! 1!

NIAID Special Ethics Review

of the PROMISE Trials

Final Report of the Expert Review Panel

Submitted to the Divisions of AIDS

October 17, 2012

! 2!

Table of Contents PROMISE Review Panel Roster 3 Acknowledgements 4 Abbreviations 5 Special Ethics Review Agenda 6 REPORT

Background and rationale for special ethics review 7 Questions to the Panel 9 Documents provided to the Panel 9 Synopsis of the presentations to the Panel 10 Gaps in knowledge that PROMISE aims to fill 13 Panel responses to questions 17 Additional comments/considerations 20 Specific recommendations 21 References 23

APPENDICES Appendix 1: Synopsis of the PROMISE trials design Appendix 2: Scientific Background for PROMISE

Appendix 3: Scientific Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment Appendix 4: PROMISE protocol modifications

Appendix 5: WHO Programmatic Update 2012 Appendix 6: Letters received by the PROMISE study teams from the Ministries

of Health of Malawi and Uganda

! 3!

PROMISE Review Panel Roster

! 4!

Acknowledgements

The Panel would like to thank the staff of the Division of AIDS and NICHD for the extensive preparatory work that was conducted to facilitate the Panel’s deliberations, along with all the presenters who provided us with extremely useful information and responded very constructively to questions from the Panel. As well, we would like to thank the DAIDS support staff for their assistance with travel and accommodations and hospitality. And finally, the Panel would like to thanks Dr. Dennis Dixon for his outstanding support and guidance through the process.

! 5!

Abbreviations

ART Antiretroviral Therapy (used before defined)

ARV Antiretroviral

AZT Azidothimidine (Zidovudine)

BF Breastfeeding

DAIDS Division of AIDS

EFV Efavirenz

HPTN HIV Prevention Trials Network

IMPAACT International Maternal-Pediatric-Adolescent AIDS Clinical Trials

MTCT Mother to Child Transmission

NIAID National Institute of Allergy and Infectious Diseases

NVP Nevirapine

PEPFAR President’s Emergency Plan for AIDS Relief

PMTCT Prevention of MTCT

PROMISE Promoting Maternal Infant Survival Everywhere

sdNVP Single-dose NVP

TDF Truvada

WHO World Health Organization

! 6!

AGENDA!

Special!Ethics!Review!for!PROMISE!

June!18E19,!2012!

!

June!18!(6700EA!Rockledge!Drive,!Room!245)!

8:30!–!9:00!Welcome!and!Introductions!–!Dr.!Emily!Erbelding,!Deputy!Director,!DAIDS!

9:00!–!9:45!Overview!of!PROMISE!Design!and!Current!Status!–!Dr.!Mary!Glen!Fowler,!Chair,!PROMISE*!

9:45!–!10:30!Scientific!Context!–!Dr.!Laura!Guay,**!George!Washington!University!and!Elizabeth!Glazer!Pediatric!AIDS!Foundation!

10:30!–!10:45!Break!

10:45!–!11:30!Operational!Context!–!Dr.!Nathan!Shaffer,!WHO(by!phone)!

11:30!–!12:15!MOH!Presentation!–!Dr.!Angela!Mushavi,**!Zimbabwe!Ministry!of!Health!and!Child!Welfare!

12:15!–!12:45!Discussion!–!Dr.!Jim!Lavery,**!University!of!Toronto,!and!Dr.!Paula!Munderi,**MRC!Uganda!Research!Unit!on!AIDS!

!

12:45!–!1:30!Lunch!

!

1:30!–!3:15!Discussion!

3:15!–!3:30!Break!

3:30!–!5:30!Discussion!

!

June!19!(6700EA!Rockledge!Drive,!Room!235)!

8:30!–!11:30!Closed!Discussion!and!Formulation!of!Recommendations!to!DAIDS!

11:30!–!11:45!Break!

11:45!–!12:30!Debriefing!

*!!Presentations!at!9:00,!9:45,!10:45,!and!11:30!on!June!18!are!30!minutes,!followed!by!15!minutes!for!questions!and!discussion!

**!Panel!CoEChair!or!Member

! 7!

Background and Rationale for the Special Ethics Review

In 2010, after several years of planning and preparation, the NIAID-NICHD sponsored International Maternal-Pediatric-Adolescent AIDS Clinical Trials (IMPAACT) network began enrolment into the set of clinical trials known as PROMISE(Promoting Maternal Infant Survival Everywhere). The PROMISE trials were designed to evaluate the optimal antiretroviral drug interventions for: the prevention of antepartum and intrapartum mother to child transmission (MTCT) of HIV; the prevention of postpartum transmission of HIV to breastfed infants; and for the preservation of maternal health beyond the period of risk for MTCT, either after delivery or following cessation of breastfeeding.

The design and management of national programs for prevention and treatment of HIV in low and middle income countries are complex policy challenges. Country guidelinesfor antiretroviral therapy for pregnant women and the prevention of mother to child transmission (PMTCT) of HIV infectiongenerally involve criteria for selecting from among available drug regimens and for starting and stopping regimens based on pregnancy and breastfeeding status and current levels of CD4 count and HIV viral load. In developing countries, guidelines are mostly adapted from World Health Organization (WHO) PMTCT and ART guidelines.

When begun, the PROMISE trials were fully compatible with treatment guidelinesin the participating countries in North and South America, Asia, and Africa. The PROMISE trials, as designed, are also compatible with the most recent, 2010 revision of WHO guidelines for antiretroviral drugs for treating pregnant women and preventing HIV infections in infants(1).The 2010 WHO guidelines emphasize the importance for maternal health, and for the PMTCT, of initiating life-long combination antiretroviral therapy (ART) for HIV-infected pregnant women with CD4 counts <350 cells/mm3, or advanced clinical stages of HIV. Women who do not yet need ART for their own health by these criteria are recommended either of two ARV prophylaxis options (Option A or Option B) for PMTCT (see Table 1.). The research population addressed by the PROMISE trials is women who do not yet meet eligibility criteria for antiretroviral therapy for their own health in the countries in which the study is conducted.

In 2011, new research findings became available indicating that initiating combination ART at the higher CD4 count threshold of 550 cells/mm3 had the added benefit of significantly reducing the risk of HIV transmission to an uninfected sexual partner(2). This new information, coupled with the growing experience of countries facing the challenges of implementing PMTCT programs, such as difficulty in identifying women who are eligible for ART for their own health, has led to the consideration of a third option (Option B+). Option B+ would provide lifelong ART to all HIV positive pregnant women at the time of diagnosis, regardless of their CD4 cell count or HIV disease stage (see Table 1). Increased availability of drugs in limited resource settings has generated optimism that wider delivery of ART might be possible. HIV-infected pregnant women are a logical target for such expanded delivery, since the women who are

! 8!

presenting for antenatal care are already linked to health care providers, and their babies remain at risk of infection through mother-to-child transmission (MTCT).

(From WHO Programmatic Update, 2012 (3, p. 2)

A recent programmatic update from the WHO(3) urges countries to consider adopting Option B+ for reasons of operational and programmatic simplicity in order to achieve the targets set by a large consortium of member states, international organizations, civil society and private sector organizations, regional bodies and agencies of the United Nations in ‘The Global Plan towards the Elimination ofnew HIV infections among children by 2015 and keeping their mothers alive’(4). The Option B+ approach is already being implemented in Malawi, and the WHO programmatic update has led other countries to consider it.

The implications of this recent WHO programmatic update for the ethical conduct of the PROMISE trials are unclear. With the WHO’s recent promotion of Option B+, it is reasonable to

! 9!

ask whether trials studying different treatment regimens remain relevant. In an effort to ensure that the PROMISE trials continue to have relevance and value scientifically and conform to the highest standards of ethical research, in late 2011 NIAID decided to assemble a panel of independent experts to review the current ethical status of the trials and offer advice, accordingly. A meeting for this purpose took place in Bethesda on June 18 and 19, 2012. This is the report of that meeting.

Questions to the Panel

NIAID’s concerns about the on-going value and ethical quality of the PROMISE trials were distilled into two main questions for the panel:

1. Given the changing landscape of PMTCT guidelines, could PROMISE be conducted (or under what conditions can PROMISE be conducted) in a country that has chosen a national policy for PMTCT that recommends initiation of antiretroviral therapy for life in all HIV-infected pregnant women regardless of CD4 count (Option B+)?

2. Given diverse implementation challenges and current policy developments in PMTCT programs, does the study still have sufficient value for informing clinical, policy, or program decisions now or in the future? If no, is there a design modification that would provide sufficient value to continue the study?

