Association between the Replication Capacity and Mother-to-Child Transmission of HIV-1, in Antiretroviral Drug–Naive Malawian Women

The replication capacity of the pol gene of HIV is a potentially important determinant of viral fitness and pathogenicity [1]. Ongoing clinical studies are evaluating the potential utility of replication capacity in the management of treatment-experienced [2–5] and treatment-naive [6, 7] patients. The present study examined the association between the replication capacity of HIV-1 and mother-to-child transmission of HIV-1, in antiretroviral drug–naive Malawian women who presented late for delivery (termed “late presenters” in our earlier report of our ongoing nevirapine-and-zidovudine [NVAZ] study [8]). Late presenters in the NVAZ study enrolled too close to delivery to be either tested for HIV-1 or counseled for predelivery antiretroviral prophylaxis. Therefore, immediately after delivery, these women were counseled, and a rapid HIV test was used to determine whether HIV-1 infection was present. Results of the rapid HIV test were confirmed by a conventional ELISA for HIV. All infants born to HIV-infected women received 1 oral dose of nevirapine at birth and were randomized to either receive or not receive an additional 1 week of daily doses of zidovudine syrup [8]

Subjects and methodsAt delivery, plasma samples were collected from the late presenters among the Malawian women enrolled in the NVAZ trial [8]; in a subsequent study, HIV-1 was subtyped on the basis of phylogenetic analysis of pol gene–region sequences [9], and all women were found to have subtype C HIV-1. All women were also antiretroviral drug–naive when the samples were collected, and they received no antiretroviral drugs during the NVAZ trial. HIV-1–infection status of the infants was determined by methods described elsewhere [8]. In the present study, a woman was classified as a transmitter if her infant was diagnosed with HIV-1 infection either at birth or during the first 6–8 weeks of life, whereas she was classified as a nontransmitter if her infant remained uninfected by 6–8 weeks of age. None of the women delivered by cesarean section, and none delivered twins. All but 1 of the women were breast-feeding at 6 weeks postpartum. The present case-control study of replication capacity included 49 transmitters and 47 nontransmitters, who were randomly selected (i.e., the selection was blinded with respect to HIV-1 load and other clinical or laboratory information) from among, respectively, the 172 transmitters and 780 nontransmitters in the NVAZ trial

HIV-1 load was determined by use of the Roche AMPLICOR Monitor Test (version 1.5; Roche Diagnostics). Of the 96 women studied, 1 had an HIV-1 load of <400 copies/mL (the lower detection limit for the assay) and was considered to have an HIV-1 load value of 200 copies/mL (because that is the midpoint between 0 and 400 copies/mL)

Replication capacity was determined by the Monogram Biosciences Clinical Reference Laboratory, which used a modification of the PhenoSense HIV assay [10, 11] and was blinded with respect to the transmission status of the women. In this assay, protease reverse-transcriptase coding sequences from the test sample are transferred to a subtype B–resistance test vector, which, together with an expression plasmid encoding a heterologous envelope protein, is cotransfected into HEK293 cells. Pseudotyped virus particles produced by the transfected cells are used to infect fresh HEK293 cells; in the absence of antiretroviral drugs, the efficiency of infection in a single replication cycle is measured on the basis of the activity of luciferase and is compared with that of a reference strain, to determine the replication capacity. A replication capacity of 100% indicates that the replication capacity of the resistance test vector is the same as the median of a subtype B, wild-type (i.e., drug-sensitive) virus population

The following statistical methods compared variables in transmitters with those in nontransmitters: t test, for equality of maternal (1) age, (2) parity, and (3) log₁₀ HIV-1 load at delivery, and (4) replication capacity of HIV-1; and Fisher’s exact test, for equal proportions of each drug regimen used to treat the infants. Multivariate logistic regression for predictors of transmission was performed by use of SAS (version 8.2; SAS Institute). The methods involved Proc Logistic, in which all of the variables considered were forced into the final multivariate model

In Malawi, the NVAZ trial was approved by the Research and Ethics Committee at the University of Malawi College of Medicine; in the United States, it was approved by the Committee on Human Research at the Johns Hopkins Bloomberg School of Public Health. All women gave written informed consent for HIV testing and enrollment in the study

