Gene Expression Profiling Identifies a Prognostically Favorable Subgroup in AML Independent of Cytogenetic Stratification.

Abstract Acute myeloid leukemia (AML) is a heterogeneous group of diseases with varying clinical outcome. So far the karyotype of the leukemic blasts as well as molecular genetic abnormalities - both abnormalities on the genomic level - have been proven to be strong prognostic markers. However, even in genetically well defined subgroups clinical outcome is not uniform and a large proportion of AML shows genetic abnormalities of yet unknown prognostic significance. Here we addressed the question whether gene expression profiles are associated with clinical outcome independent of the known genomic abnormalities. Therefore, gene expression analyses were performed using Affymetrix U133A+B oligonucleotide microarrays in a total of 403 AML treated uniformly in the AMLCG studies. This cohort was divided randomly into a training set (n=269) and a test set (n=134). The training set included 18 cases with t(15;17), 22 cases with t(8;21), 29 cases with inv(16), 14 cases with 11q23/MLL-rearrangement, 19 with complex aberrant karyotype and 167 cases with normal karyotype or “other” chromosome aberrations. The respective data for the test set were: 10 t(15;17), 8 t(8;21), 11 inv(16), 8 11q23/MLL, 19 cases with complex aberrant karyotype and 78 with normal karyotype or “other” chromosome aberrations. Based on the clinical outcome the training cohort was divided into 4 equally large subgroups. We trained support vector machines (SVM) with the training set and classified the cases of the test set with the respective most discriminating genes. Next a Kaplan-Meier analysis was performed with the test set cases assigned to prognostic groups 1 to 4 according to SVM classification. Based on the expression level of 100 genes group 1 showed an overall survival rate of 57% at 3 years. 31 of 134 (23%) patients were assigned to this favorable subgroup. They belonged to the following cytogenetic subgroups: t(15;17) n=6, t(8;21) n=4, inv(16) n=3, 11q23/MLL n=4, complex aberrant karyotype n=1 and normal karyotype or “other” chromosome aberration n=13. The overall survival rate of groups 2, 3, and 4 did not differ significantly (17%, 21%, and 19% at 3 years). Among the genes highly expressed in the favorable group were MPO and the transcription factor ATBF1, which regulates CCND1. The unfavorable groups were characterized by a higher expression of the transcription factors ETS2, RUNX1, TCF4, and FOXC1. Interestingly, 10 of the top 40 differentially expressed genes are involved in the TP53-CMYC-pathway with a higher expression of 9 of these in the unfavorable groups (SFRS1, TPD52, NRIP1, TFPI, UBL1, REC8L1, HSF2, ETS2 and RUNX1). In conclusion, gene expression profiling leads to the identification of prognostically important alterations of molecular pathways which have not yet been accounted for by use of cytogenetics. This approach is anticipated to help optimizing therapy for patients with AML.

Download Full-text

Combining Proteomics and Gene Expression Profiling Identifies Proteins/Genes Associated with Short Overall Survival in Multiple Myeloma

Blood ◽

10.1182/blood.v120.21.197.197 ◽

2012 ◽

Vol 120 (21) ◽

pp. 197-197

Author(s):

Ricky D Edmondson ◽

Shweta S. Chavan ◽

Christoph Heuck ◽

Bart Barlogie

Keyword(s):

