Results from a Thai study (PHPT-2) designed to evaluate whether greater mother-to-child transmission efficacy could be gained by adding single-dose nevirapine to standard zidovudine prophylaxis found that efficacy came with the risk of resistance.¹

In the study, 1,844 women were enrolled and mother and infant pairs randomized to three arms: single 200 mg nevirapine dose to the mother in labor and 6 mg to the baby within 72 hours of birth (the nevirapine-nevirapine arm); nevirapine dose to the mother and placebo to the infant (nevirapine-placebo) and both mother and baby receiving placebo (placebo-placebo). Additionally all mothers received zidovudine from 28 weeks of gestation and infants one week of zidovudine and formula feeding. The study endpoint was HIV infection of the infant.

Presenting author Marc Lallemant reported that the placebo-placebo arm was discontinued following the trial's first interim analysis due to the highly significant 80% reduction in transmission among those receiving the additional drug: 1.1% in the nevirapine-nevirapine arm and 6.3% in the placebo-placebo arm (p=0.00026). Subsequent analysis showed that transmission rates between the nevirapine-nevirapine and the nevirapine-placebo arms did not differ dramatically: 2.0% and 2.8%, respectively.

Subsequent Treatment Compromised

In a second presentation, Gonzague Jourdain assessed the effect of nevirapine exposure in PHPT-2 on the response to subsequent NNRTI-containing HAART regimens.² Genotypic analysis was performed on 90 randomly selected, 12-day postpartum samples from trial participants. Eighteen percent of the women sampled were found to have detectable NNRTI mutations.

Ultimately, 25 percent of the women who participated in the PHPT-2 study initiated antiretroviral therapy and subsequently received an NNRTI-containing regimen (nevirapine/3TC/d4T). Of the 255 women who started HAART, 42 had not, and 213 had, been exposed to nevirapine. At six months, 75% of the unexposed, 53% of the exposed/no detectable mutations and 34% of the women exposed/with detectable mutations had a viral load below 50 copies. Initiation of therapy six months or more after nevirapine exposure was associated with a modest but non-significant improvement in virological response.

In a comment from the floor after this presentation, John Mellors remarked that these findings echoed those from ACTG 398, in which patients receiving efavirenz-containing regimens had responded less well if previously exposed to an NNRTI, even without detectable resistance.³ These reports suggest that prior NNRTI exposure can select minor resistant variants not detectable by standard genotype assays, but which can nevertheless contribute to the failure of subsequent NNRTI-containing regimens.

Problem May Be Widespread

In a surveillance study nevirapine resistance in Kwazulu-Natal, South Africa, Gordon and colleagues examined nevirapine resistance patterns in 30 mother and infant pairs (including one set of twins) with HIV-1 subtype C who had participated in a single-dose (to mother and infant, respectively) prevention of mother-to-child transmission (PMTCT) program at a clinic in Hlabisa, South Africa.⁴

At six weeks following the nevirapine prophylaxis, 12/30 (40%) of women and 40% of infants had detectable resistance. The K103N mutation was the most common mutation in 10/12 (83%) of the mothers. Other mutations reported in the mothers included: Y181C in 3/12 (25%), Y188C in 3/12 (25%), V106M in 2/12 (17%) and G190A in 1/12 (8%) Two or more mutations were found in 4/12 (33.3%) mothers. Among the infants, the Y181N was the most common mutation and was present in 11/12 (92%) of the children (including one of the twins). Additionally 2/12 infants (17%) had the K103N and another 1/12 (8%) had a subtype C associated V106M mutation.

The investigators concluded: "Given the high rate of resistance in mothers and infants after single-dose nevirapine, the search for safer regimens to prevent MTCT should be intensified."

Persistence of Resistance

A resistance substudy of the Ditrame Plus trial in Abidjan, Côte d'Ivoire -- in which women received single-dose nevirapine in addition to short course zidovudine to reduce MTCT and the infants short course zidovudine and single-dose nevirapine syrups -- evaluated nevirapine resistance at four weeks post partum.^5,6

Baseline and four week samples were available for 63 women. The investigators reported 21/63 (33.3%) of women had developed nevirapine resistance at week four, with the K103N being the most common mutation. They also reported that mothers with infected and uninfected infants developed resistance at the same rate (33.3%), 7/21 and 14/42, respectively. No zidovudine resistance was detected in this group.

Analysis of nevirapine plasma concentrations revealed wide inter-patient variability with a median concentration of 648 (range 417-954) ng/mL. Resistance was significantly associated with a higher plasma concentration of nevirapine; among women who received two doses of nevirapine, 3/4 (75%) acquired resistance.

Additionally, 6/26 (23%) of the infected infants developed nevirapine resistance at four weeks post partum, and follow-up samples in two children -- one at 3 and one at 12 months old -- detected archived mutations.

The investigators note that the association between high nevirapine plasma concentrations and resistance suggests, "That a high level of nevirapine concentration induced a prolonged viral replication under suboptimal drug selective pressure which promotes the emergence of resistant strains." Concerning the infants they write: "Recent studies have reported a negative impact of nevirapine resistance on a subsequent treatment including nevirapine; our results raise anxiety for those very young children presenting with resistant viruses."

Resistance in HIVNET 012

A resistance substudy of the HIVNET 012 trial presented by Susan Eshleman and colleagues examined the impact of HIV subtypes A vs. D on the selection and "fading" of nevirapine associated mutations K103N and Y181C in a group of women following a single dose of nevirapine to reduce mother-to-child transmission.⁷ Genotypes were obtained at 7 days and at 6-8 weeks; paired data were available for 159 women. Of this group, 83 women had subtype A and 57 had subtype D.

