Clinical implementation of chromosomal microarray analysis: summary of 2513 postnatal cases

Abstract

Background: Array Comparative Genomic Hybridization (a-CGH) is a powerful molecular cytogenetic tool to detect genomic imbalances and study disease mechanism and pathogenesis. We report our experience with the clinical implementation of this high resolution human genome analysis, referred to as Chromosomal Microarray Analysis (CMA).

Methods and findings: CMA was performed clinically on 2513 postnatal samples from patients referred with a variety of clinical phenotypes. The initial 775 samples were studied using CMA array version 4 and the remaining 1738 samples were analyzed with CMA version 5 containing expanded genomic coverage. Overall, CMA identified clinically relevant genomic imbalances in 8.5% of patients: 7.6% using V4 and 8.9% using V5. Among 117 cases referred for additional investigation of a known cytogenetically detectable rearrangement, CMA identified the majority (92.5%) of the genomic imbalances. Importantly, abnormal CMA findings were observed in 5.2% of patients (98/1872) with normal karyotypes/FISH results, and V5, with expanded genomic coverage, enabled a higher detection rate in this category than V4. For cases without cytogenetic results available, 8.0% (42/524) abnormal CMA results were detected; again, V5 demonstrated an increased ability to detect abnormality. Improved diagnostic potential of CMA is illustrated by 90 cases identified with 51 cryptic microdeletions and 39 predicted apparent reciprocal microduplications in 13 specific chromosomal regions associated with 11 known genomic disorders. In addition, CMA identified copy number variations (CNVs) of uncertain significance in 262 probands; however, parental studies usually facilitated clinical interpretation. Of these, 217 were interpreted as familial variants and 11 were determined to be de novo; the remaining 34 await parental studies to resolve the clinical significance.

Conclusions: This large set of clinical results demonstrates the significantly improved sensitivity of CMA for the detection of clinically relevant genomic imbalances and highlights the need for comprehensive genetic counseling to facilitate accurate clinical correlation and interpretation.

Figure 1

Example of reciprocal deletion/duplication CMA ratio plots with FISH validation. A Three sets of ratio plots showing examples of reciprocal microdeletion and microduplication at PWS/AS (on V5), NF1 (on V4) and DGS/VSFS (on V5) critical regions on chromosomes 15q11.2, 17q11.2 and 22q11.2, respectively. The vertical lines represent the BAC clones interrogating targeted regions of the genome, aligned according to their chromosomal position with the signal intensity ratios observed from the combined dye-swap experiments. Gains (G) are to the right and losses (L) to the left. Red circles highlight the signal ratios revealing genomic losses whereas the blue circles indicate copy number gains consistent with genomic deletions and reciprocal duplications, respectively. B FISH validation results. Control probes were labeled with a green chromophore and clones detecting deletions (in metaphase) or duplications (in interphase) were labeled in red for PWS/AS del/dup, NF1 del/dup and DG/VCFS del/dup.

Figure 2

Ideogram of Copy Number Variations (CNVs). Big arrows indicate CNVs Type 1 (N = 119), except AZFc on the Y chromosome (N = 35). * represents V4 and for V5 for CNVs Type 2 (N = 143). A total of 262 copy number variations (CNVs) were identified by CMA. Of these, 217 were interpreted as familial variants (green) and 11 were determined to be de novo (red); 34 CNVs that await parental studies are depicted with unknown clinical significance (black). Positions showing losses are on the left side of the chromosome whereas gains are on the right side.