Developments in nevirapine use in MTCT reduction

18 April 2005. Related: Conference reports, Pregnancy, CROI 12 (Retrovirus) 2005.

Polly Clayden, HIV i-Base

A “roller coaster” was James McIntyre’s description of recent events surrounding controversies associated with nevirapine (or how “Concerns about Good Clinical Practice escalated to crimes against humanity”) in his Wednesday plenary at this conference [1]. And Bob Huff had described, “…a tsunami propagating across oceans, political shockwaves reverberated across continents” in the December edition of GMHC’s Treatment Issues [2]. Conspiracies, whistleblowing and dangerous politics to one side, this conference showed some interesting new scientific reports concerning the use of nevirapine to reduce mother to child transmission (MTCT).

Increased levels of resistance detected with real time PCR assay

Several studies presented similar findings related to concerns about resistance with single dose nevirapine exposure.

Three oral presentations in the resistance session, inconveniently scheduled at the same time as DITRAME Plus and other MTCT presentations, reported greater frequency of nevirapine mutations than are detected with standard genotyping. All three studies used real time PCR to analyse samples from South African women with subtype C HIV who had received single dose nevirapine at the onset of labour.

Conventional genotype assays are unable to reliably detect variants that comprise less than 20% of the sample virus population. Jeffrey Johnson’s group used a real-time PCR assay with a detection limit of 0.2% K103N in a background of wild type virus (a sensitivity 100-times greater than conventional sequencing) [3].

The study analysed matched samples from 50 women pre- and post- nevirapine exposure (at 6-36 weeks post partum). Using conventional genotyping, no women had detectable K103N mutation prior to receiving nevirapine and 10/50 women had detectable K103N in the post partum samples.

Results from real-time analysis also found all pre-treatment samples to have undetectable K103N and found the mutation present in all 10 samples (100%) that had detectable K103N by conventional sequencing. Additionally, the more sensitive assay found 16/40 (40%) specimens, in which no mutation was detected by standard genotype, to have detectable K103N.

Clonal sequence analysis confirmed the resistance mutation in five samples and provided K103N-positive virus frequencies of 1.1 of 11.1% of the sample virus populations.

A more complex analysis presented by Sarah Palmer revealed similar results [4]. This study used standard genotyping compared to an allele-specific RT-PCR assay for K103N and Y181C to test longitudinal samples from 17 women. Both NNRTI- associated mutations were detected as follows:

The median frequency of mutations was 3% (range 1 to 63%).

Using these findings the investigators estimated that, among their cohort, up to 69% of women could have mutations at 4 months, 32% at 5-9 months and 22% out to one year from receiving single dose nevirapine.

In a third presentation Shayne Loubser presented real-time PCR and standard genotyping results results for RNA samples at 6 weeks post-partum [5]. This study found the K103N mutation in 16 of 18 (89%) samples using real-time PCR and in 9 of 18 (50%) by conventional population sequencing. The investigators reported that generally only samples with ≥ 32% K103N by real-time PCR were detected by sequencing.

Additionally, a longitudinal analysis of samples from 16 women was performed to monitor the levels of K103N in RNA over time. At 12 months the K103N was still detectable at low levels in 4 of 16 (25%) RNA samples and in none of 15 DNA samples tested.

However, the authors noted: “We found no evidence for archiving of K103N mutations in DNA at 1 year post-single-dose NVP”.

Finally, a poster from Susan Eshleman and colleagues used a sensitive point-mutation assay to look at whether K103N could persist at low levels in women and infants for a year or more after exposure to single dose nevirapine [6].

The level of K103N-containing variants was quantified using a sensitive resistance assay, LigAmp, which has a lower limit of detection of 0.08%.

The authors reported that at 6 to 8 weeks after receiving nevirapine, K103N was detected by standard genotyping in 8 of 9 women for which samples were available and 2 of 5 infants, and was detected by LigAmp at a level above 0.1% in 8 of 9 women and 4 of 5 infants.

At 12 to 24 months, the K103N mutation was not detected by standard genotyping in any of the samples, but was detected by LigAmp above pre-nevirapine levels in 3 to 9 women (at 0.8%, 1.3%, and 3.5%) and 1 to 5 infants (at 1.5%).