Gene Expression ◽

Overall Survival ◽

Multivariate Analyses ◽

Rank Test ◽

P Value ◽

Newly Diagnosed ◽

Training Set ◽

Test Set ◽

Log Rank Test ◽

Test Sets

Abstract Abstract 197 We and others have used gene expression profiling to classify multiple myeloma into high and low risk groups; here, we report the first combined GEP and proteomics study of a large number of baseline samples (n=85) of highly enriched tumor cells from patients with newly diagnosed myeloma. Peptide expression levels from MS data on CD138-selected plasma cells from a discovery set of 85 patients with newly diagnosed myeloma were used to identify proteins that were linked to short survival (OS < 3 years vs OS ≥ 3 years). The proteomics dataset consisted of intensity values for 11,006 peptides (representing 2,155 proteins), where intensity is the quantitative measure of peptide abundance; Peptide intensities were normalized by Z score transformation and significance analysis of microarray (SAM) was applied resulting in the identification 24 peptides as differentially expressed between the two groups (OS < 3 years vs OS ≥ 3 years), with fold change ≥1.5 and FDR <5%. The 24 peptides mapped to 19 unique proteins, and all were present at higher levels in the group with shorter overall survival than in the group with longer overall survival. An independent SAM analysis with parameters identical to the proteomics analysis (fold change ≥1.5; FDR <5%) was performed with the Affymetrix U133Plus2 microarray chip based expression data. This analysis identified 151 probe sets that were differentially expressed between the two groups; 144 probe sets were present at higher levels and seven at lower levels in the group with shorter overall survival. Comparing the SAM analyses of proteomics and GEP data, we identified nine probe sets, corresponding to seven genes, with increased levels of both protein and mRNA in the short lived group. In order to validate these findings from the discovery experiment we used GEP data from a randomized subset of the TT3 patient population as a training set for determining the optimal cut-points for each of the nine probe sets. Thus, TT3 population was randomized into two sub-populations for the training set (two-thirds of the population; n=294) and test set (one-third of the population; n=147); the Total Therapy 2 (TT2) patient population was used as an additional test set (n=441). A running log rank test was performed on the training set for each of the nine probe sets to determine its optimal gene expression cut-point. The cut-points derived from the training set were then applied to TT3 and TT2 test sets to investigate survival differences for the groups separated by the optimal cutpoint for each probe. The overall survival of the groups was visualized using the method of Kaplan and Meier, and a P-value was calculated (based on log-rank test) to determine whether there was a statistically significant difference in survival between the two groups (P ≤0.05). We performed univariate regression analysis using Cox proportional hazard model with the nine probe sets as variables on the TT3 test set. To identify which of the genes corresponding to these nine probes had an independent prognostic value, we performed a multivariate stepwise Cox regression analysis. wherein CACYBP, FABP5, and IQGAP2 retained significance after competing with the remaining probe sets in the analysis. CACYBP had the highest hazard ratio (HR 2.70, P-value 0.01). We then performed the univariate and multivariate analyses on the TT2 test set where CACYBP, CORO1A, ENO1, and STMN1 were selected by the multivariate analysis, and CACYBP had the highest hazard ratio (HR 1.93, P-value 0.004). CACYBP was the only gene selected by multivariate analyses of both test sets. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

Predicting clinical outcome through gene expression profiling in stage III melanoma

Journal of Clinical Oncology ◽

10.1200/jco.2007.25.18_suppl.8502 ◽

2007 ◽

Vol 25 (18_suppl) ◽

pp. 8502-8502

Author(s):

T. John ◽

M. A. Black ◽

T. Toro ◽

C. A. Gedye ◽

I. D. Davis ◽

...

Keyword(s):

Gene Expression ◽

Clinical Outcome ◽

Gene Expression Profiling ◽

Expression Profiling ◽

Stage Iiib ◽

Stage Iii ◽

Test Set ◽

Prognostic Groups ◽

Melanoma Patients ◽

Stage Iii Melanoma

8502 Background: Melanoma patients with clinically involved regional lymph nodes (Stage IIIB&C) represent a prognostically heterogeneous population. Current prognostic factors cannot distinguish the 30% of patients who will achieve long term survival from those who will relapse early. We hypothesized that gene expression profiling could identify these different prognostic groups and provide a greater understanding of the genetic mechanisms involved. Methods: Lymph node sections from 29 patients with Stage IIIB & IIIC melanoma and divergent clinical outcome as defined by time to tumor progression (TTP), including 16 poor (TTP<6 months) and 13 good (TTP>28 months) prognosis patients, were subjected to molecular profiling using spotted oligonucleotide arrays containing 30,888 probes as an initial test set. The differentially expressed genes were determined using a Wilcoxon-Mann-Whitney t-test with the false discovery rate controlling method of Benjamini-Hochberg and validated using quantitative real-time RT-PCR. Using logistic regression, a predictive score algorithm was developed based on the 15 genes for which the correlation between the two platforms was the strongest. The score was then applied to two independent validation sets of 10 and 14 patient samples. Results: Supervised analysis using differentially expressed genes was able to distinguish the two prognostic groups in the test set. The score correlated directly with clinical outcome, with higher scores associated with improved TTP. When the score was then applied to two independent sets of Stage III melanoma patient samples, it predicted clinical outcome accurately in 90% of samples. Conclusions: Stage IIIB and IIIC melanoma can be prognostically sub-classified according to the expression of 15 genes. To our knowledge this is the first study focused on Stage III disease using ex vivo patient samples. These results are encouraging and this genetic signature is currently being validated on a larger cohort. This method will allow appropriate stratification of stage III melanoma patients in adjuvant clinical trials, ameliorating the inherent biological heterogeneity that can confound these studies. [Table: see text]

