Profiling of phospho-AKT, phospho-mTOR, phospho-MAPK and EGFR in non-small cell lung cancer

Abstract

Activation of numerous pathways has been documented in non-small cell lung cancer (NSCLC). Epidermal growth factor receptor (EGFR) has emerged as a common therapeutic target. The mitogen-activated protein kinase (MAPK) and AKT signaling pathways are downstream of EGFR and deregulated via genetic and epigenetic mechanisms in many human cancers. We evaluated selected markers in the EGFR pathway with reference to outcome. Tissues from 220 cases of NSCLC patients presented in a tissue microarray were assayed with immunohistochemistry for phosphorylated AKT, phosphorylated MAPK, phosphorylated mTOR, and EGFR and then quantified by automated image analysis. Individually, the biomarkers did not predict. Combined as ratios, p-mTOR/p-AKT, and p-MAPK/EGFR function as prognostic markers of survival (p=0.008 and p=0.029, respectively), however, no significance was found after adjustment (p=0.221, p=0.103). The sum of these ratios demonstrates a stronger correlation with survival (p<0.001) and remained statistically significant after adjustment (p=0.026). The algebraic combination of biomarkers offer the capacity to understand factors that predict outcome better than current approaches of evaluating biomarkers individually or in pairs. Our results show the sum of p-mTOR/p-AKT and p-MAPK/EGFR is a potential predictive marker of survival in NSCLC patients.

Keywords: AKT; EGFR; Image analysis; Immunohistochemistry; MAPK; Non-small cell lung cancer; Survival analysis; Tissue microarray; mTOR.

Figure 1.

(A and B) Immunohistochemical staining of representative cores from two different patients for the panel of markers investigated: phosphorylated mammalian target of rapamycin (p-mTOR), phosphorylated protein kinase B (p-AKT; T308), phosphorylated mitogen-activated protein kinase (p-MAPK), and the epidermal growth factor receptor (EGFR). EGFR showed membrane staining, whereas p-AKT, p-mTOR and p-MAPK showed cytoplasmic staining. (C) Immunohistochemical staining and quantitative imaging: strong positive pixels are red, positive pixels are orange, weak positive pixels are yellow. The region of the tissue core used for the analysis is outlined in black, with excluded internal regions shaded-out in grey. Scale bars = 100 µm.

Figure 2.

Kaplan-Meier survival analysis of non-small cell lung cancer patients. (A) Correlation of each single antibody expression with patient outcome. (B) Correlation of the ratio p-mTOR to p-AKT (p-mTOR/p-AKT) and the ratio of p-MAPK to EGFR (p-MAPK/EGFR) with patient outcome. (C) Correlation of three groups: both of the ratios were high (H/H), one was high (H/L), or both were low (L/L). (D) Correlation of double ratio with patient outcome.

Figure 3.

Hierarchical clustering of correlation coefficients for immunohistochemical expression of p-AKT, p-MAPK, p-mTOR and EGFR was performed with Weight Score. Four groups (Category 1 to 4) are defined.

Figure 4.

(A–D) Box plots of the scores of the individual biomarkers, p-AKT, p-MAPK, p-mTOR and EGFR, categorized by groups as defined by hierarchical clustering (Figure 3).