The findings from these studies suggest that only a minority of women do not develop nevirapine resistance mutations after single dose exposure, that conventional sequencing substantially underestimates the emergence of resistance, and that resistance continues long after conventional genotyping suggests it “fades”. There was inconclusive discussion around whether or not resistance is archived in DNA even when it is fading in plasma, but it was uncontroversial was that these results stress the importance of assessing the clinical implications of resistant variants.

Subtype comparison

A poster from Susan Eshleman’s group reported results from a study comparing the rate of nevirapine resistance in Malawian women with subtype C (from the NVAZ study) to Ugandan women subtypes A and B (from the HIVNET 012 study), 6 to 8 weeks after single-dose NVP exposure [7].

The authors reported a significantly higher rate of resistance among women with subtype C, 45/65 (69%), than either subtype A, 28/144 (19%, p<0.001) or D, 35/97 (36%, p<0.001). Additionally the number of women with two or more nevirapine mutations was higher among the women with subtype C, 29/65 (45%) than with subtype A, 12/144 (8%, p<0.001) or subtype D, 16/97 (17%, p<0.001).

In a multivariate analysis they found viral load and subtype to be independent predictors of nevirapine resistance: C vs.A OR =8.84 (95% CI: 4.33-18.04); C vs D OR=3.44 (95% CI: 1.67-7.07) and log 10 viral load at delivery OR=2.27/log 10 RNA copies/ml (95% CI: 1.55-3.32). Maternal age, parity and time from nevirapine dosing to 6-8 week visit were not. The investigators also observed similar associations between these factors and the detection of two or more nevirapine mutations.

Although related findings have been reported previously this is the first study to make a direct comparison across the subtypes.

DITRAME Plus

In an oral presentation, Francois Dabis reported results from the ANRS DITRAME Plus 1.1 study. This study evaluated the frequency of resistance mutations following a short course (from 32 weeks) of AZT/3TC, plus single dose nevirapine to reduce mother to child transmission, ending with three days of AZT/3TC post partum [8]. This was an in an open label study conducted in Abidjan Cote D’Ivoire.

Data were available for 329 women for which the 6 week transmission rate was 4.7% (n=16, 95% CI 2.4 – 7.0). The median baseline CD4 cell count was 293 cells/mm3 for transmitting mothers and 416 cells/mm3 for non-transmitters and the median viral load was 5.16 and 4.45 log10 copies/mL, respectively.

Of this group the overall frequency of nevirapine associated resistance mutations was 1.14% (CI: 0.03 to 6.17%) and 8.33% (CI 3.66 to 15.76%) for 3TC.

Findings from this study are consistent with interim results from the TOPS study presented by James McIntyre in Bangkok. The TOPS study showed that adding AZT/3TC to cover the nevirapine “tail” significantly reduced nevirapine resistance (9.3 vs 50%). Further analysis of 226 mothers from TOPS, where early results are expected to hold out, will be available mid-2005 [9].

Following this presentation Charles Gilks from the WHO commented that they would be convening a small expert group to consider these findings with respect to the WHO guidelines “very soon”.

Impact on second pregnancy

There has been widespread concern that resistance following the use of single dose nevirapine for MTCT reduction will impact on its efficacy in second and subsequent pregnancies.

Neil Martinson presented data from a case controlled pilot study across 13 sites in Soweto comparing women using nevirapine for the first and second time in pregnancy [10]. Baseline results were available for 106 women who had received nevirapine for the second time and compared in a 2:1 comparison to a control group of 212 women who had received nevirapine for the first time. Six week resistance data were available for 77 exposed and 140 naïve at baseline women.

Transmission rates to infants determined by DNA-PCR, for whom results were available, were 5/132 (3.8%) in the control group and 8/75 (10.7%) in the second pregnancy group. Resistance was detected in 54% of the control group and 45% of the second pregnancy group of women by standard genotyping. There were a greater number of women with two or more nevirapine associated mutations in the control group (47% vs 23%, 2 mutations and 34% vs 17%, 3 or more mutations) but these women appeared to have more advanced HIV that the authors suggested could explain this.

The differences in transmission rates were reported as not statistically significant and the investigators concluded that previous nevirapine exposure had no effect.