Download Full-text

Tumor-Infiltrating T Cells Are Not Predictive of Clinical Outcome in Follicular Lymphoma.

Blood ◽

10.1182/blood.v108.11.824.824 ◽

2006 ◽

Vol 108 (11) ◽

pp. 824-824 ◽

Cited By ~ 2

Author(s):

Wei Yun Z. Ai ◽

Debra Czerwinski ◽

Sandra J. Horning ◽

John Allen ◽

Robert Tibshirani ◽

...

Keyword(s):

Gene Expression ◽

Flow Cytometry ◽

Overall Survival ◽

T Cells ◽

Clinical Outcome ◽

Follicular Lymphoma ◽

Clinical Features ◽

Training Set ◽

Hla Dr ◽

Validation Set

Abstract Background: Follicular lymphoma (FL) has variable clinical outcomes. It has been suspected that tumor-infiltrating immune cells affect the biology and outcome of this disease. Using gene expression profiling and immunoperoxide tissue staining techniques, T cells and macrophages have been related to the survival outcome in some studies, but not others. In the current study, we used flow cytometry to analyze T cells and their subsets in follicular lymphoma biopsy specimens and determined whether these cell populations correlated with clinical features and outcomes. Methods: Two hundred and eighty-nine follicular lymphoma patients (pt) presented from 1997 to 2003 underwent an excisional lymph node biopsy prior to any treatment. The median age of pt at diagnosis was 45.7 yrs, median follow-up was 8.6 yrs for living pts, and median survival was 15.7 yrs. All but 8 patients had stage III/IV disease, 5 had stage I/II, and 3 were unknown. The histological grades were: 162 (56%) grade 1, 112 (39%) grade 2, 13 (4.5%) grade 3 and 2 (0.5%) unknown. Among the 289 patients, 41(17%) had low FLIPI score, 150 (63%) intermediate, 48 (20%) high and 50 unknown. All biopsies were analyzed for CD20, CD3, CD4, CD8 and HLA-DR expression by single-parameter flow cytometry. The 289 pts were divided into a training set of 147 and a validation set of 142, stratified by age and era of diagnosis. We used these two factors to stratify the pts because age at diagnosis is the most important prognostic factor for survival, and, in our data set, the era of diagnosis had an impact on survival and on the time from diagnosis to first treatment. For our analysis, we began with the training set and used the percentages of each immune cell population as a continuous variable in a univariate analysis in relation to clinical features and outcomes. We chose 8 phenotypic variables: CD20, CD3, CD4, CD8, HLA-DR, CD4/CD3 ratio, CD8/CD3 ratio, and activated T cells [defined as (HLA-DR-CD20)/CD3]. Five parameters were used as clinical endpoints: overall survival, FLIPI score at diagnosis, the time from diagnosis to first treatment (defined as the time from the first treatment to second treatment), response to CVP as the first treatment and the duration of the benefit from the first treatment (defined as the time interval between initiation of first treatment and initiation of second treatment). Results: The number of pt evaluable for each of the outcome parameters was as follows: 289 for time to first treatment and for overall survival, 239 for FLIPI scores, 164 for response to CVP and 129 for duration of the benefit from the first treatment., Of the 8 variables tested in the training set, only CD4/CD3 ratio and CD8/CD3 ratio were marginally significant for the survival endpoint, with p 0.034 and 0.088, respectively. None of the variables was significant for any of the other endpoints. A multivariate analysis yielded CD4/CD3 as the only significant predictor for survival. When CD4/CD3 was tested in the validation set, it yielded a p value of 0.48. Conclusion: We find no evidence that the percentage of tumor-infiltrating T cells or their subsets is predictive of clinical outcome in follicular lymphoma. Any gene expression signature involving T cells that does relate to clinical outcome could therefore be a property of the activity of the cells rather than a simple reflection of their numbers.

