Clinically relevant characterization of lung adenocarcinoma subtypes based on cellular pathways: an international validation study

Abstract

Lung adenocarcinoma (AD) represents a predominant type of lung cancer demonstrating significant morphologic and molecular heterogeneity. We sought to understand this heterogeneity by utilizing gene expression analyses of 432 AD samples and examining associations between 27 known cancer-related pathways and the AD subtype, clinical characteristics and patient survival. Unsupervised clustering of AD and gene expression enrichment analysis reveals that cell proliferation is the most important pathway separating tumors into subgroups. Further, AD with increased cell proliferation demonstrate significantly poorer outcome and an increased solid AD subtype component. Additionally, we find that tumors with any solid component have decreased survival as compared to tumors without a solid component. These results lead to the potential to use a relatively simple pathological examination of a tumor in order to determine its aggressiveness and the patient's prognosis. Additional results suggest the ability to use a similar approach to determine a patient's sensitivity to targeted treatment. We then demonstrated the consistency of these findings using two independent AD cohorts from Asia (N = 87) and Europe (N = 89) using the identical analytic procedures.

Figure 1. Overall Study Design.

We developed cellular pathway expression summaries and tested the relationship of each to pathological subtypes of lung adenocarcinoma (AD). We also tested each pathway's association with survival. Because the cellular pathways are driving the pathological differences, the relationship between pathology and prognosis is secondary to the relationship between the cellular pathway and prognosis (indicated by a thinner line with both component colors). We also directly tested the relationship between pathology and prognosis to examine the need for molecular information.

Figure 2. Hierarchical clustering yields three distinct groups.

Hierarchical clustering of all 432 lung adenocarcinomas (AD) which shows three main groups of tumors denoted by yellow, blue and orange in the dendrogram. Below is a heat map that represents the 200 genes that are most highly correlated to the left most cluster followed by the middle cluster and the right most cluster. Red indicates relative over-expression (compared to the median) while green indicates relative under-expression.

Figure 3. Clusters show survival differences.

Kaplan-Meier survival curves for 432 lung adenocarcinomas (AD) showing a significant difference between clusters (log-rank test: p = 0.000194). Abbreviations: Cluster 1, left most cluster in Figure 2; Cluster 2, middle cluster is Figure 2; Cluster 3, right most cluster in Figure 2.

Figure 4. Stage, grade, sex and pathology effects on survival.

Log-rank tests of differences between Kaplan Meier survival curves verify that the dataset is consistent with previous results. Higher stage patients do significantly poorer as compared to lower stages (A) and high grade patients (B) (poor differentiation) have increased hazards compared to low or intermediate grade patients. Gender is a marginally significant prognostic indicator; males have poorer survival (C). Additionally we examined pathology (D) defined by major lung adenocarcinoma (AD) subtype (plurality of tumor cross section). There was no significant overall difference between the four main subtypes as well as tumors that did not fall into one of these categories. Abbreviations: CIS, carcinoma in situ.

Figure 5. Stage specific survival differences of cell cycle pathway.

(A) Kaplan-Meier survival analysis of the cell cycle stimulatory (CC+) pathway shows that patients with relative over-expression of CC+ do significantly worse (p = 0.000113). This trend is consistent inside each stage however is only marginally significant (p = 0.062) in stage 1 patients (B), not significant in stage 2 patients (C) and highly significant (p = 0.0042) in stage 3 patients (D). Abbreviations: w/, with; w/o, without.

Figure 6. Survival differences for different pathologies.

(A) Kaplan-Meier survival analysis of solid lung adenocarcinoma (AD) shows that patients with any solid component do significant worse (p = 0.000166) than those with no solid component. This trend is highly significant (p = 0.00562) in stage 1 patients (B) and marginally significant (p = 0.0295) in stage 3 patients (C). Those patients with any carcinoma in situ (CIS) component (D) did significantly better (p = 0.0342) than those without any CIS component. Comparing patients with papillary component (E) to those without any papillary component showed no significant difference but in stage 3 patients (F) those with some papillary component did significantly better than those without any papillary component. Abbreviations: w/, with; w/o, without.