Author of

• 10525

  +

  P1.02 - Poster Session 1 - Novel Cancer Genes and Pathways (ID 144)

  • Event: WCLC 2013
  • Type: Poster Session
  • Track: Biology
  • Presentations: 1
  • Moderators:
  • Coordinates: 10/28/2013, 09:30 - 16:30, Exhibit Hall, Ground Level

  • 10530

    +

    P1.02-005 - Prognostic role of expression levels of FABP3, H19, TFPI2, AKR1B10 CYP3A5, SCGB3A2 genes in adenocarcinoma stage I patients (ID 3208)

    09:30 - 09:30  |  Author(s): G. Servillo
    • Abstract

    Background
    In resected lung cancer, no reliable clinical or molecular predictors are currently available for identifying patients with high risk for developing recurrent disease. In a previous study, we compared gene expression profiling from adenocarcinoma specimens and normal lung tissue with non relapse (NR) and early relapse (ER), using Affimetrix human microarray HG-U133Plus 2.0. We selected 5 genes up-regulated and 4 genes down-regulated were predictive for clustering patients in ER and NR (Siggillino et al. ECCO 2011). Here, we validate our results using an independent cohort of patients with lung adenocarcinoma stage I to identify novel genes involved in the risk of ER compared to NR disease.

    Methods
    From tissue banking of 180 consecutive resected NSCLC stage I patients at two Italian institutions, we selected 58 frozen specimens of lung adenocarcinoma tissue with corresponding normal lung. Total RNA was isolated from tumor and normal lung specimens using RNA Universal Tissue Kit and automatically purified by Biorobot-EZ1 instrument (Qiagen). Quantification of mRNA expression levels of 9 genes (5 up-regulated: CLCA2, FABP3, H19, TFPI2, AKR1B10 and 4 down-regulated: CYP3A5, ALDH3A1, SCGB3A2, SCGB1A1, were analyzed by real-time one-step RT-PCR using QuantiFast technology by RotorGeneQ instrument (Qiagen), and the results were compared considering β-actin as the internal reference gene and as calibrator the pool of normal tissues of analyzed patients. The gene expression of all evaluated genes and their association with relapse disease measures were assessed by t-test and logistic regression model was used for multivariate analysis.

    Results
    Fifteen-eight adenocarcinoma stage I patients were evaluable, 17% of which had an ER. Patients characteristics were as follows: median age was 65.8 years (38.7-81.5), 67.2% were male, 91.3% were PS 0, 70.7% were ever-smokers. The expression median values of 9 genes: CLCA2, FABP3, H19, TFPI2, AKR1B10, CYP3A5, ALDH3A1, SCGB3A2, SCGB1A1 were 0.30, 0.71, 0.34, 1.10, 0.26, 0.24, 0.42, 0.53, 0.09, respectively. Among all genes evaluated, the NR vs ER mean expression levels of two genes down-regulated (CYP3A5, 1.09 vs 0.30; SCGB3A2, 2.28 vs 0.98) and two genes up-regulated (AKR1B10, 4.53 vs 34.20; FABP3 1.25 vs 1.55) were superimposable respect to the results of previous microarray analysis. The median disease free survival (DFS) and overall survival (OS) were 21 and 23 months, respectively. In the logistic multivariate analysis the mean expression levels of all genes showed a tendency to predict the ER in the overall population (p=0.07). Nevertheless considering only the expression levels of genes (FABP3, H19, TFPI2, AKR1B10, CYP3A5, SCGB3A2) identified as significant with t-test, the covariates in multivariate analysis increased their capacity of ER prediction (p= 0.028).

    Conclusion
    Our results indicate that it is possible to define, through gene expression, a characteristic gene profiling of early relapse tumor patients with an increased risk of relapse disease. The contemporary expression levels of 6 genes (FABP3, H19, TFPI2, AKR1B10, CYP3A5, SCGB3A2) predicted a worse DFS. Such features may have important implications for future targeted therapies. We thank Italian Association for Cancer Research (AIRC) for supporting the study.