Resistance during MTCT strategies: new data on protective benefit of Combivir added to single-dose nevirapine, and impact of treatment and breastfeeding on the infant

15 July 2005. Related: Conference reports, Pregnancy, Intl Drug Resistance Workshop 14 Quebec 2005.

Polly Clayden, HIV i-Base

• Short course of 3TC and AZT reduces the emergence of nevirapine resistance
• Higher rates of resistance in TOPS using more sensitive resistance tests
• Avoiding maternal nevirapine and adding AZT to infant prophylaxis
• Resistance patterns vary by subtype in infants in HIVNET 012 and NVAZ studies
• Mutations present in majority of women: early analysis suggests a similar rate of resistance in infants
• Patterns of viral load and resistance in breast milk vary by HIV sub-type

The frequency and persistence of nevirapine associated mutations after single dose prophylaxis and strategies to preserve this class of drug for future maternal and paediatric treatment generated several reports at this meeting. It was good to see the prominence of this issue this meeting. The first four oral presentations covered this area of research.

Short course of 3TC and AZT reduces the emergence of nevirapine resistance

David Hall from Boehringer Ingelheim presented further analyses from the TOPS (Treatment Options Preservation Study), supporting interim findings that were first presented at the World AIDS Conference in Bangkok in 2004. [1]

TOPS compared nevirapine single dose (HIVNET 012 regimen) to single dose nevirapine plus either 4- or 7-days of Combivir, to mother and infant to cover the nevirapine “tail”. In the previous interim analysis of 61 women, resistance was detected in 53.3% receiving nevirapine alone and 9.3% overall in those receiving nevirapine plus Combivir (p=0.001) by standard genotyping at 6 weeks. Enrollment in the single dose nevirapine arm was closed following the interim results.

This larger analysis included 6-week follow up from 226 mothers and 228 infants [2]. This presentation by the company was unexpected at this meeting. Full reporting of the 6-week results from this trial will be at the 3rd IAS conference in Rio in July, with separate presentations covering the results in mothers and infants.

All mothers in the trial were subtype C except for one subtype B and one CFR02AG. Population sequencing (Bayer) was performed at baseline (pre propylaxis) and at 2 and 6 weeks. Maternal results show changes from baseline sequence known to be associated with nevirapine resistance in 41/68 (60%), 8/67 (12%) and 7/68 (10%) of samples in the nevirapine, nevirapine plus 4 days Combivir and the nevirapine plus seven days Combivir arms respectively. The most frequently observed nevirapine associated mutations were: K103N, Y188C, Y181C, V106M, A190G and V106A.

The investigators reported that after single dose nevirapine, mutations emerged at different rates and remained for different durations. Mutations at positions 106 and 181 emerged earliest and were present at 2 weeks. Mutations at codon 190 emerged last, and were not present until after week 2 in over 40% of samples with those mutations. By week 6, the majority of mutations at 181 and at 106A were no longer detectable by population sequencing. The investigators wrote: “The pattern of mutations observed will depend on when the samples are collected. Mutations were observed for the first time as late as 12 weeks.” Emergence of resistant virus in mothers receiving single-dose nevirapine was associated with higher baseline viral load.

The addition of Combivir prevented the emergence of detectable resistance in 80% of mothers compared to single dose nevirapine. The investigators also noted that while 30/41 (75%) of the nevirapine group had multiple mutations, only 5/15 (33%) of the mothers receiving additional Combivir had multiple mutations. There was no evidence of 3TC associated mutations at codon 184.

Of the 228 evaluable infants, 8, 7 and 7 children were infected in utero in the nevirapine, nevirapine plus 4 days Combivir and nevirapine plus 7 days Combivir arms respectively. Nevirapine resistance was detected in 7/8 (88%) infants in the single-dose nevirapine arm and 1/7 (14%) and 0/7 (0%) in the Combivir 4- and 7-day arms respectively. Mutations observed in the infant samples were: V106A/M, Y181C, Y188C, and G190A in the single dose nevirapine arm (in 5, 3, 3 and 2 samples respectively), and K103N and Y181C in the one sample with resistance in the nevirapine plus 4-day Combivir arm.

