Identification and validation of gene expression models that predict clinical outcome in patients with early-stage laryngeal cancer

Improving data interpretability with new differential sample variance gene set tests

Background

Gene set analysis methods have played a major role in generating biological interpretations from omics data such as gene expression datasets. However, most methods focus on detecting homogenous pattern changes in mean expression and methods detecting pattern changes in variance remain poorly explored. While a few studies attempted to use gene-level variance analysis, such approach remains under-utilized. When comparing two phenotypes, gene sets with distinct changes in subgroups under one phenotype are overlooked by available methods although they reflect meaningful biological differences between two phenotypes. Multivariate sample-level variance analysis methods are needed to detect such pattern changes.

Results

We use ranking schemes based on minimum spanning tree to generalize the Cramer-Von Mises and Anderson-Darling univariate statistics into multivariate gene set analysis methods to detect differential sample variance or mean. We characterize these methods in addition to two methods developed earlier using simulation results with different parameters. We apply the developed methods to microarray gene expression dataset of prednisolone-resistant and prednisolone-sensitive children diagnosed with B-lineage acute lymphoblastic leukemia and bulk RNA-sequencing gene expression dataset of benign hyperplastic polyps and potentially malignant sessile serrated adenoma/polyps. One or both of the two compared phenotypes in each of these datasets have distinct molecular subtypes that contribute to heterogeneous differences. Our results show that methods designed to detect differential sample variance are able to detect specific hallmark signaling pathways associated with the two compared phenotypes as documented in available literature.

Conclusions

The results in this study demonstrate the usefulness of methods designed to detect differential sample variance in providing biological interpretations when biologically relevant but heterogeneous changes between two phenotypes are prevalent in specific signaling pathways. Software implementation of the developed methods is available with detailed documentation from Bioconductor package GSAR. The available methods are applicable to gene expression datasets in a normalized matrix form and could be used with other omics datasets in a normalized matrix form with available collection of feature sets.

Molecular characterization, immunocorrelation analysis, WGCNA analysis and machine learning modeling of genes associated with copper death subtypes of laryngeal cancer

BACKGROUND

Laryngeal cancer is a malignant tumor that originates from the mucous membrane of the larynx. Currently, the specific involvement mechanism of copper death in laryngeal cancer patients has not been deeply studied.

OBJECTIVE

This study aims to explore the molecular characteristics and clinical survival significance of copper death-related genes in laryngeal cancer.

METHODS

Relevant transcriptomes and clinical data were retrieved and downloaded from the GEO database. Differential expression genes related to laryngeal cancer and copper death were selected, and the immune function, clinical risk correlation, and survival prognosis were analyzed.

RESULTS

The differential analysis results showed that the differential expression genes related to laryngeal cancer and Cu-proptosis included SLC31A1 and ATP7B, and there was interaction between the immune cell groups in the differential genes of copper death in laryngeal cancer. Decreasing the expression of the gene ANXA5 or increasing the expression of the gene SERPINH1 can increase the susceptibility to laryngeal cancer.

CONCLUSION

Copper death-related genes can affect the survival prognosis of laryngeal cancer patients. Detection of changes in their expression can provide new diagnostic and treatment directions for the progression of early-stage laryngeal cancer.

Oncological Outcomes of Transoral Laser Microsurgery for Early Stage Glottic Cancer with Involvement of the Anterior Commissure

OBJECTIVES

To assess the value of carbon dioxide transoral laser microsurgery (CO2 TOLMS) for early-stage glottic cancer with special regard to involvement of the anterior commissure (AC).

STUDY DESIGN

Single-center retrospective cohort study.

SETTING

Grade-A tertiary hospital.

METHODS

A retrospective analysis of patients with early-stage (Tis-T2) glottic cancer who underwent CO2 TOLMS. All patients had at least 2 years of follow-up. The univariate and multivariate survival analyses were used to identify the risk factors for recurrence and the Kaplan-Meier method was used to analyze OS and DSS rates.

RESULTS

A total of 102 patients were included in the study. Eleven patients (10.78%) had recurrence. The univariate analysis showed that the recurrence was associated with the AC classification, T staging, tumor size, and tobacco use (P < .05). However, on multivariate analysis, the AC classification was the only independent risk factor for recurrence (P < .001, HR = 3.179). AC classification were distributed as follows: 59 (57.84%) AC0, 29 (28.43%) AC1, 8 (7.84%) AC2, and 6 (5.88%) AC3, 2-year/5-year OS and DSS rates were progressively reduced in the AC0, AC1, AC2, and AC3 groups (P < .001). At the same T staging, the OS rates incrementally decreased as the level of involvement of the AC became higher (P = .004).

