Lung Squamous Cell Carcinoma mRNA Expression Subtypes Are Reproducible, Clinically Important, and Correspond to Normal Cell Types

Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, 27599, USA.
Clinical Cancer Research (Impact Factor: 8.72). 10/2010; 16(19):4864-75. DOI: 10.1158/1078-0432.CCR-10-0199
Source: PubMed


Lung squamous cell carcinoma (SCC) is clinically and genetically heterogeneous, and current diagnostic practices do not adequately substratify this heterogeneity. A robust, biologically based SCC subclassification may describe this variability and lead to more precise patient prognosis and management. We sought to determine if SCC mRNA expression subtypes exist, are reproducible across multiple patient cohorts, and are clinically relevant.
Subtypes were detected by unsupervised consensus clustering in five published discovery cohorts of mRNA microarrays, totaling 382 SCC patients. An independent validation cohort of 56 SCC patients was collected and assayed by microarrays. A nearest-centroid subtype predictor was built using discovery cohorts. Validation cohort subtypes were predicted and evaluated for confirmation. Subtype survival outcome, clinical covariates, and biological processes were compared by statistical and bioinformatic methods.
Four lung SCC mRNA expression subtypes, named primitive, classical, secretory, and basal, were detected and independently validated (P < 0.001). The primitive subtype had the worst survival outcome (P < 0.05) and is an independent predictor of survival (P < 0.05). Tumor differentiation and patient sex were associated with subtype. The expression profiles of the subtypes contained distinct biological processes (primitive: proliferation; classical: xenobiotic metabolism; secretory: immune response; basal: cell adhesion) and suggested distinct pharmacologic interventions. Comparison with lung model systems revealed distinct subtype to cell type correspondence.
Lung SCC consists of four mRNA expression subtypes that have different survival outcomes, patient populations, and biological processes. The subtypes stratify patients for more precise prognosis and targeted research.

Download full-text


Available from: Christopher R Cabanski, Dec 16, 2013
1 Follower
57 Reads
  • Source
    • "While dendrograms provide an intuitive representation for studying the results of hierarchical clustering, the researcher is still ultimately left to decide which partitions along the tree to interpret as biologically important subpopulation differences. Often, in genomic studies, the determination and assessment of subpopulations are left to heuristic or ad hoc methods (Bastien et al., 2012; Verhaak et al., 2010; Wilkerson et al., 2010). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Cluster analysis has proved to be an invaluable tool for the exploratory and unsupervised analysis of high dimensional datasets. Among methods for clustering, hierarchical approaches have enjoyed substantial popularity in genomics and other fields for their ability to simultaneously uncover multiple layers of clustering structure. A critical and challenging question in cluster analysis is whether the identified clusters represent important underlying structure or are artifacts of natural sampling variation. Few approaches have been proposed for addressing this problem in the context of hierarchical clustering, for which the problem is further complicated by the natural tree structure of the partition, and the multiplicity of tests required to parse the layers of nested clusters. In this paper, we propose a Monte Carlo based approach for testing statistical significance in hierarchical clustering which addresses these issues. The approach is implemented as a sequential testing procedure guaranteeing control of the family-wise error rate. Theoretical justification is provided for our approach, and its power to detect true clustering structure is illustrated through several simulation studies and applications to two cancer gene expression datasets.
  • Source
    • ". Tumor protein p63 (TP63) was particularly upregulated in squamous cell carcinoma, as recently reported in another study [15] [35]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Introduction The development of reliable gene expression profiling technology increasingly impacts our understanding of lung cancer biology. Here, we used RNA sequencing (RNA-Seq) to compare the transcriptomes of non-small cell lung cancer (NSCLC) and normal lung tissues and to investigate expression in lung cancer tissues. Methods We enrolled 88 male patients (mean age, 61.2 years) with NSCLC. RNA-Seq was performed on 88 pairs of NSCLC tumor tissue and non-tumor tissue from 54 patients with adenocarcinoma and 34 patients with squamous cell carcinoma. Immunohistochemistry was performed to validate differential candidate gene expression in a different NSCLC group. Results RNA-Seq produced 25.41 × 106 (± 8.90 × 106) reads in NSCLC tissues and 24.70 × 106 (± 4.70 × 106) reads in normal lung tissues [mean (± standard deviation)]. Among the genes expressed in both tissues, 335 were upregulated and 728 were downregulated ≥ 2-fold (P < 0.001). Four upregulated genes–CBX3, GJB2, CRABP2, and DSP–not previously reported in lung cancer were studied further. Their altered expression was verified by immunohistochemistry in a different set of NSCLC tissues (n = 154). CBX3 was positive in 90.3% (139 cases) of the samples; GJB2, in 22.7% (35 cases); CRABP2, in 72.1% (111 cases); and DSP, in 17.5% (27 cases). The positive rate of CRABP2 was higher in adenocarcinoma than squamous cell carcinoma (p < 0.01). Conclusions CBX3 and CRABP2 expression was markedly increased in lung cancer tissues and especially CRABP2 may be promising candidate genes in lung adenocarcinoma.
    Lung cancer (Amsterdam, Netherlands) 06/2014; 84(3). DOI:10.1016/j.lungcan.2014.03.018 · 3.96 Impact Factor
  • Source
    • "Preclinical studies suggest that signaling via the INSR isoforms IR-A and IR-B may be of critical importance in NSCLC [11,12]. After the failures of two large randomized phase III anti-IGF1R trials in NSCLC, (Figitumumab [13], hR1507 [13,14]), IR-A signaling has been postulated as one of the major mediators of resistance to anti-IGF1R therapy [11,15]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate the insulin receptor isoform mRNA expression status in non-small cell lung cancer (NSCLC) patients. RNA-seq data from 614 NSCLC [355 adenocarcinomas (LUAD) and 259 squamous cell carcinomas (LUSC) ] and 92 normal lung specimens were obtained from The Cancer Genome Atlas (TCGA) to evaluate the mRNA expression of insulin receptor isoform A (IR-A) and insulin receptor isoform B (IR-B). The differential expression status of the insulin receptor isoforms in NSCLC patients was confirmed using qRT-PCR assays with lung cancer cDNA arrays and primary tumor samples. The mRNA expression levels of IR-B were significantly lower in some NSCLC samples compared to normal lung specimens, including both LUAD and LUSC. Notably, no IR-B transcripts were detected - only the IR-A isoform was expressed in 11% of NSCLC patients. This decrease in IR-B expression contributed to an elevated IR-A/IR-B ratio, which was also associated with lower epithelial-mesenchymal transition gene signatures in NSCLC and longer patient survival under standard of care in LUSC. In addition to NSCLC, RNA-seq data from TCGA revealed a similar increase in IR-A/IR-B ratio in many other cancer types, with high prevalence in acute myeloid leukemia, glioblastoma multiforme, and brain lower grade glioma. Our results indicate a common reduction of the mRNA expression level of IR-B and an increased IR-A/IR-B mRNA ratio in NSCLC and other tumor types. The relationship of altered IR-A/IR-B ratios with cancer progression and patient survival should be prospectively explored in future studies.
    BMC Cancer 02/2014; 14(1):131. DOI:10.1186/1471-2407-14-131 · 3.36 Impact Factor
Show more