Guilherme Henrique Hencklain Fonseca, Hanadi Elayoubi, Ibrahima Diagne, Ryad Tamouza, Evandra Strazza Rodrigues, Renato Cunha, Carla Luana Dinardo, Yahia Benzerara, Annalisa Ruggeri, Francisco S. Fernandes, Indou Deme Ly, Simone Kashima, Luciana Ribeiro Jarduli, Fernanda Volt, Itaua Leston Araujo, Belinda Pinto Simões, Eliane Gluckman, Chantal Kenzey, Vanderson Rocha, Ana Cristina Silva Pinto, Julia Pavan, Barbara Cappelli, Christina Mary Mariaselvam, Sandra Fátima Menosi Gualandro, Karina Tozatto-Maio, and Robert Girot
Background: Patients (pts) with sickle cell disease (SCD) show a high phenotype variability that is not fully understood. Because inflammation plays a major role in SCD pathophysiology, we hypothesized that single nucleotide polymorphisms (SNPs) in genes of innate and adaptative inflammatory response modulate the occurrence of SCD complications. Objective: to establish associations between SNPs and clinical complications in pts with SCD. Methods: Case-control retrospective study; 500 pts were included, followed at Senegal (n=56), Brazil (n=230) and France (n=214). We analyzed the effect of 20 SNPs in 6 clinical complications: acute chest syndrome (ACS), stroke, leg ulcers, cholelithiasis, osteonecrosis and retinopathy. Using TaqMan 5'-nuclease assay, we genotyped SNPs in genes encoding Toll-like receptor (TLR) 1 (rs4833095), 2 (rs4308099, rs4308100, rs4696480), 6 (rs5743810), 10 (rs11466653, rs11096957), natural killer (NK) group 2 member D (NKG2D) receptor (rs1982536, rs2617160, rs2617169, rs2617170, rs2617171, rs1049174, rs2246809, rs2255336), human leukocyte antigen (HLA)-G (rs9380142), HLA-E (rs2517523), major histocompatibility complex class I polypeptide-related sequence A (MICA, rs1051792) and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) (rs5742909, rs231775). All SNPs had a call rate >90% and minimum allele frequency >1%. We performed analyses of correspondence for indicating associations between SNPs and number of clinical complications. Logistic regressions were used to identify associations between SNPs and each complication, using geographical origin, SCD genotype, rate of transfusions and gender for modeling adjustment; the significance was adjusted for multiple testing using false discovery rate. Comparisons of genotype frequencies with the population of African descent from 1000 genomes database were done by chi-square. Results: Pts were originally from Brazil (n=228), Sub-Saharan Africa (n=200), French West Indies (n=53), North Africa (n=7) and 12 unknown. SCD genotype (available n=498) was SS (n=402), SC (n=46), SB (n=42), SD or SE (n=4). 280 pts were female; median age was 32 years (range 0-69); 184 pts received at least 20 transfusions. 71 pts presented stroke, 200 had at least 1 episode of ACS, 69 had leg ulcers, 271 had cholelithiasis, 150 had retinopathy and 90 osteonecrosis. 97 pts did not present complications, 135 had 1 type of clinical complication, 134 had 2, 94 had 3, 34 had 4 and 6 had 5. 21 pts underwent hematopoietic stem cell transplantation. 11 pts died during follow-up, mostly from ACS and hemorrhagic stroke. Indication of association with number of complications: TLR2 rs4696480 TA, TLR2rs3804099 CC and HLA-Grs9380142 AA were associated with occurrence of 0-1 clinical complications, MICA rs1051792 AA/AG with up to 2 complications and NKG2D rs2617169 AA with 5 complications. Association between genotypes/haplotypes and each complication: no association was found between SNPs and stroke, ACS, leg ulcers and osteonecrosis. Rs9380142 was the only SNP significantly associated with cholelithiasis in the logistic regression additive model. The G allele increased the risk of cholelithiasis (AG x AA, OR 1.57, 95%CI 1.16-2.15; GGxAA, OR 2.47, 95%CI 1.34-4.64; P=0.02). For retinopathy, in the logistic regression additive model, the presence of the A allele decreased the risk of retinopathy for rs2246809 in pts of same origin (AAxGG: OR 0.22, 95%CI 0.09-0.50; AGxGG: OR 0.47, 95%CI 0.31-0.71; P=0.004), rs2617160 (ATxTT: OR 0.67, 95%CI 0.48-0.92; AAxTT: OR 0.45, 95%CI 0.23-0.84; P=0.04) and rs2617169 in pts of same SCD genotype (AAxTT: OR 0.33,95%CI 0.13-0.82; ATxTT: OR 0.58, 95%CI 0.36-0.91, P=0.049). No haplotype was associated with complications. The genotype distribution of SNPs rs4696480, rs3804099, rs1051792 and rs2617169 differed significantly from the population of African descent from the 1000 genomes database. Discussion: We have previously shown that TLR2 rs4696480 TA decreases occurrence of bacterial infections in SCD; in this study, TA is also associated with less complications. Also, HLA-G rs9380142 AA had less complications and cholelithiasis, and 3 SNPs in NKG2D modulated occurrence of retinopathy. TLR and NKG2D may be activated by heat shock proteins, released in ischemia-reperfusion injury that occurs in SCD. Our findings help to better understand the role of inflammation in phenotype heterogeneity in SCD. Disclosures No relevant conflicts of interest to declare.