When prevention of mother-to-child HIV transmission fails

Introduction In high-income and some middle-income countries, effective programs to prevent mother-to-child transmission (PMTCT) using combination antiretroviral therapy (ART) for pregnant HIV-positive women and antiretroviral prophylaxis for HIV-exposed infants have resulted in 99–100% reduction in vertical transmissions [1]. However, in Sub-Saharan Africa, the implementation of PMTCT programs has been less successful, with new pediatric infections having declined only by 21–86% between 2009 and 2015 [1]. Sub-Saharan Africa lags behind global trends because of several factors, including late or missed HIV testing, incident HIV infection in pregnant women, low uptake of or deferred treatment initiation, suboptimal ART adherence pre and postpartum, late identification and testing of HIV-exposed infants, amongst others [2–7]. The fact that children become HIV-infected despite the use of PMTCT (failing PMTCT) is not only problematic because these infections are preventable, but also because infection during antiretroviral exposure increases the risk of acquiring drug-resistant HIV variants. Consequently, despite a decrease in the overall numbers of vertical HIV infections, the proportion of children who become infected and bear drug-resistant virus early in life is very high (35–64%) in the region [8–10]. The plight of these children is dire with high risk of failure on a suboptimal standard first-line ART regimen, absence of resistance tests to inform drug choices and lack of alternative regimens in case of ART failure – all this in light of the need for lifelong effective therapy. In this opinion article, we provide arguments that increased efforts are needed to prevent HIV drug-resistance early in life in the context of ongoing PMTCT programs and limited ART options for children. Challenges in pediatric HIV management after failure of prevention of mother-to-child transmission Studies in Sub-Saharan Africa have reported that between 35 and 64% of HIV-infected infants have pretreatment drug-resistant HIV (PDR), predominantly associated with the class of nonnucleoside reverse transcriptase inhibitors (NNRTIs) [8–10]. Although WHO has recommended protease inhibitor-based first-line ART since 2013, costs and logistics have largely hindered their accessibility. In 2015, about 77% of African children were still initiated on NNRTI-based regimens [11]. The recent development of a new lopinavir–ritonavir formulation as pellets (capsulated mini-tablets sprinkle formulation) may help improve access by overcoming cold-chain requirements for the liquid formulation and are easy to administer with a simplified weight dosing, as opposed to the tablet formulations. On the other hand, limited production and high costs (∼6× that of NNRTI-based regimens) still hinders access in these settings [12]. Even if access to protease-inhibitor-based regimens would be assured for young children, there are currently no good options available for children who experience ART failure; WHO recommends switching to either two NRTIs and raltegravir or efavirenz (if age 3–10 years) or maintaining the child on the failing protease-inhibitor-based regimen [13]. These options are, however, suboptimal citing the high levels of both NNRTI and NRTI resistance in both pretreated [8–10] and treated children [14]. Prevention of pretreatment drug resistance in infants Preventing PDR in infants will require strengthening the current PMTCT strategies as described below (Table 1).Table 1: Summary of different strategies for optimization of prevention of mother-to-child transmission and prevention of pretreatment drug-resistant HIV in HIV-infected infants.Strengthening the support framework for prevention of mother-to-child transmission to ensure better virological control during pregnancy and breast-feeding Mothers with drug-resistant HIV can transmit the resistant strain to their infants during pregnancy, partus or breastfeeding [15]. Although there are limited data on the prevalence of HIV drug resistance in pregnant women, it is likely that this might be increasing with wide-access to ART in this setting as observed in the general population [16]. For example, a recent study from Botswana showed that the rate of PDR in pregnant women had increased from less than 3% in 2012 to 9.7% in 2015 [17]. Moreover, about 29–50% of mothers have been reported to have detectable viremia during their last trimester [18,19], of whom 60–90% have drug-resistant HIV [20]. A recent study also showed that about 30% of women experience incident viremia during the postpartum phase highlighting an increased risk of transmission of drug-resistant variants during the extended period of breastfeeding [21], possibly attributable to suboptimal adherence. A systematic review from 51 studies before the current era of lifelong triple ART for mothers in PMTCT programs also known as ‘option B+’, reported that nearly 25% of women had suboptimal adherence (<80%) during prepartum with an even higher proportion of 47% during postpartum [22]. A complimentary review from 26 studies conducted between 2001 and 2012 reported that about 49% of mothers are lost-to-follow up during pregnancy and 33% of infants are lost-to-follow up within the first 12 weeks of life [23]. Studies report high attrition under option B+ in the early treatment phase and increased suboptimal adherence during postpartum [2,6]. Therefore, to prevent vertical transmission of drug-resistant HIV, we propose to strengthen the PMTCT support framework in several ways. First, enhanced adherence and retention support can be achieved through peer support-based programs, such as the ‘expert mothers’, enhanced adherence counseling for newly diagnosed mothers, male-partner involvement, phone calls/text-messages reminders and cash-based interventions [24–26]. However, wide-implementation is needed. Second, increased frequency of viral-load monitoring can help to ensure viral suppression during both preterm and the breastfeeding period. The current guidelines recommend viral-load testing at 6 and 12 months postpartum [13], but this is not optimal for preventing the risk of HIV vertical transmission, citing high-levels of viral nonsuppression at time of delivery [18,19] and during the breastfeeding period [20]. The minimal frequency of testing would preferably be every 3 months during partum and until cessation of breast-feeding, in line with the recommendations in high-income countries [27]. Further research is needed to assess the cost-effectiveness of these strategies. Third, integrase-strand transfer inhibitors (INSTIs) could be used in women who have high viremia in the last trimester and during breast-feeding, where possible and appropriate, to increase the chances for viral-suppression to maximize prevention of vertical transmission. The use of