Download Full-text

Omphaloceles Associated with Chromosomal Segmental Alterations Carry a Poor Prognosis

American Journal of Clinical Pathology ◽

10.1093/ajcp/aqaa161.052 ◽

2020 ◽

Vol 154 (Supplement_1) ◽

pp. S26-S27

Author(s):

H Zhou

Keyword(s):

Overall Survival ◽

Survival Rate ◽

Chromosomal Abnormalities ◽

Prognostic Significance ◽

Normal Karyotype ◽

Chromosomal Microarray ◽

Chromosomal Microarray Analysis ◽

Overall Survival Rate ◽

Significant Difference ◽

Worse Prognosis

Abstract Introduction/Objective Omphaloceles are frequently associated with chromosomal abnormalities, including aneuploidy and segmental alterations. High resolution chromosomal microarray analysis (CMA) can detect segmental alterations < 5 Mb, which is not detectable by G-banding. However, the prognostic significance of the segmental alteration in infant with omphalocele is not elucidated. Methods To identify omphalocele cases with genetic studies, a CoPath database search (1/2000 - 7/2017) was performed with key words “omphalocele” and “genetic”. From 1/2000 to 12/2008, only G-banding was performed. From 1/2009 to 7/2017, omphalocele cases were screened with karyotyping. Cases with normal karyotype were reflexed to CMA. Copy number gains/losses and corresponding genes were analyzed by the Affymetrix Chromosome Analysis Suite. Results Follow-up data are available in 75% (67/89) cases. There is no significant difference of the overall survival rate of male and female patients (80.5% vs 76.9%; χ 2, p = 0.7645). There are 16.9% (15/89) omphaloceles with aneuploidy, 10.1% (9/89) cases with segmental alterations by CMA, and 73.0% (65/89) cases with normal CMA and/or normal karyotype. Although patients with segmental alterations have a significantly higher survival rate than those with aneuploidy (44.4% vs 0%, χ2, p = 0.0119), their overall survival rate is significantly lower than infant with normal CMA and/or normal karyotype (44.4% vs 82.8%; χ2, p < 0.0001). Infants with segmental alterations carry a significantly worse prognosis than infants with normal genetic study. Conclusion To date, this is the first study of the prognostic significance of segmental alteration in infants with omphalocele. Our data demonstrated that omphaloceles with segmental alterations carry a significantly worse prognosis than those with normal CMA and/or karyotype. It is crucial to convey the prognosis to the parents with a fetus carrying segmental alterations; so the family could make an informed decision and get ready for an infant with special needs.

Download Full-text

Gene Expression Profiling Predicts Outcome in De Novo Acute Myeloid Leukemia (AML) with Normal Karyotype: Results of Children’s Oncology Group (COG) Study POG #9421.

Blood ◽

10.1182/blood.v108.11.1915.1915 ◽

2006 ◽

Vol 108 (11) ◽

pp. 1915-1915

Author(s):

Norman J. Lacayo ◽

Maureen O’Brien ◽

Shweta Jain ◽

Soheil Meshinchi ◽

Ron Yu ◽

...

Keyword(s):