Background documents provided to the Panel

• A Synopsis of the PROMISE Trial Design (see Appendix 1)

• A comprehensive Scientific Background document for the PROMISE trials including a detailed review of the relevant clinical science, including references to earlier clinical trials and other studies that informed the design of the PROMISE trials, and review of relevant international guidelines, including WHO PMTCT Guidelines, and a range of other issues related to PROMISE(5). (see Appendix 2)

• Scientific Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment. A summary of other ongoing clinical trials including timelines for completion. (see Appendix 3)

• Current versions of the PROMISE P1077 protocols. (P1077BF, P1077FF, P1077HS)

• A document highlighting modifications in PROMISE protocols currently in process. Specifically amendments made to address i) results of HPTN 052 showing that

! 10!

antiretroviral treatment can reduce heterosexual HIV transmission, ii) changes in infant feeding patterns to longer duration (>12 months), iii) increasing use of tenofovir-based antiretroviral regimens in pregnant women without hepatitis B infection, and iv) streamlining the protocols to make them less complex for subjects and staff to implement. (see Appendix 4)

• WHO Programmatic Update: Use of Antiretroviral drugs for treating pregnant women and preventing HIV infection in infants. April 2012 statement of current thinking at the WHO, reflecting the importance that operational and implementation issues are likely to play in the next revision of WHO recommendations expected in 2013(3). (see Appendix 5)

• Letters received by the PROMISE study teams from the Ministries of Health of Malawi and Uganda supporting continuation of the study as designed. (see Appendix 6)

Synopsis of the presentations to the Panel

Dr. Emily Erbelding

Dr. Erbelding, Deputy Director, Division of AIDS (DAIDS), NIAID, welcomed the panel and invited presenters to the meeting. To inform its discussions the panel heard presentations on background, design, and current status of PROMISE, by Dr. Mary Glenn Fowler of the Johns Hopkins University School of Medicine and PROMISE chair. The broader scientific context of related ongoing clinical research was presented by Dr. Laura Guay of the Elizabeth Glaser Pediatric AIDS Foundation, and also a member of the panel. Insights about recent discussions by WHO and member countries that led to the release of the programmatic update in 2012(3), and are expected to lead to changes in the Guidelines themselves in 2013, were provided by Dr. Nathan Shaffer, PMTCT Team Leader, Department of HIV/AIDS,WHO, by teleconference. And an account of the challenges associated with implementing national guidance for ART for pregnant women and PMTCT was provided by Dr. Angela Mushavi, National PMTCT Coordinator, Ministry of Health and Child Welfare, Zimbabwe, also a member of the panel.

Dr. Mary Glenn Fowler

Dr. Fowler presented the design and current status of PROMISE, whose primary and secondary objectives relate to maternal and infant health outcomes for HIV-infected women who do not qualify for ART for their own health, according to current WHO and national treatment guidelines. The three primary research questions being addressed by PROMISE are:

! 11!

i) Which is more effective and safe to prevent MTCT during the antenatal period (pregnancy): ZDV plus intrapartum single dose nevirapine, or a triple-antiretroviral drug prophylaxis regimen?

ii) Which is more effective and safe to prevent postnatal MTCT through breast milk: provision of daily nevirapine prophylaxis to the infant, or provision of triple-antiretroviral drug prophylaxis to the mother for the duration of breastfeeding (up to 18 months)?

iii) For women who receive triple-antiretroviral drug prophylaxis during pregnancy and/or breastfeeding, is it better to stop or continue the triple-antiretroviral drug regimen after risk of MTCT has ceased (i.e., after delivery, if formula feeding, or after lactation, if breastfeeding)?

In each case the first of the pair of comparators corresponds to current WHO guidelines.

In addition to discussing these three primary comparisons, Dr. Fowler emphasized that PROMISE will also contribute important new knowledge with respect to regimen adherence, emergence of antiretroviral drug resistance, safety of Tenofovir (TDF) in pregnant women and their babies, and any complications of combination ART to mothers, for example those arising from concurrent hepatitis B infection.

Dr. Laura Guay

Dr. Guay emphasized that there are important limitations in our knowledge about all current ART regimens (A, B and B+). She summarized a large number of relevant clinical trials and observational studies, most of them very much smaller than PROMISE and less likely to produce clear answers, even though in some cases the questions are similar. She pointed out that very limited information is available regarding the magnitude of the various problems associated with the implementation any of the PMTCT options, such as the implications of non-adherence. Dr. Guay presented a summary of potential implementation, operational and clinical advantages and disadvantages for A, B, and B+. She presented priority research areas that were outlined in the April 2012 WHO programmatic update, and highlighted in the Scientific Background document for the panel, particularly around Option B+, for which data are needed, including:

• Service organization/delivery of ART in MCH and primary care settings

• Cost and sustainability

• ARV adherence and retention in care

• Referral mechanisms and transitions from the PMTCT program to HIV care and treatment programs

! 12!

• HIV drug resistance with long-term use of ART initiated in early HIV disease

• Safety of increased exposure to ARVs for the fetus/infant

• Acceptability and equity

Several of these issues are being addressed in the PROMISE study, such as ARV adherence, drug resistance, safety, and acceptability across the study arms. Dr. Guay also reviewed some key issues related to the urgent need for evidence about the acceptability and comparative program advantages of Options A, B and B+, including:

• Operational advantages of providing triple ARVs to all pregnant women (Options B and B+)

• How best to meet the programmatic and implementation requirements of these approaches

• The acceptability, effectiveness and prevention impact of Option B+

Dr. Nathan Shaffer

Dr. Shaffer reviewed WHO’s process for updating guidelines for treatment and prevention of HIV infections. As part of this review, he noted that developing countries have expressed a preference that WHO guidelines recommend a single option for a given clinical situation, rather than present alternatives. Based on recent research findings, WHO has determined that the global goal of decreasing new HIV infections in young children by 90% (from 400,000 to 40,000 annually) and decreasing population-based MTCT rates to <5% in breastfeeding settings (and <2% in non-breastfeeding settings), and thus virtually eliminating MTCT globally by 2015, is achievable. One key to the realization of that goal is simplification and harmonization of guidelines across various segments of the population. One such simplification would be the elimination of the need for CD4 cell count tests to guide the initiation and interruption of ART, which would be accomplished with the widespread adoption of B+. B+ would further simply this process by providing the same ARV regimen to all pregnant HIV infected women.

In the panel’s discussion with Dr. Shaffer, it was highlighted that the PROMISE trial and other past studies on PMTCT have not evaluated the safety profile of ART regimens containing Efavirenz (EFZ), which is currently recommended in the WHO programmatic update, even though it is contraindicated during pregnancy in the U.S. This issue was identified as a potentially valuable subject for future operational research to address the current lack of

! 13!

evidence about the safety of widespread use of Efavirenz in pregnant women in their first trimester.

Dr. Shaffer emphasized that guidelines have the greatest force when based on the highest quality evidence, preferably the results of randomized clinical trials. However, with respect to the care of HIV-infected pregnant women, WHO advocates the adoption of B+ because it seems highly feasible that the expanded use of combination ART regimens could add benefit without adding serious harm. Prospective evaluation of B+ may be several years away, and delaying implementation until these evaluations are complete would limit the expected benefits. Despite WHO’s promotion of B+ for its programmatic advantages, serious questions remain about the sustainability of the efficacy and safety of the regimen.

Dr. Angela Mushavi

The panel was anxious to learn more about the process of responding to changes in WHO guidelines at the national level. Dr. Mushavi gave a thorough description of the current approach followed in Zimbabwe. There is a standing organization of committees and working groups established by the Ministry of Health and Child Welfare and made up of expert stakeholders throughout Zimbabwe, who are charged with assessing the implications of changing national guidelines. In making their recommendations, these groups consider the five touchstone principles of avoiding harm, providing equitable access to care, improving quality of care, maximizing efficient use of resources, and implementing sustainable programs. In the past, approval of national plans has taken up to 3 years, and full implementation even longer, but the more systematic approach being followed currently is expected to reduce these times considerably.

Gaps in knowledge that PROMISE aims to fill

The Scientific Background document prepared for the panel by Dr. Lynne Mofenson(5) contained a detailed description of the current evidence, and gaps in the evidence, regarding ARVs for PMTCT. The panel has drawn on this review, supplemented by our own review of primary sources, where necessary.

92% of maternal mortality and 88% of perinatal and postnatal MTCT occur in pregnant women who meet current WHO criteria for ART(CD4 count < 350), and current guidelines recommend lifelong ART for these women. For pregnant women who do not yet meet current criteria for treatment for their own health, two options for PMTCT are offered, Option A and Option B(5). The PROMISE protocol was designed to address questions restricted to this group of women—those who do not meet current eligibility criteria for ART for their own health in the countries in

! 14!

which the research is conducted—and will directly compare the full Option A versus Option B(5). Although there have been no clinical trials to compare the full Option A vs Option B, available data indicate that the two prophylaxis options have similar efficacy for PMTCT in women not eligible for treatment(5). In the current WHO guidelines for PMTCT, “more reliable data comparing Option A and Option B prophylaxis” is identified as a research priority(1, p. 94).

PROMISE therefore offers an important opportunity to build foundational knowledge in the form of high quality trial evidence about the PMTCT regimens being delivered in the majority of PMTCT programs in high burden countries.

