228P - Significance of para-aortic lymph node dissection for advanced gastric cancer patients following DCS therapy

Date 17 December 2016
Event ESMO Asia 2016 Congress
Session Poster lunch
Topics Anti-Cancer Agents & Biologic Therapy
Gastric Cancer
Surgery and/or Radiotherapy of Cancer
Presenter Sachio Fushida
Citation Annals of Oncology (2016) 27 (suppl_9): ix68-ix85. 10.1093/annonc/mdw582
Authors S. Fushida, K. Oyama, J. Kinoshita, I. Ninomiya, T. Ohta
  • Gastroenterological Surgery, Kanazawa University, 920-8641 - Kanazawa/JP

Abstract

Background

Although JCOG0405 revealed preoperative S-1 plus cisplatin chemotherapy followed by surgery was safety and effective, it was not clear for detailed consideration, such as the correlation between the number of metastatic lymph nodes and prognosis. In this study, we demonstrated the significance of para-aortic lymph node dissection (PAND) for advanced gastric cancer patients following DCS therapy.

Methods

Between July 2005 and November 2015, we evaluated retrospectively 24 advanced gastric cancer patients with para-aortic lymph node metastasis diagnosed by thin slice helical CT scan, without hematological metastasis or disseminated metastasis (16a2b1: 14, +a: 10). Patients received two or three cycles of preoperative chemotherapy consisting of docetaxel and cisplatin (35mg/m2) on day 1 and 15 and oral S-1 (40mg/m2 twice daily) on day 1-14 every 4weeks. After chemotherapy, gastrectomy with systematic PAND (16a2b1) was performed. The lymph nodes with disappeared cancer cells due to chemotherapy were determined as follows: no evidence of cancer cells by H-E staining; lymphoid follicles were disturbed by fibrosis and granulomatous changes.

Results

Response rate was 92%, and pathological response rate was 77%. The adverse events related DCS therapy (G3/4) were 38% of neutropenia, 8% of diarrhea and 8% of anorexia. Gastrectomy with PAND required 396 min (median) of operation time and lost 1225 g of blood. Median number of dissected para-aortic lymph node was 13. Surgery related complications were anastomotic leakage in 2, chyle leakage in 2, and pancreas fistula in 1, postoperative bleeding in 1 and SSI in 1. The mean number of residual cancer cells in dissected para-aortic lymph node was 6, in which small lymph nodes without diagnosis as metastasis were contained. Three-years overall survival was 53%. The patients with 3 more than PAN metastasis or with expand lymph node metastasis beyond 16a2b1 showed significantly worse prognosis than patients without them.

Conclusions

Although PAND following DCS therapy has underwent safely, this therapy should be only performed in specialized institutions. To improve prognosis, systematic PAND or basin dissection should be recommended rather than pick up dissection.

Clinical trial indentification

UMIN000006036

Legal entity responsible for the study

N/A

Funding

N/A

Disclosure

All authors have declared no conflicts of interest.