Detection of copy number aberrations in cell free DNA from plasma of neuroblastoma patients using shallow massive parallel sequencing

Neuroblastoma (NB) is a heterogeneous disease characterized by distinct clinical and biological features. The somatic mutation burden is low in these tumors, while copy number alterations are highly recurrent. Low stage tumors usually present with numerical chromosome aberrations, while high stage tumors are characterized by the presence of segmental aberrations, the latter being associated with poor prognosis. Presently, determination of DNA copy number profiles is mandatory in most treatment protocols and determined using FISH, MLPA, array-CGH or SNP-arrays. Such analysis can be precluded through the inability to biopsy the tumor or lack of sufficient tumor material. In this study, we investigated whether cell free DNA (cfDNA) isolated from plasma samples of NB patients could offer an alternative to biopsies, and whether shallow massive parallel sequencing (MPS) could be used as a substitute for classical DNA copy number analyses. CfDNA was isolated from plasma using the QiaAmp circulating nucleic acid kit (Qiagen). Samples were preprocessed with the Ion Plus Fragment Library kit, barcoded, pooled by six and sequenced on an Ion Proton (ThermoFisher) instrument. Data analysis for CNV detection was done using the QDNAseq algorithm implemented in the online genomic data visualisation tool Vivar. The minimal number of reads per sample was set at 10 million. Comparison of the MPS data with the array-CGH profiles of the tumors from the same patient showed that all structural and numerical chromosome aberrations including MYCN amplification could be readily detected with MPS using a very low input of cfDNA (5 ng). Interestingly, MPS data show a much better signal-to-noise ratio compared with array data. In conclusion, MPS analysis of cfDNA isolated from plasma offers a cost-effective, non-invasive and rapid alternative for DNA copy number profiling in neuroblastoma