Specific PROMISE research questions PROMISE is designed to provide evidence from randomized, controlled trials to address 3 critical questions about the efficacy and safety of ART regimens being offered to HIV-infected pregnant and postpartum women who do not yet qualify for ART as treatment for their own health, and to their infants. First, what is the optimal intervention for the prevention of antepartum and intrapartum HIV MTCT? Second, what is the optimal intervention for the prevention of postpartum transmission in breastfeeding (BF) infants? And third, what is the optimal intervention for the preservation of maternal health after the risk period for MTCT ends (either at delivery or cessation of BF)? More specifically:

i) Will an antenatal maternal triple ARV regimen be more effective than AZT/sdNVP in reducing the risk of in utero and intrapartum MTCT in women with higher CD4 count (>350) and will it be safe? (Antepartum Component)

ii) Will a postpartum maternal triple ARV regimen given for the duration of

breastfeeding (to 18 months) be more effective than infant NVP prophylaxis in reducing the risk of postnatal MTCT in infants of women with higher CD4 count (>350) and will the interventions be safe? (Postpartum Component)

iii) Will stopping a maternal triple ARV regimen given solely for PMTCT in HIV-

infected women with higher CD4 counts either at delivery or at the cessation of breastfeeding be deleterious to the mothers’ health and disease progression? (Maternal Health Component)

The B+ option of initiating ART earlier for all women irrespective of CD4 count and disease stage (i.e., treating at CD4 >350) was not formally considered for the 2010 revision of the WHO guidelines due to insufficient evidence and lack of field experience with the regimen(5). At the same time, the optimal criteria for ART initiation in resource-limited settings remain controversial and while some observational studies in higher resource settings suggest a benefit to initiating ART at higher CD4 cell counts (e.g. CD4 <500 cells/mm3), others do not(6-8).

! 15!

However, important new research indicates that initiation of ART in individuals with CD4 <550 cells/mm3 significantly reduces sexual transmission of HIV among discordant couples(2), and in Malawi, after considering the 2010 WHO recommendations for PMTCT, the country decided to adopt the B+ option for reasons of programmatic simplicity (e.g., as a way to address the widespread lack of access to CD4 testing, which is a prerequisite for successful implementation of Options A and B). Subsequently, a recent programmatic update from WHO advises countries to consider the B+ option, emphasizing these reasons of programmatic and operational simplicity.

The Option B+ approach is a third regimen for PMTCT being used in the field that has not been studied for efficacy in women with high CD4 count. At the time of writing the current WHO guidelines for PMTCT, the B+ approach was identified as a priority for research(1. p.94).

Despite the current push to the B+ option, there remain important questions about its sustainability in economic terms, in programmatic terms, and in terms of important clinical questions, such as ARV drug toxicity for mothers and fetus/child, ARV drug resistance, ART failure and the availability of alternative ARV drugs.

Variations in breastfeeding practice Due to variations in practice at different IMPAACT sites globally, not all of the PROMISE trial specific research questions are relevant at all sites of the network. Therefore, three versions of the PROMISE protocol have been developed, each containing only those components relevant to the different settings of the IMPAACT network. One version, P1077BF (BF=Breastfeeding), is designed for areas of the world where HIV-infected women are advised to breastfeed. Another version, P1077FF (FF=Formula feeding), is designed for areas of the world where HIV-infected women formula-feed but where less complex antiretroviral prophylaxis regimens to reduce mother to child transmission (MTCT) are used; this protocol addresses questions 1 and 3, above. To date, there have been no direct comparisons between stopping or continuing ART in breastfeeding populations and very little is known about the relative merits of continuing ARVs for maternal health after cessation of breastfeeding, or about HIV disease progression and associated problems, such as the evolution of ARV drug resistance. The PROMISE trial will provide unique and valuable data about these questions. Additional key clinical, public health and economic questions In addition to these main protocols, PROMISE also includes modules for intensive study of possible bone and renal toxicity, study of Hepatitis B/HIV co-infected women, ARV resistance in women and in infants who become infected, and a cost effectiveness analysis. PROMISE is expected to follow women and infants for at least 96 weeks. These additional data from a

! 16!

relevant research population will be a valuable contribution to the evidence base for PMTCT program implementation. Key programmatic questions While there are important potential programmatic advantages to the Option B+ approach, there are also many important unanswered questions and an urgent need for careful monitoring and evaluation. These questions figured prominently in the presentations to the panel and in the panel’s deliberations. The Scientific Background document prepared for the panel(5) outlines the following key questions:

• Acceptability: How acceptable to women who would not otherwise be taking ART is early treatment for the purposes of prevention of HIV transmission to partners and infants when the clinical benefit to the woman herself is possible but unconfirmed?

• Adherence: To what extent are women adherent to the regimen throughout the MTCT risk period, and do they continue to be adherent after the MTCT risk period?

• Retention: How many women remain in care and treatment after the PMTCT period?

• Service delivery: Can non-physician providers (e.g. nurses) provide ART on a large scale in maternal and child health settings?

• Logistics: What is required to assure a reliable supply chain of ART and to avoid stock-outs?

• Monitoring: What are the appropriate strategies for clinical monitoring of response to treatment (including CD4 testing and, potentially, measurement of viral load)?

• Safety: What new evidence can be provided on the safety of EFV and TDF during pregnancy and breastfeeding? The most reliable data will come from well-supported, prospective pharmaco-vigilance registries.

• Effectiveness for PMTCT: What is the impact of this intervention on MTCT rates, both early transmission, at six weeks, and overall rates, at the end of breastfeeding, and on HIV-free survival?

• Effect on the mother's health: What are the benefits to maternal health in women with high CD4 counts?

! 17!

• Effectiveness for decreasing sexual transmission: What is the amount of benefit for sero-discordant couples in terms of increased prevention of sexual transmission of HIV and at the population level if a ‘treatment as prevention’ strategy is implemented?

• Cost and cost-benefit: What are the costs and cost-benefit of this approach?

• Equity: What is the impact of this approach on the ability of the country programme to serve other adults in need of ART? (5, p. 11)

Although PROMISE will not answer all of these questions comprehensively, it will provide critical insights about many of them. It is extremely likely that any high quality evidence that can be generated about any of these questions will prove valuable for PMTCT programs in the future, regardless of which regimens are adopted. But in the short-term, since PROMISE will address some of these questions for the regimens (A and B) that are currently being implemented in the field, the findings will have immediate relevance in terms of optimizing the implementation of PMTCT.

PANEL Responses to Questions

To more accurately reflect its reasoning, the Panel has decided to respond to Question 2 before responding to Question 1. We first explain why we think PROMISE continues to have scientific and social value. We then move to the case of countries that have already adopted B+ and explain why we believe it is ethically acceptable to continue PROMISE in those countries.

Why PROMISE continues to have scientific and social value

Question 2 to the Panel

Given diverse implementation challenges and current policy developments in PMTCT programs, does the study still have sufficient value for informing clinical, policy, or program decisions now or in the future? If no, is there a design modification that would provide sufficient value to continue the study?

Response to Question 2

Yes, the panel believes that PROMISE poses important and highly relevant questions and is extremely likely to produce findings that will have value for informing clinical, policy, or program decisions now or in the future.

! 18!

Rationale

The current move from A to B/B+ regimens appears to be driven largely by perceived “programmatic and operational advantages”(3, p. 4) of the regimens. However, there remain important gaps in the evidence about the relative merits of A and B and B+. Despite the momentum in favor of B/B+ regimens, the WHO itself highlights the need for precisely the type of evidence that PROMISE is designed to produce:

“There is an urgent need to assess country experience and evidence that address the preferences among options A, B, and B+.”(3, p. 4)

We believe that PROMISE will provide evidence of comparative safety between A, B and B+ and will prove to be valuable for addressing some of the evidence gaps for future clinical, policy and program decisions, although precisely how and when this value will be realized is difficult to anticipate, given the current emphasis on the implementation of B/B+ regimens. In addition to the primary analyses planned in PROMISE, the panel also found that some of the planned secondary analyses—particularly those related to some key programmatic, operational and clinical questions occupying the field—were also likely to yield evidence that could be extremely valuable for the on-going management of program implementation, including B and B+. For example, PROMISE is designed to provide evidence about various aspects of ARV adherence and retention in care, HIV drug resistance associated with the various regimens, and safety of the increased ARV exposure for the fetus/infant.

Is it ethically acceptable to continue PROMISE in countries that have adopted B+?

Question 1 to the Panel Given the changing landscape of PMTCT guidelines, could PROMISE be conducted (or under what conditions can PROMISE be conducted) in a country that has chosen a national policy for PMTCT that recommends initiation of antiretroviral therapy for life in all HIV-infected pregnant women regardless of CD4 count (Option B+)?

Response to Question 1

Yes, the Panel believes that PROMISE could be conducted ethically in countries that have chosen B+.

! 19!

