Microarray Classification of Myelodysplastic Syndrome (MDS) Identifies Subgroups with Distinct Clinical Outcomes.

Abstract The MILE (Microarray Innovations in LEukemia) study has previously shown that gene expression signatures associated with initial leukaemia classifier (LCver7) give an overall cross-validation accuracy of >95% for distinct sub-classes of pediatric and adult leukemias. However, only 50% of the 174 MDS samples in the whole-genome microarray analysis (Stage 1) of the MILE study were correctly identified; the remainder showed AML-like or non-leukemia-like gene profiles. An external morphological review (DB & HL) according to FAB and WHO criteria, of the 174 slides was performed independently (blind) which resulted in 6 samples being reclassified as AML and 4 non-leukemia cases excluded from the study. A recently improved, hierarchical based algorithm correctly identified 100% of the confirmed MDS cases. In this study, using LCver7, the confirmed 164 samples had 50% MDS classifications (Class 17), 23.8% non-leukemia classifications (Class 18), and 22.6% AML classifications (Classes 13 or 14) with the remaining 3.7% having a classification tie between 2 or 3 Classes (due to low confidence). No 5q- syndrome patients had an AML call, whilst 68.3% of RAEB2 patients had an AML classification and none were Class 18. Similarly, 95.6% of Low IPSS patients were classified as Class 17 or 18, whilst all patients (n=5) with High IPSS had an AML call. The classification was independent of blast cells: 10.2% of Class 18 calls had >5% blasts; 28.2% of AML-like cases had <5% blasts. Outcome data (132 MDS patients) was correlated with Class: significant difference (p<0.028, (Kaplan-Meier)) was seen in overall survival; with p <0.004 if AML (Classes 13 & 14) was compared to “non AML” (Classes 17 & 18). Statistically significant differences were seen for time to transformation to AML between the classes (p<0.0001) and between AML and “non AML” (p<0.00007, Kaplan-Meier)) with a probability of transformation of 44% at 18 months for the AML group compared to <8% for the “non-AML” group. A further linear classifier has been used to discriminate patients who transform to AML within 18 months (poor prognosis) with patients with no transformation after >60 months (good prognostic group). Bioinformatic analysis of molecular mapped functions and canonical pathways showed that cell signalling processes were over-represented when comparing de novo AML (n=204) with MDS, from the MILE study, whilst signal transduction pathways were deregulated when comparing non-leukemia samples (n=71) with MDS. Similar pathways and functions were also deregulated when comparing the correctly classified MDS with Class 17 call against MDS with Class 18 call and MDS with AML Classes 13 or 14 calls. In conclusion, the use of microarrays within the initial study, solely intended for diagnostic purposes, has now evolved towards a position in which novel prognostic value may be gained from distinct gene expression signatures. This has also resulted in a better molecular understanding of the progression from non-leukemia, through MDS into full blown AML.

Download Full-text

Comprehensive genomic analysis in metastatic pancreatic ductal adenocarcinoma (PDAC).

Journal of Clinical Oncology ◽

10.1200/jco.2016.34.4_suppl.285 ◽

2016 ◽

Vol 34 (4_suppl) ◽

pp. 285-285

Author(s):

Hui-Li Wong ◽

Martin Jones ◽

Peter Eirew ◽

Joanna Karasinska ◽

Kasmintan A Schrader ◽

...

Keyword(s):

