Validating the prognostic value of marker genes derived from a non-small cell lung cancer microarray study

Identification and structural prediction of the unrevealed amidohydrolase enzyme: Pterin deaminase from Agrobacterium tumefaciens LBA4404

Microbes make a remarkable contribution to the health and well-being of living beings all over the world. Interestingly, pterin deaminase is an amidohydrolase enzyme that exhibits antitumor, anticancer activities and antioxidant properties. With the existing evidence of the presence of pterin deaminase from microbial sources, an attempt was made to reveal the existence of this enzyme in the unexplored bacterium Agrobacterium tumefaciens LBA4404. After, the cells were harvested and characterized as intracellular enzymes and then partially purified through acetone precipitation. Subsequently, further purification step was carried out with an ion-exchange chromatogram (HiTrap Q FF) using the Fast-Protein Liquid Chromatography technique (FPLC). Henceforward, the approximate molecular weight of the purified pterin deaminase was determined through SDS-PAGE. Furthermore, the purified protein was identified accurately by MALDI-TOF, and the sequence was explored through a Mascot search engine. Additionally, the three-dimensional structure was predicted and then validated, as well as ligand-binding sites, and the stability of this enzyme was confirmed for the first time. Thus, the present study revealed the selected parameters showing a considerable impact on the identification and purification of pterin deaminase from A. tumefaciens LBA4404 for the first time. The enzyme specificity makes it a favorable choice as a potent anticancer agent.

Tumor-secreted GRP78 facilitates the migration of macrophages into tumors by promoting cytoskeleton remodeling

Glucose-regulated protein 78 (GRP78), an important molecular chaperone in the endoplasmic reticulum, is often over-expressed in the central region of advanced tumor and acts as a promoter of tumor progression. As main immune cells in the tumor microenvironment, infiltration of abundant macrophages into advanced tumor further facilitates growth of tumor. Although has potential association between GRP78 and infiltration of macrophages, its underlying mechanisms are poorly understood. Here, we report that secreted GRP78 facilitates recruitment of macrophages into tumors both in vitro and in vivo. Further studies reveal that secreted GRP78 transports into macrophages and bound to intracellular Ca2+, which lead to uneven distribution of Ca2+ and subsequent polarization of macrophages. The polarization of macrophages activates expression of microRNA-200b-3p. By directly targeting RhoGDI, miR-200b-3p stimulates the activity of RhoGTPase and ultimately leads to the distribution of GTP-Rac1 and GTP-Cdc42 in front protrusion and GTP-RhoA in rear contraction, which further results in migration of macrophages in a certain direction. Our results reveal a novel function of GRP78 to promote the recruitment of macrophages to tumor and provide a potential therapeutic target for malignancies.

Citron Rho-interacting kinase silencing causes cytokinesis failure and reduces tumor growth in multiple myeloma

Citron Rho-interacting serine/threonine kinase (CIT) is a serine/threonine kinase that acts as a key component of the midbody and is essential for cytokinesis. CIT has been reported to be highly expressed in some tumor tissues and to play a role in cancer proliferation; however, the significance of CIT has not been investigated in multiple myeloma (MM). Here, we identified, by protein microarray and immunohistochemistry, that CIT is 1 of the upregulated proteins in the plasma cells of MM patients compared with healthy controls. Analysis of a gene expression profile data set showed that MM patients with high CIT gene expression had significantly worse overall survival compared with MM patients with low CIT gene expression. CIT silencing in MM cell lines induced cytokinesis failure and resulted in decreased MM cell proliferation in vitro and in vivo. TP53 expression was found to be an independent predictor of CIT dependency, with low-TP53 cell lines exhibiting a strong dependency on CIT. This study provides the rationale for CIT being a potential therapeutic target in MM in future trials.

RhoGDI deficiency induces constitutive activation of Rho GTPases and COX-2 pathways in association with breast cancer progression

Rho GDP Dissociation Inhibitor (RhoGDI) is a key regulator of Rho GTPases. Here we report that loss of RhoGDI significantly accelerated xenograft tumor growth of MDA-MB-231 cells in animal models. At the molecular level, RhoGDI depletion resulted in constitutive activation of Rho GTPases, including RhoA, Cdc42, and Rac1. This was accompanied by Rho GTPase translocation from the cytosol to membrane compartments. Notably, COX-2 protein levels, mRNA expression, and biological activity were markedly increased in RhoGDI-deficient cells. The upregulated expression of COX-2 was directly associated with increased Rho GTPase activity. Further, we assessed the expression level of RhoGDI protein in breast tumor specimens (n = 165) by immunohistochemistry. We found that RhoGDI expression is higher in the early stages of breast cancer followed by a significant decrease in malignant tumors and metastatic lesions (p < 0.01). These data suggest that downregulation of RhoGDI could be a critical mechanism of breast tumor development, which may involve the hyperactivation of Rho GTPases and upregulation of COX-2 activity. Additional studies are warranted to evaluate the therapeutic potential of inhibiting Rho GTPases and COX-2 for treating breast cancers.