Rationale When the Panel convened on June 18 and 19 of 2012 only Malawi had officially adopted B+. But B+ is also being considered in several other countries, including Kenya, South Africa, Uganda, Zambia and Zimbabwe. Of these countries, PROMISE trials are actively enrolling patients in Malawi, South Africa, Uganda, Zambia and Zimbabwe(5, p. 12). Based on the information provided to the panel, we believe that no intervention arm within PROMISE, as currently designed, provides a level of care that is known to be inferior to B+. This conclusion appeared to be well supported in the presentations to the panel and in supporting documentation provided to the panel. For example, the WHO Programmatic Update of April 2012 states that:

“The two recommended prophylaxis options (A and B) are quite different programmatically, but were judged to be equally efficacious, if implemented appropriately [emphasis in the original], in reducing the risk of infant infections for women with CD4 counts >350 cells/mm3”(3, p. 1)

Similarly, the scientific background document provided to the panel stated that:

“The expected efficacy of Option B+ for PMTCT would be similar to that observed with option B since both provide a combination ARV regimen during pregnancy and breastfeeding.”(5, p. 7)

Although the Scientific Background document goes on to acknowledge that there may be advantages to B+ over B, there are an equal number of questions about the implementation and sustainability of B+ to warrant the on-going generation of high quality evidence about the regimens employed in the vast majority of PMTCT programs in high burden countries. Given these conclusions, and in the absence of any scientific evidence to challenge this claim, the panel found no reason why randomization in PROMISE should be viewed as unacceptable from the standpoint of participant rights or welfare.

Additional Considerations for Question 1

Informed Consent The core concern reflected in the first question to the panel is that PROMISE investigators and teams may find themselves in the position in which a country adopts B+ as the national policy, yet women within PROMISE could be randomized to a different drug regimen. Although the Panel has concluded that there is sufficient uncertainty about the relative merits of the different options to justify randomization, the very fact that some women will receive a regimen that is different from the national program regimen could cause confusion and uncertainty about the implications of the different regimens. To avoid these outcomes, there must be adequate

! 20!

assurance that the informed consent procedures implemented across the trial sites take into account a change in the national policy to B+. Women must be properly apprised of the implications of their choice to participate in the PROMISE trial, including the fact that they could be randomized to receive a regimen that is different than that endorsed by the national program, along with the risks and benefits associated with each option.

Additional Comments/Considerations There are strong ethical and scientific grounds for continuing the PROMISE studies.

A. There are no randomized clinical trials, aside from the PROMISE trials, that have addressed the comparative efficacy and safety of the different approaches to PMTCT in women who are not yet eligible for ART for their own health.

B. The WHO programmatic update is not a formal WHO guideline document and revised WHO PMTCT guidelines will not be available until 2013. Until these changes come into effect, from an ethical-regulatory point of view the proper reference guidance for all PROMISE research sites is the current country guideline. Currently, 21 of the 22 high burden countries identified in the Global Elimination Plan(4) continue to deliver Option A or Option B in their national programs. And in Malawi, the only country (and the only PROMISE country) that has formally endorsed option B+ in its National guidelines, and where recruitment into the PROMISE study is ongoing, the national policy makers have endorsed continued recruitment into PROMISE, because of the likelihood that the trials will produce valuable safety evidence.

C. Countries’ reactions toWHO’s anticipated endorsement of B+ is uncertain. Each country is likely to follow its own usual process of deciding what is best locally. In the near-term, this will likely result in a diversity of strategies among countries. And since a widespread transition to B+ would take a long time, in the meantime, additional data from the PROMISE trials could be extremely valuable for countries following their own internal review processes.

D. The Panel also recognized that the concept behind B+ is not universally endorsed. For example, in India, where PROMISE was not approved at a site in Chennai, there were significant concerns about the vulnerability of B+ to economic circumstances and government finance decisions. Also, culturally in many developing countries, in households where there are more than one person infected with HIV, the woman in the B+ arm may share her ARVs with her husband and children who may not have ART, resulting in sub-therapeutic doses for all. These vulnerabilities are not unique to Option B+ and they reinforce the need to continue the careful and deliberate comparative

! 21!

evaluations of these implementation challenges for Options A and B as well, as is done in the PROMISE trials.

Specific Recommendations In addition to our responses to the two main questions framing the mandate of the panel, we also offer several additional specific recommendations, based on our deliberations following the presentations and submissions to the panel.

Recommendation 1 PROMISE is not studying the EFV-based B+ regimen that is currently recommended for roll-out by WHO and is being supported by US government funding through PEPFAR. Hence NIAID/NIH should explore ways of gathering evidence about the EFV-based B+ regimen comparable to the lopinavir/ritonavir-based B+ evidence being generated by PROMISE.

Recommendation 2 PROMISE will have limits in terms of its ability to account for the impact of implementation differences between Options A, B and B+, or to project 'worst case' scenarios when systems fail for any Option. Therefore, NIAID/NIH, should consider supporting other operational research studies aimed at key operational and programmatic questions that are not currently planned in the PROMISE analyses.

Recommendation 3 NIAID/NIH should consider adding some qualitative sub-studies to the PROMISE trial to address key questions, such as women’s perceptions of, and preferences for, the various options, especially given that Option B+ requires women who may not necessarily be clinically eligible for treatment beyond pregnancy and breastfeeding to take treatment for the rest of their lives. Inadequate understanding of the personal and social implications of B+ could result in missed opportunities to improve the overall impact of the intervention. !

Recommendation 4 NIAID/NIH should review the PROMISE communication and dissemination strategies to ensure that they effectively communicate the goals and potential value of the trials to key stakeholders

! 22!

and optimize the ultimate impact of the findings for reducing MTCT and preserving maternal health.

! 23!

References

!

1. World Health Organization. Antiretroviral drugs for treating pregnant women and preventing HIV infections in infants: recommendations for a public health approach: 2010 version. Geneva, Switzerland: World Health Organization.!!!

2. Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011 Aug 11;365(6):493-505.!

3. World Health Organization. Programmatic update: use of antiretroviral drugs for treating pregnant women and preventing HIV infection in infants – executive summary. April 2012. World Health Organization, Geneva, Switzerland, 2012.

4. UNAIDS. Countdown to Zero: The Global Plan towards the Elimination of new HIV infections among children by 2015 and keeping their mothers alive. World Health Organization, Geneva, 2011.

5. Mofenson LM. Scientific background, WHO PMTCT Guidelines, and other issues related to PROMISE. Background document prepared for the PROMISE Expert Review Panel, June 2012.

6. Kitahata MM, Gange SJ, Abraham AG et al. Effect of early vs deferred therapy for HIV on survival. N Eng J Med 2009; 360: 1815-26.!

!

7. When to Start Consortium (Sterne JA et al). Timing of initiation of antiretroviral therapy in AIDS-free HIV-1-infected patients: a collaborative analysis of 18 HIV cohort studies. Lancet 2009; 373: 1352-63.!

!

8. Wright ST, Carr A, Woolley I, et al. CD4 cell responses to combination antiretroviral therapy in patients starting therapy at high CD4 cell counts. JAIDS 2011; 58: 72-9.!

! 24!

Appendix 1

Synposis of the PROMISE trials design

1

PROMISE (Promoting Maternal Infant Survival Everywhere) Design Pregnant/postpartum women who require treatment for their own health (current WHO recommendations: CD4 <350 cells/ul and/or advanced HIV disease WHO clinical stages 3 or 4) should initiate a combination antiretroviral (ARV) drug regimen as soon as possible and continue on this regimen for life. However, there is controversy about how to optimize prevention of mother to child transmission (PMTCT) as well as maternal/infant health in women who do not currently require treatment for their own health. Based on currently available data, the WHO recommends two options which appear to be equally effective for PMTCT – zidovudine (AZT) given during pregnancy with single-dose nevirapine (sdNVP) at onset of labor and daily infant nevirapine (NVP) if the infant is breastfeeding (BF) for up to 12 months (Option A), or maternal triple combination ARV regimen during pregnancy and continued if BF for up to 12 months, after which the maternal ARV regimen is discontinued (Option B). Each of these options has different advantages and disadvantages as well as cost. The PROMISE protocol (P1077) is a research protocol of the IMPAACT network designed to address in an integrated and comprehensive fashion four critical questions currently facing HIV-infected pregnant and postpartum women who do not yet require treatment for their own health, and their infants:

1. What is the optimal intervention for the prevention of antepartum and intrapartum HIV MTCT? 2. What is the optimal intervention for the prevention of postpartum transmission in BF infants? 3. What is the optimal intervention for the preservation of maternal health after the risk period for

MTCT ends (either at delivery or cessation of BF)? The PROMISE protocol began development in 2008; P1077HS opened to enrollment in January 2010, 1077BF opened to enrollment in March 2011, and P1077FF opened to enrollment in May 2011. The prolonged development process of PROMISE is reflective of changes in scientific research and in World Health Organization recommendations during the course of protocol development that resulted in modifications to the studies and delays in protocol opening. For example, in December 2009, WHO updated guidance on Option A and B for prevention of mother to child transmission (PMTCT), recommending new guidance for 6 weeks of infant NVP in infants of mothers receiving triple drug regimens; as a consequence, P1077BF and P1077FF were delayed to enable incorporation of this new guidance and infant regimen into the protocol document. The original PROMISE protocol contained a component  (‘Infant  Health’  component)  to  evaluate  whether  the  use of co-trimoxazole in HIV-exposed uninfected infants who were weaned prior to age 12 months would decrease the morbidity and mortality associated with early weaning. However, the new WHO guidance recommended at least 12 months of infant or maternal prophylaxis while breastfeeding, and as a consequence after the study opened it was observed that breastfeeding duration in the countries in which the study was being performed was now at least 12 months or longer. Given that few infants are now being weaned early, PROMISE is currently being revised to drop the Infant Health component of the protocol. The overall PROMISE protocol has 3 separate interventional components to address each of these questions. Due to variations in the standard of care for HIV-infected pregnant and postpartum women and their infants at different IMPAACT sites globally, not all of these questions are relevant at all sites of the network. Therefore, three versions of the PROMISE protocol have been developed, each containing only those components relevant to the different settings of the IMPAACT network.

