P-328 - Which is the best first approach for liver-only synchronic metastasis rectal cancer?

Date 04 July 2015
Event WorldGI 2015
Session Posters
Topics Cytotoxic agents
Surgical oncology
Colon and Rectal Cancer
Biological therapy
Radiation oncology
Presenter I. Ghanem
Citation Annals of Oncology (2015) 26 (suppl_4): 1-100. 10.1093/annonc/mdv233
Authors I. Ghanem1, M. Blazquez2, O. Higuera1, L. Lema2, N. Rodriguez1, C. Gomez-Martin2, A. Mata1, D. Lora2, A. Custodio1, J. Feliu1
  • 1University Hospital La Paz, Madrid/ES
  • 2University Hospital 12 de Octubre, Madrid/ES



Approximately 10% of patients (p) with rectal cancer have synchronic liver-only metastasis (RCSLM). Although local and systemic therapies are required for this setting, there is not a well defined sequence of treatments. We previously reported the analysis of both, the initial systemic and local approaches at a single institution. Now we present the analysis data of two institutions


We retrospectively reviewed the medical records of RCSLM p between Jan 2005 and Jan 2014. The “first chemotherapy” (Ch) approach and the “first local” (Lo) approach (surgery or radiotherapy) were reviewed in terms of patient characteristics, tumour features and survival.


Seventy-four p (43 men) with a median age of 64 years old (35-88) and ECOG 0-1 (88%) were analyzed. Most (80%) of the 51 p with known K-Ras status had wild-type. Thirty-nine p (53%) had bilobar liver metastasis (LM) with a median number LM of 3 and a median size of 2.9 cm for the largest lesion. Fifty-two p (70%) were treated with curative intention and, liver and rectal resections (LRR) were performed for 43p (58%). Basal CEA > 10.5 (p = 0.05), largest LM size (p = 0.04), non-LRR (p = 0.007) and non-R0 resection of LM (p = 0.03) were found to be predictor of poor overall survival (OS). Thirty-four p were treated with Ch and 18 p with Lo approach. Median value of CEA for these patients was 10.5 ng/ml. Table 1 shows the characteristics of both groups of p. After a median follow up of 34 months there were 20 events (38%) with a median OS of 54.7m vs. 67.6m (0.41) for Ch and Lo respectively.


Basal CEA > 10.5, largest LM size, non-LRR and non-R0 resection of LM were predictors of poor OS. The Ch approach was chosen for patients with worse prognostic. However, there were no significant differences in OS between Ch and Lo approaches. Clinical trials are needed to evaluate the best therapy strategy at this setting.

Table: P-328. Ch and Lo patients characteristics