Chromothripsis in a case of TP53-deficient chronic lymphocytic leukemia

Abstract

We describe genomic findings in a case of CLL with del(17p13.1) by FISH, in which SNP array analysis revealed chromothripsis, a phenomenon by which regions of the cancer genome are shattered and recombined to generate frequent oscillations between two DNA copy number states. The findings illustrate the value of SNP arrays for precise whole genome profiling in CLL and for the detection of alterations that would be overlooked with a standard FISH panel. This second report of chromothripsis in CLL indicates that this phenomenon is a recurrent change in this disease.

1. Introduction

Because lymphocytes of chronic lymphocytic leukemia (CLL) patients often fail to divide sufficiently in vitro, fluorescence in situ hybridization (FISH) probes have become a staple in the analysis of chromosomal aberrations in CLL (reviewed in Ref. 1). In a groundbreaking study, a FISH probe panel was used to classify CLL patients into one of five prognostic groups, with the poorest prognosis seen in patients with deletion of 17p13.1 (site of TP53), followed by loss of 11q22.3 (ATM), trisomy 12 and normal FISH results, whereas patients with del(13q14) as the only alteration had the most favorable prognosis.²

Although FISH analysis is now widely used for the cytogenetic assessment of CLL, other approaches such as oligonucleotide-based array comparative genomic hybridization and single nucleotide polymorphism (SNP) gene chips show comparable results, but also assess all chromosomal regions rather than the current standard clinical practice of identifying alterations with probes targeting only 4–5 chromosomal sites.^3,4 The high-resolution gene copy number analysis afforded by DNA arrays provides greater precision in demarcating boundaries of individual genomic imbalances and uncovers alterations overlooked by FISH analysis. In this report, we describe cytogenetic findings in a CLL case with del(17p13.1) by FISH in which SNP arrays detected profound genomic upheaval due to chromothripsis, a phenomenon by which a specific region(s) of the cancer genome is shattered into pieces and then recombined via a single devastating event, generating frequent oscillations between two DNA copy number states, that may lead to malignant transformation.⁵

2. Materials and methods

Karyotypic analysis was performed on a blood sample cultured for 24 h. FISH analysis was performed using Abbott's Vysis CLL FISH Probe Kit. DNA copy number and allele analysis were performed as described in detail elsewhere.⁶ Briefly, total genomic DNA from blood was Nsp-digested, ligated with an adaptor complimentary to PCR primer, PCR amplified, purified, fragmented, biotin labeled, and hybridized to an Affymetrix CytoScan HD array, according to the manufacturer's protocol. The hybridized array was washed and then scanned with a GeneChip Scanner 3000 7G prior to data analysis.

3. Results

Karyotyping revealed chromosomal abnormalities in 19 of 20 metaphase spreads examined. All abnormal metaphases had a del(14)(q22), and 17 metaphases were interpreted to contain a three way translocation, t(2;5;7)(q31;q21;q11.2) and loss of all or part of chromosome 6. FISH analysis showed loss of one copy of TP53 in 95 of 150 nuclei examined and loss of MYB (at 6q23.3) in 10 of 150 nuclei. No abnormalities were detected for ATM (11q22.3), D12Z3 (12cen), D13S319 (13q14.2-q14.3), or LAMP1 (13q34).

DNA copy number analysis revealed a strikingly complex pattern, including chromothripsis involving several chromosomes, numerous tiny focal deletions in other chromosomes, and larger deletions in 14q and 17p (Fig. 1). The del(17p) included TP53, consistent with the FISH analysis. As is typical of chromothripsis, multiple oscillations between two DNA copy number states were observed. In this case, the changes were almost exclusively fluctuations between one and two copies, resulting in discontinuous deletions of numerous segments in chromosomes 2, 5, 6 and 7.

Fig. 1.