CONCLUSION

CO2 TOLMS is an effective treatment for early-stage glottic cancer. AC involvement is an independent risk factors for recurrence and poor prognosis. The AC classification system may be better at grading the prognosis of patients with early-stage glottic cancer and has prognostic value independent of T staging.

Tumor-infiltrating immune cells and survival in head and neck squamous cell carcinoma: a retrospective computational study

The immune infiltration profiles of the tumor microenvironment have effects on the prognosis of head and neck squamous cell carcinoma (HNSCC). Whereas, HNSCC is a heterogeneous group of tumors, but past work has not taken this into consideration. Herein, we investigate the associations between survival and the function of immune cells in different tumorigenic sites of HNSCC. 1149 samples of HNSCC were collected from publicly accessible databases. Based on gene expression data, CIBERSORTx was applied to determine the proportion of 22 immune cell subpopulations. In the Cox regression model, the associations between overall survival, disease-free survival, and immune cells were examined, modeling gene expression and immune cell proportion as quartiles. Consensus cluster analysis was utilized to uncover immune infiltration profiles. Regardless of tumor sites, CD8+ T cells and activated CD4 memory T cells were associated with favorable survival, while eosinophils were the opposite. The survival of the hypopharynx, oral cavity, and larynx subsites was somewhat affected by immune cells, while the survival of the oropharynx subsite potentially was the most impacted. High expression of TIGIT, CIITA, and CXCR6 was linked to better survival, mainly in the oropharynx subsite. Immune cell clusters with four distinct survival profiles were discovered, of which the cluster with a high CD8+ T cell content had a better prognosis. The immune-infiltration pattern is related to the survival of HNSCC to varying degrees depending on the tumor sites; forthcoming studies into immune-mediated infiltration profiles will lay the groundwork for treating HNSCC with precision therapy.

Transcriptional subtypes of glottic cancer characterized by differential activation of canonical oncogenic programming

BACKGROUND

There is a paucity of data concerning molecular heterogeneity among glottic squamous cell carcinoma, and the clinical implications thereof.

METHODS

Data corresponding to glottic squamous cell carcinoma were derived from The Cancer Genome Atlas. The Onco-GPS computational methodology was levied to derive four patterns of transcriptional activity and three functional subtypes of glottic cancer.

RESULTS

Thirty glottic cancer samples stratified to three distinct oncogenic states (S0-S2) based on a Onco-GPS model containing four transcriptional components (F0-F3). Membership in S2 and association with transcriptional component F0 conveyed an invasive phenotype, with transcriptional activity strongly reflecting EMT programming (including TGF-B and NF-KB signaling). S2 membership also correlated with inferior disease-specific survival (HR 9.027, 95% CI 1.021-79.767), and higher incidences of extracapsular spread and perineural invasion.

CONCLUSIONS

We present a functional taxonomy of glottic cancer, with subtypes demonstrating differential upregulation of canonical oncogenic networks and survival implications.

Transcriptomics and solid tumors: The next frontier in precision cancer medicine

Transcriptomics, which encompasses assessments of alternative splicing and alternative polyadenylation, identification of fusion transcripts, explorations of noncoding RNAs, transcript annotation, and discovery of novel transcripts, is a valuable tool for understanding cancer mechanisms and identifying biomarkers. Recent advances in high-throughput technologies have enabled large-scale gene expression profiling. Importantly, RNA expression profiling of tumor tissue has been successfully used to determine clinically actionable molecular alterations. The WINTHER precision medicine clinical trial was the first prospective trial in diverse solid malignancies that assessed both genomics and transcriptomics to match treatments to specific molecular alterations. The use of transcriptome analysis in WINTHER and other trials increased the number of targetable -omic changes compared to genomic profiling alone. Other applications of transcriptomics involve the evaluation of tumor and circulating noncoding RNAs as predictive and prognostic biomarkers, the improvement of risk stratification by the use of prognostic and predictive multigene assays, the identification of fusion transcripts that drive tumors, and an improved understanding of the impact of DNA changes as some genomic alterations are silenced at the RNA level. Finally, RNA sequencing and gene expression analysis have been incorporated into clinical trials to identify markers predicting response to immunotherapy. Many issues regarding the complexity of the analysis, its reproducibility and variability, and the interpretation of the results still need to be addressed. The integration of transcriptomics with genomics, proteomics, epigenetics, and tumor immune profiling will improve biomarker discovery and our understanding of disease mechanisms and, thereby, accelerate the implementation of precision oncology.

