4.6 Detection and Analysis of TP53 and ATM Deletions by Multiplex Ligation-Dependent Probe Amplification in CLL

Abstract

Genomic copy number aberrations (CNAs) are a characteristic feature of chronic lymphocytic leukemia (CLL). The presence of specific deletions, in particular 17p13(TP53) and 11q22-23(ATM), identifies subsets of patients with a shortened overall survival and a poor response to purine analogue/alkylating agent-based therapy. Consequently, the detection of these abnormalities is pivotal to disease management. Multiplex ligation-dependent probe amplification (MLPA), a molecular technique that can simultaneously detect CNAsatmultiplelociinasinglepolymerasechainreaction,hasbeen highlighted as an accurate, low-cost, high-throughput alternative to fluorescence in situ hybridization (FISH) for the detection of these abnormalities in CLL. Currently, the number of CLL cases with deletions of TP53 and ATM analysed by MLPA are too few to reliably assess the efficacy of this technique for their detection. In this study, MLPA was assessed for the detection of TP53 and ATM deletionsinprimarytumoursamples.Unsorted,densitygradient‐separated peripheral blood mononuclear cells from 142 selected patients with CLL were screened with 2 commercially available MLPA kits (MRC Holland SALSA-P037 and P038), comprising 80 locusspecific MLPA probes. A total of 75 of 142 patients had TP53 deletion and 63 of 126 had ATM deletion, with clone sizes ranging from 6% to 97% for both abnormalities. We established a normal range for each MLPA probe as the mean ratio 2 SD from analysis of 36 normal individuals. Owing to high variability in the normal individuals, 6 of 80 probes, including 2 of 8 TP53 and 1o f 6ATM probes, were excluded from analysis of the CLL patient data. Each of the remaining 74 probes was scored as showing loss or gain if their ratio laybeloworabovethenormalrange,respectively.Sixtyof75patients with TP53 deletion detected by FISH had 4 or more of 6 TP53 probesshowingloss,and50of67patientswithoutTP53deletionby FISH had 0 of 6 probes showing loss. Using these 2 cut-offs resulted in a positive predictive value (PPV) for TP53 deletion of 60 of 60 (100%, 95% CI 93%‐100%) and for the absence of TP53 deletion of 50 of 52 (96%, 95% CI 86%‐100%). The 2 cases misclassified as having no TP53 loss by MLPA had clone sizes of 7% and 9%. The remaining 30 of 142 cases (13 TP53 deleted and 17 TP53 nondeleted) with 1 to 3 TP53 probes showing loss were unclassified by MLPA. These cases had FISH clone sizes of 9% to 25%. Altogether, 48/63 cases with deletion of 11q by FISH had 3 or more of 5 ATM probes showing loss and 56 of 63 without ATM deletion by FISH had 0 of 5 probes showing loss. Using these 2 cut-offs resulted in a PPVforATMdeletionof48of48(100%,95%CI91%‐100%)and for the absence of ATM deletion of 56 of 61 (92%, 95% CI 82%‐