CVID is characterized by humoral immune deficiency with onset after 2 years of age and usually in young adulthood and it is one the most common form of primary antibody disorders in humans with an estimated incidence of 1:25000 (20, 21). In 2012, our 25 CVID patients were screened for alterations in genes known to be associated with autosomal recessive CVID and only one patient carried heterozygous TNFRSF13B c.310T > C (p. Cys104Arg) (rs34557412) (4%) which was recorded as disease causing (20). Seven of the 25 patients in that study had consanguineous parents (28%) (20). Karaca et al (21) have reported the rate of parental consanguinity in Turkish CVID patients as 19.1%. In both US and Sweden study, known consanguineous parents were 0%, while it was 63% in Iran. In this study, we have included 100 Turkish CVID patients diagnosed and followed-up with ESID criteria and found out that parental consanguinity is 24%. In summary, the rate of consanguinity for CVID patients in our reported studies differ between 19.1% and 28.0%, very high than American and Europe population and so much less than Iranian cohort.
Our study subjects were recruited only from childhood population whereas their median age was 5.8 years when they were diagnosed as CVID and our genetic diagnostic delay was about 3.2 years. In Abolhassani’s study, median age of the US, Swedish and Iranian patients, were 44 years (range, 5–85 years), 49.5 years (range, 4–90 years) and 9 years (range, 1–65 years), respectively (5). This data shows us that Abolhassani et al recruited not only children CVID cases, but also adult ones. Besides geographical and ethnical facts, great differences in age groups may be one of the reasons of different findings between our study and Abolhassani’s study, because different genotypes cause different CVID phenotypes and different ages for the beginning of clinical and infectious symptoms. Another important difference between these two studies is about the gender of patients whereas in Turkish patients males were 65% (as it was expected for primary immune deficiencies) and in three countries from different continents males were 48.5% of the study population.
In three different countries from America, Asia and Europe, mutations leading to CVID phenotype were identified in 31–54% of patients although this rate was less than 10% before 5–10 years (5). In our study, pathogenic mutations were defined in 40% of Turkish CVID patients living in a geographical area on the bridge between Asia and Europe. In Abolhassani’s study, 68 known disease-causing genes underlying an immune defect presenting as CVID were revealed while in our study 17 different genes and 40 pathogenic mutations were identified. Abolhassani et al (5) included X-linked agammaglobulinemia patients (Bruton tyrosine kinase-BTK), immune dysregulation syndromes (FOXP3, DOCK8), WHIM syndrome cases, X-linked hyper IgM case (CD40L) and severe combined immune deficiency patients (recombination activating genes, RAG1/RAG2 and DNA ligase 4, LIG4) in their study, although these patients do not fulfill the criteria for CVID. We did not enroll any of these patients and we found less number of different genes causing CVID. Besides previously described mutations, we identified 16 novel mutations (16%) in known disease-causing genes (Table-1-2-3-4) in patients exactly diagnosed as CVID with all clinical and laboratory findings.
In our Turkish cohort, the most common gene with a surface molecular defect was TACI (15/40 in mutation identified cases and 15/100 in all cases). The second most common genes were PLCү2, LRBA, TCF3 and STAT1, all of them were 3/40 in genetically defined cases and 3/100 in all CVID patients. NFKB2 gene was the other common one with two previously defined mutations (Table-3). We had expected that Turkish and Iranian cohorts from very near geographical area with similar median ages and with a history of consanguinity higher than 20%, would be more similar, but this was not observed as the most genes in Turkey and Iran were TACI and LRBA, respectively. TACI was the predominant gene in US identified in 35% of cases while LRBA and TACI were both the most common ones (10.8% each of them) in Sweden (5). The results of our Turkish cohort were more similar to European (Sweden) and American (US) population.
The advent of TNGS has enabled rapid identification of the molecular etiology of CVID for the last 10 years. In Turkish laboratories, next-generation sequencing is highly used for genetic diagnosis of inborn errors of immunity while WES is commonly used to diagnose CVID in developed countries (22). In TNGS, gene panels are customized for specific genetic diseases, considering the high specificity and sensitivity of the entire genes with the maximum coverage because of the limited numbers of the gene in the customized panel. TNGS panel has been reported to provide a diagnosis to 80% of cases (23). TNGS revealed disease-causing mutations in 35% of our CVID patients whereas WES revealed pathogenic genes in 40% of the study population. We believe that any of the above genetic tests where they are available enable us to define the molecular defect in CVID and improves the quality of disease management and patient outcome.
In conclusion, a combination of clinical and genetic diagnosis are recently more extensively used in the management of CVID patients. However, because of the complexity and heterogeneity of the disease, genetic investigation is still a great challenge. In this presented study, the rate of disease-causing mutations was 40% in Turkish CVID patients (40/100) and 16 of these 40 mutations were not previously described and will add new information to the genetic databases. The most common gene was TACI (15/40 in mutation identified cases and 15/100 in all cases) in Turkish cohort followed by PLCү2, LRBA, TCF3, STAT1 and NFKB2 genes. Our results were more similar to European and American population rather than Iranian population suggesting us that geographical factors do not highly effect genetic alterations in CVID patients.