There were 2 perinatal infections: one in the nevirapine arm and one in the nevirapine plus 4-day Combivir arm, both without mutations at six weeks.

The investigators wrote: “A short course of Combivir reduces the emergence of nevirapine resistance associated mutations in mothers taking single dose nevirapine to prevent mother to child transmission of HIV-1. A short course of AZT and 3TC reduces the emergence of nevirapine resistance associated mutations in infants taking single dose nevirapine after intrauterine infection.”

Higher rates of resistance in TOPS using more sensitive resistance tests

Additionally, Sarah Palmer presented findings from an analysis of samples from the TOPS study, using an allele specific RT-PCR assay that quantifies variants encoding 103N or 181C variants at frequencies <0.1% [3].

This substudy analysed samples at baseline, and at 2 and 6 weeks post prophylaxis, from 32 women (10 from the single-dose nevirapine, and 11 from each of the nevirapine plus Combivir arms).

The investigators reported that using the more sensitive test, 103N or 181C variants were detected in 75% of week 6 samples from women receiving single dose nevirapine prophylaxis (mutant frequency 1-75%, median 7%) and 6/22 (27%) of women receiving single dose nevirapine plus either 4 or 7 days of Combivir (mutant frequency 0.8-8%, median 1.9% and 0.3%-1%, median 0.9%, 4 and 7 days respectively)

The investigators noted that at 6 weeks, the median fraction of plasma virus that is comprised of 103N (AAC) or 181C variants, appears to be lower in women who receive Combivir in addition to single-dose nevirapine.

They added that additional studies are needed to establish clinical significance of these results and optimum duration of Combivir. The results presented above show a modest trend towards less detectable resistance using population sequencing in mothers receiving 4 vs 7 days of Combivir (12% vs 10%). This had been counter intuitive in the interim analysis (5% vs 13%) but only evaluated a small group of mothers (20 and 23 in the 4 and 7 day Combivir groups respectively).

Avoiding maternal nevirapine and adding AZT to infant prophylaxis

Susan Eshleman presented data analysing nevirapine resistance in mother and infant pairs from the NVAZ study in Malawi, which studied four different MTCT prophylaxis regimens as follows [4]:

• Group 1: Mothers and infants each received single dose (SD) nevirapine according to the HIVNET 012 regimen.
• Group 2: Mothers received SD nevirapine; infants received SD nevirapine plus AZT BID for 7 days;
• Group 3: Mothers (late presenters in labour) no nevirapine; infants SD nevirapine;
• Group 4: Mothers (late presenters in labour) no nevirapine; infants SD nevirapine plus AZT BID for 7 days.

The overall transmission rate at 6-8 weeks was 14.1%, 16.3%, 20.9%, 15.3% in groups 1-4 respectively (so rates were similar for 1,2 and 4 but statistically higher for group 3).

The investigators analysed resistance using the ViroSeq genotype system, in evaluable samples collected at 6-8 weeks from 78 infected infants (23, 21, 19 and 15 in groups 1-4 respectively)..

Dr Esheleman reported detectable nevirapine resistance in 50/78 (64.1%) of the infants (see Table 1). There were significant differences in the frequency in groups 1-4 (p=0.001). The highest was in group 1 (as HIVNET 012 regimen) in which 20/23 87% infants had nevirapine resistance. The frequency was lower when there was infant but no maternal nevirapine dose, 14/19 (74%) in group 3 or when the infant received AZT in addition to the nevirapine, 12/21 (57%) as in group 2. The lowest frequency of nevirapine resistance was achieved combining the two strategies: no maternal nevirapine and nevirapine plus AZT 4/15 (27%) to the infant, as in group 4.

Table 1: Nevirapine resistance detected in infants by MTCT treatment strategy

In a multiple logistical regression analysis, both avoiding the maternal nevirapine dose (OR= 0.20, p<0.001) and adding AZT BID for 7 days to the infant regimen (OR=0.31, p=0.04) were independently associated with reduced infant nevirapine resistance.

