Treatment Experience and Repeat Pregnancy Impact the Effectiveness of Non-Nucleoside Reverse Transcription Inhibitor-Highly Active Antiretroviral Therapy for the Prevention of Mother to Child Transmission of Human Immunodeficiency Virus

Abstract

Non-nucleoside reverse transcription inhibitor (NNRTI)-containing antiretroviral therapy (ART) for the prevention of mother to child transmission (PMTCT) of human immunodeficiency virus (HIV) has led to dramatic reductions in perinatal HIV infection in resource-constrained settings. Nonetheless, PMTCT programs are complicated by repeat pregnancies, in which long-term or repeat exposures to PMTCT regimens over time may lead to the acquisition of HIV drug resistance mutations, and consequent treatment failure. In this study, we retrospectively assessed the effectiveness of the NNRTI-based PMTCT protocol from 2008 to 2010 in The Bahamas National HIV/AIDS Program. We show that women who had been in repeat pregnancies and those who were already prescribed ART at conception were at increased risk of virologic failure, relative to treatment-inexperienced women and primigravida, respectively (AOR 3.1, 95% CI: 1.3–7.1, p = .008 and AOR 5.0, 95% CI: 1.8–14.1, p = .002). In addition, women undergoing treatment at conception were more likely to possess HIVDR mutations relative to treatment-naive women (AOR 447.1, 95% CI: 17.9–11,173.5, p = .001). Therefore, individual treatment history is a key metric determining the effectiveness of current and future PMTCT interventions. The implications of this to PMTCT programmatic success in light of the most recent WHO guidelines are discussed.

Introduction

Until their update in 2019 to include use of integrase inhibitors for prevention of mother to child transmission (PMTCT),^1,2 the World Health Organization (WHO) guidelines for the PMTCT of human immunodeficiency virus (HIV) recommended a non-nucleoside reverse transcriptase inhibitor (NNRTI)- and nucleoside reverse transcriptase inhibitor (NRTI)-based treatment protocol. Two options were offered by the 2010 WHO guidelines: Option A consisting of antepartum AZT and peripartum sdNVP and a postpartum AZT +3TC tail; and Option B in which pregnant women with CD4 cell counts <350 cells/μL are administered highly active antiretroviral therapy (HAART) containing the NNRTI nevirapine (NVP) supported by two NRTIs and continued indefinitely, whereas women with CD4 cell counts of 350 cells/μL and above are offered pregnancy-limited HAART consisting of the NNRTI efavirenz (EFV) and two NRTIs, which is discontinued following delivery. Option B+ recommends treatment for life regardless of CD4 cell count.³

Starting in 2001, The Bahamas National PMTCT Program administered pregnancy-limited NVP-based HAART for PMTCT prophylaxis for women with CD4 count of 350 cells/μL and greater; women with CD4 cell count less than 350 cells/μL were continued indefinitely on this regimen. These guidelines were revised in 2006 to change from NVP to EFV in the prophylactic treatment arm, in response to emerging evidence of the “NVP tail” phenomenon⁴; this was in anticipation that this would decrease the risk of acquisition of NNRTI resistance following short-course pregnancy-limited NNRTI-HAART. The revised 2006 Bahamian guidelines foreshadowed and mirrored the 2010 WHO Option B guidelines described above.

Despite the proven effectiveness of NNRTI-HAART for preventing vertical HIV transmission,^1
–3 less is known about the effectiveness of this regimen over time, that is, the “durability” of the regimen,^{4

–12} and this has implications particularly for the implementation of an Option B+ “treatment for life” protocol. This is particularly important in contexts in which so-called repeat pregnancies^{13

–20} are common, as there is a major concern that prior exposure to antiretrovirals might lead to selection of drug resistance mutations (DRMs), resulting in virologic failure and an increased risk of vertical transmission in a subsequent pregnancy.^{19

–32}

In this retrospective analysis of The Bahamas PMTCT Program, we examined 3 years of medical and laboratory records of HIV-positive pregnant women enrolled in the national PMTCT program in the period from 2008 to 2010. The major objective of this study was to assess whether there was any effect of treatment history and/or repeat pregnancy on the effectiveness of NNRTI-HAART for PMTCT of HIV.

