Introduction
The recent development of short, practical and affordable anti-retroviral regimens greatly improves prospects for preventing perinatal HIV-1 transmission in countries with limited resources where the vast majority of HIV-infected mothers live [1–4]. Transmission of HIV through breast milk, however, represents a major barrier to improving HIV-free survival in children, especially where the health risks and social stigma associated with lack of breastfeeding are high. Interventions that decrease HIV transmission both perinatally and during breastfeeding offer a significant advantage over existing regimes. Passive immunization with an immune globulin product containing HIV antibody is a potential addition to antiretroviral therapy in the peripartum and postnatal periods. The long half-life of immunoglobulin may extend the period of protection into the first months of breastfeeding where the transmission risk is the greatest [5]. As with hepatitis B virus (HBV), the combination of passive immunization with HIV hyperimmune globulin (HIVIGLOB) and an effective HIV vaccine may provide long lasting protection [6].
A number of immune globulin products have been developed which demonstrate therapeutic or prophylactic efficacy against a number of pathogens [7,8]. Beasley et al. demonstrated that a single dose of hepatitis B immunoglobulin (HBIG) given to a newborn of a mother seropositive for hepatitis B surface antigen (HBsAg) reduced the HBV chronic carrier state from developing from approximately 92% to 55% [9]. A three-dose regimen with HBIG at birth, 3 and 6 months further reduced this rate to 26% [9]. Several studies in animal models using neutralizing monoclonal antibodies or immunoglobulin have shown promising results in prevention of infection in adult and neonatal animals [10–18]. In addition, intravenous administration of HIV hyperimmune globulin, inactivated HIV antibody positive plasma preparations, or monoclonal antibody to HIV infected adults demonstrated a reduction in plasma viremia and clinical improvement in some studies [19–24].
We undertook the production and phase I/II testing of a Ugandan HIVIGLOB product in preparation for an efficacy study to prevent peripartum and early breastfeeding transmission of HIV in Kampala, Uganda. The primary objectives of the trial were to assess the safety, tolerance, titer and half-life of p24 antibody pre- and post-infusion of the product in a three-arm dose escalation study. Secondary objectives included determining the virologic and immunologic changes and the rate of HIV-1 vertical transmission in the maternal–infant pairs receiving the HIVIGLOB. The association between HIV-1 vertical transmission and maternal plasma HIV-1 RNA levels, CD4 cell counts, and HIV-1 p24 antibody was also assessed.
Subjects and methods
Prior to the phase I/II trial, the procedures for the collection, shipment, and production of a Ugandan HIVIGLOB product were established and validated.
Plasma collection and processing
Collection of HIV-1 antibody positive plasma began in January 1993 at the Nakasero Blood Bank in Kampala, Uganda. Plasma (200–300 ml) collected from units with CD4 cell counts > 500 × 106/l that tested negative for HBsAg, hepatitis C virus (HCV) antibody, and HIV-1 p24 antigen were shipped to the Swedish Institute for Infectious Disease Control (SIIDC) in Stockholm, Sweden for fractionation into intravenous HIV-1 hyperimmune globulin.
Plasma fractionation
Pooled plasma units were treated with a solvent/detergent mix of 1% Tri(N-butyl) phosphate/Tween-80, as described by Cummins et al. [25]. This procedure has been shown to inactivate lipid-enveloped viruses like HIV-1, HBV and HCV [26]. The subsequent fractionation procedure, developed by Falksveden and Lundblad for preparation of hyperimmune globulin, includes multiple steps to isolate the IgG fraction [27]. These include multiple ion exchange and precipitation and extraction procedures. The unbound IgG fraction is precipitated with ethanol and extracted with glycine to protect the IgG from aggregation and salt denaturation during lyophilization. The fractionation procedure yields a pure monomeric unfragmented and undenatured human IgG. The final product is formulated as a 5% IgG isotonic solution for intravenous injection. Sterility testing is performed on the final product and on samples taken from every step during fractionation. The Uganda plasma pools contained an average of 29 g IgG/kg with after fractionation yields of about 15 g/kg (51–52%).
Characteristics of final HIVIGLOB product
HIV-1 cultures, HIV-1 DNA PCR, and HIV-1 RNA PCR were performed on the original plasma pool, the plasma pool after virus inactivation by Tri(N-butyl) phosphate/Tween-80, and the 16% hyper-HIV-1 immunoglobulin product before dilution. Only HIV-1 genomic RNA was detected in the original plasma pool at a level of 30 000 copies/ml using the Roche Amplicor Monitor assay (Roche, Emeryville, California, USA). However, all HIV-1 cultures and PCR testing were negative on the 16% hyper-HIV-1 immunoglobulin preparation and no HIV-1 RNA was detected in the final 5% IV IgG product. In addition, a provocative challenge experiment performed at each step of the fractionation process found an overall log reduction of at least 1 × 1010 50% tissue culture infectious dose (TCID50) and 1 × 109.5 in terms of HIV-1 RNA. The final HIVIGLOB product was negative for hepatitis B by HBsAg and HBV DNA PCR testing, HCV by RNA PCR testing, and human T cell lymphoma virus types I and II by PCR.
