PD-0021 - Value of KRAS, BRAF and PIK3CA Mutations and Benefits from Systemic Chemotherapy in Colorectal Peritoneal Carcinomatosis

Date 27 June 2014
Event World GI 2014
Session Poster discussion session IV - Miscellaneous
Topics Biomarkers
Colon Cancer
Rectal Cancer
Presenter Yusuke Sasaki
Citation Annals of Oncology (2014) 25 (suppl_2): ii5-ii13. 10.1093/annonc/mdu164
Authors K. Kato, Y. Yamada, H. Shoji, N. Okita, A. Takashima, Y. Honma
  • National Cancer Center Hospital, Chuo-ku/JP

Abstract

Introduction

Peritoneal carcinomatosis (PC) from colorectal cancer has been historically associated with a poor prognosis. Although survival improves after cytoreduction and hyperthermicintraperitoneal chemotherapy, systemic chemotherapy is not typically considered.

Methods

This is a retrospective review of 547 patients with metastatic colorectal cancer treated with chemotherapy at our institution from February 2006 to October 2011. Data collected included patient characteristics, sites of metastases, chemotherapy regimen, tumor location, gene variations (KRAS codon 12 and 13, BRAF exon 15 and PIK3CA exon 9 and 20), and survival duration. We performed direct sequencing of KRAS and PIK3CA, and high-resolution melting analysis of BRAF.

Results

The characteristics of the PC group (n = 116) were similar to those of non-PC patients in median age, performance status, and chemotherapy regimen as first-line treatment, but differed in sex (52.1% vs. 61.8% males, P < 0.001) and tumor location (51.2% vs. 23.5% proximal, P < 0.001). The median overall survival (OS) was 23.3 and 29.1 months for PC and non-PC patients, respectively (hazard ratio [HR] = 1.20; 95% CI, 0.92–1.57; P = 0.17). In PC patients, the only prognostic factor for OS was the addition of bevacizumab to chemotherapy in both univariate and multivariate analysis. Although chemotherapy without bevacizumab to PC patients resulted in a significantly shorter OS than to non-PC patients (HR = 1.84, P = 0.004), there was no difference in OS between PC and non-PC patients as a bevacizumab treatment (HR = 1.07, P = 0.70). Furthermore, a statistically significant higher incidence of the BRAF mutation was observed in PC patients than in non-PC patients (26.0% vs. 6.9%, P < 0.001), although KRAS and PIK3CA status showed no difference. BRAF mutations were strongly associated shorter OS in non-PC patients (13.5 vs. 30.1 months, HR 1.55), but not in PC patients (20.4 vs. 19.0 months; HR 0.89).

Conclusion

Systemic chemotherapy, especially bevacizumab, improved survival in patients with PC from colorectal cancer as well as non-PC. High frequency of BRAF mutation was observed in PC patients, but it wasn't associated with prognosis.