Multiplex Quantum Dot ISH in Tissue Microarrays Identifies HOXA9 and DNMT3A as Unfavourable Markers in Acute Myeloid Leukaemia
E Tholouli, S McDermott, JA Hoyland, C Glennie, R Swindell, JA Liu Yin, RJ Byers. Manchester Royal Infirmary, Manchester, United Kingdom; University of Manchester, Manchester, United Kingdom; Christie Hospital, Manchester, United Kingdom
Background: Validation of prognostic genes identified by microarray expression profiling is required in clinical samples. We have developed a quantum dot (QD) based multiplexed in-situ hybridization (ISH) method for quantitative localization of mRNA targets in formalin fixed paraffin embedded tissue (FFPET) which we used to identify prognostic genes in AML.
Design: Fifteen tissue microarrays (TMAs) were made using FFPET bone marrow trephine samples from 240 patients with AML treated at Manchester Royal Infirmary with standard chemotherapy between 1994 and 2005, of which 192 patients were suitable for analysis. Samples were represented in triplicate and a whole blood white cell pellet used as standard. QD-ISH was performed for nine candidate prognostic genes in AML using triplex QD-ISH for: Bcl2, survivin and XIAP; DNMT1, DNMT3A and DNMT3B; HOXA4, HOXA9 and Meis1. Signal intensity was measured by spectral imaging and scrambled oligonucleotides used to measure background staining. Background noise was corrected for by dividing expression of anti-sense probes by that of scrambled probe followed by normalization of expression values against the standard. Statistical analysis was performed using chi-square and Mann Whitney-U tests and overall survival (OS) and disease free survival (DFS) used in Kaplan-Meier analysis.
Results: Median age was 52 years and the OS was 43% at 5 years with 80% complete remission (CR) after induction. Low expression of HOXA4 was associated with improved OS (p=0.013) and DFS (p=0.025) on univariate and multivariate analysis. High expression of HOXA9 (p<0.0001) and DNMT3A (p=0.04) were associated with failure to achieve CR. High expression of Meis1 was of borderline significance for poor response to chemotherapy (p=0.05). The other 5 genes showed no correlation with CR, DFS or OS.
Conclusions: These results demonstrate the utility of a novel standardized, quantitative multiplex QD-ISH method for identification of prognostic markers in FFPET. The advantages of the method are its application to TMAs, enabling high throughput, use of archived materials and transferability across a spectrum of malignancies.
Category: Special Category for 2009 - Pan-genomic/Pan-proteomic approaches to Cancer
Monday, March 9, 2009 2:30 PM
Platform Session: Section G, Monday Afternoon