The final HIVIGLOB product contained 98% monomeric IgG. IgG subclass enzyme immuno-assays performed on the 16% HIVIGLOB solution revealed the titer of IgG1 antibody to rgp160 to be 1 : 3 200 000, to rp24 to be 1 : 10 000 000, and to LP2-V3/MN-peptide to be 1 : 100 000. Titers of IgG2, IgG3, and IgG4 were significantly lower. The two lots of the final 5% IV IgG preparation had an average p24 antibody titer of 1 : 183 000 using the Abbott p24 antibody assay (Abbott Laboratories, Delkenheim, Germany). These values are similar to an average titer of 1 : 181 000 for the AIDS Clinical Trials Group (ACTG) 185 HIVIG product when both products were tested on the same run using a specific competitive anti-p24 antibody enzyme immunoassay (Envacore, Abbott Laboratories) [28]. Repeat antibody testing on one lot stored at 2–6°C for 3.5 years revealed a similar p24 antibody titer of 1 : 177 893 compared with the initial titer of 1 : 178 555.
HIV-1 neutralization assays
HIV-1 neutralization assays using p24 antigen detection were performed on the 16% HIVIGLOB solution at the SIIDC. A 1 : 320 dilution of HIVIGLOB was able to inhibit 120 TCID50 of the HIV-1 IIIB strain on Jurkat-Tat cells and 60 TCID50 on peripheral blood mononuclear cells (PBMC). A 1 : 80 dilution of HIVIGLOB was able to inhibit 35 TCID50 of the European primary isolate 6794 on PBMC. These titers were approximately four to fivefold greater than the titers of the plasma pool.
In addition, neutralizing antibody assays against a wide range of primary subtype and laboratory isolates were performed with the 5% HIVIGLOB preparation by Quality Biologicals (Gaithersburg, Maryland, USA). Using a virus reduction assay, a 10–35-fold reduction in infectivity could be attained against a number of different primary subtype isolates, including three subtype B, one C, one E, and two Ugandan A isolates. Subtype A is the predominant subtype in Uganda with V3 loop sequences closely related to US/European V3 loop sequences [29,30]. A 10-fold reduction in infectivity could not be reached against two Ugandan subtype D isolates.
HIVIGLOB testing and infusion in Sweden
In 1995, five non-pregnant Swedish volunteers were infused with 200 mg/kg HIVIGLOB to determine safety and tolerance, half-life, and in vivo reduction in plasma HIV-1 RNA levels. These volunteers included two Ugandan females with subtype A virus and three white Swedish males with probable subtype B infection. One volunteer developed fever and another developed back pain near the end of the infusion requiring early discontinuation of the infusion. All symptoms resolved and were not considered serious adverse events.
Phase I/II Trial in Uganda
The phase I/II HIVIGLOB trial was approved by Institutional Review Boards in the USA, Sweden, and Uganda. After review of the fractionation process and Swedish patient data, an NIAID Data Safety Monitoring Board approved proceeding with the trial of HIVIGLOB in pregnant HIV-1 infected Ugandan women and their neonates. Between June 1996 and April 1997, 31 HIV infected pregnant women were enrolled. The first 10 mother–infant pairs were infused at a dose of 50 mg/kg (group A) at 37–38 weeks gestation in the mothers and within 16 h of birth for the infants without observing any infusion related unexpected serious adverse events. Subsequently, another 10 pairs were infused at a dose of 200 mg/kg (group B), and a further 10 pairs at a dose of 400 mg/kg (group C). Due to the neonatal death of an infant who did not receive HIVIGLOB in the 200 mg/kg group, an additional pair was enrolled. The last infant born in the 400 mg/kg group also died at birth before HIVIGLOB was given, but was not replaced.
Inclusion/exclusion criteria
For study eligibility, women had to be ≥ 18 years old, 28–34 weeks pregnant, live within 15 km of Mulago hospital and provide written informed consent. Eligible women had a plasma HIV-1 RNA level > 2000 copies/ml, hemoglobin > 8.0 g/dl, platelet count > 75 × 109/l, absolute neutrophil count > 1 × 1010/l, alanine aminotransferase level < 5 times the upper limit of normal, bilirubin < 2.5 mg/dl, a serum creatinine < 2.0 mg/dl, and a fetal ultrasound without congenital anomalies. Women were excluded if they received antiretroviral medication, had known or suspected hypersensitivity to immune globulin preparations, had a history of spontaneous abortion or stillbirth, and/or had evidence of clinically significant disease that would compromise their ability to complete the trial. Infants were excluded from receiving HIVIGLOB if they needed an exchange transfusion or had a condition that the on-site pediatrician believed was incompatible with life or would be exacerbated by infusion of HIVIGLOB.
Infusion protocol and pharmacokinetic evaluations
HIVIGLOB was administered via peripheral vein to mothers at 37 weeks gestation by gravity drip, starting at a rate of 0.02 ml/kg per min and increasing to 0.08 ml/kg per min if no reactions were seen. HIVIGLOB was administered to neonates within 16 h of birth via peripheral vein by autosyringe, starting at a rate of 0.01 ml/kg per min and increasing to 0.08 ml/kg per min over 1 h. After infusion, mothers and infants remained hospitalized for 24 h. Anti-p24 antibody was selected as the primary determinant of the pharmacokinetics of HIVIGLOB. Maternal blood samples for p24 antibody were obtained at enrollment (30–36 weeks gestation), pre-infusion, and 1 and 24 h post-infusion, delivery, and 6 weeks post-delivery. Infant blood samples for measurement of p24 antibody were obtained from cord blood (pre-infusion), 1 and 24 h post-infusion, 6 and 14 weeks of age.