Signature microRNAs and long noncoding RNAs in laryngeal cancer recurrence identified using a competing endogenous RNA network

The aim of the present study was to identify novel microRNA (miRNA) or long noncoding RNA (lncRNA) signatures of laryngeal cancer recurrence and to investigate the regulatory mechanisms associated with this malignancy. Datasets of recurrent and nonrecurrent laryngeal cancer samples were downloaded from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus database (GSE27020 and GSE25727) to examine differentially expressed miRNAs (DE-miRs), lncRNAs (DE-lncRs) and mRNAs (DEGs). miRNA-mRNA and lncRNA-miRNA networks were constructed by investigating the associations among these RNAs in various databases. Subsequently, the interactions identified were combined into a competing endogenous RNA (ceRNA) regulatory network. Feature genes in the miRNA-mRNA network were identified via topological analysis and a recursive feature elimination algorithm. A support vector machine (SVM) classifier was established using the betweenness centrality values in the miRNA-mRNA network, consisting of 32 optimal feature-coding genes. The classification effect was tested using two validation datasets. Furthermore, coding genes in the ceRNA network were examined via pathway enrichment analyses. In total, 21 DE-lncRs, 507 DEGs and 55 DE-miRs were selected. The SVM classifier exhibited an accuracy of 94.05% (79/84) for sample classification prediction in the TCGA dataset, and 92.66 and 91.07% in the two validation datasets. The ceRNA regulatory network comprised 203 nodes, corresponding to mRNAs, miRNAs and lncRNAs, and 346 lines, corresponding to the interactions among RNAs. In particular, the interactions with the highest scores were HLA complex group 4 (HCG4)-miR-33b, HOX transcript antisense RNA (HOTAIR)-miR-1-MAGE family member A2 (MAGEA2), EMX2 opposite strand/antisense RNA (EMX2OS)-miR-124-calcitonin related polypeptide α (CALCA) and EMX2OS-miR-124-γ-aminobutyric acid type A receptor γ2 subunit (GABRG2). Gene enrichment analysis of the genes in the ceRNA network identified that 11 pathway terms and 16 molecular function terms were significantly enriched. The SVM classifier based on 32 feature coding genes exhibited high accuracy in the classification of laryngeal cancer samples. miR-1, miR-33b, miR-124, HOTAIR, HCG4 and EMX2OS may be novel biomarkers of recurrent laryngeal cancer, and HCG4-miR-33b, HOTAIR-miR-1-MAGEA2 and EMX2OS-miR-124-CALCA/GABRG2 may be associated with the molecular mechanisms regulating recurrent laryngeal cancer.

Carbonyl reductase 1 is a new target to improve the effect of radiotherapy on head and neck squamous cell carcinoma

Background

Human carbonyl reductase 1 (CBR1) plays major roles in protecting cells against cellular damage resulting from oxidative stress. Although CBR1-mediated detoxification of oxidative materials increased by stressful conditions including hypoxia, neuronal degenerative disorders, and other circumstances generating reactive oxide is well documented, the role of CBR1 under ionising radiation (IR) is still unclear.

Methods

The formalin-fixed and paraffin-embedded tissues of 85 patients with head and neck squamous cell carcinoma (HNSCC) were used to determine if CBR1 expression effects on survival of patients with treatment of radiotherapy. Subsequently colony formation assays and xenograft tumor mouse model was used to verify the relationship between CBR1 expression and radiosensitivity in HNSCC cells. Publicly-available data from The Cancer Genome Atlas (TCGA) was analysed to determine if CBR1 expression affects the survival of patients with HNSCC. To verify CBR1-mediated molecular signalling pathways, cell survival, DNA damage/repair, reactive oxygen species (ROS), cell cycle distribution and mitotic catastrophe in HNSCC cells with modulated CBR1 expression by knockdown or overexpression were measured using by colony formation assays, flow cytometry, qRT-PCR and western blot analysis.