Gene Expression ◽

Clinical Outcome ◽

Gene Expression Profiling ◽

Expression Profiling ◽

De Novo ◽

Gene List ◽

Normal Karyotype ◽

Classification Error ◽

Classification Error Rate ◽

Number Of Genes

Abstract We previously reported a 36% event-free survival (EFS) estimate for patients with normal karyotype (NK) on the COG study POG #9421 (n=144). In addition, we hypothesized that gene expression profiling would identify signatures linked to clinical outcome and useful for retrospective risk determination. Bone marrows in a subset of patients with NK (n=58) were analyzed using a 43,760-element spotted arrays containing 41,751 unique genes and expressed sequence tags; arrays were obtained from the Stanford University Microarray Core Facility. Prediction analysis for microarrays (PAM) was used to find genes that identified samples associated-with and unassociated-with events (relapse or death); after analyzing 28,711 genes with PAM we chose a 727-gene cluster that differentiated patients with NK on the basis of clinical outcome (cumulative classification error rate 19%). The analysis was biased for a larger number of genes in order to obtain a more biologically informative gene pathways analysis. Significance analysis of microarrays (SAM) on the PAM output identified 633 genes (false-discovery rate of 0%) that differed significantly between the event-associated and event-unassociated samples. Spearman based hierarchical clustering on these genes yielded 2 clusters with statistically significant different event-free survivals: 65% (n=24) for the event-unassociated curve and 23% (n=34) for the event-associated curve with P=0.01. The patients in these clusters did not differ at diagnosis for WBC (70K vs. 100K/microL with P=0.19) and age (10.1 vs. 9.9 yrs with P=0.83) by unpaired t-test; or for sex (P=0.11) and FLT3-ITD status (P=0.76) by Fisher’s exact test. The gene list (GenBank #) and fold-change in gene expression from SAM output were analyzed using Ingenuity Pathway Analysis software (Ingenuity™ Systems, Mountain View, CA). Canonical pathways identified 33 networks associated with event-unassociated outcome using 302 eligible genes that included: underexpressed CDC73, RAD50, SPARC, PTPN12, MXD1, TNF, ABCA1, STAT4, CCNA1, TNF, BCL2A1, JUN, BCL6 and AREG; and overexpressed RUNX3, FKBP9, FKBP8, MAP2K2, CHES1, HOXA11, HRK, CDK6, MGA, MAPK3, ABL1, HDAC7A, SMARCC2, SYK, MXD4, CDC42. Several of these genes have been previously reported to be associated with improved outcome in AML. However, two of these genes (MXD4 and MXD1) previously not identified as related to outcome in AML, but identified in our analysis in two highly interacting networks related to the MYC gene, result in a difference in EFS of 51% vs. 24% (P=0.04), suggesting that a smaller number of genes may be predictive of outcome. Conclusion: Risk assignment for patients with NK may be feasible by analyzing a limited number of genes. We will validate these findings by correlating gene expression results with quantitative real-time PCR. Prospective validation of this strategy in clinical trials may be warranted.

Download Full-text

Improved personalized survival prediction of patients with diffuse large B-cell Lymphoma using gene expression profiling

BMC Cancer ◽

10.1186/s12885-020-07492-y ◽

2020 ◽

Vol 20 (1) ◽

Author(s):

Adrián Mosquera Orgueira ◽

José Ángel Díaz Arias ◽

Miguel Cid López ◽

Andrés Peleteiro Raíndo ◽

Beatriz Antelo Rodríguez ◽

...

Keyword(s):

Gene Expression ◽

Machine Learning ◽

Expression Profiling ◽

Cell Lymphoma ◽

Clinical Information ◽

B Cell Lymphoma ◽

Training Set ◽

Test Set ◽

Large B Cell Lymphoma ◽

Large B Cell

Abstract Background Thirty to forty percent of patients with Diffuse Large B-cell Lymphoma (DLBCL) have an adverse clinical evolution. The increased understanding of DLBCL biology has shed light on the clinical evolution of this pathology, leading to the discovery of prognostic factors based on gene expression data, genomic rearrangements and mutational subgroups. Nevertheless, additional efforts are needed in order to enable survival predictions at the patient level. In this study we investigated new machine learning-based models of survival using transcriptomic and clinical data. Methods Gene expression profiling (GEP) of in 2 different publicly available retrospective DLBCL cohorts were analyzed. Cox regression and unsupervised clustering were performed in order to identify probes associated with overall survival on the largest cohort. Random forests were created to model survival using combinations of GEP data, COO classification and clinical information. Cross-validation was used to compare model results in the training set, and Harrel’s concordance index (c-index) was used to assess model’s predictability. Results were validated in an independent test set. Results Two hundred thirty-three and sixty-four patients were included in the training and test set, respectively. Initially we derived and validated a 4-gene expression clusterization that was independently associated with lower survival in 20% of patients. This pattern included the following genes: TNFRSF9, BIRC3, BCL2L1 and G3BP2. Thereafter, we applied machine-learning models to predict survival. A set of 102 genes was highly predictive of disease outcome, outperforming available clinical information and COO classification. The final best model integrated clinical information, COO classification, 4-gene-based clusterization and the expression levels of 50 individual genes (training set c-index, 0.8404, test set c-index, 0.7942). Conclusion Our results indicate that DLBCL survival models based on the application of machine learning algorithms to gene expression and clinical data can largely outperform other important prognostic variables such as disease stage and COO. Head-to-head comparisons with other risk stratification models are needed to compare its usefulness.