Monitoring evaluations
For mothers, clinical examination was performed at entry, infusion, delivery, and 1, 6, and 14 weeks post-delivery. Laboratory evaluations included a complete blood count and differential, CD4 cell count, and chemistries (alanine aminotransferase, bilirubin, creatinine) at enrollment, infusion, delivery, 6 and 14 (except chemistries) weeks post-delivery. Virologic assays included acid-dissociated p24 antigen levels and plasma HIV-1 RNA levels at enrollment, pre-infusion, 24 h post-infusion, delivery, 6 and 14 (RNA only) weeks post-delivery. For infants, clinical examination was performed at birth, pre- and 24 h post-HIVIGLOB infusion, at 1, 6, 10, and 14 weeks, then at 6, 9, 12, 18, 24, and 30 months of age. Laboratory evaluations included a complete blood count and differential, CD4 cell count, and chemistries (alanine aminotransferase, bilirubin, creatinine) at birth, and 1 and 6 weeks. Detection of HIV-1 infection was assessed by HIV-1 DNA PCR (birth, 1.5, 3.5, 6 months), qualitative HIV-1 RNA RT–PCR (birth, 1.5, 3.5, 6, 12, 18, 30 months), HIV-1 cell culture (birth, 1.5, 3.5, 6, 12 months), HIV-1 enzyme immunoassay (EIA) (12, 18, 30 months), and HIV-1 Western blot of infants with a positive HIV-1 EIA at 18 and 30 months of age. An infant was considered HIV-1 infected if positive by one or more of these tests at two different time-points or was HIV-1 Western blot positive at ≥ 18 months of age.
Laboratory assays
All laboratory evaluations, except the p24 antibody levels, were performed on-site in Kampala at the Makerere University-Johns Hopkins University-Case Western Reserve University Collaborative Laboratory. Automated hematology and CD4 cell counts were performed using a Coulter T540 and EPIC Profile II flow cytometer (Coulter Co., Miami, Florida, USA). Automated chemistries were performed on a Kodak Desktop Ectachem DT60 chemistry analyzer (Eastman Kodak Co., Rochester, New York, USA). Presence of HIV-1 antibody was determined using a licensed commercial EIA (Recombigen HIV-1 EIA, Cambridge Biotech Co., Worcester, Massachusetts, USA) with confirmation using a licensed Western Blot assay (Bio-Rad, Emeryville, California, USA). HIV-1 qualitative cultures of PBMC were performed using a sensitive HIV isolation procedure [31].
Detection of HIV-1 DNA was performed using the Roche Amplicor HIV-1 kit, with the use of additional primers (SK151, SK145) to efficiently amplify HIV-1 subtypes A and D, the predominant subtypes in Uganda [32]. The measurement of HIV-1 RNA in 200 μl of EDTA-treated plasma was performed using the Roche Amplicor Monitor HIV-1 kit, also with the additional primers SK151 and SK145.
Detection and measurement of HIV-1 p24 antigen in 200 μl of plasma after acid-dissociation of immune complexes was performed using a commercial kit (Coulter Corp, Miami, Florida, USA). Measurement of HIV-1 p24 antibody was performed by Quest Laboratories (Baltimore, Maryland, USA) using a specific competitive EIA (Envacore) [28].
Statistical analysis
Results were summarized for each dosing group by descriptive statistics, including HIV-1 infection in the infant detected at birth, 6, and 14 weeks. Given the sample size and non-normal distributions of CD4 cell count, HIV-1 RNA viral load and p24 antibody level, non-parametric tests (Mann–Whitney U and Wilcoxon signed-rank) were used to determine statistical significance between dosing groups and between infant infection status.
Additional analyses recoding maternal CD4 cell count, HIV-1 p24 antigen response and viremia response was done using Chi-square contingency table analysis. Maternal CD4 cell count response was defined as an increase > 50 × 106cells/l or a 25% increase from pre-infusion to delivery whichever was greater. Viremia response (by plasma HIV-1 RNA levels) was defined as a > 0.5 log10 decrease from the pre-infusion viral titer to that at delivery.
All analyses were conducted using SPSS for Windows, (V10.0), with statistical significance set at P < 0.05. Bonferroni adjustments for multiple comparisons were done when comparing pre-infusion levels to follow-up levels.
The pharmacokinetic analyses of HIV-1 p24 antibody were performed using methods consistent with those previously reported in other trials [33,34]. The estimated terminal half-life (t1/2) of HIV-1 p24 antibody in the mothers was calculated by linear regression using the last three sample points after infusion. For infants, the t1/2 of HIV-1 p24 antibody was estimated from samples obtained at 6 and 14 weeks post-infusion. The Mann–Whitney U test was used to evaluate the association between HIV-1 vertical transmission and maternal plasma HIV-1 RNA levels, CD4 cell count, and HIV-1 p24 antibody. Values were log10 transformed to allow parametric testing.
