Chronic lymphocytic leukemia: assessing pathogenesis and prognosis by modern molecular cytogenetic studies and microRNAs expression

Chronic lymphocytic leukemia (CLL) is a B-cell clonal lymphoprolipherative disorder characterized by the accumulation of small lymphocytes in the peripheral blood, bone marrow and lymph nodes deriving from the transformation of CD5+ B-cell. Despite a homogeneous immunophenotype consisting of CD19+, CD20+, CD5+ and CD23+, CLL is clinically heterogeneous. Several adverse prognostic features have been identified including stage, CD38 positivity, the unmutated configuration of the variable region of the immunoglobulin heavy chain gene (IGHV), ZAP70 positivity, chromosome aberrations and molecular abnormalities. Detailed immunophenotypic and genetic analysis allowed for the identification of a number of markers of activation and genetic instability, some of which are gaining relevance in clinical practice to predict outcome. Cell surface CD38 is one of these markers since it is an indicator of cell activation and proliferation that may prelude clonal evolution and worse clinical outcome. We therefore studied the biological and clinical significance of the presence of genetic heterogeneity in the minor CD38+ leukemic population, in a cohort of untreated low-risk CD38-negative CLL patients, defined by the presence of <7% CD38+ cells, and by the absence of unfavourable genetic lesions. Our data showed that a significant proportion of CD38- CLL patients with low risk FISH findings presented genetic aberrations within CD38+ cells. Most of these abnormalities were high risk lesions (11q deletion and 17p deletion) and, in most of the cases, these lesions were found in different cells indicating that multiple cytogenetically unrelated minor clones may be present in the CD38+ cell fraction. Interestingly, the presence of these additional FISH lesions in the small CD38+ cell fraction was associated with shorter time to first treatment (TTT). To identify biomarkers associated with this phenomenon, we performed miRNA expression analysis. We were thus able to show a deregulated miRNA expression profile in CLL cases with additional FISH lesions in CD38+ cells. In particular, miR-125a-5p was found to be downregulated both in CD38+ and CD38- cells in patients with FISH abnormal clones as compared to patients without FISH abnormal clones. The relevance of miR-125a-5p as a biomarker of inferior outcome and genetic complexity was then validated in a prospective cohort. In this validation cohort, we were able to confirm the predictive role of miR-125a-5p down-regulation in terms of shorter TTT. In addition we found that CLL patients with lower levels of miR-125a-5p displayed an increased rate of mutations in CLL-related genes by next-generation sequencing. Several recent studies have shown that CD38 expression is higher in CLL cells in the bone marrow and lymphoid tissues, especially in the proliferation centres (PCs), which are regarded as the histologic hallmark of this disease. Indeed CLL is a disease in which the host’s microenvironment promotes leukemic cell growth, leading to sequential acquisition and accumulation of genetic alterations and proliferation centers may play an important role in the biology of CLL, as they represent its proliferative compartment. To better define the significance of proliferation centers, we studied lymph node biopsies taken from a cohort of patients by fluorescence in situ hybridization (FISH) studies using a 5-probe panel on tissue microarrays (TMA). The cases were classified into two categories: “PCsrich” and “typical”. The PCs-rich group was associated with 17p-, 14q32/IGH translocations and +12. The median survival from the time of TMA for PCs-rich and typical groups was 11 and 64 months respectively. The PCs-rich pattern was the only predictive factor of an inferior survival. These findings establish an association between cytogenetic profile and the amount of PCs in CLL, and show that this histopathologic characteristic is of value for risk assessment in patients with clinically significant adenopathy. CLL turned out to be a disease with multiple facets in its pathogenic mechanisms including genetic aberrations, antigen drive and microenvironmental interactions. In the first part of this work, we focused our attention on the correlation between CD38-positivity, proliferation centres and development of genetic aberrations. To translate this knowledge in clinical practice we planned further studies focusing i) on the correlation between chromosomal aberrations and clonal evolution and ii) on how to stratify patients into different risk-groups at diagnosis according to cytogenetic abnormalities and gene mutations. The presence of cytogenetic abnormalities is a hallmark of CLL. It was reported that a fraction of CLL patients developed new cytogenetic abnormalities at chromosome sites of known prognostic importance during the course of their disease (clonal evolution, CE). To better define the incidence and signature of CE, a cohort of patients were analysed sequentially by FISH. Recurring aberrations at clonal evolution were 14q32/IGH translocation, 17p-, 11q-, 13q- and 14q32 deletion. The development of CE occurred only in previously treated patients. Our data show that the 14q32/IGH translocation may represent one of the most frequent aberrations acquired during the natural history of CLL. CE occurs in pre-treated patients with short TTT and survival, after the development of CE with and without 14q32 translocation, is relatively short. Having assessed the incidence of chromosome aberration in CLL with evolution and/or adenopathy we next moved to CLL with an apparently “favourable” profile of cytogenetic lesions, to establish if improved cytogenetic techniques could help refine prognostication. We therefore designed a study to assess whether karyotypic aberrations in patients without FISH anomalies correlate with established clinical and prognostic parameters. The clinical and prognostic significance of karyotypic aberrations in normal FISH CLL was first evaluated in a retrospective single centre series of patients and then validated prospectively in a multicentre series of cases diagnosed and analysed for karyotype with DPS30/IL2 stimulation. Conventional karyotyping using DSP30/IL2 stimulation is an effective method for the detection of chromosome aberrations in approximately one third of CLL with normal FISH. The abnormal karyotype correlated with shorter time to first treatment and shorter survival. This set of data also showed that, in CLL patients with normal FISH, conventional cytogenetic analysis identifies a subset of cases with adverse clinical and prognostic features to be considered for the design of risk-adapted treatment strategies. In the last part of our experimental work we moved from the consideration that the cytogenetic lesions do not entirely explain the molecular pathogenesis and the clinical heterogeneity of CLL. Indeed, the advent of next‐generation sequencing (NGS) technologies has enabled exploration of the CLL genome, uncovering genetic lesions that recurrently target the leukemic cells. NGS studies have further elucidated the genomic complexity of CLL. In order to improve understanding of genetic basis of CLL and to apply NGS to CLL, we sequenced DNA samples from untreated patients affected by chronic lymphocytic leukemia with a panel of 20 genes and we correlated mutational status with clinicobiological parameters.Mutations were identified in the following genes: TP53, SF3B1, POT1 , ATM , MYD88 , FBXW7 , MAPK1 , DDX3X , KLHL6 , KRAS. The presence of mutations correlated with high risk FISH (11q- and/or 17p-) and unfavourable cytogenetic (11q-, 17p- or complex karyotype) findings. Patients carrying CLLs with gene mutations showed a significant shorter median time to first treatment in comparison to those without mutations (20 months vs not reached at 76 months). The frequency of mutations in the 20 investigated genes is in line with published data in the literature using whole exome sequencing. This study shows that the simultaneous sequencing of a panel of genes implicated in CLL is feasible. In conclusion, in this work we tried to improve our knowledge on some fundamental pathogenetic aspects of CLL, including: i) the development of genetic lesions in CD38+ activated cells, obtained from the PB in patients in an initial and indolent phase of the disease; ii) the pattern of cytogenetic lesions in lymph node microenvironment (proliferation centres) in patients in a more advanced phase of the disease; iii) to translate this knowledge in clinical practice, we assessed prognosis with modern techniques and we identified cytogenetic lesions associated with disease progression and shorter time to treatment; iv) we identified recurrent genetic lesions potentially useful for further refinement of our ability to predict prognosis and response to treatment.