The conclusion that the second pregnancy was not compromised by previous use of nevirapine raised much debate. A speaker from the floor questioned how the investigators could conclude that there was “no effect” (of previous dose) when, although not statistically significant, the results showed such a strong trend (RR = 2.3).

Further analysis of these results is ongoing and urgently needed.

Reducing resistance risk and using HAART in pregnancy

Discussions continue around the best strategies to reduce the risk of resistance.

A report from the “Mashi” study (see below) suggests that adding single dose nevirapine to maternal AZT prophylaxis offers no added benefit in transmission reduction. This contrasts with results from both DITRAME and PHPT-2 which both reported a substantial reduction in transmission with this strategy (although similar nevirapine resistance rates to single dose alone) [11, 12, 13].

There were also encouraging results from groups that follow the strategy that James McIntyre called “Treat them all”.

Women in the PMTCT-Plus programme receive triple therapy if they are eligible for HAART (WHO stage 4, WHO stage 2 or 3 with CD4 count < 350/mm3, CD4 count < 200/mm3). A poster from Besigin Tonwe-Gold’s group in Abidjan reported no transmissions, four weeks after birth, among the women receiving HAART [14].

Similar findings (approx 4% transmission rate at 1 month) were reported in an oral presentation from Leonardo Palombi from the DREAM programme in Mozambique in which 778 women had received generic HAART (nevirapine/3TC/d4T) for more than two weeks in the last stage of pregnancy [15]. Additionally, the authors reported low levels of resistance among the women (3/20) for whom genotype results were available.

Palombi noted that with the women who did not require treatment for their own HIV for whom treatment was stopped (originally after one month changing to 6 months over the course of the study), they had originally stopped the nevirapine 6 days earlier than the nucleosides but this had changed to 15 days.

A poster authored by Lyn Zijenah and colleagues reported 16 week data from 51 women receiving a generic nevirapine containing combination of AZT/3TC (Duovir) and nevirapine (Nevimune), both manufactured by Cipla. The study compared virologic response between women who had previously received single dose nevirapine (n=21) or short course AZT (n=30) to reduce mother to child transmission [16]. This Zimbabwean study also reported results for 33 men receiving the same treatment.

At week 16, the investigators reported a median increase in CD4 count from 135 cells/mm³ (IQR: 65-143) to 253 cells/mm³ (IQR: 155-302) among the women in the group. Of the women for whom viral load results were available the number with detectable virus < 500 copies/mL, was 4/30 exposed to short course zidovudine and 1/13 exposed to single dose nevirapine.

The investigators report despite “growing concern about the impact of maternal prophylaxis on subsequent treatment”, in this early analysis of women with subtype C HIV, previously exposed to single dose nevirapine they were no more likely than those receiving AZT short course to have an impaired treatment response to nevirapine containing HAART.

It is difficult though to reach any conclusion from this early analysis from a small study that contrasts with previous findings from Jourdain et al (showing an impaired response amongst previously nevirapine exposed women to subsequent nevirapine containing HAART) at six months [17]. There is a need for longer follow up.

Adverse events

There are concerns over treatment in pregnancy particularly for women with >250 cells/mm³ because of hepatoxicity risk in this group.

A poster from Timothy Thomas’s group described adverse events attributed to nevirapine among women from the Kisumu Breastfeeding Study [18].

This is an open label trial to evaluate safety, tolerability, adherence and efficacy of using AZT, 3TC and nevirapine from week 34 in pregnancy and for the first six months of breastfeeding.

The authors report a similar percentage of serious hepatic or cutaneous adverse events among women receiving nevirapine in other studies: 19/241 (7.9%) total serious adverse events 11/241 (4.6%) grade 3 or 4 serious adverse events. The authors note, however: “while other reports indicate increased risk for women with CD4 counts > 250 cells/mm³, this is not seen among pregnant HIV+ women in Kisumu to date. The data underscore need for close monitoring of women on NVP-HAART.”

In an oral presentation from the Thai Red Cross 2.1% pregnant women receiving the same combination as in the Kisumu Study were reported to have experienced grade 3 to 4 serious adverse events and this report observed > 400 CD4 cells/mm³ (p=0.018) to have increased risk [19].

It is unclear whether guidelines from FDA and Boehringer Ingleheim – the originator manufacturer of nevirapine – will be as relevant in African and other low and middle income settings.