Gene Expression ◽

Expression Analysis ◽

Gene Expression Analysis ◽

De Novo ◽

Expression Patterns ◽

Genomic Analysis ◽

Ductal Adenocarcinoma ◽

Tumor Biopsy ◽

Significant Difference ◽

Gene Expression Signatures

285 Background: In the absence of defined tumor molecular subtypes and validated predictive markers, PDAC has been largely treated as a single disease. Recent studies of molecular subtyping in PDAC reveal a complex mutational landscape with data suggesting the presence of genomic and gene expression signatures that may have prognostic and therapeutic significance. These studies predominantly focused on resected PDAC and lack data on metastatic tumors. We aim to explore the clinical utility of whole genome sequencing (WGS) and transcriptome analysis from metastatic biopsy samples in patients (pts) with advanced PDAC. Methods: Pts with incurable advanced cancers undergo tumor biopsy for in-depth WGS and RNA sequencing (RNASeq) as part of an ongoing prospective study (NCT02155621). Comprehensive bioinformatics analysis is performed to identify somatic cancer aberrations, gene expression changes and cellular pathway abnormalities. Here we describe clinical and molecular data on the subset of pts with advanced PDAC. Results: Sixteen PDAC pts have been enrolled; median age 59 years, 8 males (50%), 10 with de novo metastases (63%). Full WGS and RNASeq were completed in 11 pts (1 failed biopsy, 4 had insufficient tumor). KRAS codon 12 and TP53 mutations were present in all but one pt. CDKN2A and SMAD4 were also frequently altered (7 and 4 pts respectively). Gene expression analysis for classical and basal subtypes similar to those recently described (PMID 26343385) identified 3 and 6 pts with classical and basal expression patterns respectively, and 2 pts with mixed expression. Overall survival (OS) was significantly worse for the basal subtype vs all others (median OS 7 vs. 13.9 months (ms), p = 0.017). When separated into 3 subtypes a significant difference was still noted (median OS 7 ms in basal, 19.2 ms in classical and 11.8 ms in mixed subtype, p = 0.032). Conclusions: WGS analysis demonstrated a similar mutation pattern to that described in resectable PDAC, with no novel actionable mutations identified. Gene expression analysis demonstrated the presence of distinct gene expression signatures significantly associated with outcome, despite small pt numbers. These results need to be validated prospectively in larger cohorts. Clinical trial information: NCT02155621.

Download Full-text

S100 TREATMENT-DEPENDENCE OF HIGH-RISK GENE EXPRESSION SIGNATURES IN DE NOVO FOLLICULAR LYMPHOMA

HemaSphere ◽

10.1097/01.hs9.0000558620.24721.04 ◽

2019 ◽

Vol 3 (S1) ◽

pp. 1

Author(s):

C. Bolen ◽

W. Hiddemann ◽

R. Marcus ◽

M. Herold ◽

S. Huet ◽

...

Keyword(s):

Gene Expression ◽

High Risk ◽

Follicular Lymphoma ◽

De Novo ◽

Risk Gene ◽

Gene Expression Signatures

Download Full-text

The Expression of I-Mfa in Adult Patients with De Novo Acute Myeloid Leukemia and Its Clinical Significance

Blood ◽

10.1182/blood.v124.21.5316.5316 ◽

2014 ◽

Vol 124 (21) ◽

pp. 5316-5316

Author(s):

Bing Xu ◽

Huijuan Dong ◽

Feili Chen ◽

Yong Zhou ◽

Jiabao Liang ◽

...

Keyword(s):

Gene Expression ◽

Acute Myeloid Leukemia ◽

Clinical Significance ◽

Myeloid Leukemia ◽

De Novo ◽

Adult Patients ◽

Expression Level ◽

Significant Difference ◽

Median Expression ◽

Acute Myeloid

Abstract Background: I-mfa has been identified as an inhibitor of MyoD and other related myogenic basic helix-loop-helix proteins. I-mfa contains a cysteine-rich C-terminal domain, and has been reported to function as transcriptional regulator of different pathways including Wnt signaling, c-jun N-terminal kinase signaling, and the regulatory properties of I-mfa depend on the C-terminal domain. Furthermore, recent studies have found that the I-mfa domain may have a close correlation with the development of myeloid neoplasms, however the role of I-mfa in adult patients with de novo acute myeloid leukemia still remain unclear. Aims: The aim of this study was to determine I-mfa expression in adult patients with de novo acute myeloid leukemia and its clinical significance. Methods: BM samples form 110 adult patients with de novo AML were analyzed. Of the 110 AML patients, 66 were males and 44 were females, with a mean age of 32 years( range from 12 to 77 years). Among them, 1 out of 110 patients was M1, 49 were M2, 14 were M4, 28 were M5, 1was M6 and 17 were acute unclassified leukemia. All patients received 1 to 2 cycles of induction of standard-dose cytarabine continuous infusion×7 days with idarubicin or daunorubicin×3days, fellowed by consolidation therapy with HiDAC and then stem cell transplantation according to patient’s condition. Real-time reverse transcription-polymerase chain reaction(RT-PCR) was used to detect the expression of I-mfa gene in 110 de novo adult AML patients, and the patients were divided into high and low I-mfa expression groups accordint to the median expression of I-mfa mRNA. Comparisons were performed using Mann-Whitney U test, Chi-square test and Kaplan-Meier method. Results:Distribution of I-mfa gene expression in different FAB subtypes was with no significant differences (P=0.169). The median age of AML pateints in low and high I-mfa gene epxression groups were 35 and 40 years old(P=0.162), and the median expression of I-mfa in 44 female patients and 66 male patients was 0.018 and 0.013 separately(P=0.728). What’s more, there was no significant difference of WBC, Hb level, PLT, bone marrow blast counts between the two groups (P＞0.05), and the I-mfa expression level was also not correlated with chromosome risk stratification and the expression of CD34 (P＞0.05). High I-mfa expression group had a lower complete remission rate than that in the low expression group (81.8% vs 63.6%, P=0.032), However, the overall survival rate was with no significant difference in the low and hign I-mfa gene expression groups(76.4% vs 76.4%, P=0.471). Conclusions: Our results showed high I-mfa expression correlates with a poor treatment response, the OS rate was with no significant difference in the two groups. There is somewhat correlation between the expression level of I-mfa gene and prognosis and the expression of I-mfa may be a prognostic factor for adult patients with de novo acute myeloid leukemia. Disclosures No relevant conflicts of interest to declare.