The investigators found a significantly higher overall rate of resistance (i.e., any nevirapine mutation) at 6-8 weeks than at 7 days, 47/140 (34%) and 31/140 (22%), respectively, in women with either A or D subtypes (p=0.013). There was a higher rate of accumulation of mutations for subtype D vs. A. The K103N mutation was detected at a higher rate in the 6-8 week samples: 41/140 (29%), than the 7 day samples: 18/140 (13%), (p=0.0001) across both subtypes. Conversely the detection rate for the Y181C mutation was higher at 7 days than at 6-8 weeks overall, 26/140 (19%) and 15/140 (11%), respectively (p=0.0145). The investigators added: "Furthermore, Y181 faded quickly in subtype A with little or no fading in subtype D."

The report suggests that nevirapine mutations are better tolerated by subtype D HIV than subtype A and that HIV-1 subtype should be considered in the design and interpretation of studies to determine whether single-dose nevirapine compromises subsequent NNRTI containing treatment.

Comment

These reports signal bad news for highly active drugs with long half-lives, given as monotherapy (or effectively as monotherapy). Although nevirapine resistant variants "faded" from detection in women in HIVNET 012 by 12-24 months using population sequencing methods, resistant variants will surely still persist as minority variants and rapidly return when drug pressure is reintroduced. "Fading" is an incongruous term to use in a room full of virologists who have warned of the risks from archived resistance for many years.

Jourdain et al. showed dramatically reduced response in women receiving NNRTI containing regimens following acquisition of nevirapine resistance after receiving single-dose nevirapine to reduce MTCT (at six months 75% unexposed, 53% of exposed but with no detectable mutations and 34% of exposed with detectable resistance were below 50 copies). Additionally when Mellors, et al., evaluated the role of minor NNTRI mutations, failure to achieve viral suppression was associated with previous NNRTI experience and NNRTI mutations at baseline. Although genotyping failed to detect NNRTI mutations in 50/216 (23%) baseline samples in the NNRTI experienced patients, this group performed no better than those with detectable NNRTI resistance and much worse that the NNRTI-naive group who similarly showed no mutations.

Furthermore, as the Thai study demonstrated, adding nevirapine to background zidovudine is not associated with significantly less nevirapine resistance. The early emergence of the Y181C in HIVNET 012 may help to explain the different rates of NNRTI mutations seen in mothers compared to infants, as previously reported. The more rapid "fading" of the Y181C would seem to suggest that this mutation is "less fit" relative to both K103N and wild-type virus, at least in sub-type A virus. Better news is that no resistance was reported for zidovudine as prescribed in the DITRAME study.

These studies confirm the efficacy of nevirapine to contribute to the reduction mother-to-child transmission. But they also confirm the likelihood of rapidly selecting resistance mutations when less-than-suppressive concentrations of the drug are allowed to persist. The finding that single-dose nevirapine exposure can negatively impact future outcomes to such a dramatic extent should lead to a rapid change in policy, especially in countries rolling out combination therapy. The potential benefit of nevirapine or efavirenz as part of a subsequent regimen for the mother must be protected, and the use of nevirapine for prevention of mother-to-child transmission (PMTCT) of HIV restricted to effective combinations only and with due consideration for its prolonged clearance from the body, which varies considerably between individuals.

Polly Clayden is with HIV i-Base. This article originally appeared in: HIV Treatment Bulletin, Volume 3 and Volume 5. www.i-base.org.uk.

1. Lallemant M, Jourdain G, Le Coeur S et al. A randomised, double-blind trial assessing the efficacy of single-dose perinatal nevirapine added to a standard zidovudine regimen for the prevention of mother-to-child transmission of HIV-1 in Thailand. Program and Abstracts of the 11th Conference on Retroviruses and Opportunistic Infections. Abstract 40LB.

2. Jourdain G, Ngo-Giang-Huong N,Tungyai P et al. Exposure to intrapartum single-dose nevirapine and subsequent maternal six-month response to NNRTI-based regimens. Program and Abstracts of the 11th Conference on Retroviruses and Opportunistic Infections. Abstract 41LB.

3. Mellors J, Palmer S, Nissley D et al. Low frequency NNRTI-resistant variants contribute to failure of efavirenz-containing regimens. Program and Abstracts of the 11th Conference on Retroviruses and Opportunistic Infections, 8-11 February 2004, San Francisco. Abstract 39.

4. Gordon M, Graham N, Bland R et al. Surveillance of resistance in KZN South Africa, including mother-infant pairs six weeks after single dose nevirapine. XIII Intl Drug Resistance Workshop, Abstract 71. Antiviral Therapy 2004; 9:S80.

5. Dabis F, Ekouevi DK, Rouet F et al. Effectiveness of a short course of zidovudine and lamivudine and peripartum nevirapine to prevent HIV-1 mother-to-child transmission. The ANRS 1201 DITRAME Plus trial, Abidjan, Cote d'Ivoire. 2nd IAS Conference. Abstract 219.

6. Chaix ML, Ekouevi DK, Peytavin G et al. Persistence of nevirapine resistant virus and pharmacokinetic analysis in women who received intrapartum NVP associated to a short course zidovudine (ZDV) to prevent perinantal HIV-1 transmission: the Ditrame Plus ANNRS 1201/02 Study, Abidjan, Cote d'Ivoire. Abstract 160. Antiviral Therapy 2004; 9:S176.

7. Eshleman SH, Wang L, Guay LA et al. Distinct patterns of selection and fading of K103N and Y181C are seen in women with subtype A vs D HIV-1 after single dose nevirapine: HIVNET 012. XIII Intl Drug Resistance Workshop, Abstract 50. Antiviral Therapy 2004; 9:S59.