Comment

As the details regarding the safety of continuous nevirapine therapy initiated in pregnancy, and the frequency and impact of mutations associated with nevirapine resistance following single dose (SD) nevirapine are gradually studied and refined and whatever the concerns regarding the adherence to GCP in HIVNET 012, it is imperative that key findings are not obscured.

First, there is little concern, despite the above, that the results of HIVNET 012 are likely to be substantially inaccurate. Second, the efficacy of SD nevirapine in reducing HIV mother to child transmission has been independently shown in the Thai PMCT-2 study where the addition of SD nevirapine to third trimester zidovudine further reduced transmission from 6% to 2%. Third, SD nevirapine has an excellent safety record in mothers and infants.

Then, in terms of the real time PCR sensitive assays to detect NVP resistance, while real time PCR detected resistance in 40-88% of women who did not have resistance detected with standard genotype at 6 weeks to 12 months post-exposure, in longitudinal studies, the frequency of detection declines over time and levels of resistance detected at 12 months were low (in the study with largest numbers it was 0.8-1.5% of quasispecies).  So resistance may persist longer but still declines over time and is low level after a year.  The lack of archiving in DNA at one year post exposure was very interesting.  The data is unclear on clinical implications of this resistance as detected by sensitive assays ie what level of resistance is associated with potential lower virologic response.

And fourth, mutations, “fading” or otherwise, are of great importance if they adversely impact on future therapy but are less relevant if NNRTI containing combination therapy, or indeed any form of HAART, is inaccessible, as is currently the case for the majority of HIV-infected persons. James Mcintyre’s talk included the “reality check” that – depending on whose data we believe only about 5% or less of HIV infected pregnant women even receive SD nevirapine.

Some have suggested that the most important data from this year’s CROI for HIV infected pregnant women globally are those from Soweto looking at HIV transmission in subsequent pregnancies when SD nevirapine was administered in both pregnancies. The finding that the transmission rate rose from 3.8% in the first pregnancy to 10.7% in the second pregnancy gives cause for concern despite the lack of statistical significance (p= 0.096 – Chi2 with Yates correction). However factors other than previous nevirapine exposure, such as the longer duration of HIV infection by the time of the second pregnancy, need to be excluded and it was noted that the 3.8% transmission rate in the first pregnancy was very low.

Where HAART for maternal health is, or is likely to become, widely available, the use of SD nevirapine alone or added to a background of short course zidovudine monotherapy to prevent MTCT no longer seems tenable. Strategies to protect the NNRTI class, particularly given the current higher cost of protease inhibitors in resource-poor settings, are urgently required.

At present the co-administration of other antiretroviral therapy to cover part or all of the period of actual nevirapine exposure seems the most likely approach, but data from the Mashi study suggest that early nevirapine administration to the neonate may have equal efficacy to maternal dose. If validated this would clearly avoid the risk of nevirapine resistance developing in the mother.

Adding SD Nevirapine to third trimester dual NRTI therapy with combivir may reduce the risk of NNRTI resistance (but at the cost of resistance to 3TC as seen in DITRAME Plus 1.1). The DITRAME study is interesting but the issue of whether the 3 day “tail” was associated with this – although mirrored in the earlier TOPS findings from McIntyre et al – or whether having low viral replication at the time of delivery was associated with low rate of resistance is unclear.  Additionally, the data from DREAM conflicted – these women received HAART during pregnancy and 6 days (now increased to 15 days) of AZT/3TC “tail” coverage and they saw 15% resistance.  As NVP levels persist for up to 21 days or longer the dramatic effect of short term “tail” of 3 days provoked comment. Besides more results from TOPS, other studies are ongoing looking at different regimens for different time periods (7 vs 28 days) that should provide more definitive data.

Interestingly, the frequency of the M184V mutation in the DITRAME study (8.3%) was considerably less than that seen in the French cohort (40%) when 3TC was added at 32 weeks to ZDV monotherapy.

Finally, concerning nevirapine containing HAART in pregnancy, none of the studies at this year’s CROI (nor at Glasgow November 2004) indicate a higher risk of toxicity in pregnant women than expected. However, a number of studies have now suggested that the 250 cells/mm3 cut off may not apply in pregnancy or as findings from Thomas et al suggest may be different in different settings. Most would agree that close monitoring is important.

