Classification of the study participants
In this study, demographic data on malaria, parasite density, and disease complication variables were analysed for 600 children of matched sex and age. The 600 subjects were categorized into three different groups. Group I: The malaria-free control [MFC, n = 200 (33.3%) subjects]; included subjects without symptoms of the disease and showed negative results for blood film examination and PCR of the malaria parasites. Group II: Uncomplicated malaria [UM, n = 218 (36.3%) patients]. Group III: severe malaria [SM, n = 182 (30.3%) patients]. Group III included patients with severe malarial anaemia [n = 56 (30.8%)], cerebral malaria [n = 61 (33.5%)] and children suffering from other complications such as hypoglycaemia, jaundice, pulmonary oedema and acute respiratory distress [n = 65 (35.7%)]. The mean number of parasites in severe malaria patients was significantly higher compared to uncomplicated malaria, P<0.001 (Table 1). The body temperature was significantly different between the study population, P < 0.001 (Table 1).
Comparison between the distribution of the FcγRIIa genotype and its allelic frequencies among the different study groups
The genotype frequencies for FcγRIIa, FcγRIIa, and FcγRIIIb did not deviate from the expectations of the HWE in each genotype group (Table 2). The frequencies for the homozygotes carrying the FcγRIIa-R/R131 in UM were lower than the ones in MFC. The logistic regression analysis revealed that there was no statistically significant difference between UM and MFC amongst both homozygotes R/R131 [18.0% in UM versus 15.1% in MFC; OR=1.39, 95% CI (0.89 to 2.19) and P value = 0.15] and H/H131 [36.7% in UM versus 30% MFC; OR = 0.92, 95% CI (0.62 to 1.36) and P value = 0.66] using the heterozygotes as the reference group (Table 3). The FcγRIIa-R/R131 genotype was statistically associated with SM compared to UM [34.6% in SM versus 15.1% in UM; OR = 2.132, 95% CI (1.287 to 3.533) and P value = 0.003]. In contrast, the FcγRIIa-H/H131 genotype was negatively associated with SM compared to UM [13.7% in SM versus 36.7% in UM; OR = 0.349, 95% CI (0.206 to 0.592) and P value < 0.001] (Table 2). The frequencies for the heterozygotes carrying the FcγRIIa-H/R131 genotype were almost the same among the three groups of MFC, UM, and SM (52.0%, 48.2%, and 51.6%, respectively) (Table 2).
Comparison between the FcγRIIa genotypes distributions and their allelic distributions among the different study groups.
As shown in Table 2, the genotype frequencies for FcγRIIa, FcγRIIa and FcγRIIIb did not deviate from expectations of the HWE in each genotype group (Table 2). The frequencies for the homozygotes carrying the FcγRIIa-R/R131 in UM (group II) were lower than the ones in MFC (group I). The logistic regression analysis revealed that there was no statistically significant difference between UM and MFC amongst both homozygotes R/R131 [18.0% in UM versus 15.1% in MFC; OR=1.39, 95% CI (0.89- 2.19) and P value= 0.15] and H/H131 [For 36.7% in UM versus 30% MFC; OR= 0.92, 95% CI (0.62- 1.36) and P value=0.66] using the Heterozygotes as the reference group (Table 3). The FcγRIIa-R/R131 was statistically associated with SM patients compared to UM [34.6% in SM versus 15.1% in UM; OR= 2.132, 95% CI (1.287- 3.533) and P value= 0.003]. In contrast, the patients carrying FcγRIIa-H/H131 genotype were statistically negatively associated with SM compared to UM [13.7% in SM versus 36.7% in UM; OR= 0.349, 95% CI (0.206- 0.592) and P value <0.001] (Table 2). The frequencies for the heterozygotes carrying the FcγRIIa-H/R131 genotype were almost the same among the three groups of MFC, UM and SM (52.0%, 48.2% and 51.6%, respectively) (Table 2).
Comparison between the distribution of FcγRIIa genotype and its allelic frequencies among the different study groups
The genotype frequencies showed no statistically significant difference among the homozygotes FcγRIIa-F/F with UM compared to MFC (Table 2). The logistic regression analysis confirmed the absence of significant differences between UM and MFC among homozygotes F/F [39% in UM versus 45.5% in MFC; OR = 1.95, 95% CI (0.65 to 2.38) and P value = 0.79]. Similarly, FcγRIIa-V/V genotype showed no statistically significant association with UM compared to MFC [12.8% in UM versus 14.5% in MFC; OR = 1.72, 95% CI (1.04 to 2.82) and P value = 0.13] using the heterozygotes as a reference group (Tables 2 and 3). On the other hand, FcγRIIa-F/F genotype was statistically associated with SM compared to UM [72.5% in SM versus 39% in UM; OR = 11.51, 95% CI (6.71 to 19.77) and P value<0.001] (Tables 2 and 3). In contrast, the FcγRIIa-V/V genotype was statistically negatively associated with SM compared to UM [3.8% in SM versus 12.8%in UM; OR = 0.20, 95% CI (0.09 to 0.47) and P value <0.001] (Tables 2 and 3). The frequencies analyses also showed differences in the distributions of the heterozygotes FcγRIIa-F/V genotype among the three groups (40% in MFC, 48.2% in UM, and 23.6% in SM) (Table 2).