Results

HNSCC patients with low CBR1 had a significantly higher survival rate than the high CBR1 expression (84.2% vs. 57.8%, p = 0.0167). Furthermore, HNSCC patients with low CBR1 expression showed a good prognosis for IR compared to patients with highly expressed CBR1. Also, we found that IR upregulated CBR1 mRNA via Nrf2 activation in HNSCC cells and patients. In vitro analysis, we found that CBR1-specific siRNA or inhibitor significantly enhanced radiosensitivity after IR, while CBR1 overexpression decreased. CBR1 inhibition by siRNA or inhibitor treatment accumulated cellular ROS leading to aberrant DNA damage repair and an increase of mitotic catastrophe. Moreover, the combination of CBR1 depletion with IR dramatically inhibited primary tumour growth in a xenograft tumor mouse model.

Conclusion

Our findings indicate that CBR1 has a key role in DNA damage response through regulation of IR-mediated ROS generation. Consistently, CBR1 expression is highly correlated with patient survival after and susceptibility to radiation therapy. Therefore, CBR1 inhibition with IR might be a potent therapeutic strategy for HNSCC treatment.

Electronic supplementary material

The online version of this article (10.1186/s13046-018-0942-9) contains supplementary material, which is available to authorized users.

Gene Expression Signatures for Head and Neck Cancer Patient Stratification: Are Results Ready for Clinical Application?

OPINION STATEMENT

Head and neck squamous cell carcinoma (HNSCC) is the sixth leading cancer by incidence worldwide and considering the recent EUROCARE-5 population-based study the 5-year survival rate of HNSCC patients in Europe ranges between 69% in localized cases and 34% in patients with regional involvement. The development of high-throughput gene expression assays in the last two decades has provided the invaluable opportunity to improve our knowledge on cancer biology and to identify predictive signatures in the most deeply analyzed malignancies, such as hematological and breast cancers. At variance, till 2010, the number of reliable reports referring gene expression data related to HSNCC biology and prediction was quite limited. A critical revision of the literature reporting gene expression data in HNSCC indicated that in the last 6 years, there were new important studies with a relevant increase in the sample size and a more accurate selection of cases, the publication of a growing number of studies applying a computational integration (meta-analysis) of different microarray datasets addressing similar clinical/biological questions, the increased use of molecular sub-classification of tumors according to their gene expression, and the release of the publicly available largest dataset in HNSCC by The Cancer Genome Atlas (TCGA) consortium. Overall, also for this disease, it become evident that the expression analysis of the entire transcriptome has been enabling to achieve the identification of promising molecular signatures for (i) disclosure of the biology behind carcinogenesis with special focus on the HPV-related one, (ii) prediction of tumor recurrence or metastasis development, (iii) identification of subgroups of tumors with different biology and associated prognosis, and (iv) prediction of outcome and/or response to therapy. The increasing awareness of the relevance of strict collaboration among clinicians and translational researchers would in a near future enable the application of a personalized HNSCCs patients' treatment in the clinical practice based also on gene expression signatures.

Gene expression profiling to predict recurrence of advanced squamous cell carcinoma of the tongue: discovery and external validation

OBJECTIVES

To establish a prognostic signature for locally advanced tongue squamous cell carcinoma (TSCC) patients treated with surgery.

RESULTS

In the discovery study, unsupervised hierarchical clustering analysis identified two clusters which differentiated the Kaplan-Meier curves of RFS [median RFS, 111 days vs. not reached; log-rank test, P = 0.023]. The 30 genes identified were combined into a dichotomous PI. In the validation cohort, classification according to the PI was associated with RFS [median RFS, 754 days vs. not reached; log-rank test, P = 0.026 in GSE31056] and DSS [median DSS, 540 days vs. not reached; log-rank test, P = 0.046 in GSE42743 and 443 days vs. not reached; P < 0.001 in GSE41613]. Among genes, positive immunohistochemical staining of cytokeratin 4 was associated with favorable prognostic values for RFS (hazard ratio (HR), 0.591, P = 0.045) and DSS (HR, 0.333, P = 0.004).

