University dissertation from Division of Clinical Genetics

Abstract: Acute lymphoblastic leukemia (ALL) is characterized by an accumulation of immature lymphoid cells in the bone marrow and is the most common cancer type in children. It is an immunophenotypically, morphologically, clinically, and genetically heterogeneous disorder that comprises several distinct subtypes. Proper classification is important because determining the correct subtype plays a vital role for prognostication and treatment strategy. During the last decades, the use of polychemotherapy and implementation of risk stratification based on the presence of certain acquired genetic changes in pediatric B-cell precursor (BCP) ALL have increased the overall survival rates substantially; they are now approaching 90%. This notwithstanding, 20% of patients still relapse and only half of these survive. A considerable proportion of all relapses lacks the high risk-stratifying genetic changes included in most current ALL treatment protocols. Hence, it is important to identify novel genetic features associated with treatment failure to ensure proper therapy intensity and to detect genes and pathways that in the future can be targeted by specific drugs. To identify relapse-associated genetic aberrations in pediatric BCP ALL, single nucleotide polymorphism array analyses were performed on uniformly treated patients accrued between 1992 and 2011 from the Lund and Linköping University Hospitals (Article I). In the 191 successfully analyzed cases, deletions of IKZF1 (∆IKZF1) and SPRED1 were shown to be associated with a poor prognosis, with ∆IKZF1 being an independent risk factor for relapse. To ascertain whether ∆IKZF1 is an independent risk factor also in the context of minimal residual disease (MRD) findings, an extended cohort including all 334 Swedish pediatric BCP ALL cases with known IKZF1 status was investigated (Article II). That study confirmed that ∆IKZF1 confers a poor prognosis, revealed that such deletions are particularly common in cases with uninformative cytogenetics, and showed that the prognostic impact of ∆IKZF1 is independent of MRD stratification. However, coexisting genetic changes may play a role in modifying the pathogenetic and/or clinical impact of ∆IKZF1. Therefore, ∆IKZF1-positive cases were investigated further in Article III in order to identify additional, recurrent changes. Furthermore, targeted deep sequencing of all IKZF1 exons in 140 BCP ALL cases was performed, identifying sequence mutations (mutIKZF1) in 5.7%. Of the mutIKZF1-positive cases, one-fourth also harbored ∆IKZF1. In total, 35 cases with IKZF1 abnormalities (abnIKZF1), comprising ∆IKZF1 and/or mutIKZF1, could be analyzed with regard to other genetic anomalies. These analyses showed that CRLF2 rearrangements, caused by deletions of the pseudoautosomal region 1 (PAR1), and JAK2 mutations were significantly overrepresented in abnIKZF1-positive cases and that the presence of PAR1 deletions conferred a poor prognostic impact. Thus, in order to ascertain correctly the clinical ramifications of abnIKZF1 in pediatric BCP ALL, PAR1 deletions should possibly also be screened for.

  This dissertation MIGHT be available in PDF-format. Check this page to see if it is available for download.