Relapsed childhood high hyperdiploid acute lymphoblastic leukemia: presence of preleukemic ancestral clones and the secondary nature of microdeletions and RTK-RAS mutations

Abstract

Although childhood high hyperdiploid acute lymphoblastic leukemia is associated with a favorable outcome, 20% of patients still relapse. It is important to identify these patients already at diagnosis to ensure proper risk stratification. We have investigated 11 paired diagnostic and relapse samples with single nucleotide polymorphism array and mutation analyses of FLT3, KRAS, NRAS and PTPN11 in order to identify changes associated with relapse and to ascertain the genetic evolution patterns. Structural changes, mainly cryptic hemizygous deletions, were significantly more common at relapse (Po0.05). No single aberration was linked to relapse, but four deletions, involving IKZF1, PAX5, CDKN2A/B or AK3, were recurrent. On the basis of the genetic relationship between the paired samples, three groups were delineated: (1) identical genetic changes at diagnosis and relapse (2 of 11 cases), (2) clonal evolution with all changes at diagnosis being present at relapse (2 of 11) and (3) clonal evolution with some changes conserved, lost or gained (7 of 11), suggesting the presence of a preleukemic clone. This ancestral clone was characterized by numerical changes only, with structural changes and RTK-RAS mutations being secondary to the high hyperdiploid pattern.

Deletions of IKZF1 and SPRED1 are associated with poor prognosis in a population-based series of pediatric B-cell precursor acute lymphoblastic leukemia diagnosed between 1992 and 2011

“…Although such issues have been addressed previously, the current study provides additional and novel data pertaining to the four above-mentioned goals. First, we investigated a population-based series, not focusing solely on high risk ALL, 4 T-ALL, 15 specific cytogenetic subgroups in BCP ALL, 16,17 or excluding some subgroups. 13 Also, none of our patients was lost to follow-up in contrast to several prior studies and for some of our cases the observation time was close to 20 years with a median follow-up of 10 years, whereas the maximum follow-up time has been 10 years or less in earlier studies, 14,17,18,20,21 (Figures 1 and 2).…”

“…13 Also, none of our patients was lost to follow-up in contrast to several prior studies and for some of our cases the observation time was close to 20 years with a median follow-up of 10 years, whereas the maximum follow-up time has been 10 years or less in earlier studies, 14,17,18,20,21 (Figures 1 and 2). Furthermore, the SNP arrays used provide higher resolution (>1 M SNPs) compared with prior SNP array-based ALL studies (250K-500K), 4,[13][14][15][16][17][18] making it possible to delineate imbalances in greater detail and to identify previously unknown gene targets, as exemplified below.…”

“…Imbalances seen in remission samples or that overlapped with copy number polymorphisms listed in the Database of Genomic Variants (http://projects.tcag.ca/variation/) were excluded from further analysis, and so were deletions 6 corresponding to somatic rearrangements of the T-cell receptor and immunoglobulin loci. 16 A flowchart depicting the cases analyzed by SNP arrays is shown in Supplementary Figure 1.…”

“…A number of studies aiming at elucidating the underlying mechanisms for relapse has compared genetic features in paired diagnostic and relapse samples using conventional cytogenetic, 7,8 clone-specific polymerase chain reaction (PCR), 9,10 fluorescence in situ hybridization (FISH), 11,12 and single nucleotide polymorphism (SNP) array analyses, 4,[13][14][15][16][17] revealing that the relapsed clones can be identical to the ones seen at diagnosis, display additional changes (clonal evolution), or harbor both additional, identical as well as fewer changes (evolution from a preleukemic/ancestral clone). The two latter evolution patterns indicate outgrowth of therapy-resistant minor subclones.…”

“…The two latter evolution patterns indicate outgrowth of therapy-resistant minor subclones. 4,[9][10][11][12]16,17 In addition, high-resolution genomic profiling has identified genes, for example CDKN2A, ETV6, IKZF1, and EBF1, that are more frequently deleted in samples from patients who subsequently relapse than in samples from those remaining in fist complete remission (CR1). 4,13,14,18 However, whether these deletions serve as independent prognostic markers is still unclear.…”