CHRONIC LYMPHOCYTIC LEUKEMIA
This is the most common form of leukemia accounting for >30 % of all cases and occurs particularly in the elderly, with a peak incidence between 60 and 80 years of age. Most cases are B cell type. This type of leukemia has the highest familial risk of all leukemias, with relatives of cases at significantly increased risk for CLL RR = 7.52 (3.63–15.56), for non-Hodgkin lymphoma RR 1.45 (0.98–2.16), and Hodgkin lymphoma RR 2.35 (1.08–5.08) (Goldin et al. 2010; Yuille et al. 2000). Moreover, approximately 13 % of apparently healthy relatives of patients with familial CLL have a monoclonal blood B lymphpocyte population detectable by cell flow analysis (3 % in controls).
In contrast to acute leukemia and chronic myeloid leukemia, there is abundant evidence for a role of genetic factors in chronic lymphocytic leukemia. An association between disordered immune function and lymphoreticular malignancy is suggested by the finding that in some reports of familial chronic lymphocytic leukemia, autoimmune or immunological disorders have been noted among unaffected relatives. There is a fourfold excess of hematoproliferative malignancies among siblings of patients with chronic lymphocytic leukemia. Case-control studies have revealed relative risks in relatives ranging from 2.3 to 5.7 (reviewed by Houlston et al. 2003). Among familial cases of chronic lymphocytic leukemia, the mean age at onset was >10 years earlier than in sporadic cases, and an increased risk of second primary Tumours was noted (Ishibe et al. 2001). Among 18 affected individuals from seven pedigrees with dominantly inherited chronic lymphocytic leukemia, Horwitz et al. (1996) described evidence for anticipation in most cases, and this has been confirmed by others. The age at diagnosis in the offspring appears to be >20 years earlier than in affected parents.
Rawstron et al. (2002) detected subclinical levels of chronic lymphocytic leukemia-like cells in 14 % of relatives compared to about 1 % of normal controls suggesting that although the lifetime risk of chronic lymphocytic leukemia in familial cases is 20–30 %, there may be a significant incidence of subclinical disease.
The relative risks of lymphoproliferative disease in first-degree relatives of CLL cases have been found to be 1.8 for B cell NHL, 2.2 for indolent B cell NHL, 1.6 for follicular lymphoma, 3.3 for hairy cell leukemia, and 4 % for LPL/Waldenstroms macroglobulinemia (Goldin et al. 2010).
Rare high-penetrance and more common lower-penetrance susceptibility alleles are thought to be implicated in familial chronic lymphocytic leukemia. Recent genome-wide association studies have found several low-penetrance risk alleles for CLL (DiBernardo et al. 2008; Crowther-Swanepoel et al. 2010) and ALL each conferring small relative risks of disease (1.2–1.7 per allele), which act independently, such that the 2 % of the population who carry 13 or more risk alleles would have an eightfold increased risk of disease. Interestingly, these variants are involved in lymphoid cell development (Houlston 2010).
Germline mutations in DAPK1, a Ca2+/calmodulin-dependent serine/threonine kinase that acts as a positive regulator of apoptosis in part via phosphorylation of p53 causing loss or reduced expression of DAPK1, cause an inherited susceptibility to CLL (Lynch et al. 2002, 2008b; Raval et al. 2007). DAPK1 expression of the CLL allele may be downregulated due to increased HOXB7 binding, and promotor methylation results in additional loss of DAPK1 expression.
Somatic inactivation of the ataxia-telangiectasia gene (ATM) occurs in >20 % of chronic lymphocytic leukemia cases, and in some cases. a germline mutation is also present. However while heterozygous ATM mutations may confer an increased risk of chronic lymphocytic leukemia, it seems likely that germline ATM mutations do not make a major contribution to familial cases of chronic lymphocytic leukemia (Houlston et al. 2003).
Wiley et al. (2002) reported that a loss-of-function polymorphism in the cytolytic P2X7 receptor gene was overrepresented in patients with chronic lymphocytic leukemia compared to controls. However in another study, the P2X7 SNP was associated with survival in, but not risk of chronic lymphocytic leukemia (Thunberg et al. 2002).
Familial clustering (sibling or parent–child pairs) of hairy cell leukemia, an uncommon subtype of chronic lymphocytic leukemia with a prevalence of 1 per 150,000, has been reported in at least 30 cases (Colovic et al. 2001; Cetiner et al. 2003). Linkage to specific HLA haplotype has been suggested, but the influence of genetic and environmental factors in familial cases is unclear.