References:

1. McIntyre J. Controversies in the use of nevirapine for the prevention of mother-to-child transmission. Wednesday Plenary 12th CROI, Boston, 2005.
2. Huff B. Single Dose Theory. GMHC treatment issues. Vol 18, No 11/12, p 1.
3. Johnson J, Li JF, Morris L et al. Resistance emerges in the majority of women provided intrapartum single-dose nevirapine. 12th CROI, Boston, 2005. Abstract 100.
4. Palmer S, Boltz V, Maldarelli F et al. Persistence of NNRTI-r resistant variants after single-dose nevirapine in HIV-1 subtype-C-infected women. 12th CROI, Boston, 2005. Abstract 101.
5. Loubser S, Balfe P, Sherman G, et al. Sensitive real-time PCR quantification of 103N resistance mutants following single-dose treatment with nevirapine. 12th CROI, Boston, 2005. Abstract 102.
6. Eshleman S, Nissley D, Claasen C et al. Sensitive drug resistance assays reveal long-term persistence of HIV-1 variants with the K103N nevirapine-resistance mutation in some women and infants after single-dose NVP: HIVNET 012. 12th CROI, Abstract 800.
7. Eshleman S, Hoover D, Chen S et al. Comparison of nevirapine resistance in women with subtype C compared with subtypes A and D after single-dose NVP. 12th CROI, Boston, 2005. Abstract 799.
8. Chaix ML, Dabis F, Ekouevi D et al. Addition of 3 days of ZDV+3TC postpartum to a short course of ZDV+3TC and single-dose NVP provides low rate of NVP resistance mutations and high efficacy in preventing peri-partum HIV-1 transmission: ANRS DITRAME Plus, Abidjan, Côte d’Ivoire. 12th CROI, Boston, 2005. Abstract 72LB.
9. McIntyre J, Martinson N, Boltz V et al. Addition of short course Combivir (CBV) to single dose Viramune (sdNVP) for prevention of mother-to-child transmission of HIV-1 can significantly decrease the subsequent development of maternal NNRTI-resistant virus. XV Intl AIDS Conference, 11-16 July 2004, Bangkok. LbOrB09.
10. Martinson N, L Pumla, Morris L et al. Effectiveness of single-dose nevirapine in a second pregnancy. 12th CROI, Abstract 103.
11. Shapiro R, Thior I, Gilbert P et al. Maternal single-dose nevirapine may not be needed to reduce mother-to-child HIV transmission in the setting of maternal and infant zidovudine and infant single-dose nevirapine: results of a randomised clinical trial in Botswana. 12th CROI, Boston, 2005. Abstract 74LB.
12. 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. 11th CROI 2004. Abstract 40LB.
13. Dabis F, Ekouevi DK, Bequet L et al. A short course of zidovudine and peripartum nevirapine is highly efficacious in preventing mother-to-child transmission of HIV-1: the ARNS 1201 DITRAME Plus study. 10th CROI 2003. Abstract 854.
14. Tonwe-Gold B, Ekouevi D, Rouet F et al. Highly active antiretroviral therapy for the prevention of perinatal HIV transmission in Africa: Mother-To-Child HIV Transmission Plus, Abidjan, Côte d’Ivoire, 2003-2004. 12th CROI 2005. Abstract 785.
15. Palombi L, Germano P, Liotta G et al. HAART in pregnancy: safety, effectiveness, and protection from viral resistance: results from the DREAM cohort. 12th CROI, Boston, 2005. Abstract 67.
16. Zijenah L, Kadzirange G, Rusakaniko R et al. Community-based generic antiretroviral therapy following single-dose nevirapine or short-course AZT in Zimbabwe. 12th CROI, Abstract 632.
17. 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. 11th CROI 2004. Abstract 41LB.
18. Thomas T, Amornkul P, Mwidau J et al. Preliminary findings: incidence of serious adverse events attributed to nevirapine among women enrolled in an ongoing trial using HAART to prevent mother-to-child HIV transmission. 12th CROI, Boston, 2005. Abstract 809.
19. Phanuphak N, Apornpong T, Intarasuk S et al. Toxicities from nevirapine in HIV-infected males and females, including pregnant females with various CD4 cell counts. 12th CROI, Abstract 21.