P-0036 - Angiogenic markers of response to neoadjuvant therapy and long-term prognosis in locally advance rectal cancers - a critical analysis of available d...

Date 28 June 2014
Event World GI 2014
Session Poster Session
Topics Anti-Cancer Agents & Biologic Therapy
Rectal Cancer
Presenter Olivia Lock
Citation Annals of Oncology (2014) 25 (suppl_2): ii14-ii104. 10.1093/annonc/mdu165
Authors O. Lock1, A. Khan2, R. Glynne-Jones3
  • 1Cardiff University, Cardiff/UK
  • 2University College London, London/UK
  • 3Mount Vernon Cancer Center, Northwood, United Kingdom, /



Prognostic heterogeneity exists in similarly staged locally advanced rectal cancer (LARC). Variability in individual tumour biology, particularly angiogenesis, is thought to determine disease progression. Analysis of molecular markers could improve accuracy of future prognoses and help tailor individual treatment protocols. Phase II trials are currently investigating the use of novel biological agents to target molecular markers implicated in angiogenesis. The aim of this review is to identify the available data regarding the integrity of VEGF, KRAS and EGFR as markers of tumour response to neoadjuvant chemoradiotherapy (NCRT) and long-term prognosis.


A literature search using Pubmed/Medline database was performed for studies analysing markers of angiogenesis (VEGF, EGFR, KRAS) in LARC. The former were cross-referenced in separate searches of studies in the English language, with the following keywords: “rectal cancer”, “response”, “downgrading”, “tumour regression grade”, “complete response”, “R resection”, “sphincter saving” and “survival”. Phased trials testing drugs, reviews, duplicate articles and studies missing information, were all excluded.


A total of 38 papers were included. Authors reported VEGF (11/15), VEGF gene (2/15) and VEGFR (2/15) expression. VEGF expression predicted response and prognosis in 5/7 and 6/8 studies, respectively. High VEGF gene and mRNA expression was associated with poor response and shortened disease-free survival (DFS). In the presence of VEGFR, one study found a better response whilst another recorded poor DFS. With regards to EGFR, authors reported EGFR expression (10/14), EGFR gene polymorphism (3/14) and EGFR gene copy number (1/14). The EGFR expression was correlated to response and prognosis in 4/6 and 6/7 studies, respectively. A positive/low EGFR expression was associated with tumour regression and high EGFR expression with poor DFS. The EGFR gene polymorphisms were observed in good response and poor DFS, whereas EGFR gene copy number was not associated with response prediction or prognosis. The KRAS mutation was reported in 16 studies half of which used Cetuximab in the NRCT regime. KRAS mutation was reported to predict response in only 6/14 and prognosis in 1/10 studies. The presence of KRAS mutation was associated with poor response in 5 studies. One study reported poor DFS in patients with mutated KRAS, however the majority did not find KRAS status to be a useful marker of prognosis.


The majority of studies correlated high VEGF and EGFR expression with poor response and shortened DFS and failed to show significant associations with KRAS status. A pooled statistical analysis on available data could not be performed due to the variability in methods used to quantify markers and study end-points. The review of current literature demonstrates the need for larger trials using a standardised method of analysis, to facilitate the quantification of biomarkers as indicators of prognosis and targeted therapy.