The authors also reported that group 4 not only had the lowest infant nevirapine resistance but that transmission rates were comparable to groups 1 and 2 and lower than group 3. None of the infants developed resistance to AZT.

The authors concluded that of the four mother to child transmission prophylaxis regimens evaluated avoiding maternal nevirapine and adding AZT to the infant nevirapine dose as in group 4 was optimal as maternal nevirapine resistance was avoided and infant nevirapine resistance was reduced.

They also noted that the transmission risk using this regimen was not statistically different to the HIVNET 012 regimen in this analysis (15.3% vs 14.1% for groups 1 and 4 respectively).

Resistance patterns vary by subtype in infants in HIVNET 012 and NVAZ studies

In a previous comparison of frequency of maternal resistance across subtypes – using samples from the NVAZ and HIVNET012 trials – Susan Eshleman and co-workers revealed that resistance was more frequent in mothers with subtype C (45/65, 69%), than with subtype D (37//97, 36%, p=0.0001) or subtype A (28/144, p=0.0001). [5] This finding was independent of maternal viral load, age, parity or time between dose and resistance testing.

The same group evaluated nevirapine resistance in infants who had become HV infected at 6-8 weeks in these two trials despite nevirapine prophylaxis being given to the mothers. [6] Plasma samples were available for 24/37 (65%) of infants in HIVNET 012, and 26/59 (44%) of infants in NVAZ. Analysis was performed using ViroSeq HIV Genotyping System. Genotypes were obtained for 24 infants in HIVNET 021 and 23 infants in NVAZ.

Analysis of subtypes in HIVNET 012 infants found 9 with subtype A, 1 with subtype C, 9 with subtype D and 5 with recombinant HIV. All NVAZ infants were subtype C.

The investigators reported that 20/23 (87%) infants in NVAZ had detectable resistance, compared to 11/24 (46%) of NIVNET 012 infants (p=0.005). (A similar number of infants had two or more mutations in both groups with 6/23 vs 2/24 in NVAZ and HIVNET respectively (p=0.49). A significant difference was reported comparing the 23 NVAZ infants to the infants in HIVNET with subtype A or D only (p=0.016); with subtype D only (p=0.23) or with recombitant HIV only (p=0.008). There was a higher proportion of infants in NVAZ with resistance than HIV 012 infants with subtype A but this did not reach statistical significance due to the small sample size, p=0.076. Additionally the infant in HIVNET 012 with subtype C had nevirapine-associated resistance.

The investigators concluded that nevirapine resistance after single dose was more frequent in the Malawian infants with subtype C than in the Ugandan infants of whom the majority were subtype A or D at 6-8 weeks post dose These comparative results are concordant with those reported for the mothers in these studies. Further analysis is needed to evaluate whether these findings are due to the effect of HIV subtype or other factors.

Mutations present in majority of women: early analysis suggests a similar rate of resistance in infants

Studies reported at CROI this year, suggested that as in that: only a minority of women exposed to single dose nevirapine do not develop nevirapine resistance; that conventional sequencing genotyping significantly underestimates this; and that resistance persists after it is no longer detectable using conventional genotypic resistance tests. [7, 8, 9]

At this meeting, Jeffery Johnson presented further analysis from a study evaluating matched samples collected pre and post single dose nevirapine exposure (6-36 weeks, median 7 weeks post partum) from 50 South Africa mothers with subtype C. They compared results using real-time PCR assays that have detection limits of 0.2% for K103N and 0.3% for Y181C to population sequencing genotyping (that requires 10-20% of resistant virus to be present for detection). [10]

The investigators found that 17//40 (43%) mothers who sequenced negative for nevirapine resistance had detectable mutations using real time PCR: 12 with K103N, 1 with Y181C and 4 with both mutations. They noted that Y181C was not found after 13 weeks post dose, but that K103N was detected throughout the 36 week study.

Resistance emerged in at least 65% of women. This was a 62% increase in the incidence reported with conventional genotyping.

Shayne Loubser also presented further sequence genotype and real time PCR results (sensitivity 0.2%) for mothers, analysing samples collected at different time points between 6 weeks and 12 months. [11]

This study found K103N mutations in 16/31 (52%) of plasma samples using population sequencing and in 27/31 (87%) using real time PCR. Standard genotyping failed to detect mixtures of K103N below a frequency of 10%.