MATERIALS AND METHODS

We conducted gene expression profiling of 26 clinicopathologically homogeneous advanced TSCC tissue samples using cDNA microarray as a discovery study. Candidate genes were screened using clustering analysis and univariate Cox regression analysis for relapse-free survival (RFS). These were combined into a prognostic index (PI), which was validated using three public microarray datasets of tongue and oral cancer (123 patients). Some genes identified in discovery were immunohistochemically examined for protein expression in another 127 TSCC patients.

CONCLUSION

We identified robust molecular markers that showed significant associations with prognosis in TSCC patients. Gene expression profiling data were successfully converted to protein expression profiling data.

Virus-encoded microRNA contributes to the molecular profile of EBV-positive Burkitt lymphomas

Burkitt lymphoma (BL) is an aggressive neoplasm characterized by consistent morphology and phenotype, typical clinical behavior and distinctive molecular profile. The latter is mostly driven by the MYC over-expression associated with the characteristic translocation (8;14) (q24; q32) or with variant lesions. Additional genetic events can contribute to Burkitt Lymphoma pathobiology and retain clinical significance. A pathogenetic role for Epstein-Barr virus infection in Burkitt lymphomagenesis has been suggested; however, the exact function of the virus is largely unknown.

In this study, we investigated the molecular profiles (genes and microRNAs) of Epstein-Barr virus-positive and -negative BL, to identify specific patterns relying on the differential expression and role of Epstein-Barr virus-encoded microRNAs.

First, we found significant differences in the expression of viral microRNAs and in selected target genes. Among others, we identified LIN28B, CGNL1, GCET2, MRAS, PLCD4, SEL1L, SXX1, and the tyrosine kinases encoding STK10/STK33, all provided with potential pathogenetic significance. GCET2, also validated by immunohistochemistry, appeared to be a useful marker for distinguishing EBV-positive and EBV-negative cases. Further, we provided solid evidences that the EBV-encoded microRNAs (e.g. BART6) significantly mold the transcriptional landscape of Burkitt Lymphoma clones.

In conclusion, our data indicated significant differences in the transcriptional profiles of EBV-positive and EBV-negative BL and highlight the role of virus encoded miRNA.

Brain Metastasis Prediction by Transcriptomic Profiling in Triple-Negative Breast Cancer

BACKGROUND

Triple-negative breast cancer (TNBC) lacks expression of steroid hormone receptors (estrogen receptor α and progesterone) and epidermal growth factor receptor type 2. This phenotype shows high metastatic potential, with particular predilection to lungs and brain. Determination of TNBC transcriptomic profiles associated with high risk of brain metastasis (BM) might identify patients requiring alternative, more aggressive, or specific preventive and therapeutic approaches.

PATIENTS AND METHODS

Using a cDNA-mediated annealing, selection, extension, and ligation assay, we investigated expression of 29,369 gene transcripts in primary TNBC tumor samples from 119 patients-71 in discovery cohort A and 48 in independent cohort B-that included best discriminating genes. Expression of mRNA was correlated with the occurrence of symptomatic BM.

RESULTS

In cohort A, the difference at the noncorrected P < .005 was found for 64 transcripts (P = .23 for global test), but none showed significant difference at a preset level of false-discovery rate of < 10%. Of the 30 transcripts with the largest differences between patients with and without BM in cohort A, none was significantly associated with BM in cohort B.

CONCLUSION

Analysis based on the primary tumor gene transcripts alone is unlikely to predict BM development in advanced TNBC. Despite its negative findings, the study adds to the knowledge on the biology of TNBC and paves the way for future projects using more advanced molecular assays.

Identification of key target genes and pathways in laryngeal carcinoma

The purpose of the present study was to screen the key genes associated with laryngeal carcinoma and to investigate the molecular mechanism of laryngeal carcinoma progression. The gene expression profile of GSE10935 [Gene Expression Omnibus (GEO) accession number], including 12 specimens from laryngeal papillomas and 12 specimens from normal laryngeal epithelia controls, was downloaded from the GEO database. Differentially expressed genes (DEGs) were screened in laryngeal papillomas compared with normal controls using Limma package in R language, followed by Gene Ontology (GO) enrichment analysis and pathway enrichment analysis. Furthermore, the protein-protein interaction (PPI) network of DEGs was constructed using Cytoscape software and modules were analyzed using MCODE plugin from the PPI network. Furthermore, significant biological pathway regions (sub-pathway) were identified by using iSubpathwayMiner analysis. A total of 67 DEGs were identified, including 27 up-regulated genes and 40 down-regulated genes and they were involved in different GO terms and pathways. PPI network analysis revealed that Ras association (RalGDS/AF-6) domain family member 1 (RASSF1) was a hub protein. The sub-pathway analysis identified 9 significantly enriched sub-pathways, including glycolysis/gluconeogenesis and nitrogen metabolism. Genes such as phosphoglycerate kinase 1 (PGK1), carbonic anhydrase II (CA2), and carbonic anhydrase XII (CA12) whose node degrees were >10 were identified in the disease risk sub-pathway. Genes in the sub-pathway, such as RASSF1, PGK1, CA2 and CA12 were presumed to serve critical roles in laryngeal carcinoma. The present study identified DEGs and their sub-pathways in the disease, which may serve as potential targets for treatment of laryngeal carcinoma.