Download Full-text

Global analysis of gene expression signature and diagnostic/prognostic biomarker identification of hepatocellular carcinoma

Science Progress ◽

10.1177/00368504211029429 ◽

2021 ◽

Vol 104 (3) ◽

pp. 003685042110294

Author(s):

Jihan Wang ◽

Yangyang Wang ◽

Jing Xu ◽

Qiying Song ◽

Jingbo Shangguan ◽

...

Keyword(s):

Gene Expression ◽

Hepatocellular Carcinoma ◽

Overall Survival ◽

Survival Rate ◽

Candidate Genes ◽

Global Analysis ◽

Disease Diagnosis ◽

Overall Survival Rate ◽

Ppi Network ◽

Tumorigenesis And Progression

Hepatocellular carcinoma (HCC) is one of the most common cancers in the world. The landscape of HCC’s molecular alteration signature has been explored over the last few decades. Even so, more comprehensive research is still needed to improve understanding of tumorigenesis and progression of HCC, as well as to identify potential biomarkers for the malignancy. In this research, a comprehensive bioinformatics analysis was conducted based on the publicly available databases from both the Cancer Genome Atlas (TCGA) program and the gene expression omnibus (GEO) database. R/Bioconductor was used to analyze differentially expressed genes (DEGs) between HCC tumor and normal control (NC) samples, and then a protein-protein interaction (PPI) network of DEGs was established through the STRING platform. Finally, the application of specific candidate genes as diagnostic or prognostic biomarkers of HCC was explored and evaluated by ROC and survival analysis. A total of 310 DEGs were detected in the HCC tumor samples. Thirty-six hub DEGs in the PPI network and 10 candidates of the 36 genes showed significant alterations in tumor expression, including CDKN3, TOP2A, UBE2C, CDC20, PBK, ASPM, KIF20A, NCAPG, CCNB2, CYP3A4. The 10-gene signature had relatively significant effects when distinguishing tumors from normal samples (sensitivity >70%, specificity >70%, AUC >0.8, p < 0.001). Eight candidate genes were negatively correlated with the overall survival rate of the patients ( p < 0.05) and were all up-regulated in HCC tumor samples. The age and gender factors had no significant impact on the overall survival rate of HCC patients ( p > 0.05), and the TNM stage status factor had a significant negative prognosis correlation ( p < 0.05). This research provides evidence for a better understanding of tumorigenesis and progression of HCC and helps to explore candidate targets for disease diagnosis and treatment.

Download Full-text

Gene Expression Profiling and FLT3 Status Correlate with Outcome in De Novo Acute Myeloid Leukemia (AML) with Normal Karyotype: Results of Children’s Oncology Group (COG) Study POG #9421.

Blood ◽

10.1182/blood.v106.11.2372.2372 ◽

2005 ◽

Vol 106 (11) ◽

pp. 2372-2372

Author(s):

Norman Lacayo ◽

Soheil Meshinchi ◽

Susana Raimondi ◽

Chitra Saraiya ◽

Maureen O’Brien ◽

...