Methods

Study design

We performed a retrospective cross-sectional analysis of all HIV-positive pregnant women enrolled in The Bahamas National PMTCT of HIV Program, between January 1, 2008 and December 31, 2010.

Medical records were reviewed to collect key indicator data, including age, nationality, pregnancy status (“first time” or “repeat”), HAART regimen and start date, delivery date, type of delivery, CD4 cell count, and viral load. In addition, we divided the women into three groups representing differing treatment history experiences: (1) treatment-naive women; (2) women who were administered short-course HAART in a previous pregnancy and restarted in the pregnancy under review; and (3) women who were on treatment at conception (consisting of both repeat and first-time pregnancies).

Laboratory testing

Viral load measurements (HIV RNA copies/mL) were performed using the Roche^® AmpliPrep™/TaqMan^® COBAS™ v1 platform at the Ministry of Health Reference Lab, Nassau Bahamas. Viral load measurements from the last trimester of pregnancy to 6 months postdelivery were recorded.

HIV genotyping was performed on a convenience sample of 29 stored plasma specimens derived from HIV-positive pregnant women from the period March 1, 2008 to September 30, 2009; these samples were selected on the basis of (1) having been drawn from within a pregnancy in the study cohort, (2) their availability following viral load testing, (3) having been in storage for less than 6 months following collection date, and (4) having a viral load of 1,000 copies/mL or greater.²⁴ These samples represent 29/103 (28.2%) of the total deliveries of HIV-positive women recorded during this 18-month period. Specimens were shipped to the British Columbia Centre for Excellence in HIV in Vancouver, Canada, where HIV genotyping was performed using an “in-house” system as described previously.²⁸

Statistical analyses

All statistical analyses were done in SPSS statistical program (IBM^®), version 21. Unadjusted odds ratios (ORs) for selected factors were calculated using logistic regression, and adjusted odds ratios (AORs) were calculated with correction for age.

Ethical approval

This study was approved by The Bahamas National Ethics Review Board.

Results

Study demography

There were 252 HIV-positive pregnancies recorded during the time period 2008–2010. The median age at antenatal HIV clinic presentation was 29 years. There were 2/252 (0.8%) HIV-positive births in the 2-year period; both women were repeat pregnancies and neither sought treatment to prevent vertical transmission. Of the available medical records, most (64%) of the pregnancies from the HIV program were Bahamian, and 36% were non-Bahamian. 39.7% were repeat pregnancies, and 60.3% were first-time pregnancies, with the majority (91.1%) originating from New Providence, the island where the capital city and the majority of the Bahamian population are located. In most cases, NNRTI-backbone HAART was administered, containing either NVP (50%) or EFV (35.1%), with a protease inhibitor (PI)-based regimen having been administered in a minority of cases (3.3%); in 11.6% of cases for which data were available, antenatal treatment was refused or not sought. Overall, in 51.9% of cases, the women were antiretroviral naive; 25.4% had short-course HAART in a previous pregnancy, discontinued, and then recommenced HAART in the current pregnancy; and 22.8% were on treatment before pregnancy and continued those ARVs during pregnancy. Viral suppression (<500 RNA copies/mL) was achieved in 87 of 123 pregnancies (69.9%). Average time from treatment commencement to delivery was 171 days (±241 days) and to DRM analysis was 42 days (±127 days), and average time from delivery to viral load was 23 days (±62 days). See Table 1.

Table 1.