Download Full-text

De Novo Generation of Hit-like Molecules from Gene Expression Signatures Using Artificial Intelligence

10.26434/chemrxiv.7294388.v1 ◽

2018 ◽

Cited By ~ 4

Author(s):

Oscar Mendez-Lucio ◽

Benoit Baillif ◽

Djork-Arné Clevert ◽

David Rouquié ◽

Joerg Wichard

Keyword(s):

Gene Expression ◽

Artificial Intelligence ◽

De Novo ◽

Molecular Design ◽

Gene Expression Signature ◽

Cellular Level ◽

Generative Models ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Gene Expression Signatures

Finding new molecules with a desired biological activity is an extremely difficult task. In this context, artificial intelligence and generative models have been used for molecular <i>de novo</i> design and compound optimization. Herein, we report the first generative model that bridges systems biology and molecular design conditioning a generative adversarial network with transcriptomic data. By doing this we could generate molecules that have high probability to produce a desired biological effect at cellular level. We show that this model is able to design active-like molecules for desired targets without any previous target annotation of the training compounds as long as the gene expression signature of the desired state is provided. The molecules generated by this model are more similar to active compounds than the ones identified by similarity of gene expression signatures, which is the state-of-the-art method for navigating compound-induced gene expression data. Overall, this method represents a novel way to bridge chemistry and biology to advance in the long and difficult road of drug discovery.

Download Full-text

De Novo Generation of Hit-like Molecules from Gene Expression Signatures Using Artificial Intelligence

10.26434/chemrxiv.7294388 ◽

2018 ◽

Author(s):

Oscar Mendez-Lucio ◽

Benoit Baillif ◽

Djork-Arné Clevert ◽

David Rouquié ◽

Joerg Wichard

Keyword(s):

Gene Expression ◽

Artificial Intelligence ◽

De Novo ◽

Molecular Design ◽

Gene Expression Signature ◽

Cellular Level ◽

Generative Models ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Gene Expression Signatures

Download Full-text

Retrospective review of subsequent treatments (tx) for esophagogastric adenocarcinomas (EGA) refractory to FOLFOX.

Journal of Clinical Oncology ◽

10.1200/jco.2017.35.4_suppl.123 ◽

2017 ◽

Vol 35 (4_suppl) ◽

pp. 123-123

Author(s):

Ciara Marie Kelly ◽

Yelena Yuriy Janjigian ◽

David Paul Kelsen ◽

Marinela Capanu ◽

Joanne F. Chou ◽

...

Keyword(s):

