanfattning: The principal aim of this thesis was to address factors that may affect survival after esophageal cancer surgery. Surgical resection remains the only established potentially curative treatment for patients with resectable esophageal cancer. It is an extensive surgical procedure that often combines surgery of the abdomen, chest, and neck and is associated with a considerable risk of major postoperative complications and in-hospital mortality. Moreover, even after successful esophageal resection, only a minority of the patients are cured. In this research project, we have conducted nationwide, population-based studies to evaluate the short-term and long-term postoperative outcomes in relation to possible effects of calendar period, neoadjuvant therapy, hospital volume, and surgeon volume. In studies I-III we identified all residents in Sweden diagnosed with primary esophageal cancer and treated with esophageal resection during the period January 1, 1987 through December 31, 2000 by means of data from the Cancer Register and In-patient Register. Data regarding tumor characteristics and preoperative treatment were collected retrospectively through manual reviews of histopathological records. The patients were followed up with respect to death or emigration through population registers. In study IV, details regarding tumor and patient characteristics, and surgical procedures were collected prospectively during the period April 2, 2001 to December 31, 2005, by using the Swedish Esophageal and Cardia Cancer Register, where most surgically treated esophageal cancer patients in Sweden are registered. The patients were followed up in the Total Population Register until April 2, 2006. According to study I, the long-term and short-term survival after esophageal cancer surgery has improved substantially in Sweden since 1987. The short-term mortality has been significantly reduced and is currently lower than 5%. The 5-year survival improved from 20% during the period 1987-1991 to 31% during the period 1997-2000, an improvement that was not explained by changes in patient or tumor characteristics. In study II, the overall postoperative survival was found to be similar in patients with and without neoadjuvant therapy (HR 0.99, 95% CI 0.86-1.16). Only patients with a complete histopathological response after neoadjuvant treatment (27% of all patients in the neoadjuvant group) had an improved prognosis (HR 0.71, 95% CI 0.53-0.94). The impact of hospital volume on long-term survival after esophageal cancer surgery was addressed in study III. After adjustment for several confounders, including tumor stage and comorbidity, no difference in long-term survival was found between patients operated on at highvolume and low-volume hospitals (HR 0.99, 95% CI 0.84-1.18). In study IV, the short-term prognosis after esophageal cancer surgery seemed to be more favorable in patients operated on by higher-volume surgeons compared to those operated on by low-volume surgeons (30-day mortality OR 0.39, 95% CI 0.09-1.70, 90-day mortality 0.42, 95% CI 0.10-1.80). There was no tendency, however, to further survival improvement with increasing surgical workload among the experienced esophageal cancer surgeons. In conclusion, the chance of cure after surgery for localized esophageal cancer has increased during recent years. Since neoadjuvant treatment and hospital volume seem to have no or only limited influence, the reasons for the improved prognosis remain to be identified. Patients operated on at high-volume hospitals by experienced esophageal cancer surgeons have a better short-term prognosis.