Descriptive Statistics of the Prevention of Mother to Child Transmission Cohort 2008–2010

	Total available	n/Total (%)
No. of HIV-positive pregnancies recorded, 2008–2011	252	252/252 (100)
Pregnancy outcome data available	233	233/252 (92.4)
Stillbirths/spontaneous abortions		27/233 (11.6)
HIV-negative live births		204/233 (87.6)
HIV-positive live births		2/233 (0.9)
Nationality data available	164	164/252 (65.1)
Bahamian		105/164 (64)
Non-Bahamian		59/164 (36)
Pregnancy status data available	252	252/252 (100)
First time		152/252 (60.3)
Repeat		100/252 (39.7)
Island of origin data available	224	224/252 (88.9)
New providence		204/224 (91.1)
Family Island		20/224 (8.9)
Antenatal treatment regimen data available	242	242/252 (96)
NVP-HAART		121/242 (50)
EFV-HAART		85/242 (35.1)
PI-HAART		8/242 (3.3)
No treatment		28/242 (11.6)
Treatment history data available	189	189/252 (75)
ARV naive		98/189 (51.9)
Previous short-course HAART, restarted		48/189 (25.4)
On treatment at conception		43/189 (22.8)
Viral load data available	123	123/252 (48.8)
0–499 RNA copies/mL		86/123 (69.9)
500 RNA copies/mL or greater		37/123 (30.1)
Median maternal age at pregnancy (years)	29 (IQR: 25–33)
Mean time between HAART start and delivery (days)	171 (±241 days)
Mean time between delivery and viral load (days)	23 (±62 days)
Mean time between HAART start and HIV genotyping (days)	42 (±127 days)

NVP, nevirapine; EFV, efavirenz; PI, protease inhibitor; HAART, highly active antiretroviral therapy; ARV, antiretroviral.

Risk factors for virologic failure

We performed logistic regression analysis to examine risk factors for virologic failure, defined here as having a viral load of 500 RNA copies/mL and greater, in the 123 pregnancies for which a viral load was available. In this analysis, age, nationality, and antenatal regimen were not associated with virologic failure. On the contrary, treatment history was a predictor of virologic failure in that those women who had been on treatment before conception were significantly more likely in the multivariate analysis to have had virologic failure than women who were antiretroviral naive before pregnancy (AOR 5.0, 95% CI: 1.8–14.1, p = .002); this was in contrast to women whom had been recommenced on treatment in the pregnancy under investigation, having been administered short-course HAART in a previous pregnancy—there was no significantly increased risk in this treatment category versus HAART-naive women (AOR 2.5, 95% CI: 0.9–7.5, p = .10). In addition to treatment history, pregnancy status (i.e., “first time” or “repeat”) was also associated with increased risk of virologic failure, as women in “repeat” pregnancies were at a significantly greater risk of virologic failure versus “first-time” pregnancies (AOR 3.1, 95% CI: 1.3–7.1, p = .008). See Table 2.

Table 2.

Predictors of Virologic Failure, Defined as Greater than 500 RNA Copies/ml)

Factor	Virologic failure
Factor	n(%)	OR (95% CI)	AOR (95% CI)	p value
Age
<29 years	18 (30.0)	1.0 (0.5–2.2)
29+ years	19 (29.7)	Reference
Nationality
Bahamian	16 (31.4)	1.5 (0.5–4.5)
Non-Bahamian	5 (23.1)	Reference
Island
New Providence	27 (24.8)	0.3 (0.1–1.7)
Family Island	3 (50.0)	Reference
Antenatal regimen
NVP	22 (33.3)	1.0 (0.2–6.0)
EFV	12 (23.5)	0.6 (0.1–3.8)
PI	3 (42.9)	Reference
Treatment history
Previous short course, recommenced	9 (34.6)	2.6 (0.9–7.6)	2.5 (0.9–7.5)	.10
On treatment before pregnancy	14 (50.0)	4.9 (1.7–13.7)	5.0 (1.8–14.1)	.002
ARV naive	9 (16.7)	Reference
Pregnancy status
Repeat	21 (42.0)	3.0 (1.4–6.7)	3.1 (1.3–7.1)	.008
First time	15 (19.7)	Reference

Bold values represent statistical significance at or above the 95% confidence level.