De Novo ◽

Complete Response ◽

Outcome Data ◽

Similar Proportion ◽

Kaplan Meier ◽

Response Status ◽

Poorly Differentiated ◽

Ecog Ps ◽

Next Generation Sequencing Ngs ◽

Ngs Data

123 Background: FOLFOX is a preferred 1st-line tx for advanced EGA. We sought to characterize outcomes on subsequent tx and to see if MSK-IMPACT, a 410-gene next generation sequencing (NGS) platform, increases tx options. Methods: We retrospectively identified patients (pts) with advanced, Her2-negative EGA treated with 1st-line FOLFOX between Jan 2012 to Dec 2014. Clinicopathologic, tx and outcome data were analyzed. Overall survival (OS) was calculated from start of FOLFOX using Kaplan-Meier methods. Landmark analysis was used to compare OS and response status. Results: 185 pts were identified. The majority were Caucasian (82%), male (76%), ECOG PS 1 (67%), with poorly differentiated histology (72%) and de novo metastatic disease (84%). Median age was 64 years. The disease-control rate (DCR, partial response + stable disease) of FOLFOX was 80% [95%CI: 74%-85%]; 19% were FOLFOX primary refractory (FR). Median time-to-progression (TTP) on FOLFOX was 7 and 2 months (mo) for FOLFOX sensitive (FS) and FR pts, respectively. There was a higher proportion of females (26% vs. 14%, P = 0.18), gastric (43% vs. 23%, P = 0.051) and moderately differentiated tumors (26% vs. 12%, p = 0.113) in the FS vs. FR group. Six mo survival from the landmark time of 2 mo after initiation of FOLFOX was 83% [95%CI: 76%-89%], and 38% [95%CI: 20%-56%] for FS and FR pts, respectively (p < 0.01). A similar proportion of FS and FR pts received 2nd-line tx (65% vs. 69%). The DCR was similar in both groups (31% vs 29%). 2nd-line tx included: irinotecan- (51%) and taxane-based regimens (32%) or a clinical trial (CT) (13%). The median TTP on 2nd-line tx was similar in FS and FR groups (2.5 vs 2 mo). Ramucirumab was given in 14% of 2nd line regimens. 3rd-line chemo use was similar in both groups (37% vs 31%) but the DCR was lower in FR patients (18% vs. 9%). 51 pts had IMPACT; 1 pt (2%) enrolled onto a genotyped-matched CT. 14 pts received immunotherapy; 1 FS Pt has ongoing complete response 1+ year. Conclusions: Surprisingly, FS and FR pts derive similar, marginal benefit from 2nd-line tx, emphasizing the appropriateness of CT options in this setting. NGS rarely expanded tx options. Updated and in-depth NGS data will be presented.

Download Full-text

Bioinformatic analysis of the inner heterogeneity within MYCN non-amplified pediatric neuroblastoma sub-group

10.21203/rs.3.rs-48689/v1 ◽

2020 ◽

Author(s):

Haiwei Wang ◽

Xinrui Wang ◽

Liangpu Xu ◽

Ji Zhang ◽

Hua Cao

Keyword(s):

Gene Expression ◽

Overall Survival ◽

Clinical Outcomes ◽

Cox Regression ◽

Prognostic Significance ◽

Young Patients ◽

Bioinformatic Analysis ◽

Mycn Amplification ◽

Expression Levels ◽

Kaplan Meier

Abstract Background: Pediatric neuroblastoma is divided into MYCN amplified and MYCN non-amplified sub-groups. However, the extent of heterogeneity within MYCN amplified or non-amplified pediatric neuroblastoma is unclear.Methods: The prognostic significance of age and MYCN amplification was determined through multivariate cox regression and Kaplan-Meier survival analysis. MYCN non-amplified pediatric neuroblastoma patients were divided into different sub-consensuses using non-negative matrix factorization (NMF) based on the gene expression profiling. Genes particularly expressed in MYCN non-amplified young neuroblastoma patients were identified using Therapeutically Applicable Research to Generate Effective Treatments (TARGET) and Gene Expression Omnibus (GEO) datasets. The prognostic effects of ALCAM, CACNA2D3, DST, EPB41L4A and KIFIB in MYCN non-amplified pediatric neuroblastoma patients were determined by Kaplan-Meier survival.Results: Age and MYCN amplification were independent prognostic factors in pediatric neuroblastoma. MYCN non-amplified pediatric neuroblastoma comprised young and old two distinct sub-groups. Compared with MYCN non-amplified old neuroblastoma patients, MYCN non-amplified young neuroblastoma patients had better clinical outcomes. MYCN non-amplified pediatric neuroblastoma was divided into three sub-consensuses through NMF assay and each sub-consensus was with significantly different clinical outcomes. However, MYCN amplified pediatric neuroblastoma was relatively homogeneous, and could not divide into sub-groups with different clinical outcomes by age or by NMF assay. ALCAM, CACNA2D3, DST, EPB41L4A and KIFIB were highly expressed in MYCN non-amplified young neuroblastoma patients. Moreover, the high expression levels of ALCAM, CACNA2D3, DST, EPB41L4A or KIFIB were associated with the favorable prognosis of MYCN non-amplified neuroblastoma patients. We also found that DST was an independent prognostic factor in MYCN non-amplified neuroblastoma patients and MYCN non-amplified neuroblastoma young patients with high DST expression levels had the best clinical overall survival.Conclusions: MYCN non-amplified neuroblastoma was a heterogeneous disease and could be divided into sub-groups based on age or the expression levels of ALCAM, CACNA2D3, DST, EPB41L4A or KIFIB. MYCN non-amplified neuroblastoma young patients with high DST expression levels had the best clinical overall survival.