Prognosis value of HIF-1α expression in patients with non-small cell lung cancer

BACKGROUND

Hypoxia-inducible factor 1 (HIF-1) is a transcription factor that plays a critical role in the development and progression of tumors. Various studies evaluating the prognostic value of HIF-1α in patients with lung cancer (LC) remain controversial. To comprehensively and quantitatively summarize the evidence on the effect of HIF-1α expression on the survival of patients with LC, a meta-analysis was carried out.

MATERIAL AND METHODS

Electronic databases were used to identify published studies before August 31st, 2013. Studies were assessed for quality using REMARK. Data were collected comparing overall survival in patients with high HIF-1α expression with those with low expression.

RESULTS

Totally, 13 papers including 1420 patients were subjected to final analysis. The combined hazard ratio (HR) was 1.60 (95% CI: 1.14-2.25, P=0.007), suggesting that high expression of HIF-1α was an indicator of poor prognosis. Further, when stratified by LC histological type (SCLC and NSCLC), study region (Asia and Europe), cut-off values (10%), tumor stage (I-III and I-IV), antibody for IHC (H1α67 and ESEE 122), and HR estimated method (univariate/multivariate analysis), most of the results were statistically significant.

CONCLUSIONS

Taken together, this meta-analysis revealed that HIF-1α overexpression might be a predicative factor of poor prognosis for NSCLC particularly in Asia.

Suppression of KIF14 expression inhibits hepatocellular carcinoma progression and predicts favorable outcome

The mitotic kinesin superfamily protein KIF14 is essential for cytokinesis and chromosome segregation and increased KIF14 expression is related to a variety of human cancers. In this study, we investigate KIF14 expression in association with clinical variables and the role of KIF14 during tumorigenesis. We found that KIF14 is overexpressed in most primary hepatocellular carcinoma (HCC) tissues compared with the adjacent normal liver tissues and KIF14 overexpression is associated with tumor grade (P = 0.002), stage (P = 0.013) and poor survival (P < 0.001). Downregulation of KIF14 decreased the capacity of proliferation both in vitro and in vivo. Furthermore, suppression of KIF14 not only decreases cancer cell migration but also induces apoptosis of cells with inactivation of the phosphatidylinositol 3-kinase-Akt signaling pathway. Therefore, our current study indicates that KIF14 promotes HCC carcinogenesis and may serve as a potential therapeutic target for human HCC.

False-positivity of mediastinal lymph nodes has negative effect on survival in potentially resectable non-small cell lung cancer

OBJECTIVES

It has been shown that increased metabolic activity of primary tumour has a negative effect on survival in non-small cell lung cancer (NSCLC) staged with positron emission tomography integrated computed tomography (PET/CT). We hypothesized that an increased metabolic activity of mediastinal lymph nodes would have worse survival even if it is false.

METHODS

Three hundred and twenty-eight consecutive patients with NSCLC histology were imaged with PET/CT within 90 days of surgery between September 2005 and March 2009. Patients who had neoadjuvant chemotherapy (n = 22), patients with prior history of NSCLC (n = 9) or other malignancies within 5 years (n = 11) were excluded from the study. Patients with negative mediastinoscopy underwent resection. Pathological results were revised according to the seventh tumor-node-metastasis staging system. Kaplan-Meier test was used for survival. Log-rank and Cox analyses were used for comparisons.

RESULTS

A total of 286 patients (262 male; mean age: 58.5 years) were evaluated. There were 22 (6.7%) operative deaths and none of the patients were lost to follow-up. The median follow-up in the remaining 264 patients was 26 months (range, 2-61 months). Tumour size, nodal spread and stage were all strongly associated with survival from NSCLC (P < 0.001). There were 63 true-positive, 65 false-positive (FP), 152 true-negative (TN) and six false-negative findings on mediastinal staging after PET/CT. The maximum standardized uptake value of primary tumour was significantly higher in FP patients than in TN patients (P = 0.012). After excluding pN2-positive patients, TN patients had better survival than FP patients (P = 0.006). Multivariate analysis showed that false-positivity of mediastinal lymph nodes was independently associated with worse survival (hazard ratio = 0.63; P = 0.02). There were 146 patients with pT1-4, pN0 treated with R0 surgical resection. Disease-free survival and overall survival were also significantly better for TN patients in completely resected group (P = 0.009 versus 0.016).

