Gene Polymorphisms of Costimulatory Molecules: CTLA-4/CD28/ICOS Are Associated with B-Cell Chronic Lymphocytic Leukemia (B-CLL).

There are strong evidences that altered immunological function entails an increased risk of cancer. Cytotoxic T-lymphocyte antigen 4 (CTLA-4), CD28, and Inducible co-stimulator (ICOS) are costimulatory molecules provide regulatory signals for T-cell activation. CD28 and ICOS potently enhance T-cell functions, while activation of T-cells is subsequently downregulated by CTLA-4. The abnormal expression of costimulatory molecules may be caused by polymorphisms of genes encoding these molecules. The extended study was undertaken to evaluate the association between the polymorphisms: CTLA-4+49A/G, CTLA-4-319C/T, CTLA-4(AT)_n, CD28+17C/T, ICOS(GT)n and susceptibility to B-CLL in a Polish population. All together 124 B-CLL patients and 202 controls were studied. Single nucleotide polymorphisms were typed by minisequencing. The microsatellite polymorphisms were studied by PCR and fuorescence based technique. The distribution of -319T/C alleles and genotypes differed significantly between groups. The frequency of T allele and T phenotype was increased in patients with respect to controls (p=0.01, OR=1.74, 1.13∼95%CI∼2.68; p=0.01, OR =1.85, 1.13∼95%CI∼3.04, respectively). CTLA-4+49A/G and CTLA-4 (AT)_n did not correlate with B-CLL. The analysis of the CD28+17C/T showed that the presence of C allele and C phenotype increased the OR of B-CLL by 1.97 and 2.07, respectively (p=0.002, 1.28∼95%CI ∼3.03; p=0.003, 1.27∼95%CI ∼3.38). For ICOS(GT)n gene polymorphism analysis we combined the alleles into 3 groups: (s)- short-alleles with 8 and 9 repeats, (m) -middle- with 10 and 11 and (l)- long- 12 and more repeats. The global distribution of alleles and genotypes was significantly different in patients and controls (p=0.00008 and p=0.00006, respectively). Individuals possessing (s) alleles were 2.75 times more prone to B-CLL than others (p=0.0000, OR=2.75, 1.67∼95%CI∼4.55), while carrying (l) alleles were protected to B-CLL (p=0.004, OR=0.47, 0.29∼95%CI∼0.79). The genotype (s)/(m) was overrepresented in patients (p=0.004, OR=2.29, 1.28∼95%CI∼2.09), but (l)/(l) individuals appeared more frequently in controls (p=0.002, OR=0.07, 0.00 ∼95%CI∼0.44). The haplotype association study of three polymorphic sites: CTLA-4-319C/T and CD28+17C/T and ICOS (s)/(m)/(l), which correlated with B-PBL in univariante analysis, showed that the distribution of haplotypes was different in cases and control, global p value was p= 3 ×108. Remarkable, the haplotypes T/C/(s) and T/T/(s) were presented only in B-CLL (p=0.0004, OR=62072, 3540∼95%CI∼1088238; p=0.00003, OR=3447, 203∼95%CI∼58334, respectively) while the haplotype C/T/(l) was significantly more frequent in control group as compared to B-CLL (p=0.000003, OR=0.3, 0.18∼95%CI∼0.51). B-CLL patients and controls polymorphic features for CTLA-4-319C/T, CD28IVS3+17C/T, ICOS (GT)n were subjected to multivariate forward stepwise logistic regression analysis. It appeared that C phenotype at CD28 T17int3C site increased twice risk of B-CLL (p=0.004, OR=2.09; 1.26∼95%CI∼3.48), while genotype (m)/(l) or (l)/(l) for ICOS gene protected from disease (p=0,0009, OR=0,41; 0,24∼95%CI∼0,69). We reported for the first time that the gene polymorphism of costimulatory molecules: CTLA-4/CD28/ICOS contribute to susceptibility to B-CLL.