The genotype analyses showed similar frequencies between UM and MFC for homozygotes carrying the FcγRIIIb-NA1/NA1 of (Table 2). This was confirmed by logistic regression analysis which revealed the lack of statistically significant difference between these two groups in the homozygotes NA1/NA1 [15.1% in UM 17.0% in MFC; OR = 0.79, 95% CI (0.48-1.30) and P value = 0.354] (Tables 2 and 3). Furthermore, NA2/NA2 genotype was not statistically significantly different among UM patients compared to MFC [36.7% in UM versus 37.0% in MFC; OR = 1.24, 95% CI (0.85-1.79) and P value = 0.263] using the heterozygotes as a reference group (Tables 2 and 3). Similarly, there was no statistical differences between patients with SM and UM for FcγRIIIb-NA1/NA1 genotype [9.9% in SM versus 15.1% UM; OR = 0.82, 95% CI (0.43 to 1.57) and P value = 0.545] (Tables 2 and 3). However, patients carrying the FcγRIIIb-NA2/NA2 genotype were significantly associated with SM compared to UM [51.6% in SM versus 36.7% in UM; OR= 1.76, 95% CI (1.15- 2.70) and P value= 0.009] (Table 2 and 3). The frequencies analyses also showed differences in the distributions of the heterozygotes FcγRIIa-NA1/NA2 genotype among the three groups (46% in MFC, 48.2% in UM, and 38.5% in SM) (Table 2).
Specific IgG subclass reactivity in the different study groups
The antibody responses for the P. falciparum blood-stage antigen AMA-1 was analysed in the different study groups. Our results showed statistically significant differences among the anti-malarial IgG subclasses antibody levels in the different study groups; the overall P value < 0.001 (Table 4). In general, the median value of IgG1 and IgG3 subclasses were expressed at higher levels than IgG2 and IgG4 antibodies in the UM group when compared to both SM and MFC subjects (Table 4). To investigate the potential association between the antimalarial IgG subclass responses and protection against infections, we first, used a logistic regression model to compare the levels of IgG subclasses between the UM infection and MFC (Table 5). The results showed that a higher level of IgG3 against the AMA-1 antigen was associated with UM patients compared to MFC subjects [OR = 2.6; 95% CI (2 to 3.3) and P value < 0.001]. In addition, the levels of AMA-1-specific IgG4 were significantly lowered in UM patients compared MFC [OR = 0.7; 95% CI (0.5 to 1.9) and P value < 0.001] (Table 5). There was no observed association for the AMA-1-specific IgG1 and IgG2 responses in UM compared to MFC (Table 5). The same logistic regression model confirmed that the apparent anti-malarial IgG1 to IgG4 antibodies were significantly associated with SM when compared to UM patients (Table 5). Collectively, the levels of AMA-1-specific IgG1, IgG2, IgG3 and IgG4 were significantly higher in SM compared to UM patients [for IgG1: OR = 3.4; 95% CI (2.1 to 4.7) and P value < 0.001, for IgG2: OR = 2.7; 95% CI (1.9 to 3.5) and P value < 0.001, for IgG3: OR = 1.8; 95% CI (1.1 to 4.6) and P value < 0.001) and for IgG4: OR = 1.5; 95% CI (1.2 to 1.8) and P value < 0.001] (Table 5).
The results indicated that patients carrying the FcγRIIa-H/H131 genotype are significantly associated with higher expression levels of the antimalarial IgG1, IgG2 and IgG4 antibodies, but not IgG3 in UM patients [for IgG1: OR = 0.3; 95% CI (0.2 to 0.6) and P value <0.001, for IgG2: OR = 0.5; 95% CI (0.3 to 0.8) and P value= 0.006 and for IgG4: OR = 0.5; 95% CI (0.3 to 0.8) and P value= 0.006] (Table 6). Comparatively, patients harboring the FcγRIIa-R/R131 genotype show significantly increased levels of antimalarial IgG2 antibodies and associated with SM compared to UM [OR = 3.7; 95% CI (2.0 to 6.7) and P value < 0.001] (Table 6). However, patients carrying the genotype FcγRIIa-R/R131 are statistically negatively associated with higher levels of AMA-1-specific IgG3 [OR = 0.4; 95% CI (0.2 to 0.6) and P value < 0.001] (Table 6). Independently, the model of the multivariate logistic regression analysis of individuals carrying the FcγRIIa-F/F genotype is significantly associated with higher levels of AMA-1-specific IgG2 and IgG4 antibodies in SM compared to UM patients [for IgG2: OR = 3.9; 95% CI (2.4 to 6.4) and P value < 0.001 and for IgG4: OR= 3.2; 95% CI (2.1 to 5.3) and P value < 0.001] (Table 6). These results together clearly show that the FcγRIIa-F/F genotype is negatively associated with higher expression levels of AMA-1-specific IgG3 among SM compared to UM patients [OR = 0.2; 95% CI (0.1 to 0.4) and P value < 0.001] (Table 6). Similarly, our data show that the FcγRIIa-V/V genotype is negatively associated with higher levels of AMA-1-specific IgG4 in SM compared to UM subjects [OR = 0.4; 95% CI (0.2 to 0.6) and P value < 0.001] (Table 6).
Furthermore, our analyses show that the FcγRIIIb-NA2/NA2 genotype is significantly associated with a higher level of AMA-1-specific IgG4 among SM compared to UM group [OR = 1.7; 95% CI (1.1 to 2.7) and P value= 0.011] (Table 6). Our results indicate that the FcγRIIIb genotypes are not associated with the independent action of the three IgG subclasses (IgG1, IgG2, and IgG3) of antibodies, maybe due to the absence of interaction in the logistic regression model.