CONCLUSIONS

We have shown that false-positivity of mediastinal lymph nodes had yielded worse survival in surgically staged or resected NSCLC patients staged with PET/CT. This result may help to allocate patients with potentially poor prognosis for considered additional therapy.

Targeting S100P inhibits colon cancer growth and metastasis by Lentivirus-mediated RNA interference and proteomic analysis

S100P was recently found to be overexpressed in a variety of cancers and is considered a potential target for cancer therapy, but the functional role or mechanism of action of S100P in colon cancer is not fully understood. In the present study, we knocked down the gene expression of S100P in colon cancer cells using lentivirus-mediated RNA interference. This step resulted in significant inhibition of cancer cell growth, migration and invasion in vitro and tumor growth and liver metastasis in vivo. Moreover, S100P downstream target proteins were identified by proteomic analysis in colon cancer DLD-1 cells with deletion of S100P. Knockdown of S100P led to downregulation of thioredoxin 1 and β-tubulin and upregulation of Rho guanosine diphosphate (GDP) dissociation inhibitor α (RhoGDIA), all potential therapeutic targets in cancer. Taken together, these findings suggest that S100P plays an important role in colon tumorigenesis and metastasis, and the comprehensive and comparative analyses of proteins associated with S100P could contribute to understanding the downstream signal cascade of S100P, leading to tumorigenesis and metastasis.

Prognostic implications of hypoxia-inducible factor-1α in epidermal growth factor receptor-negative non-small cell lung cancer

Hypoxia-inducible factor (HIF)-1α is a regulatory subunit of HIF-1 that is stabilized and activated under hypoxic conditions. In non-small cell lung cancer (NSCLC), over-expression of HIF-1α has been associated with poor overall survival. However, there is conflicting data on the role of HIF-1α as a prognostic factor. Some studies have demonstrated close association between the HIF-1α signal pathways and epidermal growth factor receptors (EGFRs). We evaluated the prognostic significance of HIF-1α expression in 178 NSCLC patients using tissue microarray in the context of EGFR gene copy number and protein expression status. EGFR gene copy number was evaluated using fluorescent in situ hybridization (FISH), and EGFR protein expression was determined using immunohistochemistry (IHC). The difference in overall survival (OS) between HIF-1α-positive and HIF-1α-negative groups was statistically significant in patients with low EGFR gene copy number and negative EGFR expression (log-rank test, P = 0.03). In univariate and multivariate analyses, HIF-1α was a significant worse prognostic factor for OS in patients with low EGFR gene copy number and negative EGFR expression (hazard ratio = 2.992; 95% CI, 1.113-8.045; P = 0.03 in univariate analysis and hazard ratio = 8.127; 95% CI, 1.874-35.251; P < 0.01 in multivariate analysis). The results demonstrated that the prognostic significance of HIF-1α should be validated in the context of EGFR status in NSCLC patients, and the gene and protein status of EGFR and HIF-1α will be important to help select patients most likely to derive the greatest clinical benefit from EGFR or HIF-1α targeted therapies.

Reinventing diagnostics for personalized therapy in oncology

Human cancers are still diagnosed and classified using the light microscope. The criteria are based upon morphologic observations by pathologists and tend to be subject to interobserver variation. In preoperative biopsies of non-small cell lung cancers, the diagnostic concordance, even amongst experienced pulmonary pathologists, is no better than a coin-toss. Only 25% of cancer patients, on average, benefit from therapy as most therapies do not account for individual factors that influence response or outcome. Unsuccessful first line therapy costs Canada CAN$1.2 billion for the top 14 cancer types, and this extrapolates to $90 billion globally. The availability of accurate drug selection for personalized therapy could better allocate these precious resources to the right therapies. This wasteful situation is beginning to change with the completion of the human genome sequencing project and with the increasing availability of targeted therapies. Both factors are giving rise to attempts to correlate tumor characteristics and response to specific adjuvant and neoadjuvant therapies. Static cancer classification and grading systems need to be replaced by functional classification systems that not only account for intra- and inter- tumor heterogeneity, but which also allow for the selection of the correct chemotherapeutic compounds for the individual patient. In this review, the examples of lung and breast cancer are used to illustrate the issues to be addressed in the coming years, as well as the emerging technologies that have great promise in enabling personalized therapy.