Developing Robust Predictive Models for Head and Neck Cancer across Microarray and RNA-seq Data

Increased understanding of the transcriptomic patterns underlying head and neck squamous cell carcinoma (HNSCC) can facilitate earlier diagnosis and better treatment outcomes. Integrating knowledge from multiple studies is necessary to identify fundamental, consistent gene expression signatures that distinguish HNSCC patient samples from disease-free samples, and particularly for detecting HNSCC at an early pathological stage. This study utilizes feature integration and heterogeneous ensemble modeling techniques to develop robust models for predicting HNSCC disease status in both microarray and RNAseq datasets. Several alternative models demonstrated good performance, with MCC and AUC values exceeding 0.8. These models were also applied to discriminate between early pathological stage HNSCC and normal RNA-seq samples, showing encouraging results. The predictive modeling workflow was integrated into a software tool with a graphical user interface. This tool enables HNSCC researchers to harness frequently observed transcriptomic features and ensembles of previously developed models when investigating new HNSCC gene expression datasets.

Use of formalin-fixed paraffin-embedded samples for gene expression studies in breast cancer patients

To obtain gene expression profiles from samples collected in clinical trials, we conducted a pilot study to assess feasibility and estimate sample attrition rates when profiling formalin-fixed, paraffin-embedded specimens. Ten matched fresh-frozen and fixed breast cancer samples were profiled using the Illumina HT-12 and Ref-8 chips, respectively. The profiles obtained with Ref 8, were neither technically nor biologically reliable since they failed to yield the expected separation between estrogen receptor positive and negative samples. With the use of Affymetrix HG-U133 2.0 Plus chips on fixed samples and a quantitative polymerase chain reaction -based sample pre-assessment step, results were satisfactory in terms of biological reliability, despite the low number of present calls (M = 21%±5). Compared with the Illumina DASL WG platform, Affymetrix data showed a wider interquartile range (1.32 vs 0.57, P<2.2 E-16,) and larger fold changes. The Affymetrix chips were used to run a pilot study on 60 fixed breast cancers. By including in the workflow the sample pre-assessment steps, 96% of the samples predicted to give good results (44/46), were in fact rated as satisfactory from the point of view of technical and biological meaningfulness. Our gene expression profiles showed strong agreement with immunohistochemistry data, were able to reproduce breast cancer molecular subtypes, and allowed the validation of an estrogen receptor status classifier derived in frozen samples. The approach is therefore suitable to profile formalin-fixed paraffin-embedded samples collected in clinical trials, provided that quality controls are run both before (sample pre-assessment) and after hybridization on the array.

Prognostic significance of the Wnt pathway in squamous cell laryngeal cancer

OBJECTIVES

We sought to determine the prognostic significance of the Wnt signaling pathway in operable squamous cell carcinoma of the larynx.

MATERIALS AND METHODS

In an annotated cohort of 289 operable laryngeal cancers we evaluated the prognostic impact of E-cadherin, P-cadherin and β-catenin protein expression with immunohistochemistry, as well as the mRNA expression of 7 key effectors of the Wnt pathway including secreted frizzled-related protein 4 (SFRP4), SNAI2 (SLUG) and WNT5A with qPCR (relative quantification [RQ]).

