Integrated molecular portrait of non-small cell lung cancers

Abstract

Background: Non-small cell lung cancer (NSCLC), a leading cause of cancer deaths, represents a heterogeneous group of neoplasms, mostly comprising squamous cell carcinoma (SCC), adenocarcinoma (AC) and large-cell carcinoma (LCC). The objectives of this study were to utilize integrated genomic data including copy-number alteration, mRNA, microRNA expression and candidate-gene full sequencing data to characterize the molecular distinctions between AC and SCC.

Methods: Comparative genomic hybridization followed by mutational analysis, gene expression and miRNA microarray profiling were performed on 123 paired tumor and non-tumor tissue samples from patients with NSCLC.

Results: At DNA, mRNA and miRNA levels we could identify molecular markers that discriminated significantly between the various histopathological entities of NSCLC. We identified 34 genomic clusters using aCGH data; several genes exhibited a different profile of aberrations between AC and SCC, including PIK3CA, SOX2, THPO, TP63, PDGFB genes. Gene expression profiling analysis identified SPP1, CTHRC1 and GREM1 as potential biomarkers for early diagnosis of the cancer, and SPINK1 and BMP7 to distinguish between AC and SCC in small biopsies or in blood samples. Using integrated genomics approach we found in recurrently altered regions a list of three potential driver genes, MRPS22, NDRG1 and RNF7, which were consistently over-expressed in amplified regions, had wide-spread correlation with an average of ~800 genes throughout the genome and highly associated with histological types. Using a network enrichment analysis, the targets of these potential drivers were seen to be involved in DNA replication, cell cycle, mismatch repair, p53 signalling pathway and other lung cancer related signalling pathways, and many immunological pathways. Furthermore, we also identified one potential driver miRNA hsa-miR-944.

Conclusions: Integrated molecular characterization of AC and SCC helped identify clinically relevant markers and potential drivers, which are recurrent and stable changes at DNA level that have functional implications at RNA level and have strong association with histological subtypes.

Figure 1

Flowchart of DGS algorithm. Driver and targets are identified in a three step process, as shown. Candidate drivers are firstly selected from genes/miRNAs that reside in copy-number altered (CNA) regions and filtered by various procedures, for example based on fold-change and consistency between expression level and copy number status. The rests of genes/miRNAs are candidate targets, which are grouped based on correlation with the candidate drivers. Correlation between all drivers and targets in each module is highlighted using sparse canonical correlation analysis (SCCA).

Figure 2

Differential genomic regions for AC vs LCC vs SCC populations aCGH profiles. The three upper panels display the average profiles of AC, LCC and SCC subpopulations as their respective frequencies of gains (green, from 0 to 100%) and losses (red, from 0 to −100%) along the human genome. Darker green bars correspond to the frequencies of amplifications, defined as regions with a log2 (ratio) above 1.0. The lowest panel shows the significance of the ANOVA tests displaying the minus log10-transformed raw (lighter blue) and BH-adjusted (darker blue) p-values. The horizontal red line corresponds to a BH-adjusted p-value < 1.0E-05. Arrows point to the two most significant differential regions: 3q26.2-3q29 and 22q12.1-22q13.1.

Figure 3

Principal component plots of the mRNA data. The first and second principal component plot (left) and the first and third principal component plot (right) of the mRNA data revealed the separation of squamous-cell carcinoma (S) from the adeno-carcinoma (A) and large-cell carcinoma (L).

Figure 4

Network links between 23 genes in pathway of mismatch repair and three driver genes. Network links between 23 genes in pathway of mismatch repair and driver genes MRPS22, NDRG1, RNF7. Links shown include physical interactions, metabolic and signaling links from the functional coupling network (http://FunCoup.sbc.su.se).