Customized oligonucleotide array-based comparative genomic hybridization as a clinical assay for genomic profiling of chronic lymphocytic leukemia

Abstract

Chromosome gains and losses used for risk stratification in chronic lymphocytic leukemia (CLL) are commonly assessed by multiprobe fluorescence in situ hybridization (FISH) studies. We designed and validated a customized array-comparative genomic hybridization (aCGH) platform as a clinical assay for CLL genomic profiling. A 60-mer, 44,000-probe oligonucleotide array with a 50-kb average spatial resolution was augmented with high-density probe tiling at loci that are frequently aberrant in CLL. Aberrations identified by aCGH were compared with those identified by a FISH panel, including locus-specific probes to ATM (11q22.3), the centromeric region of chromosome 12 (12p11.1-q11), D13S319 (13q14.3), LAMP1 (13q34), and TP53 (17p13.1). In 100 CLL samples, aCGH/FISH concordance was seen for 89% of FISH-called aberrations at the ATM (n=18), D13S319 (n=42), LAMP (n=12), and TP53 (n=22) loci and for chromosome 12 (n=14). Eighty-four percentage of FISH/aCGH discordant calls were in samples either at or below the limit of aCGH sensitivity (10% to 25% FISH aberration-containing cells). Therefore, aCGH profiling is a feasible routine clinical test with comparable results to multiprobe FISH studies; however, it may be less sensitive than FISH in cases with low-level aberrations. Further, a customized array design can provide comprehensive genomic profiling with additional accuracy in both identifying and defining the extent of small aberrations at target loci.

Figure 1

aCGH data for FISH-validated loci in CLL. A single case is shown that has deletions spanning the ATM locus at chr 11q 22.3 (91%), the D13S319 locus at chr 13q14.3 (94%), and the TP53 locus at chr 17p13.1 (87%). The thin blue line represents a 50 megabase moving average. Green/Red dots indicate less or greater than Log2 ratio change of 0.5. The horizontal blue bar represents regions called statistically significant by CGH Analytics with an ADM-2 threshold of 6.0.

Figure 2

Discordance between aCGH calls and FISH results largely relate to differential sensitivity of the assays. A case with 22% of cells demonstrating a D13S319 loss by FISH that had the aberration called by low (top array panel) but not high (bottom array panel) stringency analysis settings. The thin blue line represents a 50 megabase moving average. Green/Red dots indicate less or greater than Log2 ratio change of 0.5. The horizontal blue bar represents the region called statistically significant by CGH Analytics with an ADM-2 threshold of 4.0. CGH Analytics settings used in the bottom panel are as described in Figure 1.

Figure 3

A: A case with 96% of cells exhibiting an ATM gene deletion at 11q22.3 (not illustrated), 57% of cells exhibiting three green CEP12 signals (3G), 37% of cells exhibiting a deletion of one red signal (1R) at the D13S319 locus, and two aqua signals (2A) representing the presence of two 13q34/LAMP1 loci by multiprobe FISH. B: The conventional karyotype for this case was: 39–46,XY,del(1)(q22), add(2)(p25), −7, del(11)(q22q25), add(12)(p11.2), −18,+1–2mar[cp8], and diploid male karyotype 46, XY CP5. Note in this representative karyotype, some of the karyotypic alterations, including the trisomy 12 seen by FISH, are not clearly identified; however, the possibility of additional chromosome 12 material, including the centromeric portion of the chromosome, in the marker chromosomes cannot be excluded. C: Although the 11q22.3 deletion and the D13S319 deletion were detected at both high and low stringency analysis settings, no detectable aberration was seen in chromosome 12 by aCGH at either the low or high stringency analysis settings. CGH Analytics settings for this analysis are as described in Figure 1.

Figure 4

Mapping of the extent of deletions at chr 13q14 in CLL. Penetrance plot of the 41 cases in both the pilot and validation studies that had detectable deletions spanning the D13S319 locus at chr 13q14.3. Using Nexus software settings detailed in the Materials and Methods, variable sized deletions are observed. The area of analysis spans chromosome 13 from 13q14.11 to 13q14.3. The percentage penetrance for all cases within the area of analysis is illustrated by the solid red area. The open boxes and solid bars below the percentage penetrance represent genes within the area of analysis. The horizontal pink lines represent copy number variants within the area of analysis. The horizontal red lines at the bottom of the figure represent the extent of deletion within the area of analysis for each individual case. Please refer to Table 5 for the genomic coordinates and deletion size for each individual case.