Download Full-text

Pulmonary vein isolation for AF: does technique influence outcome?

EP Europace ◽

10.1093/europace/euab116.060 ◽

2021 ◽

Vol 23 (Supplement_3) ◽

Author(s):

SA Reddy ◽

SL Nethercott ◽

BV Khialani ◽

MS Virdee

Keyword(s):

Pulmonary Vein ◽

Pulmonary Vein Isolation ◽

De Novo ◽

Outcome Data ◽

Adverse Event Rate ◽

Single Shot ◽

Funding Sources ◽

Significant Difference ◽

Single Centre Study ◽

Unselected Population

Abstract Funding Acknowledgements Type of funding sources: None. Background Over the last 20 years various techniques have been developed striving for safer and more durable pulmonary vein isolation (PVI). The three most commonly used tools are pulmonary vein ablation catheter (PVAC) and cryoballoon (‘single-shot’ techniques), and point-by-point (PBP) radiofrequency ablation using 3D electroanatomical mapping (EAM). Objective Evaluate the safety and efficacy of the different techniques in an unselected population undergoing de-novo ablation for persistent or paroxysmal atrial fibrillation (AF) at a single high-throughput centre. Method Retrospective, single-centre study of consecutive AF ablations between March 2017 and April 2018. Demographic, procedural and outcome data were analysed. Results Over the study period 329 first-time PVI procedures were performed. 37.4% were performed using PBP, 39.8% using cryoballoon and 22.8% using PVAC. There was no significant difference in age or sex between different ablation technique groups. 238 procedures were performed for paroxysmal AF and 91 for persistent AF. A higher proportion of the persistent cases were performed using point-by-point techniques compared to paroxysmal cases (58.2% vs 29.0%, p < 0.05). Procedural times were significantly longer in the group undergoing PBP ablation compared to cryoballoon or PVAC. However, there was no statistically significant difference in 12-month freedom from symptomatic AF or procedural complications between the groups. Conclusions PBP, PVAC and cryoballoon AF ablation all appeared equally efficacious in an unselected population, though PVAC and cryoballoon procedures were shorter. All procedures were associated with a low adverse event rate. Prospective examination is required to substantiate this finding. Table 1CARTOn= 123Cryoballoonn= 131PVACn = 75p-valueAge/years61.7 ± 9.259.5 ± 10.661.7 ± 9.70.14Male92 (74.8)88 (67.2)49 (61.3)0.80Paroxysmal AF70 (56.9)106 (78.6)62 (82.7)0.14Cardiovascular risk factors Hypertension Diabetes Ischaemic heart disease Cerebrovascular disease Heart failure Dyslipidaemia73 (59.3)23 (18.7)40 (32.5)2 (1.6)0 (0) 12 (9.8)79 (60.3)19 (14.5)45 (34.4)0 (0)1 (0.8)16 (12.2)43 (57.3)16 (21.3)22 (29.3)1 (1.3)0 (0)10 (13.3) 0.58 0.24 0.62 - - 0.71Left atrial diameter/cm4.2 ± 0.74.1 ± 0.73.9 ± 1.00.69Procedure time/mins191.3 ± 39126.7 ± 24117.4 ± 30<0.056 month success Paroxysmal Persistent50/66 (75.8)32/51 (62.7)78/103 (75.7)18/24 (75.0)48/61 (78.6)10/12 (83.3) 0.99 0.80Complications9 (7.3)3 (2.3)1 (1.3)0.07Patient demographics, procedural characteristics and outcomes for Carto, cryoballoon and PVAC cases. Values presented as mean ± SD or n (%)Abstract Figure. Time to arrhythmia recurrence

Download Full-text

Reductive regulation of BECN1 gene in adult Egyptian patients with do novo AML

Egyptian Journal of Medical Human Genetics ◽

10.1186/s43042-020-00087-z ◽

2020 ◽

Vol 21 (1) ◽

Author(s):

Manal Fawzy Ghozlan ◽

Botheina Ahmed Thabet Farweez ◽

Nesma Ahmed Safwat ◽

Noha Bassiouny Hassan ◽

Walaa Ali Elsalakawy

Keyword(s):