Understanding prognostic gene expression signatures in lung cancer

In non-small-cell lung cancer (NSCLC), molecular profiling of tumors has led to the identification of gene expression patterns that are associated with specific phenotypes and prognosis. Such correlations could identify early-stage patients who are at increased risk of disease recurrence and death after complete surgical resection and who might benefit from adjuvant therapy. Profiling may also identify aberrant molecular pathways that might lead to specific molecularly targeted therapies. The technology behind the capturing and correlating of molecular profiles with clinical and biologic endpoints have evolved rapidly since microarrays were first developed a decade ago. In this review, we discuss multiple methods that have been used to derive prognostic gene expression signatures in NSCLC. Despite the diversity in the approaches used, 3 main steps are followed. First, the expression levels of several hundred to tens of thousands of genes are quantified by microarray or quantitative polymerase chain reaction techniques; the data are then preprocessed, normalized, and possibly filtered. In the second step, expression data are combined and grouped by clustering, risk score generation, or other means, to generate a gene signature that correlates with a clinical outcome, usually survival. Finally, the signature is validated in datasets of independent cohorts. This review discusses the concepts and methodologies involved in these analytical steps, primarily to facilitate the understanding of reports on large dataset gene expression studies that focus on prognostic signatures in NSCLC.

Gain of miR-151 on chromosome 8q24.3 facilitates tumour cell migration and spreading through downregulating RhoGDIA

Recurrent chromosomal aberrations are often observed in hepatocellular carcinoma (HCC), but little is known about the functional non-coding sequences, particularly microRNAs (miRNAs), at the chromosomal breakpoints in HCC. Here we show that 22 miRNAs are often amplified or deleted in HCC. MicroRNA-151 (miR-151), a frequently amplified miRNA on 8q24.3, is correlated with intrahepatic metastasis of HCC. We further show that miR-151, which is often expressed together with its host gene FAK, encoding focal adhesion kinase, significantly increases HCC cell migration and invasion in vitro and in vivo, mainly through miR-151-5p, but not through miR-151-3p. Moreover, miR-151 exerts this function by directly targeting RhoGDIA, a putative metastasis suppressor in HCC, thus leading to the activation of Rac1, Cdc42 and Rho GTPases. In addition, miR-151 can function synergistically with FAK to enhance HCC cell motility and spreading. Thus, our findings indicate that chromosome gain of miR-151 is a crucial stimulus for tumour invasion and metastasis of HCC.

Three immunomarker support vector machines-based prognostic classifiers for stage IB non-small-cell lung cancer

PURPOSE

Approximately 30% of patients with stage IB non-small-cell lung cancer (NSCLC) die within 5 years after surgery. Current staging methods are inadequate for predicting the prognosis of this particular subgroup. This study identifies prognostic markers for NSCLC.

PATIENTS AND METHODS

We used computer-generated random numbers to study 148 paraffin-embedded specimens for immunohistochemical analysis. We studied gene expression in paraffin-embedded specimens of lung cancer tissue from 73 randomly selected patients with stage IB NSCLC who had undergone radical surgical resection and evaluated the association between the level of expression and survival. We used support vector machines (SVM)-based methods to develop three immunomarker-SVM-based prognostic classifiers for stage IB NSCLC. For validation, we used randomly assigned specimens from 75 other patients.

RESULTS

We devised three immunomarker-SVM-based prognostic classifiers, including SVM1, SVM2, and SVM3, to refine prognosis of stage IB NSCLC successfully. The SVM1 model integrates age, cancer cell type, and five markers, including CD34MVD, EMA, p21ras, p21WAF1, and tissue inhibitors of metalloproteinases (TIMP) -2. The SVM2 model integrates age, cancer cell type, and 19 markers, including BCL2, caspase-9, CD34MVD, low-molecular-weight cytokeratin, high-molecular-weight cytokeratin, cyclo-oxygenase-2, EMA, HER2, matrix metalloproteinases (MMP) -2, MMP-9, p16, p21ras, p21WAF1, p27kip1, p53, TIMP-1, TIMP-2, vascular endothelial growth factor (VEGF), and beta-catenin. The SVM3 model consists of SVM1 and SVM2. The three models were independent predictors of overall survival. We validated the classifiers with data from an independent cohort of 75 patients with stage IB NSCLC.