Depiction of DNA copy number and allele peaks in the abnormal chromosomes identified in blood sample from CLL patient. For each chromosome shown, the y axis depicts DNA copy number (Upper) and allele peaks (Lower). Allele peak panels normally show three distinct “bands,” representing all homozygous (Top and Bottom bands) and heterozygous (Middle band) allele calls. Note multiple oscillations between two DNA copy number states in chromosomes 2, 5, 6 and 7, indicative of chromothripsis. Numerous discontinuous deletions are indicated by blue arrows, with allele peak panels showing only two bands (loss of heterozygosity) at chromosomal segments corresponding to each deletion. In addition to these focal deletions, larger single deletions can be seen in chromosome arms 14q and 17p, the latter encompassing the TP53 locus (green arrow). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

4. Discussion

While the DNA copy number data were in agreement with the FISH findings, the genome-wide DNA copy number analysis also uncovered many other abnormalities indicative of chaotic genomic disruption. Chromothripsis is thought to occur during a single cellular catastrophe and may occur in at least 2%–3% of all cancers, across many subtypes.⁵ Interestingly, the very first reported evidence of chromothripsis was found via a genome-wide sequencing screen of 10 CLL patients, one (a previously untreated 62-yr-old woman) showing massive rearrangement of chromosome arm 4q and focal alterations of chromosomes 1, 12, and 15.⁵ The copy number changes in 4q alternated between 1 and 2, with regions of copy number 1 showing loss of heterozygosity (LOH), whereas regions of copy number 2 retained heterozygosity. The whole genome sequencing revealed that the many discontinuous regions of copy number 1 in 4q were not caused by simple deletions but instead were the result of a series of complex rearrangements spanning 4q. In that same report, subsequent study uncovered a second CLL patient with losses of single copies of CDKN2A and miR-15a/16-a due to a cluster of interchromosomal rearrangements involving chromosomes 4, 9, and 13. In our patient, the discontinuous regions of LOH in chromosomes 2, 5, 6 and 7 are also separated by regions retaining heterozygosity (Fig. 1). Notably, the clinical course in the initial CLL patient reported by Stephens et al.⁵ showed rapid deterioration and quickly relapsed. Sequence analysis of a relapse specimen collected 31 months after diagnosis revealed no new genomic rearrangements, suggesting that the process generating this complex genomic remodeling had occurred before the patient was diagnosed and was not indicative of further genomic instability.

As noted above, while FISH evidence for a deletion of chromosome 17p13.1 (TP53) is generally associated with a poor prognosis, CLL patients with del(17p13.1) exhibit clinical heterogeneity. Some TP53-deficient CLL patients have a relatively indolent course. One reason for such heterogeneity may be that not all cases with loss of one copy of TP53 have a mutation in the remaining allele. Thus, there is insufficient evidence for rigid therapeutic recommendations to be made based on deletion of 17p alone, although survival may be predicted using other risk factors for progressive disease including the presence of unmutated immunoglobulin heavy-chain variable region genes (IGHV).⁷ As in other hematological malignancies, the genetic profile of CLL may evolve during the disease course. For example, while del(17p13.1) is uncommon at the time of diagnosis (5%–10%), it is observed in about 45% of patients with relapsed or refractory CLL.⁷ In del(17p13.1) CLL cases, clonal evolution with acquisition of other genomic imbalances is associated with unmutated IGVH, resistance to therapy, and short survival.⁸ As array-based DNA copy number analysis becomes more widely used in the clinical setting, it will be important to determine if chromothripsis represents yet another risk factor of progressive disease, as we predict. Unfortunately, we do not know whether or not the disease course was rapid in our CLL patient, because the DNA was from a de-identified sample that was used for a validation study of a DNA copy number assay.

The deletion of 14q seen in our patient included the IGH locus at 14q32.33. This is noteworthy, because loss of 14q has been reported as a recurrent aberration acquired during the natural history of CLL.⁹ The acquisition of a del(14) may thus be indicative of CLL in transformation. In a study of a large German cohort, del(14q) was found in ∼2% of newly diagnosed CLLs, and these patients typically required more immediate therapy.¹⁰ The investigators used a panel of 14q probes to determine that the del(14) was usually interstitial and heterogeneous in size, with a breakpoint cluster at the centromeric site in 14q24.1; in most cases the breakpoint at the telomeric site was within the IGH locus while ∼25% showed a deletion of the entire IGH locus, as was the case in our patient.

In conclusion, this report illustrates the value of SNP arrays for precise whole genome profiling of chromosomal alterations in CLL and the detection of complex alterations that would be overlooked with a standard FISH panel. The identification of this third case of chromothripsis in CLL indicates that this phenomenon is a recurrent change in this disease, and future studies may determine if its occurrence is associated with distinctive clinical features.

Conflict of interest

The authors have no conflicts of interest to declare.