|
|
Failure after Neoadjuvant Chemoradiation and Surgery for Esophageal Cancer: Do Complete Responders Recur Differently than Non-Complete Responders?
R. A. Meguid1, C. M. Hooker1, J. T. Taylor1, L. R. Kleinberg1, S. M. Cattaneo1, M. Sussman1, S. C. Yang1, R. F. Heitmiller2, A. A. Forastiere1, M. V. Brock1. 1Johns Hopkins University School of Medicine, Baltimore, MD, 2Union Memorial Hospital, Baltimore, MD,
BACKGROUND: Since the early 1990s, neoadjuvant chemoradiation and surgery has become the mainstay of therapy for localized esophageal cancer in our hospital. We hypothesize that current radiation therapy and surgery lead to superb local control with most relapses being outside of the surgical and radiation fields. In addition, there are few data that characterize recurrence patterns and time to recurrence in patients who undergo esophagectomy after complete response to neoadjuvant chemoradiation therapy.
METHODS: A retrospective review of 269 patients who underwent neoadjuvant chemoradiation and definitive surgical therapy was performed at a single institution. Sites of recurrence were identified as esophagus, chest (excluding esophagus), liver, abdomen (excluding liver), brain, bone and other. Pathology of recurrence was confirmed when available.
RESULTS: Of the 269 patients, 82 (30.5%) were complete responders (CR) to neoadjuvant therapy, with no residual disease on surgical pathology after initial resection. The remaining 187 (69.5%) were partial or non-responders (non-CR). Of 269 patients examined, 84 (31.2%) had recurrence of esophageal cancer. Of those who recurred, median time from surgery to recurrence was10.8 months. Of those who were CR, 20/82 (24.4%) developed recurrence, whilst 64/187 (34.2%) of the non-CR patients developed recurrence (p=0.109). Non-CR patients were1.96 times more likely to recur than CR patients (p=0.009). A comparison ofthe patterns of, and associated times to recurrence for CR versus non-CR patients are shown in the accompanying table. The Kaplan-Meier curves comparing survival of CR recurrences versus the non-CR recurrences are shown in the accompanying figure. These survival curves diverge at 15 months. The median survival of CR patients with thoracic recurrence was 52.1 months, as compared to 30.1 months for CR patients with extra-thoracic recurrence, and 24.6 months for non-CR patients with thoracic recurrence, and 18.3 months for non-CR patients with extra-thoracic recurrence.
CONCLUSIONS: In the vast majority of cases, recurrences are systemic, not local. Extra-thoracic recurrences occur sooner than thoracic recurrences, regardless of CR status. Despite CR and non-CR patients having similar patterns and timing of thoracic and extra-thoracic recurrences, CR patients with thoracic metastases have a longer median survival.
| Site | Recurrence in CR group (% of total recurrence) | Median time to recurrence for CR in months | Recurrence in non-CR group (% of total recurrence) | Median time to recurrence for non-CR in months | Log Rank Test p-value comparing median time to recurrence | | Abdomen (excluding liver) | 4 (20.0) | 9.8 | 12 (18.7) | 8.7 | 0.650 | | Liver | 1 (5.0) | Not yet reached | 11 (17.2) | 5.8 | 0.297 | | Brain | 0 (0.0) | - | 4 (6.3) | 10.6 | - | | Bone | 1 (5.0) | Not yet reached | 8 (12.5) | 10.8 | 0.465 | | Chest (excluding esophagus) | 6 (30.0) | 14.4 | 17 (26.6) | 14.5 | 0.924 | | Esophagus | 3 (15.0) | 9.3 | 7 (10.9) | 7.1 | 0.303 | | Other | 2 (10.0) | 28.2 | 2 (3.1) | 17.0 | 0.433 | | Multiple Sites | 3 (15.0) | 37.7 | 3 (4.7) | 3.4 | 0.025 |
Back to 2008 Annual Meeting
Back to Program Outline
|