CONCLUSION

The three immunomarker-SVM-based prognostic characteristics are closely associated with overall survival among patients with stage IB NSCLC.

Lung cancer: developmental networks gone awry?

High-throughput genomic data for both lung development and lung cancer continue to accumulate. Significant molecular intersection between these two processes has been hypothesized due to overlap in phenotypes and genomic variation. Examining the network biology of both cancer and development of the lung may shed functional light on the individual signaling modules involved. Stem cell biology may explain a portion of this network intersection and consequently studying lung organogenesis may have relevance for understanding lung cancer. This review summarizes our understanding of the potential overlapping mechanisms involved in lung development and lung tumorigenesis.

Prediction of recurrence-free survival in postoperative non-small cell lung cancer patients by using an integrated model of clinical information and gene expression

PURPOSE

One of the main challenges of lung cancer research is identifying patients at high risk for recurrence after surgical resection. Simple, accurate, and reproducible methods of evaluating individual risks of recurrence are needed.

EXPERIMENTAL DESIGN

Based on a combined analysis of time-to-recurrence data, censoring information, and microarray data from a set of 138 patients, we selected statistically significant genes thought to be predictive of disease recurrence. The number of genes was further reduced by eliminating those whose expression levels were not reproducible by real-time quantitative PCR. Within these variables, a recurrence prediction model was constructed using Cox proportional hazard regression and validated via two independent cohorts (n = 56 and n = 59).

RESULTS

After performing a log-rank test of the microarray data and successively selecting genes based on real-time quantitative PCR analysis, the most significant 18 genes had P values of <0.05. After subsequent stepwise variable selection based on gene expression information and clinical variables, the recurrence prediction model consisted of six genes (CALB1, MMP7, SLC1A7, GSTA1, CCL19, and IFI44). Two pathologic variables, pStage and cellular differentiation, were developed. Validation by two independent cohorts confirmed that the proposed model is significantly accurate (P = 0.0314 and 0.0305, respectively). The predicted median recurrence-free survival times for each patient correlated well with the actual data.

CONCLUSIONS

We have developed an accurate, technically simple, and reproducible method for predicting individual recurrence risks. This model would potentially be useful in developing customized strategies for managing lung cancer.

Lung adenocarcinoma: modification of the 2004 WHO mixed subtype to include the major histologic subtype suggests correlations between papillary and micropapillary adenocarcinoma subtypes, EGFR mutations and gene expression analysis

The histologic heterogeneity of lung adenocarcinoma creates a variety of complex challenges to pathologists in analyzing the various subtypes. Current classification schemas do not take into account the major subtype. We analyzed 100 cases for clinical, pathologic, and molecular features using a modification of the 2004 World Health Organization (WHO) classification to record the major component in the mixed subtype tumors. The tumors were analyzed for KRAS mutation and epidermal growth factor receptor (EGFR) by mutation, chromogenic in situ hybridization, and immunohistochemistry for EGFR and phosphorylated EGFR. Gene expression analysis was performed using HG-U133A Affymetrix oligonucleotide microarrays revealing 3 gene clusters. The most common major histologic subtype was papillary (37%) followed by acinar (30%), solid (25%) and bronchioloalveolar (7%) carcinoma (BAC), although no pure BACs were seen. Sixteen tumors harbored EGFR mutations and 14 KRAS mutations. Papillary adenocarcinoma strongly correlated with EGFR mutation (P<0.001) and gene profile Cluster 1 (P=0.006) with weaker correlations with low grade (P=0.038) and favorable behavior in Stage 1 patients (P=0.047). Micropapillary subtype correlated strongly with EGFR mutation (P<0.001) and weakly with Cluster 1 (P=0.030). Solid adenocarcinoma strongly correlated with gene profile Cluster 3 (P=0.001) and worse survival (P=0.001). BAC as a major component strongly correlated with gene Cluster 2 (P=0.001). Cluster 1 correlated strongly with less smoking (P<0.001), papillary histology (P<0.001), and EGFR mutations (P<0.001). Cluster 3 strongly correlated with heavier smoking (P<0.001), larger tumor size (P<0.001), solid subtype (P<0.001), and poor grade (P=0.004); weak correlations were found with KRAS mutation (P=0.025). No correlation was found with BAC and EGFR by mutation, chromogenic in situ hybridization or immunohistochemistry. Higher stage (P<0.001), grade (P<0.001), and solid subtype (P=0.001) correlated with shorter survival. Our data suggest that EGFR mutations are associated with papillary adenocarcinoma and gene profile Cluster 1. We discovered this only because we used a comprehensive approach examining in detail all histologic subtypes and we modified the 2004 WHO mixed subtype to include the major histologic subtype. As we do not know the major genetic pathways of 30% to 70% of lung adenocarcinomas, the comprehensive histologic subtyping we propose gives advantage for recognition of unanticipated histologic-genetic correlations that might not be detected using classification systems that focus primarily on specific aspects of adenocarcinomas such as BAC or EGFR mutations. Such an approach should be considered in future studies for validation in other datasets.