Results
Study population
Thirty-one HIV infected pregnant Ugandan women were infused with HIVIGLOB. Ten women received 50 mg/kg (dosing group A), 11 received 200 mg/kg (group B), and 10 received 400 mg/kg (group C). Two infants died prior to HIVIGLOB infusion, resulting in 10 infants in group A who received a 50 mg/kg HIVIGLOB infusion, 10 in group B receiving 200 mg/kg, and nine in group C receiving 400 mg/kg. Characteristics of study women and their infants are shown in Table 1. No differences at baseline between the three sequential dose escalating study arms were observed in CD4 cell count or HIV-1 RNA viral load; however, a significant difference in HIV-1 p24 antibody levels was observed between group A and group C (P = 0.035). Delivery occurred at a median of 15 days after receipt of the HIVIGLOB infusion (range, 3–46 days). Infants were infused a median of 5.92 h after birth (range, 1.67–16.50 h).
Safety and tolerance
HIVIGLOB infusions were well tolerated by both mothers and infants, with only one grade 3 maternal adverse event in group B that was thought to be related to the HIVIGLOB infusion. This consisted of persistent fever despite treatment with diphenhydramine and aspirin leading to discontinuation of the infusion with 15 of 225 ml remaining. In addition, the same woman had one episode of fetal heart rate variability with a decrease to 105 beats/min and instant resolution to 195 beats/min. There were five asymptomatic grade 1–2 infusion related blood pressure changes: two in group A, one in group B and two in group C. These mild events resolved either without any therapy or infusion changes (three) or with diphenhydramine and/or acetaminophen treatment (two). In addition, there was one grade 2 fever with chills and thigh pain in group C that resolved with diphenhydramine, acetaminophen, and chloroquine.
Three maternal deaths were observed. Causes of death included tuberculosis and wasting syndrome in one woman with AIDS, gastroenteritis and pneumonia in the second woman, and persistent gastroenteritis and wasting in the third. All three deaths were determined to be related to HIV disease and not to the HIVIGLOB infusion.
One neonate in group A had hematemesis during and after the infusion, possibly due to swallowed maternal blood at delivery. This was classified as a possibly related grade 3 toxicity since some contribution by the infusion could not be completely ruled out. In addition, two group A infants had grade 2 fever (39–39.5°C) 12–14 h after the infusion that were considered to be possibly related to the HIVIGLOB. However, there was no growth on bacterial culture of the HIVIGLOB product that each infant received. The fevers resolved after 2 days of intravenous ampicillin and gentamicin and 5–8 days of oral ampicillin.
There have been eight deaths among the 31 study infants (one in group A, five in group B, and two in group C) during the 30-month study period. Death occurred in two infants within 24 h of birth; these infants were HIV negative on birth specimens and never directly received HIVIGLOB. These deaths were due to sepsis after prolonged rupture of membranes and prematurity with respiratory distress. Neither death was thought to be related to the HIVIGLOB infusions received by their mothers 12 and 6 days prior to delivery. The other six deaths occurred in HIV infected infants, three of whom were HIV positive at birth. Deaths occurred from 5 to 28 months of age and were due to pneumonia, tuberculosis, gastroenteritis with dehydration, fever, or marasmus. Based on the documented HIV infection in these infants, the history of symptomatic HIV infection, and the elapsed time since the HIVIGLOB infusion, the on-site investigators judged that these infant deaths were not related to the HIVIGLOB infusion, but were rather due to rapidly progressing HIV disease.
HIVIGLOB pharmacokinetics
Mothers
The median maternal pre-infusion anti-p24 antibody titer for all mothers was 1 : 18 282 (range, 1 : 25 to 1 : 6 576 925) with medians of 190 723 (group A), 18 282 (group B), and 6579 (group C) (Table 2). The median p24 antibody titers over time by dosing group are shown in Table 2. The increase in titer post-infusion was not significant in group A, but was significant at 1 h post-infusion in group B and 24 h post-infusion in group C. The t1/2 of anti-p24 antibody in the mothers in group A (n = 8), group B (n = 5), and group C (n = 7) were 27, 15, and 29 days respectively for an overall median t1/2 of 28 days (range, 9–195 days). Eleven women did not contribute to the analysis of HIVIGLOB half-life, four due to missing specimens and seven due to increasing p24 antibody titers over time.
Infants
The median anti-p24 antibody titer in cord blood was 1 : 14 191 (range, 1 : 1437 to 1 : 2 348 010) with an overall cord blood : maternal ratio at the time of delivery of 0.47 : 1 (range, 0.15 : 1 to 2.22 : 1) (Table 2). Post-infusion (1 and 24 h) increase in p24 antibody titer was significant only in group C. There were significant decreases in p24 antibody titer by 6 to 14 weeks post-infusion as the infused antibody disappeared. The majority of infants were not infected and thus not producing their own antibody. The median t1/2 of HIV-1 p24 antibody in the infants was 30 days in group A (n = 10), 32 days in group B (n = 4), and 29 days in group C (n = 9), for an overall t1/2 of 30 days (range, 11 to 264 days). Eight infants (seven in group B, one in group C) were unevaluable for the half-life measurement: two died, four had insufficient specimens, and two infected infants had an increased p24 antibody level at 14 weeks compared to 6 weeks.