2

One version, P1077BF (BF=Breastfeeding), is designed for areas of the world where HIV-infected women are advised to BF, and addresses all four questions. Another version, P1077FF (FF=Formula feeding), is designed for areas of the world where HIV-infected women formula-feed but less complex antiretroviral prophylaxis regimens to reduce mother to child transmission (MTCT) are used; this protocol addresses above questions 1 and 3. The third version, P1077HS (HS=HAART  Standard”), is designed for countries such as the U.S., where use of triple drug regimens during pregnancy for PMTCT as well as formula feeding is standard; this protocol addresses question 3 only. In addition to these main protocols, PROMISE also includes modules for intensive study of possible bone and renal toxicity, study of Hepatitis B/HIV co-infected women, ARV resistance in women and in infants who become infected, and a cost effectiveness analysis. PROMISE is expected to follow women and infants for at least 96 weeks. The study will provide first-line regimens to all participants either randomized to triple ARV or who require treatment during the course of follow-up, and has several second-line regimens available for those who need them.

P1077BF in BF populations, has 3 components to address each of the 3 research questions, and involves 3 sequential randomizations of women with entry CD4 >350 who are first enrolled while pregnant (or who are late presenters and enroll at delivery). If during the course of study follow-up a woman qualifies for treatment based on country guidelines, she is provided with treatment and does not undergo further randomization. Data from P1077BF and P1077FF will be combined for analysis.

Antepartum Component: The primary objectives are to compare the efficacy of maternal antepartum triple ARV prophylaxis versus antepartum AZT + sdNVP to reduce antepartum and intrapartum HIV transmission at age 1 week and to assess/compare safety and tolerability of these ARV regimens, including adverse pregnancy outcomes (e.g., stillbirths, prematurity, low birth weight and birth defects). Data from the Antepartum Component of P1077BF and P1077FF will be combined for analysis of this question.

Postpartum Component: The primary objectives are to compare the efficacy of maternal triple ARV versus infant NVP prophylaxis during BF to reduce cumulative HIV transmission from BF and to assess the safety/tolerability of these ARV regimens for mother and infant.

Maternal Health Component: The primary objective is to compare the rate of progression to AIDS-defining illness or deaths between women who received triple ARV prophylaxis during BF, and are randomized to stop or continue the regimen after cessation of BF. Data from the Maternal Health Component of P1077BF and P1077FF will be combined for analysis of this question.

3

P1077FF in formula feeding populations, has 2 components to address 2 of the research questions, and involves 2 sequential randomizations of women with entry CD4 >350 who are first enrolled while pregnant. If during the course of study follow-up a woman qualifies for treatment based on country guidelines, she is provided with treatment and does not undergo further randomization.

Antepartum Component: The primary objectives are to compare the efficacy of maternal antepartum triple ARV prophylaxis versus antepartum AZT + sdNVP to reduce antepartum and intrapartum HIV transmission 1 week (7-12 days) of age and to assess and compare the safety and tolerability of these ARV regimens, including adverse pregnancy outcomes (e.g., stillbirths, prematurity, low birth weight and birth defects).

Maternal Health Component: The primary objective is to compare the rate of progression to AIDS-defining illness or death between study arms between women who received triple ARV prophylaxis and are randomized to stop or continue the regimen after delivery of their infants.

4

P1077HS in areas where combination ARV for PMTCT is standard, addresses only the maternal health question regarding stopping or continuing the ARV regimen after delivery in women with CD4 >400 prior to starting ARV for PMTCT. If during the course of study follow-up a woman qualifies for treatment by current guidelines, she is provided with treatment and does not undergo further randomization. Data from P1077HS will be analyzed separately from P1077BF and P1077FF.

Maternal Health Component: The primary objective is to compare the rate of progression to AIDS-defining illness or death between study arms between women who received triple ARV prophylaxis and are randomized to stop or continue the regimen after delivery of their infants.

The sample sizes for each component are shown in the slides below:

5

As of April 30, 2012, 1,150 mother/infant pairs have been enrolled into 1077BF and FF: 1013 mother-infant pairs enrolled in 1077BA (antepartum - breastfeeding) 88 mother-infant pairs registered in 1077BL (late presenter - breastfeeding) 544 mother-infant pairs proceeded to 1077BP (postpartum - breastfeeding) 56 mothers proceeded to 1077BM (maternal health - breastfeeding)

6

137 mother-infant pairs enrolled in 1077FA (antepartum – formula feeding) 27 mothers proceeded to 1077FM (maternal health – formula feeding)

Current enrollment into P1077BF and FF is ~115 mother/infant pairs per month, with 12 of 14 sites

actively enrolling (2 additional sites in Tanzania and Zambia will be open soon). Depending on monthly enrollment, accrual is estimated to be completed by August 2013 (if increase enrollment to 200/month) to June 2014 (if stay at 125/month). The first interim efficacy analysis for Antepartum Component (combined analysis of Antepartum BF and FF components) is expected in November 2012.

Current enrollment into P1077HS is 610 (of planned 2000 mothers). Enrollment is open at sites in Argentina, Botswana, Brazil, China, Haiti, Peru, Thailand and the United States. Currently, 42 women per month are being enrolled, with accrual estimated to be completed January 2015. If accrual increases to 50 women per month, accrual will be completed in August 2014.

! 25!

Appendix 2

Scientific Background for PROMISE

Scientific Background, WHO PMTCT Guidelines, and Other Issues Related to PROMISE

Background

To provide background for discussion of the PROMISE protocol, this paper will review the World Health Organization (WHO) guidelines for prevention of mother to child HIV transmission (PMTCT), summarize the evidence basis for the 2010 guidelines, discuss the rationale for a newly proposed approach of life-long antiretroviral therapy (ART) for all pregnant women  (known  as  “Option B+”)  and some of the uncertainties and key questions with this approach, and discuss current state of country implementation of PMTCT guidelines.

PMTCT is a dynamic and rapidly changing field. In 2010, research developments led to revision in the WHO guidelines for antiretroviral (ARV) drug use by HIV-infected pregnant and breastfeeding women for maternal health and PMTCT in resource-limited countries (WHO 2010). The availability of new highly effective interventions to prevent perinatal and postpartum MTCT through breast milk led the UNAIDS to announce the Global Plan Towards the Elimination of New HIV Infections Among Children by 2015 and Keeping Their Mothers Alive - a bold and challenging new global goal of decreasing new HIV infections in young children by 90% (from 400,000 to 40,000 annually) and decreasing population-based MTCT rates to <5% in breastfeeding settings (and <2% in non-breastfeeding settings), and thus virtually eliminating MTCT globally by 2015 (UNAIDS 2010) (http://www.unaids.org/en/media/unaids/contentassets/documents/unaidspublication/2011/20110609_JC2137_Global-Plan-Elimination-HIV-Children_en.pdf). The revised 2010 WHO guidelines emphasized the importance of initiation of life-long combination ART for HIV-infected pregnant women with CD4 counts <350 cells/mm3or WHO stage 3 or 4 clinical disease (approximately 40-50% of all HIV-infected pregnant women) for maternal health as well as for the prevention of infant HIV infection. For women who do not yet need ART for their own health, a choice of two ARV prophylaxis options is recommended for PMTCT: antepartum zidovudine (AZT) plus intrapartum single-dose nevirapine (sdNVP) and daily infant NVP through 12 months of breastfeeding (Option A), or a maternal triple drug regimen during pregnancy and 12 months of breastfeeding (Option B).