HIV drug resistance

Of a convenience sample of 29 stored plasma samples from women enrolled in the PMTCT program from March 2008 to September 2009, 16 were antiretroviral-naive before pregnancy, 6 had short-course treatment in a previous pregnancy and then recommenced in the current pregnancy, and 7 were on treatment before pregnancy. HIV genotyping of the samples revealed that 7/29 (24%) contained HIV DRMs: in 6 of these cases, there was resistance to both NNRTI and NRTI class ARVs, and in 1 case, there was resistance to NNRTIs alone. The major NNRTI resistance mutations seen included K103N, Y181C, K101E, K101P, G190A, and V106A; in addition, the major NRTI mutations M184V, Y181C, K219E, M41L, T215Y, T215S, L74V, D67N, and K70R were seen. Women on treatment before conception were significantly more likely to possess HIV DRMs when compared with HAART-naive women (AOR 447.1, 95% CI: 17.9–11,173.5, p = .001), although the same was not the case with those women who had been recommenced on treatment in the pregnancy under review (AOR 5.9, 95% CI: 0.4–79.0, p = .18). Neither pregnancy history (first time or repeat), age, nor antenatal regimen was associated with increased risk of DRM detection; however, the risk of DRM in repeat versus first-time pregnancies was significant (p < .05) in the unadjusted OR, and very near to significance after adjustment for age (AOR 5.5, 95% CI: 1.0–32.0, p = .06). See Table 3.

Table 3.

Risk Factors Associated with Drug Resistance Mutations

	HIV drug resistance mutations (NNRTI/NRTI)
	n(%)	OR (95% CI)	AOR (95% CI)	p value
Treatment history
Previous short course, recommenced	1 (16.7)	6.4 (0.5–78.2)	5.9 (0.4–79.0)	.18
On ART before pregnancy	6 (85.7)	192.0 (15.9–2,316.8)	447.1 (17.9–11,173.5)	.001
ART naive	0 (0)	Reference	Reference
Antenatal treatment regimen
NVP-HAART	6 (40)	10.1 (1.1–97.0)
EFV-HAART	1 (7.1)	Reference
Pregnancy status
Repeat	5 (50)	5.7 (1.0–32.4)	5.5 (1.0–32.0)	0.06
First time	2 (10.5)	Reference	Reference

Bold values represent statistical significance at or above the 95% confidence level.

Discussion

We sought in this study to assess the impact of treatment experience and repeat pregnancy on the effectiveness of the NNRTI-based PMTCT HAART regimen. In a retrospective study cohort from 2008 to 2010 in The Bahamas National HIV/AIDS Program, the effectiveness of NNRTI-HAART was demonstrated by a very low vertical transmission rate of 0.8% (Table 1). Indeed, it is of note that the two instances of vertical HIV transmission in the study period were due to confounding socioeconomic factors that led to a failure to link to care, rather than a failure of the regimen itself.

Despite these successes, there is concern that the NNRTI-HAART might lose effectiveness over time and serial pregnancies, due to the acquisition of DRMs at both the individual and population levels; this is particularly the case with NVP which has a low genetic barrier to resistance.⁴ In further support, in our study, we found that women on NNRTI-HAART at conception (i.e., those on long-term treatment) were significantly more likely to experience virologic failure and DRM acquisition than treatment-naive counterparts (Tables 2 and 3).

However, the same was not the case for women whom had previously been prescribed short-course HAART, and this latter finding was inconsistent with a previous large-scale study showing a weakly significant association between a history of pregnancy-limited short-course NNRTI-HAART and detectable viral load in a subsequent pregnancy.¹⁹ A number of studies also report virologic failure and significant rates of DRM detection following cessation of short-course HAART.^{20

–27} However, we speculate that this difference might be due to sampling size; it is notable that in our study, the increased risk of virologic failure and HIVDR for women on previous short-course NNRTI-HAART was near to the 95% confidence level in our analysis (Tables 2 and 3; p = .10 and p = .18, respectively), and it is possible that the comparative risk may have been significant in a larger sampling, and/or more sensitive genotyping methodology.

We also speculate that our finding of a strongly significant correlation between long-term treatment and negative clinical outcome (virologic failure and DRM emergence), by contrast to those women whom had previously undergone short-course HAART, might be explained by the fact that many of those on long-term treatment were already at a more advanced stage of disease than those previously administered short-course HAART (given the dependence of Option B treatment guidelines on CD4 cell count).

