DEEP SEQUENCING IDENTIFIES TP53 MUTATIONS BEFORE THEIR CLONAL SELECTION BY THERAPY IN CHRONIC LYMPHOCYTIC LEUKEMIA

 Sborník

Background. Defects in the TP53 gene are associated with a particularly adverse prognosis in chronic lymphocytic leukemia (CLL). The p53 mutation and/or 17p deletion effect a poor chemotherapy response and shorter clinical responses to chemoimmunotherapy or alemtuzumab. The frequency of p53 inactivation is substantially higher in pretreated cohorts compared to those investigated before first-line therapy. This raises questions on the role of treatment in TP53 mutagenesis and specifically whether therapy only selects mutations or induces them de novo. Aims. The aims were to analyze i) frequency of new clonal TP53 mutations in untreated and treated CLL patients ii) in cases with new clonal TP53 mutation emerging after therapy, to analyze its presence before therapy using the ultra-sensitive deep sequencing. Methods. Clonal p53 mutations were identified by the yeast functional analysis (FASAY) and/or by direct Sanger sequencing. Presence of 17p- was assessed by I-FISH using probe from Abott-Vysis. The pyrosequencing on the GS Junior 454 platform (Roche) was performed to identify low-proportion (yet unselected) TP53 mutations. For the positions of the studied mutations, we achieved a median coverage of 6 657 reads (range 5 212-16 427). Mutations were reliably detected with the sensitivity reaching 0.2%. Results. We have repeatedly analyzed p53 status in 248 CLL patients which had an intact TP53 gene (no mutation and no 17p-) at first investigation; the median follow up was 33 months during which 60% of patients received therapy. The repeated investigations identified 31 patients with new p53 mutation (12.5%). All these mutations have emerged exclusively in patients that received treatment prior to detection of the p53- mutated clone; approximately each fifth treated patient acquired a new p53 mutation. Remarkably, missense mutations located in the p53 DNA-binding motifs (DBMs) - well defined and structurally important parts of the DNA-binding domain - were preferentially selected by therapy. Deep sequencing of pretherapy samples from 9 patients that subsequently developed p53 mutation has revealed that all mutations were present as minor clones undetectable by standard techniques (range 0.24% -13.03% of mutated reads) (P < 0.001); all mutations were identified from both the forward and reverse primers. After the therapy, all analyzed mutations were clearly detectable by deep-sequencing in a much larger proportion of cells. Our data shows that therapy selects “ready-touse” mutations; by other words, there is probably none or negligible de novo induction of new TP53 mutations by therapy in CLL. Summary and Conclusions. We demonstrate that selection of very minor clones of CLL cells harboring TP53 mutations is likely connected to the relapse of the disease and is a key event in evolution of certain refractory CLL cases. The detection of minor clones of CLL cells with p53 mutation prior to therapy could be of potential relevance for therapy selection and subsequent management of the disease relapse. This work was supported by the projects CZ.1.05/1.1.00/02.0068, CZ.1.07/2.3.00/20.0045, CZ.1.07/2.4.00/17.0042, and MSM0021622430.

Haematologica, 2012; 97(s1):  234