The cause of CLL

The extreme differences in geographical variation worldwide ought to give us a clue as to the etiology of CLL. Prevalence rates show a 40-fold difference between white Europeans and North Americans, and Asians, but unlike other with malignancies, Asians migrating to the USA retain their low incidence [ 38 Pan et al Cancer causes Control 2002; 13:791-5]. This suggests that genetic factors rather than environmental ones are responsible. Despite the low incidence in Asians, their disease bears a close resemblance to that occurring in Caucasians [39 Irons et al Leuk Res 2009 epub]. The evidence for environmental factors playing a role in the etiology is weak and inconsistent, with some studies having found a link to agriculture (pesticides, herbicides, animal exposure) and some to exposure to benzene and the rubber industry, but others have not found such links [40 Goldin, Slager Hematology 2007; 1:339]. Not unexpectedly, there was no association with exposure to radiation fall-out from the Japanese atomic bombs, since CLL is very rare in Japan, but more recently the role of radiation exposure has been raised again [41 Hamblin Leuk Res, 2007].

There are certainly genetic elements that are important in the pathogenesis of CLL. Somewhere between 5 and 15% of patients have a family history of CLL. Rawstron et al found a prevalence of unsuspected MBL in 13.5% among 59 first-degree relatives in 21 CLL kindreds [42 Rawstron Blood 2002b] and Marti et al found an even higher prevalence of 18% [43 Marti Cytometry B Clin Cytom 2003; 52:1-12]. Familial cases of CLL do not differ significantly from sporadic cases apart from fitting into the more benign categories owing to selection bias [40]. Linkage studies looking at the co-inheritance of genetic markers and CLL have so far been unsuccessful in identifying any genetic defects that make a family member prone to CLL [40]

Searches for single nucleotide polymorphisms (SNP) among candidate genes or molecular pathways have similarly been largely unsuccessful [40], but Raval et al [41 Cell 2007 129, 879-90] identified an SNP of the Death Associated Protein Kinase 1 (DAPK1) gene on chromosome 9 which segregated with the disease in a single large family with CLL and was associated with the downregulation of the enzyme’s expression. DAPK1 is an actin-filament-associated, calcium calmodulin-dependent, serine/threonine kinase that promotes apoptosis in response to various stimuli. Although there were many polymorphisms found, most could be eliminated, but one, an A to G switch at position c.1-6531, was not found in 383 control samples from the US and Northern Europe. Investigation of this polymorphism demonstrated that the A to G switch reduces DAPK1 expression by increasing the affinity of DAPK1 for HOXB7, a transcription factor that normally opposes the expression of DAPK1.
That this might be important can be gathered from the fact that DAPK1 is normally silenced by demethylation in sporadic CLL as it was in this study in 60 out of 62 cases. However, the SNP was found in the germ-line of only one among 263 cases of sporadic CLL from the US and Northern Europe but not among other familial kindreds.

Another polymorphism of possible relevance was discovered by Pekarsky et al [37 PNAS 2008; 105: 19643-8] They demonstrated two mutant sequences of TCL1 among 600 CLL patients studied, one of which was present in the germ line. Activator protein 1 (AP-1) –dependent transcription, which includes both c-Jun and c-Fos, induces apoptosis by transactivating proapoptotic genes. Wild type TCL1 inhibited AP-1 dependent transactivation around 2.5 fold, whereas the mutant forms inhibited it around 100 fold.

It is likely that no simple genetic defect is responsible for the occurrence of CLL in families, but rather that a large number of components of molecular pathways appear in variant forms which together influence the rates of proliferation and apoptosis of B lymphocytes.