Abstract

Abstract 2007

Background:

The clinical course of patients with mantle cell lymphoma (MCL) is highly variable and there is an urgent need for individual patient risk stratification to facilitate accurate interpretation of clinical trials and rational development of alternative treatments for those with high risk disease. Despite the identification of potential prognostic biomarkers through various screening studies including gene expression profiling, these findings have yet to be translated into a clinical application. Array-based quantitative nuclease protection assay (qNPA) can assess gene expression in formalin-fixed paraffin-embedded (FFPE) tissue in a simple, robust manner and may have potential to become a clinical laboratory test. The aim of this study was to determine the ability of qNPA to identify prognostic mRNA biomarkers in routinely processed FFPE diagnostic biopsies of MCL.

Design:

The expression of 42 genes with potential prognostic significance was analyzed using qNPA in a plate based, low density array format on FFPE tissue from 57 patients with MCL diagnosed between 1999 and 2010. Gene expression was normalized to two housekeeping genes, TBP and B2M.

Results:

56 (98%) of the cases were successfully analyzed with an average co-efficient of variation (CV) of gene expression between triplicate specimens for all genes in each case ranging from 7% to 16% and median CV of 11%. By univariate analysis with the Cox proportional hazard model, expression of the MYC, TNFRSF10B, CDKN2A, and CCND1 genes were significantly associated (p < 0.05) with progression free survival.

Of relevance, MYC and TNFRSF10B were amongst 5 genes out of 33 screened by Hartmann et al. (J Clin Oncol 26:4966-72, 2008) that produced the best survival predictor in a previous qRT-PCR-based assay performed on both fresh-frozen and FFPE MCL tumor samples. In addition, homozygous deletions of the CDKN2A locus on 9p21 has been implicated as a key component in aggressive disease manifestation and is postulated to be a result of the loss of two key regulatory elements encoded at this locus, the cell cycle inhibitor INK4a and the p53 regulator ARF. Finally, expression of a more stable, truncated form of CCND1 lacking the 3’UTR, was first described by Wiestner et al. (Blood 109:4599-4606, 2007) and has been associated with shorter survival. We designed the CCND1 probe in the qNPA assay to target the distal 3’UTR in order to specifically measure non-truncated CCND1. By qNPA, 7 of the 57 MCL cases lacked CCND1 expression and the 3-year overall survival for these cases was 42% compared to 70% for positive cases.

Conclusion:

qNPA can reliably and reproducibly assess the expression of specific genes from FFPE tissue with relatively small CVs and identified four genes of possible prognostic significance in archived MCL specimens, suggesting this test may have potential to be developed into a clinical assay to guide individualized therapy in patients with MCL.

Table 1.

Association between gene expression and progression free survival by Cox proportional hazard model

GeneHazard Ratio95% lower confidence limit95% upper confidence limitp-value
MYC 2.17 1.35 3.51 <0.01 
TNFRSF10B 0.46 0.22 0.96 0.04 
CDKN2A 0.65 0.43 1.00 0.05 
CCND1 0.72 0.52 0.98 0.04 
GeneHazard Ratio95% lower confidence limit95% upper confidence limitp-value
MYC 2.17 1.35 3.51 <0.01 
TNFRSF10B 0.46 0.22 0.96 0.04 
CDKN2A 0.65 0.43 1.00 0.05 
CCND1 0.72 0.52 0.98 0.04 
Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.