RESULTS

Using median immunoreactive scores as a pre-defined cut-off, patients whose tumors overexpressed both cytoplasmic E-cadherin and β-catenin experienced longer median OS as compared to those whose tumors overexpressed β-catenin only (median OS 124 vs. 72 months, p=0.0301) and patients whose tumors overexpressed both cytoplasmic and membranous E-cadherin experienced longer DFS as compared to those whose tumors overexpressed cytoplasmic E-cadherin only (median 118 vs. 91 months, p=0.0106). Upon hierarchical clustering of SFRP4, SNAI2 and WNT5A RQ values, profiles including co-expression of all 3 genes but also profiles with under-expression of SNAI2 and WNT5A were associated with worse outcome as compared to profiles not related to the Wnt pathway. In multivariate analysis, clustering was an independent predictor for DFS (p=0.0221) and OS (p=0.0077).

CONCLUSION

We identified gene expression profiles and IHC patterns associated with aberrant Wnt signaling conferring aggressive clinical behavior in operable squamous cell carcinoma of the larynx. Prospective validation of these results will determine whether targeting the Wnt pathway merits investigation in this disease.

Identification and validation of a multigene predictor of recurrence in primary laryngeal cancer

Purpose

Local recurrence is the major manifestation of treatment failure in patients with operable laryngeal carcinoma. Established clinicopathological factors cannot sufficiently predict patients that are likely to recur after treatment. Additional tools are therefore required to accurately identify patients at high risk for recurrence. This study attempts to identify and independently validate gene expression models, prognostic of disease-free survival (DFS) in operable laryngeal cancer.

Materials and Methods

Using Affymetrix U133A Genechips, we profiled fresh-frozen tumor tissues from 66 patients with laryngeal cancer treated locally with surgery. We applied Cox regression proportional hazards modeling to identify multigene predictors of recurrence. Gene models were then validated in two independent cohorts of 54 and 187 patients (fresh-frozen and formalin-fixed tissue validation sets, respectively).

Results

We focused on genes univariately associated with DFS (p<0.01) in the training set. Among several models comprising different numbers of genes, a 30-probe set model demonstrated optimal performance in both the training (log-rank, p<0.001) and 1^st validation (p = 0.010) sets. Specifically, in the 1^st validation set, median DFS as predicted by the 30-probe set model, was 34 and 80 months for high- and low-risk patients, respectively. Hazard ratio (HR) for recurrence in the high-risk group was 3.87 (95% CI 1.28–11.73, Wald's p = 0.017). Testing the expression of selected genes from the above model in the 2^nd validation set, with qPCR, revealed significant associations of single markers, such as ACE2, FLOT1 and PRKD1, with patient DFS. High PRKD1 remained an unfavorable prognostic marker upon multivariate analysis (HR = 2.00, 95% CI 1.28–3.14, p = 0.002) along with positive nodal status.

Conclusions

We have established and validated gene models that can successfully stratify patients with laryngeal cancer, based on their risk for recurrence. It seems worthy to prospectively validate PRKD1 expression as a laryngeal cancer prognostic marker, for routine clinical applications.

Cyclin D1, EGFR, and Akt/mTOR pathway. Potential prognostic markers in localized laryngeal squamous cell carcinoma

INTRODUCTION

EGFR (epidermal growth factor receptor), cyclin D1 and Akt/mTOR pathways are active in head and neck cancer. The aim of this study was to explore biomarker expression, their correlations with clinicopathological parameters and their prognostic utility in a cohort of patients with localized squamous laryngeal carcinoma.

PATIENTS AND METHODS

We assessed relative messenger RNA expression of EGFR, Akt1, 2, and 3, mTOR and CCND1, copy number variants of the EGFR and CCND1 genes and immunohistochemical protein expression of EGFR, p-Akt308, p-Akt473, pmTOR, PTEN, p53 and cyclin D1 in paraffin-embedded tissue samples of localized laryngeal carcinomas.

RESULTS

In 289 patients with T3-4 (77.8%), node-negative (84.1%) tumors of the larynx, high EGFR and CCND1 mRNA correlated with no or ex-smoking, (p = 0.003 and p = 0.029, respectively), while low Akt3 mRNA correlated with alcohol abuse, N0 stage, total laryngectomy, and absence of neck dissection. At a median follow-up of 74.5 months, high mTOR mRNA expression was marginally associated with shorter disease-free survival (hazard ratio [HR] = 1.54; p = 0.093) and high Akt3 mRNA with shorter overall survival (HR = 1.49; p = 0.0786), in univariate analysis. In multivariate analysis, node-positive status, subglottic-transglottic location, surgery other than total laryngectomy and mTOR/CCND1 mRNA interaction with a hazard ratio of 2.16 (p value for interaction:  0.0010) were independent predictors of relapse, while node-positive status and subglottic-transglottic location were associated with higher risk for death.