The first-degree relative of an affected person may have an eightfold increased risk of developing CLL themselves, and a 2.6 relative risk of any other lymphoproliferative disorder, but considering the population risk of these conditions is low, the absolute risk is low, and early detection of CLL is not likely to affect outcome, therefore surveillance would not be recommended.
Other subtypes of leukemia which may occur commonly in certain genetic conditions include juvenile myelomonocytic leukemia (JMML), where about 10 % of cases arise in children with neurofibromatosis type 1 (NF1), and in Noonan syndrome, a condition due to germline gain of function mutations in one of the oncogenes in the RAS signaling pathway, including PTPN11, SOS1, RAF1, KRAS and CBL (Pandit et al.
2007). Children with Noonan syndrome (NS) are at increased risk of developing juvenile myelomonocytic leukemia or a myeloproliferative disorder associated with NS (MPD/NS) resembling JMML in the first weeks of life. JMML constitutes about 30 % of childhood cases of myelodysplastic syndrome and 2 % of leukemia. Whereas JMML is an aggressive disorder requiring hematopoietic stem cell transplantation, MPD/NS may resolve without treatment, and cases with spontaneous remission have also been reported (Bastida et al. 2011).
CBL is an E3 ubiquitin protein ligase responsible for the inactivation of protein tyrosine kinases by tagging them for degradation. Somatic mutations in this gene, as with PTPN11, KRAS, and NRAS, are well characterized in a variety of leukemia, particularly Juvenile myelomonocytic leukemia (JMML) and CML. Germline mutations in the CBL gene have been identified in a very small proportion of Noonan syndrome patients with a predisposition to
JMML (Niemeyer et al. 2010). These children have impaired growth, developmental delay, and cryptorchidism.
Legius syndrome, due to germline mutations in the SPRED1 gene, is a syndrome characterized by café-au-lait macules, axillary and inguinal freckling, and neurofibromas and schwannomas. There is an increased risk of JMML in this condition also.
In Down syndrome, there is a 500-fold increased risk of acute megakaryoblastic leukemia, with an overall risk of developing leukemia of about 2 %, and about 10 % of DS patients have transient myeloproliferative disorder (TMD) at birth. TMD is the clonal proliferation of myeloid blasts, usually with megakaryoblastic features and presents almost exclusively in DS patients (Klusmann et al. 2007; Izraeli et al. 2007). The symptoms vary from asymptomatic leukocytosis to severe disease causing multiorgan failure. The disorder regresses in about 80 % cases, but about 20 % of cases will develop AML, usually by 5 years of age (Rabin and Whitlock 2009). TMD is a preleukemic state, and additional mutations of genes regulating the proliferation of the megakaryocyte progenitors such as GATA-1 and RUNX-1 need to occur for disease progression (Izraeli 2005). GATA1 mutations are almost universally found in the AML of children with DS, and these may arise very early in the child’s life.
Children with DS are also predisposed to ALL but much less so than to acute megakaryocytic leukemia. Twenty to thirty-three percent of DS ALL cases are found to have a point mutation in exon 14 of JAK2, compared to 0 out of 41 JAK2 mutations in non-DS ALL patients (Kearney et al. 2009).
Patients with a JAK2 mutation usually present earlier and with a higher WBC count than other ALL patients (Tigay 2009). JAK2 mutations confer a cell advantage by promoting cell growth via the JAK/STAT signaling pathway (Mulligan 2008), and the mutations found in DS patients appear to be specific for DS ALL (Bercovich et al. 2008; Rabin and Whitlock 2009).
In non-DS patients, the most common genetic association seen in ALL is the TEL-AML1 translocation between chromosomes 12 and 21, which accounts for about 25 % of childhood ALL (Armstrong and Look 2005). The mechanism by which the fusion protein formed by this translocation causes leukemogenesis is unclear, but the roles of both TEL and AML1 have been shown to be vital in hematopoiesis. The translocation disrupts parts of AML1 known as the core-binding factor, causing disruption of normal differentiation of B cell progenitors (Ford et al. 2009). The translocation can
be found in the blood cells of patients at birth, which is many years before the age at which ALL patients typically present, as it takes so many years for the disease to develop sufficiently to present clinically (Armstrong and Look 2005).