Genomic Mechanisms of 17p / TP53 Loss in Primary “ultra High-risk” and Refractory Chronic Lymphocytic Leukemia: Results from the CLL2O Trial

Background: Unraveling the cytogenetic background helped to decipher the molecular basis of many hematologic cancers and to develop specific therapies. Recently, using chromosome banding analysis (CBA), jumping translocations were identified as a cause of 17p loss in multiple myeloma, providing new insights into the origin of clonal evolution and copy number alterations (CNA) (Sawyer et al, Blood 2014). In chronic lymphocytic leukemia (CLL) the genomic mechanisms leading to 17p loss are not fully understood.

Aims: Characterization of underlying mechanisms of 17p loss using CBA and correlation with other clinicobiological features in “ultra high-risk” CLL.

Methods: Samples from 112 patients (pts.) with refractory and/or 17p- CLL enrolled in the multicenter CLL2O trial were screened for CNAs by Affymetrix 6.0 SNP array analysis of CD19 sorted CLL cells and for chromosomal abnormalities by CBA using CpG oligonucleotide and interleukin-2 stimulation.

Results: Considering both CBA and SNP data, 728 aberrations resulted in a mean of 6.5/case. 89 (79%) pts. had 17p deletion and 83 (74%) TP53 mutation. Regarding the origin of 17p/TP53 loss, 6 distinct types of rearrangements could be delineated: 1) whole arm translocations (WAT) 2) jumping translocations (JT) 3) dicentric chromosomes (DC) 4) cytogenetically balanced translocations (CBT) 5) other unbalanced translocations and 6) interstitial 17p deletions. WAT were identified in 33/112 (30%) cases and 30/33 (91%) involved chromosome 17 leading to 17p loss. Chromosomes involved ≥ 2 times in an unbalanced WAT were der(17;18)(q10;q10) (8, 24%), der(8;17)(q10;q10) (5, 15%), der(15;17)(q10;q10) (4, 12%), i(17)(q10) (4, 12 %), der(17;22)(q10;q10) (2, 6%). JT were identified in 11 (10 %) cases, 6 showing jumping WAT with 17q as donor chromosome, 1 case with breakpoints located in the pericentromeric regions of chromosome 17p11 (donor chromosome) and the receptor chromosomes 4p14 and 16p11. In 4 cases, initially a WAT involving 17q occurred and subsequently the partner chromosome “jumped off” leaving a 17p deletion behind. DC were detected in 19 pts., 8 with breakpoint in 17p11, 7/8 with TP53 mutation. Of note, all cases had the breakpoint on chromosome 17 in 17p11 indicating a fragile site affecting the pericentromeric region. Interestingly, of a total of 382 translocations observed by CBA, only 32 were CBT and except for those involving the IGH and IGK/L loci (n=6) all were random. 17p involvement in CBT was detected in 4 cases, 3 had TP53 deletion and all were TP53 mutated. Of the unbalanced translocations, der(17)t(8;17) was identified in 5 pts. simultaneously generating 8q gain. Nevertheless, breakpoints on chromosome 17p covered cytobands 17p11-13 and on chromosome 8, 8q11-22, one case having the breakpoint telomeric to the TP53 locus and no TP53 mutation, pointing to other putative candidate genes on 17p. In 36/112 (32%) cases, 17p deletion was induced by random rearrangements. Interstitial 17p deletions were identified in only 9/112 (8 %) cases. According to the inclusion criteria of the trial, 36/112 (32%) pts. had 17p deletion and were treatment-naïve while 76/112 (68%) were relapsed or refractory to fludarabine or bendamustine based therapy, 53/76 (70%) having a 17p deletion. Treatment naïve pts. had a mean of 7.36 aberrations/case and pretreated pts. 6.09/case. Focusing on WAT and JT, 18/33 (54%) pts. with WAT and 7/11 (63%) pts. with JT were pretreated whereas 57/78 (73%) pts. in the other cytogenetic subgroups had prior therapy exposure. Considering other genomic features, WAT and JT occurred almost exclusively within complex karyotypes (≥3 chromosomal aberrations), 31/33 WAT and 10/11 JT, were IGHV unmutated, 30/33 WAT and 11/11 JT and harbored TP53mutations, 29/33 WAT and 10/11 JT.

Conclusions: “Ultra high-risk” CLL pts. are characterized by a high genomic complexity as compared to standard risk treatment-naïve CLL pts. (CLL8 trial with 1.8 CNAs/case). Previous genotoxic therapy had no influence on the total number of aberrations or the underlying mechanism, suggesting an intrinsic genomic instability of the tumor cells with TP53 alterations. WAT and JT emerged as nonrandom aberrations involved in 17p loss. Given the strong association of TP53 deletion with TP53 mutations of the remaining allele, one may speculate that TP53 mutations precedes TP53 deletion by disrupting the normal DNA repair mechanisms permitting incorrect recombinations.