In addition, most studies agree that adherence decreases over time on treatment^33

–36; this means that, because adherence is inversely correlated with DRM detection and virologic failure (especially in the case of NNRTI-HAART), the risk for DRM emergence increases over time on treatment.^34
–36 It is indicative that in the case of Option B+ lifelong treatment programs, in which HIV-positive women are offered NNRTI-HAART regardless of CD4 cell count, there is evidence for suboptimal adherence, virologic failure, and DRM emergence from as early as 6 months postpartum.^37

–42 Therefore, notwithstanding the long-term health benefits of Option B+ versus Option B (which is CD4 dependent),^1,2 women undergoing pregnancy-limited HAART may be more adherent than those prescribed treatment for life.

Collectively, this evidence indicates that women historically prescribed short-course NNRTI-HAART for PMTCT, as well as those prescribed lifelong NNRTI-HAART, may be at increased risk of having acquired DRMs and may consequently experience virologic failure in a subsequent pregnancy, potentially compromising present and future PMTCT success.

In The Bahamas, HIV-positive repeat pregnancies are relatively common; indeed, in our cohort, 39.7% of the pregnancies were repeat pregnancies (Table 1), similar to that seen elsewhere.^{12

–20} Importantly, we found that women in repeat pregnancies were at increased risk of virologic failure, relative to first-time pregnancies (Table 2), suggesting that prior exposure to the PMTCT regimen was a risk factor for virologic failure. This was consistent in previous large cohort studies done in the United Kingdom and Ireland showing that NNRTI-HAART-experienced women in repeat pregnancies were at greater risk of having a detectable viral load than treatment-naive primigravidae.^19,20 We infer that repeat pregnancies are correlated to virologic failure because of historic exposure of these women to NNRTI-HAART (both long-term and short-course treatment) and the probable emergence of DRMs as a result.

There are several limitations to this study. First, our convenience sample of stored plasma specimen used for HIV genotyping was small, consisting of only 29 samples derived from an 18-month period in 2008/2009, and therefore, may not be fully representative of the 3-year 2008–2010 cohort. That being said, this resistance data support the more robust data on virologic failure; together, this underpins our hypothesis that in the real-world context of a national PMTCT program, the effectiveness of the WHO Option B regimen is inversely correlated to both parity and treatment exposure in HIV-positive women.

In addition, the data here are historical, and do not necessarily depict the current situation in The Bahamas PMTCT Program; indeed the purpose of our retrospective approach is to analyze risk of treatment failure in a real-world context of a continuous national program in which women commonly have multiple HIV-positive pregnancies. Here, we have used a “Sankofa” approach (to paraphrase—analyzing the past to illuminate the present),⁴³ and in this regard our major relevant finding is that the historical NNRTI-based regimen, although highly effective for PMTCT, resulted in DRM emergence in a significant proportion of women during the period of analysis, and this has implications for current and future treatment success in this population of women.

In their most recent PMTCT guidelines, the WHO recommends use of either an integrase inhibitor (dolutegravir [DTG])- or alternatively EFV-based HAART, along with two NRTIs.² Our finding in our study population of both NNRTI- and NRTI- associated DRMs (Table 3) indicates that treatment-experienced women (including those in repeat pregnancies) are at potential risk of virologic failure if prescribed according to the most recent WHO PMTCT guidelines, whether EFV-based HAART, or a DTG-based regimen that includes NRTIs, is selected⁴⁴ (although recent evidence indicates that DTG can maintain virologic suppression even in combination with NRTIs against a background of NRTI resistance⁴⁵).

Critically, implementation of the new WHO PMTCT guidelines will take place in the context of national programs such as The Bahamas' in which there are likely to be a significant proportion of multiparous HIV-positive women (“repeat pregnancies”) who would have been prescribed a historical HAART regimen, including both NRTIs and NNRTIs. We speculate from our present results that there may be a significant proportion of women who are likely to have acquired resistant HIV strains as a result of prior treatment experience, and are reappearing in the present for treatment under current guidelines. In this light, analysis of treatment history and HIV genotyping before PMTCT treatment initiation may help to identify a personalized treatment regimen in a PMTCT population that includes a significant proportion of treatment-experienced individuals. Therefore, these findings may be useful in the future for devising cost-efficient tests and treatment algorithms for national PMTCT programs.