In a cross-sectional analysis of plasma RNA samples collected at 6 weeks, 3 months, 7 months and 12 months; 27/31 (87%), 18/27 (67%), 13/37 (35%), 7/ 54 (13%) respectively had detectable K103N variants. DNA analysis from samples collected at 2 weeks and 12 months found that 23/43 (53%) and 2/48 (4.2%) samples had detectable mutations. Longitudinal analysis of plasma RNA in samples collected 6 weeks to 12 months for 15 women showed similar rates of decay, but mothers with higher levels resistance detected earlier had longer persistence of K103N variants.

Dr Loubser’s group found an additional 73% (n=15) of samples contained K103N variants than identified by standard genotyping, which declined over time but were present in a minority at 12 months.

Finally Dr Troyer presented data from a detailed phylogenetic analysis using a quantitative radiolabelled oligonucleotide DNA ligation assay (OLA) performed on samples collected in an observational mother to child transmission cohort in Uganda [12].

This study looked at mother and infected infant pairs receiving: AZT (n=90) or nevirapine (n=61) and no prophylaxis at 6 weeks post dose.

This analysis revealed 75% of samples from the nevirapine receiving pairs to have detectable resistance with this assay 6 weeks post dose.

The clinical implications of this very low of level resistance, and its potential to prejudice future NNRTI treatment response, are unclear.

Patterns of viral load and resistance in breast milk vary by HIV sub-type

The majority of breastfeeding transmission takes place in the first six weeks postpartum. Although single dose nevirapine has been much discussed and studied in terms of maternal and infant resistance patterns, less is known about the effect of single dose nevirapine on breast milk viral load, or about the prevalence of nevirapine-associated mutations in breast milk compared to blood plasma.

Dr Lehman presented findings from a sub study of a Kenyan trial comparing single-dose nevirapine (HIVNET 012 regimen) to AZT monotherapy (Thai short-course regimen) [13]. Breast milk samples from 30 women in the nevirapine arm were collected every 2-3 days for the first 6 weeks postpartum, and plasma samples were collected during pregnancy, at delivery and at one-month post partum,.

Breast milk HIV-1 RNA was quantified using the Gen-Probe HIV-1 assay. Blood samples, collected at one month post partum, were quantified using an allele-specific PCR assay (sensitive to 0.1%) to detect the K103N mutation in subtypes A and D.

The investigators reported comparable levels of HIV viral load in breast milk from women receiving single dose nevirapine to those found in women receiving the Thai short-course AZT regimen. 404 breast milk samples were evaluated (a median of 14 samples per woman) collected during the first 6 weeks postpartum.

Women receiving single dose nevirapine had significantly greater log reductions in breast milk viral load during days 3-21 postpartum: 3 to 7 days (-2.42 vs –1.98, p=0.1); 8 to 14 days (-2.48 vs –1.78 p=0.005); and 15 to 21 days (-2.97 vs -1.90, p=0.003).

Additionally they reported that 40% plasma samples collected at one-month post partum had detectable K103N mutation using the allele specific assay. They noted that these are preliminary findings and that breast milk samples are currently being analysed.

The investigators wrote: “Compared to the Thai CDC short-course zidovudine regimen, single-dose nevirapine results in sustained suppression of breast milk viral loads during weeks 2-6 postpartum. However the benefits of this suppression may be counterbalanced with a high prevalence of drug resistance.”

This group are currently completing a detailed temporal analysis of nevirapine resistance in breast milk during the first 6 months post partum, from women receiving single dose nevirapine. They will also analyse viral load and resistance from samples from women receiving either short-course AZT pus nevirapine vs short-course HAART during breastfeeding.

Comment

Although single dose nevirapine was a tremendous step forward it is certainly not the future.