ResultsThe present study compared the replication capacity of HIV-1 in Malawian women in the NVAZ trial who did or did not transmit HIV-1 to their infants. Replication capacity was 1.3%–95% in transmitters and 0.32%–76% in nontransmitters; the mean ± SD replication capacity for all maternal plasma samples was 32.5% ± 19.3% and was higher in transmitters (37.2% ± 19.3%) than in nontransmitters (27.5% ± 18.3%) (P=.01, by t test) (table 1 and figure 1). The mean log₁₀ maternal HIV-1 load at delivery also was higher in transmitters than in nontransmitters (table 1). Mean maternal age and mean parity were similar in the 2 groups (table 1). In the NVAZ trial, transmission of HIV-1 was statistically more frequent in infants who received 1 dose of nevirapine alone than it was in infants who received 1 dose of nevirapine plus 1 week of daily doses of zidovudine [8]. In this randomly selected subset of 96 women from the NVAZ trial, the proportion of infants who received 1 dose of nevirapine was slightly (albeit not statistically significantly) higher for the transmitter mothers (32/49 [69.3%]) than for the nontransmitter mothers (27/47 [57.4%]) (table 1). In multiple logistic regression with adjustment for the maternal HIV-1 load, the drug regimen used to treat the infants, maternal age and parity, and the time of rupture of membranes, a higher replication capacity of maternal HIV-1 was independently associated with transmission of HIV-1 (odds ratio, 1.42 [95% confidence interval, 1.11–1.90], for each 10% increase in replication capacity; P=.006) (table 2)

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Figure 1

Comparison of replication capacity of HIV-1: transmitters vs. nontransmitters. For each of the 2 columns of data, the mean and SD values are indicated by horizontal and vertical lines, respectively

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Table 1

Characteristics of study subjects, and comparison of replication capacity of HIV-1 in transmitters and nontransmitters

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Table 2

Risk factors associated with HIV-1 transmission in a multiple logistic-regression model

DiscussionPrevious studies have found associations between mother-to-child transmission of HIV-1 and a number of factors, including maternal HIV-1 load, birth order, mode of delivery, and breast-feeding, and also have shown that the replication capacity of the pol gene of HIV is a potentially important determinant of viral fitness and pathogenicity [1]. To our knowledge, the present report is the first to document an association between the replication capacity of HIV-1 and the frequency of mother-to-child transmission of HIV-1. In the present study’s cohort of antiretroviral drug–naive pregnant women in Malawi, a higher replication capacity of maternal HIV-1 was associated with a higher frequency of mother-to-child transmission of HIV-1 when the infant received either 1 dose of nevirapine or 1 dose of nevirapine plus 1 week of daily doses of zidovudine. This association persisted after the analysis was controlled for maternal HIV-1 load and other factors

It is noteworthy that the median replication capacity of the samples from these drug-naive, subtype C–infected subjects was low (32.5%). This finding is consistent with data from some other non–subtype B samples lacking known drug-resistance mutations in protease or reverse transcriptase (N.P., unpublished observations); whether it represents an assay effect due to creation of intersubtypic chimeras (because the test vector’s backbone is subtype B) or an inherent biological property of non–subtype B viruses remains to be determined and is the subject of ongoing investigation. Nevertheless, because the samples studied here are all subtype C and did not involve direct comparisons with other subtypes, there is only a minimal possibility that an assay effect could alter the conclusion that viruses with a lower replication capacity are less likely to be transmitted

The present study suggests that the replication capacity of HIV-1 may be an important determinant of mother-to-child transmission of HIV-1. Further studies are necessary to confirm the association between the replication capacity and transmission of HIV-1, in diverse HIV-1 subtypes, in different clinical settings, and for different routes of transmission of HIV-1. It is not immediately obvious whether/how the measurement of replication capacity may be used in clinical decision making in the setting of prevention of mother-to-child transmission of HIV-1. Perhaps if the replication capacity of the maternal virus were high, the impetus toward aggressive antiretroviral therapy would be strengthened. Therefore, the replication capacity of HIV-1 might represent an additional piece of information to be considered by the physician and patient when they make decisions about treatment

The results of the present study do suggest that determinants in the gag/pol-gene region of HIV-1 influence mother-to-child transmission of HIV-1, because, in the PhenoSense HIV assay, this is the only region of the patient’s HIV-1 genome that is inserted into the resistance test vector. This finding may help to direct future studies seeking to identify and define those determinants, thereby expanding our basic understanding of the transmission of HIV-1 in this and other settings. It is possible that such studies might lead to simpler, less expensive methods to identify women at higher risk for transmission, who might benefit from more-intensive intervention. The clinical utility of such assays would be likely to depend on the resources available for prevention of mother-to-child transmission of HIV-1

• Received November 9, 2005.
• Accepted January 11, 2006.

• © 2006 by the Infectious Diseases Society of America