CONCLUSION

In localized laryngeal cancer, clinicopathological parameters and an interaction of high mTOR and CCND1 mRNA expression were found to be associated with poor patient outcome.

Quality assessment metrics for whole genome gene expression profiling of paraffin embedded samples

BACKGROUND

Formalin fixed, paraffin embedded tissues are most commonly used for routine pathology analysis and for long term tissue preservation in the clinical setting. Many institutions have large archives of Formalin fixed, paraffin embedded tissues that provide a unique opportunity for understanding genomic signatures of disease. However, genome-wide expression profiling of Formalin fixed, paraffin embedded samples have been challenging due to RNA degradation. Because of the significant heterogeneity in tissue quality, normalization and analysis of these data presents particular challenges. The distribution of intensity values from archival tissues are inherently noisy and skewed due to differential sample degradation raising two primary concerns; whether a highly skewed array will unduly influence initial normalization of the data and whether outlier arrays can be reliably identified.

FINDINGS

Two simple extensions of common regression diagnostic measures are introduced that measure the stress an array undergoes during normalization and how much a given array deviates from the remaining arrays post-normalization. These metrics are applied to a study involving 1618 formalin-fixed, paraffin-embedded HER2-positive breast cancer samples from the N9831 adjuvant trial processed with Illumina's cDNA-mediated Annealing Selection extension and Ligation assay.

CONCLUSION

Proper assessment of array quality within a research study is crucial for controlling unwanted variability in the data. The metrics proposed in this paper have direct biological interpretations and can be used to identify arrays that should either be removed from analysis all together or down-weighted to reduce their influence in downstream analyses.

Insulin-like growth factor 1 receptor (IGF1R) expression and survival in operable squamous-cell laryngeal cancer

INTRODUCTION

Prognosis of patients with operable laryngeal cancer is highly variable and therefore potent prognostic biomarkers are warranted. The insulin-like growth factor receptor (IGFR) signaling pathway plays a critical role in laryngeal carcinogenesis and progression.

PATIENTS AND METHODS

We identified all patients with localized TNM stage I-III laryngeal cancer managed with potentially curative surgery between 1985 and 2008. Immunohistochemical (IHC) expression of IGF1R-alpha, IGF1R-beta and IGF2R was evaluated using the immunoreactive score (IRS) and mRNA levels of important effectors of the IGFR pathway were assessed, including IGF1R, IGF-binding protein 3 (IGFBP3), suppressor of cytokine signaling 2 (SOCS2) and members of the MAP-kinase (MAP2K1, MAPK9) and phosphatidyl-inositol-3 kinase (PIK3CA, PIK3R1) families. Cox-regression models were applied to assess the predictive value of biomarkers on disease-free survival (DFS) and overall survival (OS).

RESULTS

Among 289 eligible patients, 95.2% were current or ex smokers, 75.4% were alcohol abusers, 15.6% had node-positive disease and 32.2% had received post-operative irradiation. After a median follow-up of 74.5 months, median DFS was 94.5 months and median OS was 106.3 months. Using the median IRS as the pre-defined cut-off, patients whose tumors had increased IGF1R-alpha cytoplasm or membrane expression experienced marginally shorter DFS and significantly shorter OS compared to those whose tumors had low IGF1R-alpha expression (91.1 vs 106.2 months, p = 0.0538 and 100.3 vs 118.6 months, p = 0.0157, respectively). Increased mRNA levels of MAPK9 were associated with prolonged DFS (p = 0.0655) and OS (p = 0.0344). In multivariate analysis, IGF1R-alpha overexpression was associated with a 46.6% increase in the probability for relapse (p = 0.0374). Independent predictors for poor OS included node-positive disease (HR = 2.569, p<0.0001), subglottic/transglottic localization (HR = 1.756, p = 0.0438) and IGF1R-alpha protein overexpression (HR = 1.475, p = 0.0504).

CONCLUSION

IGF1R-alpha protein overexpression may serve as an independent predictor of relapse and survival in operable laryngeal cancer. Prospective evaluation of the IGF1R-alpha prognostic utility is warranted.