Comprehensive Profiling of Disease-Relevant Copy Number Aberrations Improves Risk Assessment and Unveils the Clonal Origin of Relapse in Pediatric Acute Lymphoblastic Leukemia

Introduction: Acute lymphoblastic leukemia (ALL) is the most common pediatric malignancy characterized by a heterogeneous genomic landscape. Copy number aberrations (CNA) emerge during the development, progression and treatment resistance of ALL, and can serve as genomic markers for prognostic classification of patients or for scrutinizing clonal evolution associated with relapse. While identification of distinct CNAs with well-characterized prognostic significance has its own value, uncovering the co-segregation of driver aberrations in individual patient samples could allow for a more personalized risk assessment and treatment response prediction.

Methods: Disease-relevant CNAs were profiled in children with B- or T-cell precursor ALL using a next-generation sequencing based digital multiplex ligation-dependent probe amplification (digitalMLPA) assay containing 598 probes specific for 54 genes with key relevance in ALL. Besides the diagnostic samples of 91 patients treated according to the BFM protocols, 14 matching samples drawn at the time of first or second relapse were comparatively analyzed. Clonal relationship between B-cell precursor cell populations prevailing at different time points during the disease course was also investigated by screening immunoglobulin heavy-chain gene rearrangements in matching diagnostic and relapse samples using Illumina deep-sequencing with >20,000x coverage.

Results: Whole chromosome gains and losses, subchromosomal CNAs as well as alterations conferring intrachromosomal gene fusions were simultaneously identified by digitalMLPA with results available within 36 hours. Aberrations were observed in 96% of diagnostic patient samples and increased numbers of CNAs were detected in individual samples at the time of relapse as compared to diagnosis. DigitalMLPA results were successfully validated by conventional MLPA, FISH and PCR data.

Comparative scrutiny of 24 matching diagnostic and relapse samples from 11 patients harboring CNAs revealed three different patterns of clonal relationships with (i) one patient displaying identical CNA profiles at diagnosis and relapse, (ii) six patients showing clonal evolution with all lesions detected at diagnosis being present at relapse and (iii) four patients displaying conserved as well as lost or gained CNAs at the time of relapse, suggestive of the presence of a common ancestral cell compartment giving rise to clinically manifest leukemia at different time points during the disease course. Time between diagnosis and first relapse of T-ALL patients displaying altered CNA profiles suggested a prolonged time requirement of clonal evolution, and of the development of manifest leukemia from an ancestral clone compared to the quick return of an identical clone at the time of relapse.

Comparison of the IGH gene rearrangements identified at diagnoses and relapse revealed identical compositions of the most abundant clonotypes in all but one B-ALL patients analyzed; hence, IGH repertoire did not reveal an additional depth of clonal history in our cohort, e.g. by demonstrating the presence of an ancestral clone as the major source of clonal expansion at disease progression in a patient with altered CNA profiles suggesting direct clonal evolution between diagnosis and relapse.

Copy number profiles acquired by digitalMLPA were used for determining CNA-based risk groups (Table 1) which were combined with karyotyping and molecular cytogenetic data in order to establish an extended prognostic classifier for patients with B-cell precursor ALL. This novel classifier distinguished four combined genetic risk groups showing significantly different 5-year survival rates (GR: 97%, IR: 84%, IHR: 63% and PR: 13%).

Conclusions: DigitalMLPA allows for a rapid, scalable and highly optimized copy number profiling of genomic regions recurrently altered by driver aberrations in pediatric ALL. Based on the comparison of CNA profiles at diagnosis and relapse, clonal evolution and emergence of relapse from an ancestral clone are the predominant driving mechanisms of disease progression. Comprehensive copy number profiling by digitalMLPA identifies distinct prognostic groups for risk assessment in B-cell precursor ALL.

Supporting grants: LP95021, K_16 #119950, NVKP_16-1-2016-0004, KH17-126718, BO/00320/18/5, FK_19 #131476, ÚNKP-19-4-SE-77

View largeDownload slide

View largeDownload slide

Close modal