Evidence for efficient vertical transfer of maternal HIV-1 envelope-specific neutralizing antibodies but no association of such antibodies with reduced infant infection

Abstract

: Little is known about the efficiency of vertical transfer of HIV-1-specific antibodies. Antibody levels in plasma from 60 mother-infant pairs near the time of birth, including 14 breast-feeding transmission pairs, were compared. The envelope-binding titers were strongly correlated (r = 0.91, P < 0.0001) and similar (1.4-fold greater in maternal plasma) between a mother and her corresponding infant as were the neutralizing antibody (Nab) levels (r = 0.80, P < 0.0001; 1.3-fold higher), suggesting efficient transfer. There was no significant difference in Nab responses between transmitting and nontransmitting mothers, although there was a trend for transmitting mothers to have higher HIV-1-specific Nabs.

Fig.1

Comparison of maternal and infant HIV-1 Envelope-specific antibodies. For panels A-E, maternal and infant antibody levels are shown on the x- and y-axis, respectively. Each dot represents the data from one mother-infant pair. The line represents regression best fit line and the r is Pearson’s product moment correlation coefficient. The p-values shown above each graph are based on the Student’s t approximation for the distribution of the corresponding standardized test statistic. The panels include: A) Envelope binding EPTs; B) average log₂IC50 for the 3 viruses; C) Q461.d1 log₂IC50; D) QD435.A4 log₂IC50; E) QD842.d16 log₂IC50, and F) the association between EPTs and Nab average log₂IC50 values for both mothers () and infants ().

Fig. 2

Comparison of the average log₂IC50 values of the 46 non-transmitting (NTM) and 14 transmitting mothers (TM). The dot plots summarize the distribution of the average log₂IC50 for the 3 viruses, with mean and 95% confidence intervals for the log₂IC50 indicated by the horizontal bars for each. The unadjusted estimate for the fold-difference in IC50 and corresponding p-value are based on Welsh’s t-test, and the adjusted fold-difference estimate is by linear regression controlling for log₁₀ viral load, with the corresponding Wald p-value.