Footnotes

Acknowledgments

We gratefully acknowledge Dr. Ismae Whyms, Mr. Keith McConnell, nurses Marva Jervis and Branishka Lewis, and Dr. Nikkiah Forbes.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Funding was provided by The Bahamas Ministry of Health.

References

World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. Recommendations for a public health approach- Second edition. Geneva, Switzerland: World Health Organization, 2016.

World Health Organization. Update of recommendations on first- and second-line antiretroviral regimens. Geneva, Switzerland: World Health Organization, 2019.

World Health Organization. Antiretroviral drugs for treating pregnant women and preventing HIV infection infants. Recommendations for a public health approach. Geneva, Switzerland: World Health Organization, 2010.

Paredes

, Marconi

, Lockman

, et al.: Impact of antiretroviral drugs in pregnant women and their children in Africa: HIV resistance and treatment outcomes. J Infect Dis, 2013; 207 (Suppl 2):S93–S100.

Palella Jr

, Chmiel

, Moorman

, Holmberg

: Durability and predictors of success of highly active antiretroviral therapy for ambulatory HIV infected patients. HIV Outpatient Study Investigators. AIDS, 2002; 1612:1617–1626.

Cuzin

, Pugliese

, Bugnon

, et al.: Triple nucleoside reverse transcriptase inhibitor- vs. nonnucleoside reverse transcriptase inhibitor-containing regimens as first-line therapy: Efficacy and durability in a prospective cohort of French HIV-infected patients. HIV Med, 2005; 6:388–395.

Ungsedhapand

, Srasuebkul

, Cardiello

, et al.; HIV-NAT 002 and HI, V-NAT 003 Study Team: Three-year durability of dual-nucleoside versus triple-nucleoside therapy in a Thai population with HIV infection. J Acquir Immune Defic Syndr, 2004; 36:693–701.

Benator

, Elmi

, Rodriguez

, et al.: True durability: HIV virologic suppression in an urban clinic and implications for timing of intensive adherence efforts and viral load monitoring. AIDS Behav, 2015; 19:594–600.

Desmonde

, Eboua

, Malateste

, et al.; IeDEA Pediatric West African Working Group: Determinants of durability of first-line antiretroviral therapy regimen and time from first-line failure to second-line antiretroviral therapy initiation. AIDS, 2015; 29:1527–1536.

10.

Sheth

, Ofotokun

, Buchacz

, et al.: Antiretroviral Regimen Durability and Success in Treatment-Naive and Treatment-Experienced Patients by Year of Treatment Initiation, United States, 1996–2011. J Acquir Immune Defic Syndr, 2016; 71:47–56.

11.

Xing

, Ruan

, Li

, et al.: HIV drug resistance and its impact on antiretroviral therapy in Chinese HIV-infected patients. Chinese National HIVDR Surveillance and Monitoring Network. PLoS One, 2013; 8:e54917.

12.

Billioux

, Nakigozi

, Newell

, et al.: Durable Suppression of HIV-1 after Virologic Monitoring-Based Antiretroviral Adherence Counseling in Rakai, Uganda. PLoS One, 2015; 10:e0127235.

13.

Floridia

, Tamburrini

, Masuelli

, et al.: Rate, correlates and outcomes of repeat pregnancy in HIV-infected women. Italian Group for Surveillance of Antiretroviral Treatment in Pregnancy. HIV Med, 2017; 18:440–443.

14.

French

, Cortina-Borja

, Thorne

, Tookey

: Incidence, patterns, and predictors of repeat pregnancies among HIV-infected women in the United Kingdom and Ireland, 1990–2009. J Acquir Immune Defic Syndr, 2012; 59:287–293.

15.

Kreitchmann

, Megazzini

, Melo

, et al.: Repeat pregnancy in women with HIV infection in Latin America and the Caribbean. AIDS Care, 2015; 27:1289–1297.