Effect of HIVIGLOB on maternal HIV-1 RNA level, CD4 cell count, and HIV-1 p24 antigen
The median pre-infusion maternal plasma HIV-1 RNA level for 30 of 31 mothers was 36 001 copies/ml (range, undetectable to 526 346 copies/ml) (Table 2). Median HIV-1 RNA copy numbers over time by dosing group are shown in Table 2. None of these changes were significant. For individual study women, two out of 10 in group A, one out of nine in group B, and two out of 10 women in group C had a > 0.5 log10 decrease in viral load from pre-infusion to delivery. All mothers were negative for HIV-1 p24 antigen pre-infusion despite the high levels of plasma HIV-1 RNA. The median pre-infusion maternal CD4 cell count for the 31 mothers was 523 × 106cells/l (range, 16 × 106–1035 × 106 cells/l). The median CD4 cell count over time in each arm is shown in Table 2. None of these changes were significant. For individual study women, two out of 10 in group A, three out of nine in group B, and one out of seven women in group C had an increase in CD4 cell count > 50 × 106/l or 25% from pre-infusion to delivery.
HIV-1 infant infection status
The last 30-month study visit was completed in January 2000. All children had stopped breast feeding before their 30-month final visit. At the completion of the study, a total of 11 out of 29 (37.9%) infants were confirmed to be HIV infected by HIV RNA RT–PCR, HIV-1 DNA PCR, and/or culture on two different specimens. Four infants had HIV detected at birth (one in group A, three in group B). One additional infant in each dosing group was identified as infected at 6 weeks age for a total of 2 out of 10 in group A, four out of 10 in group B, and one out of nine in group C. Thus, the overall transmission rate for the time that the HIVIGLOB would be expected to have an impact was 21%. HIV infection through breast feeding was later detected in one infant in group B at 3 months of age, in two additional infants at 12 months of age (one in group A and one in group C), and one additional infant in group A at 18 months of age. Sixteen infants have no laboratory evidence of HIV-1 infection through 30 months of age. Two infants had a single positive plasma HIV-1 RNA result (one at birth, one at 3 months of age), but have tested negative for HIV-1 RNA, DNA, and/or HIV-1 culture on four subsequent visits and were both EIA and HIV-1 plasma RNA negative at 30 months of age.
Association between HIV-1 vertical transmission and maternal plasma HIV-1 RNA level, CD4 cell count and anti-p24 antibody
Only maternal plasma HIV-1 RNA levels up to delivery were significantly associated with HIV-1 infection in the infant by 6 weeks of age (pre-infusion, P = 0.001; 1 h post-infusion, P = 0.014; 24 h post-infusion, P = 0.003; delivery, P = 0.054;Fig. 1). Median plasma HIV-1 RNA levels (copies/ml) for transmitting and non-transmitting mothers were 112 700 (range, 47 940–526 346) and 12 535 (range, undetectable to 426 977) at baseline, respectively, and 79 571 (range, 26 299–565 019) and 28 506 (range, undetectable to 327 124) at delivery. Maternal CD4 cell counts (not shown) and anti-p24 antibody levels (data not shown) in the mother and baby were not significantly associated with HIV-1 vertical transmission at any time point. Similar results were obtained when evaluating the infant infection status at 14 weeks.
Discussion
Based on the HBV and animal SIV models, passive and active immunotherapy could play an important role in preventing vertical HIV transmission. In particular, the possible protective role of active and passive immunization against early breast feeding transmission needs to be examined. As a first step, this phase I/II study using an HIV hyperimmune globulin preparation from Ugandan donors demonstrates the safety, tolerance and pharmacokinetics of the HIVIGLOB product.
Previous studies have identified the protective effect of maternal neutralizing antibody on vertical transmission [35–37]. Several animal studies have shown protection against vertical transmission of SIV, SHIV, and FIV with passive immunization of immunoglobulin or monoclonal antibody preparations with neutralizing capability [10–18]. Results from early animal studies were inconsistent, with some studies showing protection from infection with use of immunoglobulin or monoclonal antibody preparations [10–18] and others showing no protection [38–40]. Many of the early studies, however, challenged animals intravenously with high doses of highly pathogenic virus [38–40]. More recent animal studies have simulated perinatal transmission scenarios using lower dose oral, vaginal, or peritoneal inoculation of virus, mimicking the potential oral/mucosal routes of exposure for infants born to HIV infected women [11,12,16,17]. The majority of these studies have found significant protection offered by these polyclonal immunoglobulin or monoclonal antibody products.
There has been one other perinatal trial of HIV hyperimmune globulin—ACTG protocol 185 investigated the efficacy of zidovudine plus HIV hyperimmune globulin (HIVIG) versus zidovudine plus standard immunoglobulin lacking antibody to HIV (IVIG) for the reduction of HIV vertical transmission [34,41]. This randomized double blind controlled phase III efficacy trial resulted in lower than expected vertical transmission rates that were not significantly different, 4.1% in the HIVIG group and 6.0% in the IVIG group [41]. There was, however, a trend to decreased transmission rates in the HIVIG arm for women with CD4 cell counts < 200 × 106/l (HIVIG, 5.6%; IVIG, 14.9%;P = 0.12) [41]. Unfortunately, the trial stopped recruitment prior to reaching a sample size adequate for reaching statistical significance in this subset of patients.
In this Ugandan study, HIVIGLOB was well tolerated by both mothers and infants at all dosing levels with no unexpected serious adverse events due to the HIVIGLOB infusions. Minor adverse events observed were similar to experiences with other immunoglobulin preparations [7,19–23,33,34,41,42]. Infant study deaths were related to neonatal sepsis, prematurity or progression of HIV disease, and not to receipt of HIVIGLOB.