While WHO increased its immunologic threshold for initiation of ART in HIV-infected adults from CD4 <200 to <350 cells/mm3 in 2010, the optimal criteria for ART initiation in resource-limited settings remains controversial; some observational studies in higher resource settings suggest a benefit to initiation of ART at higher CD4 cell counts (e.g., CD4 <500 cells/mm3) while others do not (Kitahata MM. NEJM 2009; Sterne JA. Lancet 2009; Wright ST. JAIDS 2011). Important new research indicates that initiation of ART in individuals with CD4 <550 cells/mm3

significantly reduces sexual transmission of HIV among discordant couples (Cohen MS. NEJM 2011). These data, along with considerations regarding program simplification, have led to a consideration of a third option, Option B+, which would be to provide life-long ART in all HIV-infected pregnant women regardless of CD4 cell count. This approach has been adopted in one country (Malawi) and is being considered in others.

2

However, in resource-limited countries there are significant limitations on the number of available ARV drugs, and the challenge in these settings is to balance the benefits of starting ART early in pregnant women with risks of ARV drug toxicity to the mother and fetus/infant in pregnancy and breastfeeding; costs and health system burden; issues related to adherence, resistance, ART failure and availability of future treatment options; and assuring treatment availability for all individuals who meet current treatment guidelines.

There is no debate regarding the need to provide ART to women who meet current WHO criteria for treatment; 92% of maternal mortality and 88% of perinatal and postnatal MTCT occur in this group of women (Kuhn L. AIDS 2010). However, there remain critical research gaps about optimal ARV  use  for  PMTCT  in  women  who  don’t  meet  current WHO ART eligibility criteria. The PROMISE protocol was designed to address questions restricted to this group of women - those who do not meet current eligibility for ART for their own health in the countries in which the research is conducted, and will directly compare the full Option A versus Option B.

Summary of and Evidence Basis for 2010 WHO PMTCT Guidelines Guidelines for the use of ARV drugs to treat pregnant women and prevent HIV infection in infants, along with the HIV and infant feeding guidelines, were last revised in 2010, in coordination with updating of the adult ART guidelines (WHO PMTCT guidelines 2010; WHO adult ARV guidelines 2010). The 2010 revision reflected new research findings at that time indicating:

the benefits of starting ARV prophylaxis for PMTCT earlier in pregnancy (Hoffman RM. JAIDS 2010);

decreased HIV transmission when mother or infant receives postpartum ARV prophylaxis during breastfeeding (Chasela CS. NEJM 2010; Kesho Bora. Lancet Infect Dis.2011); and

the benefit of ART for improved health and survival when it is started at a CD4 count <350 cells/mm3 instead of waiting for a decline to <200 cells/mm3 (Severe P. NEJM 2010). This finding resulted in a change in guidance, calling for initiation of ART in all HIV-infected  individuals  with  CD4  counts  of  ≤350  cells/mm3 (or WHO stage 3 or 4 clinical disease), including pregnant women.

The revised 2010 WHO PMTCT guidelines recommend starting pregnant women eligible for treatment (i.e. with CD4 counts  of  ≤350  cells/mm3) on lifelong ART. The recommendation calls for giving treatment priority to pregnant women as part of treatment scale-up. ART for pregnant women meeting the eligibility criteria would address the great majority of mothers and infants at risk. A study in Zambia found that 68% of the HIV-infected women studied met WHO criteria for ART, and 92% of maternal mortality and 88% of perinatal and postnatal MTCT occurred in these women (Kuhn L. AIDS 2010). A study of more than 6,000 women in nine countries, all but one in sub-Saharan Africa, found that 48% qualified for treatment under current WHO criteria (Carter RJ. JAIDS 2010). The HIV and infant feeding guidelines recommend that, when the breastfeeding mother or infant is receiving ARVs, breastfeeding  should  continue  for  the  infant’s  first 12 months and beyond, with complementary foods introduced at six months.

For women not eligible for ART for their own health (CD4 count >350 cells/mm3), the 2010 guidelines offer countries a choice of two effective prophylaxis options for PMTCT (Table 1).

3

Option A expands the previous (2006) recommended approach to prophylaxis for women not eligible for ART, which consisted of maternal zidovudine (AZT) during pregnancy, single-dose nevirapine (sdNVP) plus AZT plus lamivudine (3TC) at delivery and continued for one week postpartum. The revised 2010 recommendation extended prophylaxis with daily NVP to the infant throughout breastfeeding.

Option B introduced a new approach: a maternal triple ARV regimen (with the regimen based  on  the  country’s  recommended  first-line ART) given all HIV-infected women during pregnancy and until cessation of breastfeeding (if CD4 count >350 cells/mm3) or for  life  (if  CD4  count  ≤350  cells/mm3).

For both options prophylaxis is recommended to start as early as 14 weeks gestation. If a woman is breastfeeding, prophylaxis is recommended to continue throughout the breastfeeding period (infant or maternal antiretroviral prophylaxis with exclusive breastfeeding for the first 6 months followed by continued breastfeeding with complementary foods through at least 12 months).

Table 1. WHO 2010 Guidelines for Prophylaxis of MTCT (adapted from Table 2 of the 2010 guidelines) (Note in both options, all women with CD4 <350 cells/mm3 or WHO Stage 3/4 disease should initiate life-long ART)

Option A Maternal AZT/sdNVP + infant NVP

prophylaxis

Option B Maternal triple ARV prophylaxis

Mother

Antepartum: AZT starting as early as 14 weeks of gestation. Intrapartum*: sdNVP at onset of labour and initiation of daily AZT/3TC until 7 days postpartum. * If maternal AZT was provided for >4 weeks antenatally,  sdNVP  and  AZT/3TC  “tail”  can  be  omitted, and only AZT given in labour.

Mother

Antepartum: Triple ARV prophylaxis starting as early as 14 weeks of gestation. Regimen choice: AZT+3TC+LPV/r or AZT+3TC+ABC or AZT+3TC or TDF+3TC (or FTC) + EFV Intrapartum: Same regimen Postpartum: Same regimen continued until 1 week after exposure to breast milk has ended.

Infant

Breastfeeding infant Daily NVP from birth until 1 week after exposure to breast milk has ended (or a minimum of 4–6 weeks following birth) Infants receiving replacement feeding only Daily NVP or single-dose NVP plus daily AZT from birth until age 4–6 weeks

Infant

Daily NVP or daily AZT from birth until age 4–6 weeks (irrespective of mode of infant feeding).

Although there have been no clinical trials to compare the full Option A vs Option B, available data indicate two prophylaxis options have similar efficacy for PMTCT in women not eligible for treatment (see below). Countries were advised to choose one of the two options based on operational and programmatic considerations. The guidelines acknowledged that access to CD4 testing is necessary to determine whether a woman needs prophylaxis or lifelong treatment, and that the requirement for CD4 testing can be a barrier to successfully implementing Option A.

4

CD4 testing also is needed to determine which women should continue on lifelong treatment if started on Option B.

Brief Review of Data on Efficacy of Option A versus Option B for PMTCT While there are not comparative studies of the full antepartum-postpartum Option A vs Option B, data from studies looking at antepartum component of Option A vs B and other studies looking at postpartum component of Option A vs B are available.

The Kesho Bora trial compared maternal triple ARV prophylaxis (the Option B approach) with AZT plus sdNVP prophylaxis (the antepartum component of Option A), both starting at 28 weeks gestation or later in African women with CD4 cell counts of 200–500 cells/mm3 (deVincenzi I. Lancet Inf Dis 2011). The women randomized to the triple ARV arm continued the drugs until the early cessation of breastfeeding at six months, while those in the AZT arm did not receive any postnatal prophylaxis after one week—a departure from the Option A regimen. Transmission rates at birth, reflecting the efficacy of the interventions for preventing MTCT during pregnancy, were similar in the two arms—1.8% (95% CI 0.9–3.7%) with the triple drug regimen and 2.5% (95% CI 1.3–4.6%) with AZT plus sdNVP. Of note, for women with CD4 counts of 350–500 cells/mm3 (women who were not eligible for treatment by current standards), the transmission rates at delivery were identical—1.7% in each arm.

The two postnatal prophylaxis regimens, infant NVP (postpartum component of Option A) and maternal triple ARV prophylaxis (postpartum component of Option B) have been compared in only one trial. In the Malawi Breastfeeding and Nutrition (BAN) trial, women with baseline CD4 counts of >250 cells/mm3 were enrolled at the time of delivery, and breastfeeding mother-infant pairs were randomized to one of three arms: a control arm of intrapartum sdNVP plus one week AZT/3TC, or this control regimen plus six months of infant NVP (the postpartum component of Option A), or maternal triple ARV prophylaxis (the postpartum component of Option B) (Chasela CS. NEJM 2010). Rates of postnatal MTCT at six months were significantly lower in the infant NVP and maternal triple drug prophylaxis arms than in the control arm but were similar in the two experimental arms—2.9% (95% CI 1.9–4.4%) with triple drugs and 1.7% (1.0–2.9%) with infant NVP, although the study was not designed or powered to compare the two experimental arms. It was noted that, if maternal prophylaxis with triple ARV regimens is started in the postpartum period or late in pregnancy, it may be less effective than infant prophylaxis because it takes several weeks to months before full viral suppression in breast milk is achieved; thus, efficacy against early breast milk MTCT may be decreased in this situation, if no early infant prophylaxis is provided.