Keyword(s):

Gene Expression ◽

Clinical Outcome ◽

Gene Expression Profiling ◽

Expression Profiling ◽

Gene List ◽

Normal Karyotype ◽

Error Rates ◽

Pcr Analysis ◽

Classification Error ◽

Wild Type

Abstract The event-free survival (EFS) estimate for patients with normal karyotype (NK) on COG study POG #9421 (n=144) was 36%. We previously reported a subgroup of patients (n=68) with AML and NK that could be divided into 2 groups whose clinical outcomes correlated with abnormalities of FLT3 [internal tandem duplications (ITD) or activating loop mutations]. EFS estimates were 13% for patients with mutant FLT3 and 61% for children with wild-type FLT3 (P=0.01). We hypothesized that gene expression profiling would identify signatures that are linked to clinical outcome and can be used for risk determination. Cytogenetic testing was carried out in clinical laboratories at the institutions in which AML was diagnosed and then centrally reviewed. We analyzed bone marrow from 45 patients with NK on 43,760-element spotted arrays containing 41,751 unique genes and expressed sequence tags; arrays were obtained from the Stanford University Microarray Core Facility. FLT3 status (mutant or wild type) was determined by RT-PCR analysis of RNA from these 45 samples (exon 11 for ITDs, exon 17 for point mutations): 18 expressed mutant FLT3, 27 expressed wild-type FLT3. Using prediction analysis for microarrays (PAM) to find the minimum number of genes that identified samples associated with and unassociated with events (relapse or death), we identified a 128 gene cluster that differentiated patients with NK on the basis of clinical outcome (classification error rates were 15% for samples associated with events and 12.5% for event-unassociated samples). Significance analysis of microarrays (SAM) identified, with a false-discovery rate of 1.25%, 82 genes in the cluster whose expression differed significantly between the event-associated samples and the event-unassociated samples. Hierarchical clustering based on these 82 genes yielded 2 signatures: one event-associated and one event-unassociated. FLT3 Status Event-Associated Signature Event-Unassociated Signature Wild-type EFS=44% (n=15) EFS=90% (n=12) Mutant EFS=7% (n=13) EFS=60% (n=5) The median WBC counts at the time of diagnosis were 68 x 109/L in the event-associated group and 61 x 109/L in the event-unassociated group (P=0.27). The gene list and d-scores from SAM analysis were analyzed using Ingenuity Pathway Analysis software (Ingenuity™ Systems, Mountain View, CA). Canonical pathways associated with poor outcome included apoptosis signaling (↑BCL2A1, ↓BAK1), ERK/MAP signaling (↑RAC2), cell cycle (↓ABL1), SAPK/JNK signaling (↑RAC2, ↓CDC42), integrin signaling (↑RAC2, ↓BCAR3, ↓ABL1, ↓CDC42), and IL6 signaling (↑IL6R). We conclude that risk assignment for patients with NK can be more precisely determined by combining FLT3 analysis and gene expression profiling. Such an approach identified 4 distinct groups with different outcomes. We will validate these findings by analyzing additional diagnostic samples with normal karyotype. Prospective validation of this strategy in clinical trials may be warranted.

Download Full-text

Identification of Prognostic Markers by Gene Expression Profiling In Myelodysplastic Syndrome Hematopoietic Stem Cells

Blood ◽

10.1182/blood.v116.21.298.298 ◽

2010 ◽

Vol 116 (21) ◽

pp. 298-298 ◽

Cited By ~ 1

Author(s):

Andrea Pellagatti ◽

Mario Cazzola ◽

Aristoteles Giagounidis ◽

Janet Perry ◽

Luca Malcovati ◽

...

Keyword(s):