The pharmacokinetics of the HIVIGLOB showed considerable variability between study subjects. This variability as well as the small sample size in each of the dosing arms made determination of statistical significance difficult for each of the parameters evaluated. The t1/2 of the single 200 mg/kg HIVIGLOB dose to the mother and the infant in our study (mother, 15 days; infant, 32 days) were similar to those reported after the first infusion of 200 mg/kg HIVIG in ACTG 185 (mother, 15 days; infant, 30 days) [34]. The first infusion in ACTG 185, however, was given at 20–30 weeks gestation compared to the 37–38 week infusion in our study. Interestingly, the median cord : maternal ratio in this study was 0.47 : 1 (range, 0.15 : 1 to 2.21 : 1) while the mean ratio in ACTG 185 was 1.16 : 1 with 55% coeficient of variation. Possible explanations for this significant difference include: the extremely high pre-infusion p24 antibody levels in the Ugandan mothers (median, 18 282; range, 25–6 576 925 pg/ml) compared to those in the USA (median, 22; range, 5–87 516) ([34], C. Fletcher personal communication); the extreme variability in cord blood : maternal p24 antibody levels in the Ugandan cohort; or the cumulative effect of multiple antibody transfusions over time in the fetus in the 185 study compared to the effect of a single infusion only in the Ugandan study.
Similar to other studies, we found no decrease in maternal viral load as measured by plasma HIV RNA or increase in CD4 cell counts after HIVIGLOB infusion. Pregnant Ugandan women have very low levels of p24 antigenemia as determined in previous cohorts [43] so it was not surprising that there were no women in this small sample who were p24 antigen positive at baseline. All previous studies using HIV immunoglobulin have shown rapid clearance of p24 antigenemia with little effect on other measures of HIV viral load. Despite lack of change in maternal HIV RNA levels, there may have been a decrease in infectious virus titer, which we did not attempt to measure. Significant increases in p24 antibody titer post-infusion in mothers and infants were detected mainly in the 400 mg/kg arm, but the increased polyclonal antibody titer in the infant may provide enhanced protection against HIV transmission.
Due to the phase I/II design, the sample size for this study was insufficient for any determination of product efficacy. The relatively low proportion of infected infants at 6 weeks in the 400 mg/kg HIVIGLOB dosing arm is, however, encouraging. Interestingly, the 400 mg/kg dosing group, which had the lowest p24 antibody titers at baseline, delivery, and in cord blood had the lowest transmission rate. Although the overall estimated vertical HIV transmission rate was relatively high, it should be noted that only women with plasma HIV RNA levels > 2000 copies/ml were eligible for the study, excluding half of the women screened. Therefore, the women in this phase I/II study had more advanced disease than women in our natural cohort studies where the transmission rate was similar (26%) [44]. In addition, four of the 11 HIV infected infants were infected through breast feeding after 6 weeks of age, when the HIVIGLOB is unlikely to have any effect.
Data from this Ugandan phase I/II study show that an HIV immune globulin product derived from HIV infected Ugandan donors is safe, well tolerated, and has pharmacokinetic properties consistent with a similar HIVIG product manufactured in the USA. Results suggest that a 400 mg/kg dose of HIVIGLOB would be the most appropriate dose for an efficacy trial due to the significant increases in antibody titer following infusion in women and their infants and the low early transmission rate. Results of the recent animal studies, as well as the trend to decreased transmission for women with < 200 × 106 CD4 cells/l in the HIVIG arm of ACTG 185, offer support for the concept of using passive immunotherapy to decrease HIV transmission from mother to child. Therefore, a phase III trial of HIVIGLOB in combination with antiretroviral therapy is warranted to assess whether the addition of passive antibody is capable of further decreasing HIV transmission compared to antiretroviral therapy alone.
Acknowledgments
The authors wish to thank: J. Hinkula, J. Albert, M. Thorhagen, J. Christer Höglund, B. Grynblat, Department of Virology, the Swedish Institute for Infectious Disease Control, Stockholm; G. Bruse, Department of Bacteriology, the Swedish Institute for Infectious Disease Control, Stockholm; K. Lidman, A. Aleus, Department of Infectious Diseases, Danderyd Hospital, Stockholm; E. Sandström, G. Bratt, Department of Venhälsan, Södersjukhuset, Stockholm; K. Hanngren, SBL Vaccin AB, Solna, Sweden; E. Piwowar-Manning, Department of Pathology, Johns Hopkins University, Baltimore, MD, USA, J. Goddard, G. Anyang, Nakasero Blood Bank, Kampala, Uganda (who all made the fractionation, preparation, safety testing, virologic and immunologic testing, and control work for the HIVIGLOB product possible). The authors also thank A. Murarka, Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University, Baltimore MD, USA, and C. Ndugwa, Department of Pediatrics, Makerere University, Kampala Uganda (for the care of the mothers and children in the trial), and J. Nakakande, Mulago Hospital, Kampala, Uganda (for administering the maternal HIVIGLOB infusions). We also thank F. Robbins, Department of Epidemiology and Biostatistics, Case Western Reserve University for his support and advice for this project.
References
1. Shaffer N, Chuachoowong R, Mock PA.
et al. Short-course zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial. Lancet 1999, 353: 773–780.