Pituitary tumor-transforming 1 increases cell motility and promotes lymph node metastasis in esophageal squamous cell carcinoma

Human pituitary tumor-transforming 1 (PTTG1)/securin is a putative oncoprotein that is overexpressed in various tumor types. However, the involvement of PTTG1 in gastrointestinal cancer development and progression remains unclear. In this study, we investigated the clinical significance and biological effects of PTTG1 in esophageal squamous cell carcinoma (ESCC). Immunohistochemical studies performed on 113 primary ESCC specimens revealed a high prevalence of PTTG1 overexpression (60.2%), which was significantly associated with lymph node metastasis (regional, P = 0.042; distant, P = 0.005), advanced tumor stage (P = 0.028), and poorer overall survival (P = 0.017, log-rank test; P = 0.044, Cox proportional hazard model). Eleven ESCC cell lines expressed PTTG1 protein at levels 2.4 to 6.6 times higher than those in normal esophageal epithelial cells (HEEpiC). PTTG1 protein expression was confined to the nucleus in HEEpiC cells but present in both the cytoplasm and nucleus in ESCC cells. Two small interfering RNAs (siRNA) inhibited PTTG1 mRNA and protein expression in three ESCC cell lines by 77% to 97%. In addition, PTTG1 down-regulation by these siRNAs significantly reduced cell motility in all three ESCC cell lines (P < 0.01) in vitro, as well as popliteal lymph node metastases of ESCC cells in nude mice (P = 0.020). Global gene expression profiling suggested that several members of the Ras and Rho gene families, including RRAS, RHOG, ARHGAP1, and ARHGADIA, represented potential downstream genes in the PTTG1 pathway. Taken together, these findings suggest that PTTG1 overexpression promotes cell motility and lymph node metastasis in ESCC patients, leading to poorer survival. Thus, PTTG1 constitutes a potential biomarker and therapeutic target in ESCCs with lymph node metastases.

Three-gene prognostic classifier for early-stage non small-cell lung cancer

PURPOSE

Several microarray studies have reported gene expression signatures that classify non-small-cell lung carcinoma (NSCLC) patients into different prognostic groups. However, the prognostic gene lists reported to date overlap poorly across studies, and few have been validated independently using more quantitative assay methods.

PATIENTS AND METHODS

The expression of 158 putative prognostic genes identified in previous microarray studies was analyzed by reverse transcription quantitative polymerase chain reaction in the tumors of 147 NSCLC patients. Concordance indices and risk scores were used to identify a stage-independent set of genes that could classify patients with significantly different prognoses.

RESULTS

We have identified a three-gene classifier (STX1A, HIF1A, and CCR7) for overall survival (hazard ratio = 3.8; 95% CI, 1.7 to 8.2; P < .001). The classifier was also able to stratify stage I and II patients and further improved the predictive ability of clinical factors such as histology and tumor stage. The predictive value of this three-gene classifier was validated in two large independent microarray data sets from Harvard and Duke Universities.

CONCLUSION

We have identified a new three-gene classifier that is independent of and improves on stage to stratify early-stage NSCLC patients with significantly different prognoses. This classifier may be tested further for its potential value to improve the selection of resected NSCLC patients in adjuvant therapy.

KIF14 messenger RNA expression is independently prognostic for outcome in lung cancer

PURPOSE

The mitotic kinesin KIF14 is overexpressed in multiple cancers including lung cancer. Therefore, we investigated KIF14 expression in association with clinical variables and the effect of KIF14 on in vitro colony formation in non-small-cell lung carcinoma.

EXPERIMENTAL DESIGN

RNA was extracted from 129 untreated, resected tumors and KIF14 expression was quantified by real-time reverse transcription-PCR. Associations with clinical variables were determined by standard statistical methods. KIF14 expression was knocked down by small interfering RNA in H1299 and HeLa cells; proliferation and growth in soft agar were assayed.