There are limited published data on implementation of the WHO 2010 guidelines and on operational evaluations that compare implementation of Options A and B. With roll-out of Option A, South Africa reported a preliminary population-based early MTCT rate of 3.5% (95% CI 2.9-4.1%) based on a 2010 national surveillance study of 6-week infants attending immunization clinics compared to an historical 22% with no ARVs (Goga IAS 2011); these data provide only an early indication of Option A effectiveness because full implementation of the Option A program was not in place until 2011. A comparison of MTCT rates before and after South Africa switched from the 2006 guidelines to the 2010 Option A guidelines found that the infection rate among infants around six weeks of age dropped from 3.4% (95% CI 2.6–5.0%) to

5

1.5% (95% CI 0.9–2.9%) – similar to that reported with triple ARV drugs (Rundare et al., CROI 2012 Ab 1003). Similar results have been reported in an intensive surveillance and quality improvement project in Kwazulu-Natal province (Horwood J Trop Med 2010). These studies confirm the effectiveness of Option A when implemented with high coverage and access to CD4 testing and ART for eligible women. In contrast, in an analysis of program implementation in Botswana between February 2009 and April 2010 (93% of infants were formula fed), MTCT rates were 5.5% (95% CI 2.6-10.6%) in the AZT/sdNVP group (without extended infant prophylaxis) and 0.4% (95% CI 0-2.2%) in the triple ART (women with CD4 <250 cells/mm3) group (Dryden-Peterson S. JAIDS 2011).

Computer modelling based on Zimbabwe suggests that Options A and B are roughly equivalent in preventing MTCT (Ciaranello AL. PLosMed 2012). The models estimated rates of MTCT among infants age 12 months under three different treatment schemes—sdNVP, Option A and Option B. All models assumed that women identified as eligible for ART were referred for treatment. At a rate of uptake of PMTCT services of 56%, the actual rate in Zimbabwe in 2009, the projected transmission rate with the sdNVP regimen was 18% compared with 14.4% with Option A and 13.4% with Option B. The level of PMTCT uptake had greater impact on MTCT rates than the choice of regimen. In a model in which CD4 assays were not available and treatment eligibility was based only on clinical assessment, at 95% uptake the projected 12-month transmission rates were 9.6% with Option A and 5.9% with Option B. In a separate model in which CD4 assays were available for all women and ART was provided for all women with CD4 <350 cells/mm3, at 95% uptake the projected 12-month transmission rates were 7.4% with Option A and 5.8% with Option B. These results demonstrate CD4 assay availability is more important to optimize the efficacy of Option A than Option B.

Benefits and Uncertainties/Risks of Option A vs Option B

There are not yet randomized clinical trial data directly comparing the full Option A and Option B and their relative benefits in terms of efficacy, safety, feasibility and cost-effectiveness. Choice among these options is complex; there are multiple objectives of ARV use in pregnancy and lactation that are sometimes conflicting. These include optimizing PMTCT, optimizing maternal and infant health, minimizing side effects to the woman and her infant, preserving future ART options for the woman (and infected infants), and maximizing overall benefit to the population given fixed and limited resources. There are advantages and disadvantages of both options from program point of view.

Option A is the less costly option in terms of initial drug cost, but it is more complex, is not harmonized with adult ART guidelines, and requires CD4 counts to differentiate women who require ART for their own health from those needing only prophylaxis. While point-of-care CD4 testing is a promising new approach, at this point, access to CD4 testing continues to be an important barrier to identifying and providing ART to women eligible for treatment. In the absence of CD4 counts, provision of Option A to all pregnant women would inappropriately provide AZT single-drug prophylaxis women with CD4 <350 cells/mm3 who require treatment.

Option B has higher initial drug costs but provides the potential opportunity to integrate PMTCT with first-line adult treatment recommendations (if non-NVP ART regimens are used). Since in

6

Option B all pregnant women receive triple ARV regimens, prophylaxis could be initiated pending (or without) CD4 assay results. Thus, with Option B all women in need of treatment would receive triple ARVs. However, CD4 assays would still be needed to make decisions regarding stopping or continuation of the regimen after MTCT risk for women who require ART for their own health. Additionally, while access to CD4 testing is not required for initiation of ART, access to testing is optimal to monitor ongoing response to therapy, as for any individual on treatment.

There is likely similar complexity in program implementation. Baseline evaluations, prescription of drugs and refills, and adherence counseling are similar. With Option A, maternal CD4 count is needed to determine whether the woman should receive ART or prophylaxis, although the regimen could be initiated pending CD4 results. With Option B, maternal CD4 count is needed to determine whether the woman should continue the triple ARV regimen after risk of MTCT has ceased, although the regimen could again be initiated pending CD4 results. Option A requires the additional prescription and administration of intrapartum sdNVP and 1 week AZT/3TC  “tail”,  which  adds  complexity.  For  Option  B,  the  choice  of  triple  ARV  regimen is complicated by NVP toxicity in women with CD4 >250 cells/mm3 and limited data on safety in pregnancy and breastfeeding of drugs other than AZT and 3TC, particularly for tenofovir (TDF) and efavirenz (EFV). In the U.S., pregnant women with CD4 count >250 cells/mm3 receive protease inhibitor (PI)-based regimens. Postpartum, infant ARV prescription is required for both options. If the triple drug regimen includes EFV and is continued postpartum, family planning must be addressed.

A New Approach - Option B+ - Lifelong ART for All Women

A number of countries are moving towards an approach of starting all pregnant women on ART, irrespective of clinical and immunologic staging, and continuing ART for life—an approach similar  to  “test  and  treat”  and  known  as  “Option  B+”.   Table 2 shows the proposed approach.

Table 2. Option B+: Lifelong ART for all pregnant women

Option B+ Maternal ART for life

Mother

Antepartum and postpartum: ART starting as early as 14 weeks of gestation and continued for life

Regimen choice (proposed to harmonize with adult ART guidelines but avoid the risk of NVP toxicity in women with high CD4 count; use a simplified, fixed-dose combination regimen):

TDF + 3TC (or FTC) + EFV (one pill, once daily regimen) AZT + 3TC + EFV

Infant

Daily NVP from birth until age 4 to 6 weeks (irrespective of mode of infant feeding)

The Option B+ approach was first proposed by Malawi, based on their concerns that they could not effectively implement either Option A or B because of limited access to CD4 testing, high

7

fertility rates—and thus the prospect of repeatedly stopping and starting ARV—and the need to integrate PMTCT more fully with the national adult ART program, based on one simplified regimen (Schouten EJ. Lancet 2011). Because of its greater operational simplicity, Option B+ circumvents these problems. While Option B+ was not formally considered in the 2010 WHO PMTCT guidelines, due to insufficient evidence on clinical benefit of ART in individuals with CD4 counts >350 cells/mm3 and lack of field experience, it was identified as a high-priority research question. A recent programmatic update from WHO has more formally endorsed consideration of this approach (WHO Program Update April 2012).

The expected efficacy of Option B+ for PMTCT would be similar to that observed with Option B, as both provide triple ARV regimens during pregnancy and breastfeeding. However, Option B+ may have several advantages over Options A and B: protection of serodiscordant partners, including protection after MTCT risk has ceased; potential clinical benefit for the women of starting ART early; avoidance of a pattern of starting, stopping and restarting ARV in women who become pregnant again soon; and potential protection of subsequent pregnancies from MTCT from conception.

Treatment as prevention (HPTN 052 and WHO couples HIV testing and counselling guidance)

The results of HPTN 052 demonstrate the strong benefit of starting ART early (i.e. in those with CD4 counts of 350 to 550 cells/mm3) to prevent heterosexual transmission in serodiscordant couples (Cohen MS. NEJM 2011). Option B+ (as well as Option B during the period of triple ARV administration) has the potential to prevent sexual transmission as well as to provide a highly effective intervention for PMTCT. WHO has provided guidance recommending couples HIV testing and counselling, with support for disclosure, and that infected partners in serodiscordant couples who do not yet require ART (CD4 >350 cells/mm3) should be offered ART to reduce sexual transmission, noting that in situations/countries with limited or inadequate resources, individuals who require ART for their own health should be given priority (WHO Couples Testing/Counselling April 2012). In ANC settings, rates of serodiscordance range from 10% to 50%. It is important to note prevention of sexual transmission depends on maintaining suppression of viral load. Adherence to the daily ART regimen is essential to maintain suppression. Therefore, in order to realize the full public health benefits of this approach there must be continuous supply of drugs without stock out and strong support for adherence.

