Pharmaco-immunological-virological dynamics in intrapartum HIV-1 transmission (PIVD study)

Abstract

Background: Multiple factors contribute to mother-to-child transmission (MTCT) of HIV-1, including virological, obstetric and biological factors. Other possible contributory determinants for high MTCT rates include immunological factors such as host genetics and viral genetic variations. Despite several therapeutic, prophylactic and obstetric interventions to reduce the proportion of infants infected during labour and delivery, mechanisms for intrapartum HIV-1 transmission remain elusive and current interventions, could, therefore remain sub-optimal. Much controversy has surrounded the correlation of HIV-1 RNA (viral load) in the systemic and genital compartments of women. The influence of short-term antiretroviral (ARV) drugs on genital tract HIV-1 is also unclear. At the time the present study was initiated, a regimen of maternal intrapartum and neonatal postpartum single-dose Nevirapine (sdNVP) was the standard of care for the prevention of mother-to-child transmission (PMTCT). In most low and middle-income countries, including South Africa, sdNVP has been documented as effective intrapartum HIV-1 prevention based on plasma pharmacokinetic levels, decreased viral loads (HIV-1 RNA) and reduced rates of intrapartum transmission, yet operational studies continue to report high intrapartum transmission rates despite the administration of sdNVP. As a result perinatal HIV-1 transmission remains a significant public health concern in several African countries. Aim: The primary aim of this study was to describe the pharmacological dynamics of Nevirapine in association with virological and immunological risk factors for intrapartum HIV-1 transmission in a South African PMTCT programme where sdNVP was the standard of care. Methods: Following regulatory approval from the Biomedical Research Ethics Committee at the University of KwaZulu-Natal (UKZN), one hundred and twenty pregnant HIV-infected women who received the sdNVP regimen for prevention of mother-to-child HIV-1 transmission were enrolled between April-December 2006 at King Edward VIII Hospital (KEH) in Durban. Blood and cervicovaginal lavage (CVL) samples were collected from women at pre-NVP (during pregnancy) and post-NVP dosing (during labour/delivery). In addition to infant blood sampling at birth (post-NVP), postnatal infants were assessed at four and six weeks postnatally. Pharmacological laboratory investigations involved measurement of NVP drug concentration by Tandem Mass spectrophotometry. Virological investigations comprised HIV-1 RNA (viral load) quantitation, HIV-1 drug resistance testing (HIV-1 transmitting women only) and HIV-1 DNA PCR testing (infants only). Immunological investigations were only undertaken in a selected case-control subset of HIV-1 transmitting women and their infants. In this component, laboratory investigations included the determination of CCL3 and CCL3-L1 gene copy numbers, identification of single nucleotide polymorphisms (SNP’s) and haplotype characterisation of the CCL3 gene. All women were also screened for the presence of sexually transmitted infections (STI’s) during pregnancy. Results: One hundred and twenty women were enrolled onto this study. Of these, 110 women delivered 117 live infants (103 singletons and 7 twin pairs). Twelve (10.9%) women transmitted HIV-1 to their infants, while 95 (86.0%) were classified as non-transmitters. As a result of seven twin deliveries, the infant cohort comprised of 117 infants in total. Following two separate DNA PCR tests, HIV-1 infection was identified in 14 (11.9%) of study infants while the remaining 90 (76.9%) were exposed-uninfected. HIV infection status remained unknown for 13 infants due to infant demise (1.7%), lost to follow-up (7.7%) or study withdrawal (1.7%). During active labour (sampling that was best representative of the intrapartum phase) and within 20 hours of dosing, the median NVP concentration of 1070 ng/ml in the maternal systemic compartment was almost 44 times higher than the NVP levels detected in the genital compartment [24.5 ng/ml] (p < 0.001). NVP drug levels were below the 100 ng/ml therapeutic target in seven (13.7%) of 51 plasma and in all 39 CVL samples. While no significant association was found between NVP concentration in the systemic compartment and HIV-1 transmission (p = 0.4), this association was statistically significant in the genital compartment(p = 0.02). The median plasma NVP level detected among infants at birth was 83 times above the IC50 WT (10 ng/ml) and eight times higher than the 100 ng/ml therapeutic target for NVP. More than 71.0% of the infants achieved NVP drug levels above the therapeutic target. In general, higher levels of HIV-1 RNA (viral load) were observed in maternal plasma when compared to CVL. Following intrapartum sdNVP dosing, reduction in HIV-1 RNA levels did occur, however R80.0% of the women experienced no change to their HIV-1 RNA levels in both systemic and genital compartments during active labour. These findings were further supported by the strong correlation observed when comparing pre and post-NVP HIV-1 RNA levels in both maternal systemic [r = 0.81, p < 0.0001] and genital compartments [r = 0.80, p < 0.0001] during active labour. HIV-1 transmitting women had significantly higher viral loads than their non-transmitting counterparts in systemic and genital compartments, before and after intrapartum sdNVP administration. In terms of perinatal transmission this observation was only statistically significant for plasma (p = 0.02) and not CVL (p = 0.7). Maternal viral load was inversely correlated with maternal CD4 cell counts in both systemic and genital compartments. Almost 40.0% of women in this study had at least one type of STI detected during pregnancy. Maternal STI’s were detected in four (66.6%) intrapartum transmitting women and in 38 (38.8%) of non-transmitting women. No significant association was observed between the presence of maternal STI’s and the risk for intrapartum MTCT (p = 0.2,RR: 2.90, 95% CI: 0.60-15.40). The presence of maternal STI’s was associated with higher median viral loads in both systemic and genital compartments of all women, independent of intrapartum HIV-1 transmission. Despite trial-like conditions and optimal sdNVP dosing, the overall MTCT rate in this exclusively formula-fed cohort was 11.9%, of which 50.0% were in utero and 50.0% were intrapartum HIV-1 transmissions. In utero and intrapartum MTCT rates were 5.9% and 5.9% respectively. Discussion/Conclusion: Detectable CVL HIV-1 RNA that correlated well with plasma HIV-1 RNA, in conjunction with sub-optimal NVP drug concentration in maternal CVL during active labour, suggests that intrapartum HIV-1 infected women continue to act as reservoirs for both vertical and horizontal HIV-1 transmission throughout the duration of pregnancy. These findings confirm that the role of sdNVP in PMTCT was primarily one of infant prophylaxis. This was further supported by relatively unchanged maternal HIV-1 RNA (viral load) during active labour, in both systemic and genital compartments. Early identification of women who need highly active antiretroviral therapy (HAART), and initiation of such therapy as early as possible during pregnancy, not only benefits maternal health but remains the best prophylaxis against mother-to-child HIV-1 transmission. Universal access to HAART and improving strategies to optimize coverage of the current dual ARV regimen sdNVP and Zidovudine for PMTCT remain urgent research priorities in several resource-limited settings. Ongoing STI counseling, intensive screening/testing of women and their partners together promotion of condom usage, safer sex practices and aggressive STI treatment are simple interventions with tremendous impact for PMTCT in resource-limited settings.