16.

Watts

, Huang

, Cohn

, Smith

, Hitti

: Repeat pregnancies among HIV-infected women enrolled in clinical trial PACTG1022. J Acquir Immune Defic Syndr, 2009; 51:231–233.

17.

Akelo

, McLellan-Lemal

, Toledo

, et al.: Determinants and Experiences of Repeat Pregnancy among HIV-Positive Kenyan Women—A Mixed-Methods Analysis. PLoS One, 2015; 10:e0131163.

18.

Suryavanshi

, Erande

, Pisal

, et al.: Repeated pregnancy among women with known HIV status in Pune, India. AIDS Care, 2008; 20:1111–1118.

19.

French

, Tookey

, Cortina-Borja

, de Ruiter

, Townsend

, Thorne

: Influence of short-course antenatal antiretroviral therapy on viral load and mother-to-child transmission in subsequent pregnancies among HIV-infected women. Antivir Ther, 2013; 18:183–192.

20.

French

, Thorne

, Tariq

, Cortina-Borja

, Tookey

: Immunologic status and virologic outcomes in repeat pregnancies to HIV-positive women not on antiretroviral therapy at conception: A case for lifelong antiretroviral therapy?. AIDS, 2014; 28:1369–1372.

21.

Geretti

, Fox

, Johnson

, et al.; INSIGHT Strategies for Management of Antiretroviral Therapy (SMART) Study Group: Sensitive assessment of the virologic outcomes of stopping and restarting non-nucleoside reverse transcriptase inhibitor-based antiretroviral therapy. PLoS One, 2013; 8:e69266.

22.

Lyons

, Coughlan

, Byrne

, Hopkins

, Hall

, Mulcahy

: Emergence of antiretroviral resistance in HIV-positive women receiving combination antiretroviral therapy in pregnancy. AIDS, 2005; 19:63–67. Erratum in: AIDS 2007 21:1671.

23.

Paredes

, Cheng

, Kuritzkes

, Tuomala RE; Women and Infants Transmission Study (WITS) Group: Postpartum antiretroviral drug resistance in HIV-1-infected women receiving pregnancy-limited antiretroviral therapy. AIDS, 2010; 24:45–53.

24.

Ellis

, Huang

, Hitti

, Frenkel

; P1022 Study Team : Selection of HIV resistance associated with antiretroviral therapy initiated due to pregnancy and suspended postpartum. J Acquir Immune Defic Syndr, 2011; 58:241–247.

25.

Duran

, Losso

, Salomón

, et al.; NISDI Perinatal Study Group: Drug resistance among HIV-infected pregnant women receiving antiretrovirals for prophylaxis. AIDS, 2007; 21:199–205.

26.

Pérez

, Vignoles

, Laufer

, et al.: Low rate of emergence of nevirapine and lamivudine resistance after post-partum interruption of a triple-drug regimen. Antivir Ther, 2008; 13:135–139.

27.

Soto-Ramirez

, Rodriguez-Diaz

, Durán

, et al.; NISDI Perinatal Study Group: Antiretroviral resistance among HIV type 1-infected women first exposed to antiretrovirals during pregnancy: Plasma versus PBMCs. AIDS Res Hum Retroviruses, 2008; 24:797–804.

28.

Harrigan

, Hogg

, Dong

, et al.: Predictors of HIV drug-resistance mutations in a large antiretroviral-naive cohort initiating triple antiretroviral therapy. J Infect Dis, 2005; 191:339–347.

29.

Rupérez

, Noguera-Julian

, González

, et al. HIV drug resistance patterns in pregnant women using next generation sequence in Mozambique. PLoS One, 2018; 13:e0196451.

30.

Adeniyi

, Obi

, Goon

, et al.: HIV-1 drug resistance surveillance among parturient women on anti-retroviral therapy in the Eastern Cape, South Africa: Implications for elimination of mother-to-child transmission. J Clin Virol, 2019; 117:89–95.

31.