the recommended raltegravir-based regimen may be limited by high cost, although in the near future this could be replaced with low-cost dolutegravir as more data on its safety in pregnancy becomes available. Preliminary findings from the nationwide operational research in Botswana showed a safety profile comparable with efavirenz at conception, with long-term data expected in 2018 [28]. Overall, these efforts complement the recent WHO guidelines seeking to increase the likelihood of favorable pregnancy outcomes, through close monitoring of the mothers during the antenatal period [29]. Triple-drug combination antiretroviral prophylaxis in infants Approximately 60–70% of HIV infections occur either in utero or intrapartum. However, due to late diagnosis, perinatally infected infants often receive subtherapeutic regimens as prophylaxis, which increases the risk of selecting resistant strains [30]. Moreover, infants can also acquire resistance from ingestion of suboptimal doses of maternal regimens during breastfeeding [31]. Most of the current regimens given to mothers are known to enter breast-milk and are passed to their infants in suboptimal but therapeutically active doses [32]. The use of triple drug prophylaxis has a number of potential advantages. First, improved prevention of PDR, as the use of the currently recommended extended nevirapine monoprophylaxis or nevirapine/zidovudine dual-prophylaxis brings the risk of selecting for NNRTI resistance in infants [13]. This is due to the low genetic barrier to resistance of nevirapine, which requires only a single mutation for resistance to occur [33]. By contrast, the use of triple-drug prophylaxis has been shown to effectively prevent resistance in early PMTCT studies [34]. Second, some studies suggest that triple-drug prophylaxis could be more efficacious for PMTCT than monotherapy or dual-therapy [13,27,34], although data are conflicting. Indeed, based on expert opinion, most high-income countries now recommend the use of triple-drug prophylaxis (zidovudine/lamivudine/nevirapine) in high-risk infants born to women presenting late in care or with viremia by time of delivery [27]. WHO already recommends this strategy as an alternative to the preferred zidovudine/nevirapine dual therapy when there is complexity in dosing of the latter [13]. In absence of good record keeping, coupled with lack of viral-load tests for pregnant and lactating mothers, it may not be feasible for programs to distinguish between low-risk and high-risk infants. In our view, a standard triple-drug prophylaxis is probably a better standard strategy for PMTCT among HIV-exposed infants in these settings. Third, the use of triple-drug prophylaxis may serve as very early treatment to the infected, yet undiagnosed infants, which has the potential benefits of inducing long-term virological remission [35]. This may especially be advantageous in cases in which birth testing is not feasible, and for the vertically infected infants who are missed by early infant diagnosis (EID) tests during the early window of infection [36]. Fourth, triple-drug prophylaxis of zidovudine/lamivudine/nevirapine is 40% cheaper than the standard WHO-recommended dual prophylaxis of zidovudine/nevirapine, and only 40% more expensive than nevirapine monoprophylaxis [11,12]. Lastly, there are no major safety concerns with triple-drug prophylaxis of zidovudine/lamivudine/nevirapine relative to zidovudine/nevirapine dual-prophylaxis [37,38]. High genetic barrier antiretroviral prophylaxis in infants The use of prophylactic drugs with a high genetic barrier to resistance, specifically protease inhibitors or INSTIs, provides an alternative strategy. However, there are safety concerns for use of protease inhibitors among preterm infants and those less than 2 weeks of age [13]. Investigations are still ongoing for use of pediatric INSTI-based regimens as prophylaxis or treatment. Initial findings suggest a good safety profile for raltegravir in infants 6 weeks old or less [39]. Further studies are needed to assess the efficacy, safety and resistance of INSTI-based prophylaxis in infants. Timely diagnosis of HIV infection and treatment initiation in infants Prevention of PDR can also be enhanced by the timely identification of HIV-infected infants and prompt treatment initiation, especially in cases in which it may not be feasible to implement the aforementioned infant prophylaxis strategies. The rate of EID and prompt treatment initiation is low in sub-Saharan Africa [7,40]. In 2015, only 51% of HIV-exposed infants were diagnosed within the first 2 months of life as per the previous WHO recommendations [7]. Potential remedies include the use of point-of-care testing, birth-testing, provider-initiated counseling and testing for all children and lactating mothers presenting in hospital for other ailments, use of text messages to speed relay of EID results, family-centered care and decentralization of pediatric ART services. However, more programmatic efforts are needed to ensure the effective implementation of these strategies [7,40]. In conclusion, high rates of PDR in infants who become HIV-infected despite the use of PMTCT are worrisome, especially considering the very limited ART options available for children in case of treatment failure. This could be prevented by strengthening the PMTCT framework, which not only prevents HIV transmissions but can also help prevent resistance in the infected infants. In our view, this could include, first, increasing the frequency of viral load monitoring in pregnant mothers to at least once every 3 months during pregnancy and breast-feeding; second, provision of integrase inhibitors to high-risk mothers with viremia in their last trimester or during breast-feeding; third, using triple-drugs or integrase inhibitors as infant prophylaxis; fourth, patient or programmatic tailor-made adherence and retention support; and fifth, timely identification of infected infants with prompt initiation of treatment. Operational research is urgently needed to assess cost-effectiveness and programmatic challenges of these strategies. Acknowledgements S.C.I. is supported by a grant from the European Union through the Erasmus Mundus program. R.L.H. is supported by a grant from the Netherlands Organization for Scientific Research through the Innovational Research Incentives Scheme. The authors acknowledge the support of the Amsterdam Institute for Global Health and Development, Emma Children's Hospital-Academic Medical Center of the University of Amsterdam, US Centers for Disease Control and Prevention, Joint Clinical Research Centre, and College of Medicine of the University of Lagos. The views expressed in this article are those of the authors and might not necessarily reflect those of the institutions for which they work. Conflicts of interest There are no conflicts of interest.