Immunoglobulin G passive transfer from mothers to infants: total IgG, IgG subclasses and specific antipneumococcal IgG in 6-week Malawian infants exposed or unexposed to HIV

The prevalence of HIV infection is high in women of childbearing age in sub-Saharan Africa but, due to efficient health policies that expanded lifelong access to antiretroviral treatment in the past decade, the number of cases of pediatric HIV is constantly decreasing [1]. However, as a consequence, the number of infants born from women living with HIV who do not acquire the infection (HIV-exposed uninfected, HEU) has increased by more than half a million between 2018 and 2020 in the region [1]. HEU children are at an increased risk of morbidity, especially due to infectious causes, and have a 2–threefold higher mortality rate compared to their counterparts not exposed to HIV [2, 3]. This increased vulnerability is probably correlated to maternal HIV infection since, despite optimal antiretroviral therapy (ART), the functional immunological defects caused by HIV can not be completely reversed [4]. The maternal IgG transplacental transfer can have a crucial role in modulating the immune system of the infants and in protecting them from infections during the first 3–6 months of life, when infants have to rely only on maternal-derived immunoglobulins because of the inability to synthesize their own. Hypergammaglobulinemia, which is common in African mothers living with HIV [5, 6] has a significant impact on IgG transfer from the mother to the fetus [7]. Although the underlying mechanisms are not fully elucidated, it has been suggested that IgG levels over 15 g/l can cause the saturation of FcRN receptors, which mediate the IgG transplacental passage [8], and indeed, in clinical studies, maternal IgG levels over 15 g/l have been associated with abnormal IgG concentrations in HEU infants [9, 10]. Impairments of IgG transfer in HEU infants have been observed by some authors [5, 11‐14], but others reported similar IgG levels in HIV-exposed and -unexposed infants [15, 16], or described a differential IgG passage depending on the characteristics of pathogen-specific antibodies [14]. The difficulty to obtain consistent results can be due to different causes, such as the timing of blood sampling (most of the mother/infant pairs studies have used cord blood whose composition mostly reflect fetal rather than neonatal condition [17]), or the lack of reference levels in healthy African infants. Moreover, IgG subclasses structures, glycosylation patterns, and polymorphisms are all factors potentially affecting transplacental passage [18, 19]. In a previous study, we reported the dynamic development of IgG in HEU infants from 1 to 24 months, finding abnormalities in concentrations and subclasses distribution. In particular, we observed persistently low IgG2 levels, but the lack of a proper control group and of long-term antiretroviral therapy of the mothers (the study was performed in the pre-Option B + era), prevented us to draw definite conclusions [20].

The primary objective of this study was to determine whether HEU infants born to ART-treated mothers had immunoglobulins (including specific anti-PCP IgG) levels and patterns, different from those of HUU infants. Our results suggest that 6-week old HEU infants still present immunoglobulins abnormalities in concentrations and distribution including low levels of specific anti-pneumococcal IgG and IgG2.

The main determinants of infant IgG profile, at 6 weeks of life, are the maternal IgG levels and distribution [14, 18]. In our study, we did not evaluate maternal samples, but it is well known that hypergammaglobulinemia and maternal immunological dysfunctions in the B compartment are common in African pregnant women living with HIV [5, 6, 10], and that the polyclonal B cell activation persist even after prolonged antiretroviral treatment [24, 25]. As we hypothesized in our previous work [6], maternal hypergammaglobulinemia could be responsible of the high levels of IgG in their infants. Indeed, in HEU we observed significantly higher levels of IgG and IgG1 compared to HUU infants, with a similar IgG1/IgG ratio in the two groups (HEU: 78.6%; HUU: 76.9%), indicating that the IgG1 distribution reflected that of total IgG. IgG1 is the most diffuse IgG subclass (about 60–70% in the general population [26]) and the preferential class in the transplacental transfer efficiency hierarchy, followed by IgG4, IgG3, and IgG2 [27]. In this study, we did not find significant differences in IgG3 and IgG4 concentrations between the two groups of infants, although the IgG3 proportion was slightly higher in the HUU group. Both subclasses have a not well-defined role in response to diseases, and abnormalities in their levels and distribution are often reported in combination with other IgG subclasses [28].

The significantly lower concentration of IgG2 in HEU infants (22% lower levels compared to HUU infants) and their lower representation out of the total IgG (-35% compared to the counterparts) is one of the relevant findings of this study; although mothers had received ART for a median of 19 months, the level of IgG2 was comparable to the one observed in HEU infants born to mothers treated with short-term ART [5, 29]. These findings suggest that the immunological functional defects due to HIV infection persist in women under continuous ART and interfere with the delicate balance of the maternal–fetal unit. The lack of correlation between total IgG and IgG2 in the HEU population (but not in HUU) is suggestive of a selective impairment of IgG2 passage through the placenta in mothers living with HIV.

The deficit of IgG2 has been associated with increased vulnerability to bacterial diseases [30, 31] since this subclass is responsible for the response against bacterial capsular antigens. We, therefore, extended our analyses to antigen-specific IgG against S. pneumoniae, which is a leading cause of lower respiratory infection among children younger than 5 years in low-income countries [32]. Vaccination against S. pneumoniae has been introduced in Malawi in 2011 and mothers of the study had not received this vaccine. HEU infants showed significantly lower levels of specific anti-PCP IgG compared to HUU infants, with a similarly low proportion of anti-PCP IgG2. To note that the anti-PCP IgG2/anti-PCP IgG ratio was similar in HEU and HUU infants (0.282 vs 0.286), suggesting that the transplacental passage of anti-pneumococcus IgG was globally lower in HEU infants, without a selective reduction of the specific anti-PCP IgG2.

Generally, higher total IgG correlates with higher pathogen-specific antibody levels [33]. In our study the lack of correlations between total IgG and anti-PCP IgG (and of IgG2 and anti-PCP-IgG2) was observed in both groups, suggesting a differential efficiency in the mechanism of IgG transplacental passage. Differences in the efficiency of transfer based on IgG subclasses structures and types of antigen-specific immunoglobulins has been previously reported [14, 27]. Contrasting results have been obtained in different studies on HEU infants: some reported reduction in transfer of specific antibodies against H. influenzae, diphtheria, pertussis, pneumococcus, measles, tetanus, and Plasmodium falciparum [10, 12, 16, 29, 34, 35], but others did not find any effect [15, 36]. The discrepancies between studies could be multifactorial, including methodology, different maternal ART coverage, different exposure to pathogens, the geographical areas, and the timing of infants analysis. One of the strengths of this study is that we determined IgG concentrations from the blood of 6-week old infants, which significantly differ from cord blood, the source used in most of the studies in this field [5, 7, 10, 12, 14‐16, 34, 35]. Cord blood samples composition mostly reflects fetal rather than neonatal characteristics [17, 37, 38]; for instance, in cord blood generally, there are higher IgG levels than in mothers [39].

Our study has some limitations, including the limited numbers of infants in both groups and the lack of detailed information on the viro-immunological conditions in ART-treated mothers that could bias our results. However, considering the lifelong ART strategy, the available data on HIV-RNA for the study, and the high rate of treatment adherence observed in mothers it is highly likely that most of the mothers arrived at delivery with controlled HIV infection. Another important limit is that we could not correlate the levels of IgG anti-PCP found in infants with clinical data and the possible subsequent development of pneumonia. Lastly, the possibility to analyze maternal samples could have helped us in the interpretation of the results.