2. Wiktor SZ, Ekpini E, Karon JM.
et al. Short-course oral zidovudine for prevention of mother-to-child transmission of HIV-1 in Abidjan, Cote d'Ivoire: a randomised trial. Lancet 1999, 353: 781–785.
3. Dabis F, Msellati P, Meda N.
et al. 6-month efficacy, tolerance, and acceptability of a short regimen of oral zidovudine to reduce vertical transmission of HIV in breastfed children in Cote d'Ivoire and Burkina Faso: a double-blind placebo-controlled multicentre trial. Lancet 1999, 353: 786–792.
4. Guay L, Musoke P, Fleming T.
et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomized trial. Lancet 1999, 354: 795–802.
5. Nduati R, John G, Mbori-Ngacha D.
et al. Effect of breastfeeding and formula feeding on transmission of HIV-1: a randomized clinical trial. JAMA 2000, 283: 1167–1174.
6. Wong VC, Ip HM, Reesink HW.
et al. Prevention of the HBsAg carrier state in newborn infants of mothers who are chronic carriers of HBsAg and HbeAg by administration of hepatitis-B vaccine and hepatitis-B immunoglobulin: Double blind randomized placebo controlled study. Lancet 1984, 1 (8383): 921–926.
7. Sacher RA, King JC. Intravenous gamma-globulin in pregnancy: a review. Obstet and Gynecol Survey 1988, 44: 25–34.
8. Stiehm ER. Role of immunoglobulin therapy in neonatal infections: where we stand today. Rev Infect Dis 1990, 12 (suppl 4): S439–S442.
9. Beasley RP, Hwang LY, Stevens C.
et al. Efficacy of Hepatitis B immune globulin for prevention of perinatal transmission of the Hepatitis B virus carrier state: final report of a randomized double-blind, placebo-controlled trial. Hepatology 1983, 3: 135–141.
10. Putkonen P, Thorstensson R, Ghavamzadeh L.
et al. Prevention of HIV-2 and SIV
sm infection by passive immunization in cynomolgous monkeys. Nature 1991, 352: 436–438.
11. Van Rompay K, Berardi C, Dillard-Telm S.
et al. Passive immunization of newborn rhesus macaques prevents oral simian immunodeficiency virus infection. J Infect Dis 1998, 177: 1247–1259.
12. Mascola J, Stiegler G, VanCott T.
et al. Protection of macaques against vaginal transmission of a pathogenic HIV-1/SIV chimeric virus by passive infusion of neutralizing antibodies. Nature Med 2000, 6: 207–210.
13. Mascola J, Louder M, VanCott T.
et al. Potent and synergistic neutralization of human immunodeficiency virus (HIV) type 1 primary isolates by hyperimmune anti-HIV immunoglobulin combined with monoclonal antibodies 2F5 and 2G12. J Virol 1997, 71: 7198–7106.
14. Mascola J, Lewis M, Stiegler G.
et al. Protection of macaques against pathogenic simian/human immunodeficiency virus 89.6PD by passive transfer of neutralizing antibodies. J Virol 1999, 73: 4009–4018.
15. Emini EA, Schleif WA, Nunberg JH.
et al. Prevention of HIV-1 infection in chimpanzees by gp120 V3 domain-specific monoclonal antibody. Nature 1992, 355: 728–730.
16. Pu R, Okada S, Little E, Xu B, Stoffs W, Yamamoto JK. Protection of neonatal kittens against feline immunodeficiency virus infection with passive maternal antiviral antibodies. AIDS 1995, 9: 235–242.
17. Baba TW, Liska V, Hofmann-Lehmann R.
et al. Human neutralizing monoclonal antibodies of the IgG1 subtype protect against mucosal simian-human immunodeficiency virus infection. Nature Med 2000, 6: 200–206.
18. Prince AM, Reesink H, Pascual D.
et al. Prevention of HIV infection by passive immunization with HIV immunoglobulin. AIDS Res Hum Retroviruses 1991, 7: 971–973.
19. Jacobson J, Colman N, Ostrow N.
et al. Passive immunotherapy in the treatment of advanced human immunodeficiency virus infection. J Infect Dis 1993, 168: 298–305.
20. Karpas A, Hill F, Youle M.
et al. Effects of passive immunization in patients with AIDS related complex and AIDS. Proc Natl Acad Sci USA 1988, 85: 9234–9237.
21. Jackson GG, Perkins JT, Rubenis M.
et al. Passive immunoneutralisation of human immunodeficiency virus in patients with advanced AIDS. Lancet 1988, 2: 647–652.
22. Levy J, Youvan T, Lee M, and the Passive Hyperimmune Therapy Study Group. Passive hyperimmune plasma therapy in the treatment of acquired immunodeficiency syndrome: results of a 12-Month multicenter double-blind controlled trial. Blood 1994, 84: 2130–2135.
23. Vittecoq D, Mattlinger B, Barre-Sinoussi F.
et al. Passive immunotherapy in AIDS: a randomized trial of serial human immunodeficiency virus-positive transfusions of plasma rich in p24 antibodies versus transfusions of seronegative plasma. J Infect Dis 1992, 165: 363–368.
24. Hinkula J, Bratt G, Gilljam G.
et al. Immunological and virological interactions in patients receiving passive immunotherapy with HIV-1 neutralizing monoclonal antibodies. J AIDS 1994, 7: 940–951.