Gene Expression ◽

Bone Marrow ◽

Peripheral Blood ◽

De Novo ◽

Quantitative Polymerase Chain Reaction ◽

Control Group ◽

Significant Difference ◽

Egyptian Patients ◽

Blast Cells ◽

De Novo Aml

Abstract Background Acute myeloid leukaemia (AML) is a clonal haematopoietic disease characterized by the proliferation of immature blast cells in the bone marrow and peripheral blood. Autophagy is an inherent cellular route by which waste macromolecules are engulfed within autophagosomes prior to their fusion with cytoplasmic lysosomes for degradation. The BECN1 gene encodes the Beclin-1 protein, which regulates autophagy. Few reports have investigated BECN1 gene expression and its value in AML patients. Results This randomized case-control study included 50 newly diagnosed AML patients, in addition to 20 subjects as a control group. BECN1 gene expression was assessed using real-time quantitative polymerase chain reaction (qRT-PCR). The median level of BECN1 gene expression in AML patients was 0.41 (IQR 0.29–1.03) in comparison to 1.12 (IQR 0.93–1.26) in the control group (P = 0.000). Seventy-two percent of AML patients showed reduced BECN1 gene expression, which was highly significantly associated with intermediate and adverse cytogenetic risk. Reduced BECN1 gene expression was associated with older age, higher total leukocyte counts, the presence of peripheral blood blast cells, a higher percentage of bone marrow blast cells, and higher expression of CD34 and CD117. FLT3-ITD mutation was detected in 14 patients (38.9%), all of whom showed reduced BECN1 gene expression (P = 0.006). BECN1 gene expression was also reduced in non-responder AML patients, with a highly statistically significant difference (P = 0.002). Conclusion A reduction in BECN1 gene expression might indicate a poor prognosis in adult Egyptian patients with de novo AML. Decreased BECN1 gene expression is associated with a higher risk of resistance to treatment. Targeting autophagy pathways may help in the treatment of AML patients.

Download Full-text

Gene Expression Profiling in the Classification of Acute Leukemia Brazilian Patients

Blood ◽

10.1182/blood.v124.21.5315.5315 ◽

2014 ◽

Vol 124 (21) ◽

pp. 5315-5315

Author(s):

Carolina Pereira de Souza Melo ◽

Catharina Brant Campos ◽

Alvaro Pimenta Dutra ◽

Angelo Atalla ◽

Mara Albonei Dudeque Pianovski ◽

...

Keyword(s):

Gene Expression ◽

Acute Leukemia ◽

De Novo ◽

Conflicts Of Interest ◽

Bioinformatic Analysis ◽

Gene Signature ◽

World Health ◽

Prediction Algorithm ◽

Lymphoid Leukemia ◽

Number Of Patients

Abstract In the past few decades, genetic data has become increasingly important for acute leukemia diagnosis and patients stratification. Indeed, the present World Health Organization (WHO) leukemia classification system is largely based upon genetically defined subgroups. Gene expression profile (GEP) may correctly predict most genetic leukemia subtypes, but so far no GEP report has evaluate patients from Latin America. In the present study, we used gene expression microarray data to build an acute leukemia classifier. Bone marrow samples were collected from 231 individuals at diagnosis, 110 presented de novo acute myeloid leukemia (AML), 97 had de novo acute lymphoid leukemia (ALL) and the remaining 24 were controls who had other conditions including chronic leukemias or non-hematological diseases. GEP was evaluated based on mRNA expression signatures obtained with the Sure Print G3 Human GE (60k) system (Agilent Technologies). k-nearest neighbors prediction algorithm was applied and the top 60 informative genes were selected for each of the most prevalent genetic subtypes (T-ALL, B-ALL BCR-ABL, B-ALL ETV6-RUNX1, B-ALL TCF3-PBX1, AML PML-RARa, AML RUNX1-RUNX1T1, AML FLT3-ITD, AML NPM1mut). The less prevalent groups such as MLL rearranged and CBFB-MYH11 were not included in the classifier because of the low number of patients carrying these aberrations in our cohort. Performance of each prediction model was assessed by leave-one-out crossvalidation through the GenePattern platform (Broad Institute). The average classifier accuracy was 94.75%. Higher accuracy and precision were achieved for T-ALL (99%/96%) and AML PML-RARa (97%/97%). However, for ALL BCR-ABL, AML FLT3-DIT and AML NPM1mut the gene signature had low precision rates (74%, 66% and 80%, respectively). The data presented here confirm that a single platform of gene expression followed by bioinformatic analysis can correctly classify genetic subgroups, but a refinement of the classifier developed is needed in order to improve the detection of heterogeneous entities such as BCR-ABL or FLT3-ITD carriers. Disclosures No relevant conflicts of interest to declare.

Download Full-text