RESULTS

Squamous cell carcinoma had the highest KIF14 level, followed by large-cell undifferentiated carcinoma, then adenocarcinoma (P = 0.002). KIF14 level decreased with differentiation (P = 0.01) but was not associated with pathologic stage, T or N stage, or sex. When dichotomized about the median, KIF14 overexpression significantly decreased disease-free survival (Kaplan-Meier log-rank, P = 0.01) and trended toward decreasing overall survival (P = 0.08). In a univariate Cox proportional hazard regression, increasing KIF14 expression decreased disease-free survival [P = 0.01; hazard ratio, 1.44 (95% confidence interval, 1.09-1.91)]. In a multivariate Cox regression, including stage, differentiation, histology, and tumor purity as covariates, KIF14 overexpression remained an independent prognostic factor for disease-free survival [P = 0.01; hazard ratio, 1.45 (95% confidence interval, 1.09-1.94)]. Knockdown of KIF14 in non-small-cell lung carcinoma and cervical carcinoma cell lines decreased proliferation and colony formation in soft agar.

CONCLUSIONS

KIF14 expression is independently prognostic for disease-free survival in lung cancer and knockdown decreases tumorigenicity in vitro, showing that it is a clinically relevant oncogene and an exciting therapeutic target for further study.

Gene expression profiling on lung cancer outcome prediction: present clinical value and future premise

DNA microarray has been widely used in cancer research to better predict clinical outcomes and potentially improve patient management. The new approach provides accurate tumor classification and outcome predictions, such as tumor stage, metastatic status, and patient survival, and offers some hope for individualized medicine. However, growing evidence suggests that gene-based prediction is not stable and little is known about the prediction power of gene expression profiles compared with well-known clinical and pathologic predictors. This review summarized up-to-date publications in microarray-based lung cancer clinical outcome prediction and conducted secondary analyses for those with sufficient sample sizes and associated clinical information. Among the most commonly used analytic approaches, unsupervised clustering mainly recaptures tumor histology and provides variable degrees of prediction for tumor stage, lymph node status, or survival. Overall, most studies lack an independent validation. Supervised learning and testing generally offer a better prediction. Noted is that when conventional predictors of age, gender, stage, cell type, and tumor grade are considered collectively, the predictive advantage of the gene expression profiles diminishes. We conclude that outcome prediction from gene expression signatures selected by current analytic approaches can be mostly explained by well-known conventional predictors, particularly histologic subtype and grade of differentiation. A strategy for establishing independent or more accurate signatures is commented.

Lung cancer staging in the genomics era

The search for clinically applicable biologic markers or tumor signatures sufficiently powered as prognosticators of tumor behaviors or responses to therapeutic interventions has significantly advanced in scope and sophistication in the last 10 years. The TNM system, examining of tumor tissues to identify histopathologic features that could be correlated with tumor biology and outcome, could be improved by the immunohistochemical assessment of individual marker proteins or painstaking sequencing of candidate genes (one at a time) from tumor tissues. Large-scale investigation of the gene or protein expression profiles using genomics or proteomics technology may further improve risk stratification and assessment of therapeutic response. Although the gene expression profiling studies summarized in this article are exciting and initially serve as proofs of concept that large-scale mining of the genome and the transcriptome yields clinically useful data, the technology is still evolving and standardization is still needed for large-scale studies and data validation. As a proof of principle, studies have been performed to demonstrate that it is feasible to perform complete tumor microarray analysis, from tissue processing to hybridization and scanning, at multiple independent laboratories for a single study, and to demonstrate significant, albeit incomplete, agreement of gene expression patterns related to lung cancer biology and predictive of treatment outcomes via cross-study comparative analysis. Leading the concerted efforts of molecular characterization of lung cancer is the National Cancer Institute Director's Challenge Program: Toward A Molecular Classification of Cancer. The ultimate goal of molecular staging, envisioned as a combination of traditional TNM classification bolstered with gene/protein unique expression signatures, is to classify patients who have lung cancer on the basis of tumor biology, for better risk stratification and treatment using targeted patient-tailored therapeutics based on unique genotypes of individual tumors.