25. Cummins LM, Weinhold KJ, Matthews TJ.
et al. Preparation and characterization of an intravenous solution of IgG from human immunodeficiency virus-seropositive donors. Blood 1991, 77: 1111–1117.
26. Piet MPJ, Chin S, Prince AM. Inactivation of viruses in plasma on treatment with tri(N-butyl) phosphate (TNBP) detergent mixtures.
XI International Congress on Thrombosis and Haemostasis. Brussels, July 1987 [abstract 1350].
27. Falksveden L, Lundblad G. Ion exchange and polyethylene glycol precipitation of immunoglobulin G. In
Methods of Plasma Protein Fractionation. Edited by Curling JM. New York: Academic Press; 1980: 93–105.
28. Allain JP, Laurian Y, Paul DA, Senn D. Serological markers in early states of human immunodeficiency virus infection in hemophiliacs. Lancet 1986, ii: 1233–1236.
29. Oram JD, Downing RJ, Roff M, Clegg JC, Serwadda D, Carswell JW. Nucleotide sequence of a Ugandan HIV-1 provirus reveals genetic diversity from other HIV-1 isolates. AIDS Res Hum Retroviruses 1990, 6: 1073–1078.
30. Albert J, Franzen L, Jansson M.
et al. Ugandan HIV-1 V3 loop sequences closely related to the U.S./European Consensus. Virol 1992, 190: 674–681.
31. Jackson JB, Coombs RW, Sannerud K.
et al. Rapid and sensitive viral culture method for human immunodeficiency virus type 1. J Clin Microbiol 1988, 26: 1416–1418.
32. Triques K, Coste J, Perret JL.
et al. Efficiencies of four versions of the AMPLICOR HIV-1 MONITOR test for quantification of different subtypes of human immunodeficiency virus type 1. J Clin Microbiol 1999, 37: 110–116.
33. Fletcher C, Goodroad BK, Cummins LM, Henry K, Balfour H, Rhame F. Pharmacokinetics of hyperimmune anti-human immunodeficiency virus immunoglobulin in persons with AIDS. Antimicrob Agents Chemother 1997, 41: 1571–1574.
34. Lambert J, Mofenson L, Fletcher C.
et al. Safety and pharmacokinetics of hyperimmune anti-human immunodeficiency virus (HIV) immunoglobulin administered to HIV-infected pregnant women and their newborns. J Infect Dis 1997, 175: 283–291.
35. Rossi P, Moschese V, Broliden PA.
et al. Presence of maternal antibodies to human immunodeficiency virus 1 envelope glycoprotein correlates with the uninfected status of children born to seropositive mothers. Proc Natl Acad Sci USA 1989, 86: 8055–8058.
36. Devash Y, Calvelli TA, Wood DG, Reagan KJ, Rubinstein A. Vertical transmission of human immunodeficiency virus is correlated with the absence of high-affinity/avidity maternal antibodies to the gp120 principal neutralizing domain. Proc Natl Acad Sci USA 1990, 87: 3445–3449.
37. Scarlatti G, Rossi AJ, Hodara V, Biraghi P, Muggiasca L, Fenyo EM. Mother-to-child transmission of human immunodeficiency virus type 1: correlation with neutralizing antibodies against primary isolates. J Infect Dis 1993, 168: 207–210.
38. Prince AM, Horowitz B, Baker L.
et al. Failure of a human immunodeficiency virus immune globulin to protect chimpanzees against experimental challenge with HIV. Proc Natl Acad Sci USA 1988, 85: 6944–6948.
39. Siegel F, Kurth R, Norley S. Neither whole inactivated virus immunogen nor passive immunoglobulin transfer protect against SIV
agm infection in the African green monkey natural host. J Acquir Immune Defic Syndr Hum Retrovirol 1995, 8: 217–226.
40. Kent KA, Kitchin P, Mills KHG.
et al. Passive immunization of Cynomolgus Macaques with immune sera or a pool of neutralizing monoclonal antibodies failed to protect against challenge with SIVmac251. AIDS Res Hum Retroviruses 1994, 10: 189–194.
41. Stiehm ER, Lambert J, Mofenson L.
et al. Efficacy of zidovudine and human immunodeficiency virus (HIV) hyperimmune immunoglobulin for reducing perinatal HIV transmission from HIV-infected women with advanced disease: Results of Pediatric AIDS Clinical Trials Group Protocol 185. J Infect Dis 1999, 179: 567–575.
42. Stiehm ER, Fletcher CV, Mofenson LM.
et al. Use of human immunodeficiency virus (HIV) human hyperimmune immunoglobulin in HIV type 1-infected children (Pediatric AIDS clinical trials group protocol 273). J Infect Dis 2000, 181: 548–554.
43. Jackson JB, Kataaha P, Hom D.
et al. Beta2-microglobulin, HIV-1 p24 antibody and acid-dissociated HIV-1 p24 antigen levels: predictive markers for vertical transmission of HIV-1 in pregnant Ugandan women. AIDS 1993, 7: 1475–1479.
44. Guay L, Hom D, Mmiro F.
et al. The detection of HIV-1 DNA and p24 antigen in breast milk of HIV-1 infected Ugandan women and vertical transmission. Pediatrics 1996, 98: 438–444.