Milne

, Silverman

, Beck

, et al.: Minority and majority pretreatment HIV-1 drug resistance associated with failure of first-line nonnucleoside reverse-transcriptase inhibitor antiretroviral therapy in Kenyan women. AIDS, 2019; 33:941–951.

32.

Martin-Odoom

, Brown

, Odoom

, et al.: Emergence of HIV-1 drug resistance mutations in mothers on treatment with a history of prophylaxis in Ghana. Virol J, 2018; 15:143.

33.

Tam

, Chui

, Brumme

, et al.: The relationship between resistance and adherence in drug-naive individuals initiating HAART is specific to individual drug classes., J Acquir Immune Defic Syndr, 2008; 49:266–271.

34.

Wilson

, Bangsberg

, Shen

, et al.; Multisite Adherence Collaboration on HIV 14 Investigators: Heterogeneity among studies in rates of decline of antiretroviral therapy adherence over time: Results from the multisite adherence collaboration on HIV 14 study. J Acquir Immune Defic Syndr, 2013; 64:448–454.

35.

Meresse

, March

, Kouanfack

, et al.; Stratall ANRS 12110/ESTHER Study Group: Patterns of adherence to antiretroviral therapy and HIV drug resistance over time in the Stratall ANRS 12110/ESTHER trial in Cameroon. HIV Med, 2014; 15:478–487.

36.

, Paredes

, Ribaudo

, et al.: Relationship between minority nonnucleoside reverse transcriptase inhibitor resistance mutations, adherence, and the risk of virologic failure. AIDS, 2012; 26:185–192.

37.

Hosseinipour

, Nelson

JAE

, Trapence

, et al.; PURE Malawi Consortium: Viral Suppression and HIV Drug Resistance at 6 Months Among Women in Malawi's Option B+ Program: Results From the PURE Malawi Study. J Acquir Immune Defic Syndr, 2017; 75(Suppl 2):S149–S155.

38.

Machnowska

, Hauser

, Meixenberger

, et al.: Decreased emergence of HIV-1 drug resistance mutations in a cohort of Ugandan women initiating option B+ for PMTCT. PLoS One, 2017; 12:e0178297.

39.

Gill

, Hoffman

, Bobrow

, et al.: Detectable Viral Load in Late Pregnancy among Women in the Rwanda Option B+ PMTCT Program: Enrollment Results from the Kabeho Study. PLoS One, 2016; 11:e0168671.

40.

Ngarina

, Kilewo

, Karlsson

, et al.: Virologic and immunologic failure, drug resistance and mortality during the first 24 months postpartum among HIV-infected women initiated on antiretroviral therapy for life in the Mitra plus Study, Dar es Salaam, Tanzania. BMC Infect Dis, 2015; 15:175.

41.

Haas

, Msukwa

, Egger

, et al.: Adherence to Antiretroviral Therapy During and After Pregnancy: Cohort Study on Women Receiving Care in Malawi's Option B+ Program. Clin Infect Dis, 2016; 63:1227–1235.

42.

Tesfaye

, Hibistu

, Abebo

, et al.: Option B plus antiretroviral therapy adherence and associated factors among HIV positive pregnant women in Southern Ethiopia. BMC Pregnancy Childbirth, 2019; 19:82.

43.

Galloway

“African Tradition, Proverbs, and Sankofa”, The Spirituals Project at the University of Denver. Available at https://web.archive.org/web/20110423095008/http://ctl.du.edu/spirituals/literature/index.cfm (2004), accessed Jan, 2020 .

44.

Dorward Dorward

, Lessells

, Drain

, et al.: Dolutegravir for first-line antiretroviral therapy in low-income and middle-income countries: Uncertainties and opportunities for implementation and research. Lancet HIV, 2018; 5:e400–e404.

45.

Demarest

, Underwood

, St Clair

, Dorey

, Brown

, Zolopa

: Short Communication: Dolutegravir-Based Regimens Are Active in Integrase Strand Transfer Inhibitor-Naive Patients with Nucleoside Reverse Transcriptase Inhibitor Resistance. AIDS Res Hum Retroviruses, 2018; 34:343–346.