Breastfeeding infants with HIV may develop drug resistance from ARVs in breast milk

Early diagnosis and more regular testing during breastfeeding could reduce risk

Two thirds of breastfeeding infants infected after birth, born to mothers on antiretroviral therapy (ART), developed resistance to one or more antiretroviral drugs according to Clement Zeh and colleagues in a secondary analysis of the Kisumu Breastfeeding Study (KiBS) published in PLoS Medicine this month.

However, drug resistance appears to have developed as a result of exposure to antiretrovirals in breast milk or exposure to antiretroviral prophylaxis immediately after infection, not through transmission of drug-resistant virus from the mother.

Drug resistance mutations in HIV-infected infants developed between two weeks and six months after birth increasing over time: from 30% by week six, to 63% by week 14 and 67% by six months.

No resistance was found among infants infected before two weeks of age or after six months when the mothers stopped ART and stopped breastfeeding.

The authors note this pattern suggests resistance was transmitted through exposure to maternal ART through breast milk, not through mother-to-child transmission (MTCT) of drug-resistant virus.

In resource-poor settings MTCT continues to cause significant death and disease.

Approximately one-third of the estimated 450,000 children infected each year are infected through breastfeeding. Safe alternatives are often not feasible for the majority of women in such settings. The risks for infant death and disease linked to not breastfeeding are greater than the risks associated with HIV infection.

In these settings breastfeeding is the norm. Proven strategies to reduce the risks of MTCT through breast milk include giving mothers ART during the (recommended) six-month breastfeeding period. What happens is that maternal viral load in breast milk is reduced, and infants also get an indirect prophylaxis of ART through breast milk.

Studies have shown that antiretroviral drugs (including zidovudine, lamivudine, nevirapine and efavirenz) when given to nursing mothers are present in breast milk.

However, antiretroviral drugs in breast milk can also have a negative effect for infants infected just before or during the breastfeeding period. Low levels of drug in the infant can encourage the evolution of drug-resistant virus if infection occurs despite prophylaxis.

The authors assessed this risk in a secondary analysis of the KiBS trial findings by looking at those infants who became HIV-infected during this time.

The parent trial, a single-arm open label prevention of mother-to-child transmission trial, looked at the safety and efficacy of giving zidovudine, lamivudine, and either nevirapine or nelfinavir to HIV-infected women from 34 weeks of pregnancy through six months of breastfeeding.

500 women were enrolled between July 2003 and November 2006 and gave birth to 502 infants. By the end of two years 32 (6%) infants were HIV-infected of which 24 (75%) were infected within the first six months of life. 

Of the 24, nine were exposed to mothers on a nelfinavir-based regimen and the other 15 exposed to mothers on a nevirapine-based regimen throughout the breastfeeding period.

All nine (100%) of those exposed to a nelfinavir-based regimen and seven (47%) of those exposed to a nevirapine-based regimen developed drug resistance (95% CI: 28-78%, p=0.0095).

The most common mutations, M184V and K103N, conferred resistance to lamivudine and nevirapine, respectively.

All infants were given single-dose nevirapine within 48 hours of birth.

Blood samples from mothers and infants were taken at various stages and analysed for drug resistance.

None of the eight (of the 24 HIV-infected children) who were HIV-positive by two weeks of age had any signs of development of drug resistance. Similarly no mutations were seen in the eight infants infected after the breastfeeding period of six months.

Among the mothers of the 24 children the majority (84%) had no drug resistance mutations. Only one mother-child pair had overlapping mutation patterns that might imply the transmission of drug-resistant virus.

The authors note that the timing of the development of drug resistance suggests this happened either through single-dose nevirapine given to the infants or indirectly through ART in breast milk. Findings from the SWEN study support this suggestion. (Infants given nevirapine prophylaxis while their mothers were on ART developed nucleoside reverse transcriptase inhibitor resistance in that study).

One major limitation, the authors note, is that HIV drug resistance genotyping was done on viral isolates from maternal plasma and not on breast milk isolates.

In light of the increasing use of infant prophylaxis during breastfeeding, it is likely that drug resistance will be seen more frequently.

The authors suggest improved early infant diagnosis and treatment may help alleviate the problem.

The authors conclude that close monitoring for the development of resistance in infants in PMTCT programmes who are exposed to ART through breast milk is needed, so that treatment can be tailored accordingly.