Applicability of microarray technique for the detection of noro- and astroviruses

Noroviruses and astroviruses are widespread viral agents causing gastroenteritis. Noroviruses are actually a diverse group of viruses. A new microarray-based detection method is presented for both noro- and astroviruses. This method enables simultaneous identification of genogroups and types of noro- and astroviruses. For pre-amplification, the monoplex- and multiplex-RT-PCR assays were applied to test sensitivity. The 202 specimens were used to determine the specificity, precision, and accuracy of the microarray. Stool samples representing 13 of norovirus GI and GII genotypes were assayed. The microarray detected all but one genotype. In a panel of 74 stool samples, 45 tested positive for norovirus by microarray, against 35 by a conventional PCR-method. In addition, microarray detected some double infections. In a panel of archival samples, astroviruses of genotype 4 frequently emerged in Finland from 1977 to 1997. This study demonstrates that microarray is useful for simultaneous monitoring of several viruses and their subtypes. Microarray provides a convenient tool for the detection of RNA viruses for which frequent changes in the panel of detection primers may be required.

How will lung cancer be treated in the future?

Lung cancer is still considered the leading cause of cancer-related deaths in industrialized countries. Despite improvements in cancer diagnosis and therapy, the present 5-year survival rate of 14% is only slightly higher than the survival rate of 8% in the early 1960s. At present, a survival plateau has been reached by the current management of lung cancer, including surgery, chemo- and radiotherapy. Although prevention and early diagnosis are undoubtedly important, more innovative treatment approaches and new agents that affect alternate mechanisms in tumor cells are needed. Since molecular targets of tumorigenesis and metastasis are currently the focus of lung cancer research, this review will summarize the most important molecular approaches and clinical trials of novel therapeutics.

The septin-binding protein anillin is overexpressed in diverse human tumors

Anillin is an actin-binding protein that can bind septins and is a component of the cytokinetic ring. We assessed the anillin expression in 7,579 human tissue samples and cell lines by DNA microarray analysis. Anillin is expressed ubiquitously but with variable levels of expression, being highest in the central nervous system. The median level of anillin mRNA expression was higher in tumors than normal tissues (median fold increase 2.58; 95% confidence intervals, 2.19-5.68, P < 0.0001) except in the central nervous system where anillin mRNA levels were lower in tumors. We developed a sensitive reverse transcription-PCR strategy to show that anillin mRNA is expressed in cell lines and in cDNA panels derived from fetal and adult tissues, thus validating the microarray data. We compared anillin with Ki67 mRNA expression and found a significant linear relationship between anillin and Ki67 mRNA expression (Spearmann r approximately 0.6, P < 0.0001). Anillin mRNA expression was analyzed during tumor progression in breast, ovarian, kidney, colorectal, hepatic, lung, endometrial, and pancreatic tumors and in all tissues there was progressive increase in anillin mRNA expression from normal to benign to malignant to metastatic disease. Finally, we used anti-anillin sera and found nuclear anillin immunoreactivity to be widespread in normal tissues, often not correlating with proliferative compartments. These data provide insight into the existence of nonproliferation-associated activities of anillin and roles in interphase nuclei. Thus, anillin is overexpressed in diverse common human tumors, but not simply as a consequence of being a proliferation marker. Anillin may have potential as a novel biomarker.

KIF14 is a candidate oncogene in the 1q minimal region of genomic gain in multiple cancers

Gain of chromosome 1q31-1q32 is seen in >50% of retinoblastoma and is common in other tumors. To define the minimal 1q region of gain, we determined genomic copy number by quantitative multiplex PCR of 14 sequence tagged sites (STSs) spanning 1q25.3-1q41. The most frequently gained STS at 1q32.1 (71%; 39 of 55 retinoblastoma) defined a 3.06 Mbp minimal region of gain between flanking markers, containing 14 genes. Of these, only KIF14, a putative chromokinesin, was overexpressed in various cancers by real-time RT-PCR. KIF14 mRNA was expressed in 20/22 retinoblastoma samples 100-1000-fold higher than in retina (t-test P=0.00002); cell lines (n=10) had higher levels than tumors (n=12) (P=0.009). KIF14 protein was overexpressed in retinoblastoma tumors and breast cancer cell lines by immunoblot. KIF14 was expressed in 4/4 breast cancer cell lines 31-92-fold higher than in normal breast tissue, in 5/5 medulloblastoma cell lines 22-79-fold higher than in fetal brain, and in 10/22 primary lung tumors 3-34-fold higher than in normal lung. Patients with lung tumors that overexpress KIF14 showed a trend toward decreased survival. KIF14 may thus be important in oncogenesis, and has promise as a prognostic indicator and therapeutic target.