Gene Expression ◽

Stem Cells ◽

Bone Marrow ◽

Molecular Markers ◽

Principal Components ◽

Expression Profiling ◽

Training Set ◽

Test Set ◽

Hematopoietic Stem ◽

Expression Levels

Abstract Abstract 298 The myelodysplastic syndromes (MDS) are a heterogeneous group of clonal hematopoietic stem cell malignancies that are characterized by ineffective hematopoiesis resulting in peripheral cytopenias and a hypercellular bone marrow. Approximately 40% of patients with MDS will develop an acute myeloid leukemia. It is important to establish prognosis of MDS patients since the treatment options vary from supportive care to bone marrow transplantation. In order to determine the relationship of gene expression levels to prognosis and so identify new molecular markers, we have used gene expression profiling to study the transcriptome of the hematopoietic stem cells of 125 MDS patients with a minimum 12 month follow up. The CD34+ cells obtained from MDS patients and healthy individuals were analyzed using Affymetrix U133 Plus2.0 arrays. The patients were split randomly in a training set (n=84) and a test set (n=41). Supervised principal components analysis was used to identify genes correlated with survival. Using the 84 patients in the training set, the Cox scores were computed for each gene, and the principal components calculated on the genes with the highest Cox scores. The first of the principal components was then used to generate a regression model to predict the survival in the test set. Finally, for each probe set an importance score was calculated equal to its correlation with the supervised principal component predictor. This approach returned a list of 150 top ranked probe sets correlated with survival. Patients in the training set were split into tertiles based on the predictor (low, medium and high score) and patients in the test set were assigned to their predicted class, and Kaplan-Meier plots were generated for both training and test set. The differences in survival for both training and test set were statistically significant (Figure 1). Top ranked genes showing lower expression levels in patients with shorter survival include CDH1, LEF1 and AKAP12/Gravin. Top ranked genes showing higher expression levels in patients with shorter survival include IL23A, WT1 and PTHR2. Figure 2 shows survival of patients divided into tertiles of expression for the individual genes CDH1, LEF1 and WT1. It is probable that the genes identified in this study will become the first validated molecular markers for MDS prognosis. Multivariate analysis is currently being performed. Figure 1 Figure 1. Figure 2 Figure 2. Disclosure: No relevant conflicts of interest to declare.

Download Full-text

Reduced integrin alpha3 expression as a factor of poor prognosis of patients with adenocarcinoma of the lung.

Journal of Clinical Oncology ◽

10.1200/jco.1998.16.3.1060 ◽

1998 ◽

Vol 16 (3) ◽

pp. 1060-1067 ◽

Cited By ~ 49

Author(s):

M Adachi ◽

T Taki ◽

C Huang ◽

M Higashiyama ◽

O Doi ◽

...

Keyword(s):

Gene Expression ◽

Lung Cancer ◽

Overall Survival ◽

Survival Rate ◽

Poor Prognosis ◽

Prognostic Significance ◽

Lymph Node Status ◽

Overall Survival Rate ◽

Rt Pcr ◽

Integrin Alpha3

PURPOSE We investigated the possible association between integrin alpha3 and motility-related protein (MRP-1), cluster of differentiation antigen 9 (CD9) gene expression in non-small-cell lung cancer (NSCLC) and evaluated the prognostic significance of integrin alpha3 expression. PATIENTS AND METHODS We performed a retrospective study of integrin alpha3 and MRP-1/CD9 expression in resected tumor tissues from 151 NSCLC patients using quantitative reverse-transcriptase polymerase chain reaction (RT-PCR) and immunohistochemistry. RESULTS The ratio of integrin alpha3/beta-actin expression ranged from 0 to 5.87 (mean was 0.80; median, 0.70). Using the cutoff value of 0.7, there were 78 (52%) integrin alpha3-positive tumors and 73 (48%) tumors with reduced integrin alpha3 expression. The immunohistochemical results agreed well with those of the RT-PCR assays, and 88% had no discrepancy. In case of discrepancy, the results of RT-PCR were used in specimen classification. Integrin alpha3 gene expression was independent from MRP-1/CD9 gene expression. No significant association was found between integrin alpha3 expression and the patients' clinical characteristics. The overall survival rate of patients with integrin alpha3-positive NSCLCs was only slightly better than that of individuals whose tumors had reduced integrin alpha3 expression (55.9% v 47.1%; P = .085). By comparison, the overall survival rate of patients with integrin alpha3-positive adenocarcinomas was strikingly greater than in those whose tumors had reduced gene expression (54.4% v 35.2%; P = .004). Multivariate analysis with the Cox regression model of NSCLC and adenocarcinoma indicated that integrin alpha3 expression correlated better (P = .0188 and P = .0008, respectively) with the overall survival rate than other variables, except lymph node status. CONCLUSION No significant association was found between integrin alpha3 and MRP-1/CD9 gene expression in lung cancer. However, reduced integrin alpha3 expression is a poor prognosis factor in patients with adenocarcinomas.

Download Full-text