226P - Clinical significance of elevated serum levels of CC thymus and activation-related chemokine in primary gastric cancer: potential for a serum diagn...

Date 28 September 2014
Event ESMO 2014
Session Poster Display session
Topics Gastric Cancer
Translational Research
Basic Principles in the Management and Treatment (of cancer)
Presenter Hye Won Chung
Citation Annals of Oncology (2014) 25 (suppl_4): iv58-iv84. 10.1093/annonc/mdu326
Authors H.W. Chung1, J. Lim2
  • 1Department Of Internal Medicine, Division Of Gastroenterology, International ST. Mary’s Hospital, 404-834 - Incheon/KR
  • 2Department Of Laboratory Medicine, Yonsei University College of Medicine, 120-752 - Seoul/KR



Chemokines and their receptors are involved in various processes of tumor cell biology. CC chemokine receptor 4 (CCR4), a receptor for thymus and activation-related chemokine (TARC, CCL17), is overexpressed in solid tumors including gastric cancer (GC), and its ligand, TARC, is considered to be a potent stimulator of GC cell proliferation and migration. Here, we evaluate the clinical significance of elevated serum TARC in GC patients and validate its potential as a biomarker for GC.


Serum levels of TARC, MDC (CCL22), MCP-1 (CCL2), and SCF were measured by chemiluminescent immunoassay, and compared among 4 disease groups; normal, high-risk, early GC (EGC), and advanced GC (AGC) groups (each, n = 25) along the gastric carcinogenic sequences (one-way ANOVA) in training set. Serum levels of TARC and SCF were re-evaluated in following independent validation set (90 normal, 30 high-risk, 50 EGC, 50 AGC) and compared with serum carcinoembryonic antigen (CEA) levels. To evaluate the diagnostic potential of serum TARC for GC, receiver operating characteristic (ROC) curve was generated and logistic regression analysis was performed. Correlations of serum TARC with clinicopathological parameters of GC were evaluated by Spearman's correlation.


In training set, serum TARC and SCF were significantly correlated with each other and different between cancer and non-cancer conditions (each, p < 0.001), while serum MDC and MCP-1 were not different between cancer and non-cancer conditions. In validation set, serum TARC levels were increased along the GC carcinogenic sequences, and were significantly higher in AGC groups (167.2 ± 111.1) than EGC (109.1 ± 67.7), high-risk (66.2 ± 47.7) and normal (67.5 ± 36.1) groups (post-hoc Bonferroni, each, p < 0.001), respectively. ROC curve and logistic regression analysis demonstrated a noticeable diagnostic potential of serum TARC both as a single-marker (72.0% sensitivity and 71.1% specificity at cut-off point, 0.37; logistic regression) and as a multiple-markers panel (72.6% and 88.2% at cut-off point, 0.54). Clinicopathologically, serum TARC levels were closely correlated with tumor size (ɣs = 0.227, p = 0.028), T-stage (ɣs = 0.340, p = 0.001), N-stage (ɣs = 0.318, p = 0.002), and distant metastasis (ɣs = 0.346, p = 0.001).


Serum TARC is a potential serologic biomarker for GC superior to CEA.


All authors have declared no conflicts of interest.