The latest data from TOPS show a significant reduction in nevirapine resistance in the mother if the prolonged maternal exposure following single dose nevirapine is covered by the addition Combivir for 4-7 days. As expected, given the known rate of plasma clearance of nevirapine, the proportion of mothers with resistance, as detected by population based sequencing, was lowest with 7 days cover. This appears to be a sensible strategy to limit the spread of NVP resistance, and to preserve maternal therapy options, in settings where short term HAART for PMTCT is neither available or feasible. Although it is disappointing that neonatal AZT plus 3TC for one week did not effect transmission, this largely reflects the efficacy of single dose nevirapine to reduce transmission during labour and the apparent preservation of treatment options for the infected infants may justify the continuation of this approach.

The alternative approach to avoid maternal nevirapine resistance is to avoid maternal nevirapine exposure. This was first suggested by the MASHI study, in which single dose nevirapine given to the infant, was of equal efficacy, to the dual maternal and infant single dose nevirapine on a background of AZT monotherapy from 34 weeks and continued in the infant for a least 1 month [14]. Although the transmission rates in all four arms are high, the data presented from NVAZ appear to indicate that 7 days of neonatal ZDV ‘makes up’ for the absent maternal nevirapine dose in terms of transmission, and significantly reduces nevirapine resistance in the infected infants.

But as to What is to be done? may be becoming clearer there are practical questions around the feasibility of implementation. James McIntyre says that he believes that the addition of a 7 day tail cover would not provide an insurmountable barrier to the use of the TOPS strategy to reduce the selection of NNRTI resistant virus. At current prices the additional cost would be minimal in relation to the benefits, especially as good quality generic AZT/3TC combinations are available. What is required is some rapid innovative thinking to develop ways to achieve this: such as co-packaging of the drugs and access mechanisms.

As we go to press the WHO are meeting with opinion leaders to discuss their pregnancy and MTCT guidelines – we hope that some of these new data will motivate some changes.

References:

Unless stated otherwise, all abstracts are from the 14th International HIV Drug Resistance Workshop (14th IHDRW), 7-11 June 2005, Quebec City, Canada.

1. 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, Bangkok. LbOrB09
2. McIntyre JA, Martinson N, Gray GE et al. Single dose nevirapine combined with a short course of Combivir for prevention of mother to child transmission of HIV-1 can significantly decrease the subsequent development of maternal and infant resistant virus. 14th IHDRW. Abstract 2.
3. Palmer S, Boltz V, Maldarelli F et al. Short course Combivir single dose nevirapine reduces but does not eliminate the selection of nevirapine resistant HIV-1:improved detection by allele-specific PCR. 14th IHDRW. Abstract 3.
4. Eshleman SH, Hoover DR et al. Infant nevirapine resistance can be substantially reduced after single dose nevirapine by avoiding maternal dosing and providing infants with nevirapine maternal nevirapine dosing and providing infants with zidovudine in addition to single dose nevirapine after birth. 14th IHDRW. Abstract 1.
5. Eshleman SH, Hoover DR, Chen C et al. Nevirapine resistance in women with HIV-1 subtype C compared with subtypes A and D, after the administration of single-dose nevirapine. JID, 192:30-36; 2005.
6. Eshleman SH, Hoover DR, Hudelson SE et al. Resistance after single dose nevirapine prophylaxis varies by subtype in infants from sub-Saharan Africa. 14th IHDRW. Abstract 10.
7. 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.
8. 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.
9. 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..
10. Johnson JA, Li JF and Morris L et al. Resistance mutations arise in the majority of women provided single dose nevirapine and appear to differ in emergence and persistence. 14th IHDRW. Abstract 11.
11. Troyer R, Lalonde M, Kyeyune F et al. High frequency of nevirapine resistant mutations in the HIV quasi species found in NVP-treated participants of an MTCT Ugandan cohort. 14th IHDRW. Abstract 12.
12. Loubser S, Balfe P, Sherman G et al. Increases sensitivity of detection of K103N resistance variants by real time PCR in RNA and DNA after single dose nevirapine. 14th IHDRW. Abstract 13.
13. Lehman DA, Chung MH, Richardson BA et al. Patterns of viral load and drug resistance in breast milk and blood from women treated with single dose nevirapine to reduce mother to child transmission of HIV-1. 14th IHDRW. Abstract 4.
14. 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.