Clinical benefit of starting ART earlier

Although not conclusive, there is evidence that starting ART earlier could have clinical benefit (Kitahata MM. NEJM 2009; Sterne JA Lancet 2009; Wright ST. JAIDS 2011; CASCADE Arch Intern Med 2011; Rhame FS. Curr Infect Dis Rep 2011). For example, in the HPTN 052 study initiation of ART when CD4 count was <550 cells/mm3 (rather than waiting until CD4 decreased to <250 cells/mm3) reduced the incidence of extrapulmonary (but not pulmonary) tuberculosis. There was no significant difference in mortality between early and deferred ART arms but median follow-up was only 1.7 years (Cohen MS. NEJM 2011). Randomized clinical trial data on the clinical benefit of ART initiation in individuals with CD4 >350 cells/mm3 are not yet available. However, the NIAID-sponsored START (Strategic Timing of Antiretroviral Treatment) Study being conducted in 35 countries is an ongoing randomized controlled trial comparing initiation of ART when CD4

8

falls to <350 cells/mm3 compared to immediate ART when CD4 is 500 cells/mm3 or above, and will provide critical data about the benefits and risks of early treatment.

Approximately 48%-68% of HIV-infected pregnant women have CD4 cell count <350 cells/mm3 when they are first seen in antenatal care (Carter RJ. JAIDS 2010; Kuhn L. AIDS 2010). For women with higher CD4 counts, the time to decline to the current treatment threshold for treatment of ≤350  cell/mm3 after delivery varies based on baseline CD4. In the Kesho Bora study, among women with CD4 cell count of 350–500 cells/mm3 at study entry in the AZT/sdNVP group, 34% dropped to a CD4 count of <350 cells/mm3 within two years postpartum (Kesho Bora Study Group. IAS 2010 Ab.ThLBB105). Similarly, in the HPTN 046 clinical trial of infant NVP to prevent transmission during breastfeeding, within one year postpartum 36.9% of women with CD4 counts between 400 and 549 cells/mm3 at delivery dropped to CD4 counts of <350 cells/mm3. In contrast, at one year postpartum only 7.4% of women with baseline CD4 count >550 cells/mm3

at delivery had dropped to <350 cells/mm3 (Fowler MG. CROI 2012 Ab 1015).

Avoiding stopping and starting ART

Studies of CD4-guided ART interruption found that interruption significantly increased the risk of opportunistic infections or death as compared with continuous ART in individuals meeting treatment criteria, largely as a consequence of lowering CD4 count and increased viral load (SMART Study Group. NEJM 2006). The relevance of these data to women with high CD4 count who receive triple ARV prophylaxis is unknown. Limited data from clinical trials in which women with higher CD4 count receive triple ARV for prophylaxis and stop following delivery or breastfeeding have not shown increased rates of disease progression compared to those receiving other prophylaxis such as AZT alone (Kesho Bora Study Group. IAS 2010 Ab.ThLBB105, Tungsiripat M JAIDS 2006).

However, many countries with HIV high burdens also have high fertility rates; women may become pregnant again soon after they stop breastfeeding. Lifelong treatment under Option B+ would avoid the stop-start pace of prophylaxis. If the woman continues taking ARVs properly, assuming continued viral suppression, they also will protect the next pregnancy starting at conception rather than waiting until the woman presents for antenatal care, which in many countries may be late in pregnancy. However, some data suggest an increase in prematurity or other adverse pregnancy outcomes in women receiving ART, particularly in women receiving ART from before conception and continued throughout pregnancy (Powis KM. JID 2011; Cohan D. CROI 2012 Abs 1027; Chen J. CROI 2012 Abs 1028, Shapiro R. CROI 2012 Abs 1031). Prematurity may be more problematic in low- than high-resource settings.

It is important to note that many pregnancies in low-resource countries are unintended (Homsy J. PLosOne 2009; King R. BMC Public Health 2011; Jhangri GS. J Fam Plan Reprod Health Care 2012). Provision of adequate family planning services is an essential component of efforts to eliminate perinatal transmission. Modelling by UNAIDS has demonstrated that provision of ART alone will not reach the goal of eliminating perinatal transmission. To achieve a 90% reduction in new paediatric infections, the analysis found, programmes must achieve 90% or higher coverage with effective ARV prophylaxis, reduce the incidence of new infections in women by 50%, meet

9

unmet need for family planning, and limit breastfeeding to 12 months (Mahy M. Sex Transm Infect 2010).

Choice of ART drug regimen

An ideal first-line regimen to provide to all HIV-infected pregnant women is not yet available. Key requirements include: low cost, available as a fixed-dose combination, safe for pregnant women and breastfeeding infants, minimal monitoring requirements, low resistance profile, and compatibility with other drugs included as part of basic care. The choice of ART regimen is more complex in women with higher CD4 counts (i.e. >350 mm3). The most common first-line drug, NVP, is not recommended for use in women with high CD4 counts because of potential increased hepatic toxicity. In well-resourced countries a protease-inhibitor-based regimen is used in pregnant women with high CD4 counts. In low-resource settings use of an EFV-based regimen is proposed. The regimen of TDF, 3TC and EFV is available as a fixed-dose combination of one tablet a day, is one of the recommended first-line regimens for adults, and has recently been recommended as the best currently available regimen for treatment optimization (WHO Tx 2.0 2011). It is also effective against the hepatitis B virus and can be used without routine laboratory monitoring.

A critical issue for any new regimen for pregnant and breastfeeding women, as well as for women who might become pregnant, is safety. To date, there are limited data on EFV and TDF use in pregnancy. Early data suggesting neural tube birth defects in primates with in utero EFV exposure and some isolated case reports and retrospective clinical data in humans have led to concern about use of EFV in women of childbearing age. Both the US Food and Drug Administration and the European Medicines Agency classify EFV as a potential teratogen, and while recognizing that the teratogenic risk is low and primarily early in pregnancy, advise against its use in pregnant women unless the potential benefits outweigh the risk (Bristol-Myers Squibb FDA EFV drug label Dec 2011, EMA EFV drug label). However, a computer simulation model projected improved survival among women initiating EFV-based instead of NVP-based ART in Cote  d’Ivoire,  with  a  small  potential  number  of  birth  defects (Ouattara EN. AIDS 2012). Evaluation of prospectively collected data on infants born to mothers taking EFV in the first trimester in humans is reassuring (Ford N. AIDS 2011). The number of first trimester EFV exposures in prospective studies remains too small, however, to allow a definitive conclusion regarding the risk of rare outcomes such as neural tube defects in women who conceive while on EFV, although it is clear that the relative risk, if any, is low. Women who desire pregnancy could be offered an alternative ART regimen while attempting to conceive and until the second trimester of pregnancy. This would increase the complexity of the Option B+ approach, however. Additionally, as noted, the majority of pregnancies in resource-limited settings are unplanned.

The optimal triple ARV regimen in pregnancy and breastfeeding and the safety of these regimens in women and infants need to be determined. The risk of toxicity or congenital defects may be affected by nutritional deficiency and other conditions more common in developing than developed countries. ART use has been associated with adverse birth outcomes in studies in Botswana and Uganda (Cohan D et al. CROI 2012 Abs 1027; Chen J et al. CROI 2012 Abs 1028, Shapiro R et al. CROI 2012 Abs 1031). With increasing use of the triple ARV regimen in pregnant and lactating

10

women, at a minimum a pharmacovigilance surveillance system needs to be in place to evaluate pregnancy outcomes and birth defects.

Adherence, acceptability, and retention in care

There is evidence that women receiving ART may require special assistance with adherence and retention in care postpartum; recent studies have shown a significant decrease in ART adherence during the postpartum period and also possibly an increased risk of loss-to-follow-up following delivery (Nachega J CROI 2012 Ab 1006, Myer L CROI 2012 Ab 22, Clouse K CROI 2012 Ab 1004). In a meta-analysis of 51 studies on ART adherence among pregnant and postpartum women in high-, mid- and low-resource countries, only 73.5% (95% CI 69-78%) of women had adequate adherence to their ART regimen (defined as >80% adherence); the point estimate of the proportion of women with >80% ART adherence was 75.7% (95% CI 72-80%) during the antepartum period and 50.3% (95% CI 33-73%) during the postpartum period (p=0.009) (Nachega J. CROI 2012). Virtually no data are published on patient acceptability of various PMTCT regimens outside of a clinical trial setting.

Loss-to-follow-up is a major obstacle to effective delivery of PMTCT programs in diverse contexts. In a study of 925 patients in a South African ART clinic, pregnant women with baseline CD4 counts <200 cells/mm3 initiating ART were six times more likely to be lost to follow-up than men (Wang B. S Afr Med J 2011). In another South African study, pregnant women were substantially more likely to be lost to follow-up than non-pregnant women for both pre-ART care and while on ART (Kaplan R. AIDS 2008).

Studies reporting on individual barriers to accessing PMTCT care cite issues such as transportation, lack of partner support, lack of awareness, stigma, and delivery outside of health care facilities (Peltzer K. BMC Public Health 2011, Peltzer K. Afr J Reprod Health 2007, Mbonye AK. Sex Trans Infect 2009, Mbonye AK. J Biosoc Sci 2010). There are no comparative data on which programmatic PMTCT approach might be more acceptable or accessible to patients.

Resource allocation considerations concerning access to ART

Given that resources for treatment remain limited and that worldwide 10 million people who require treatment still lack access to ART, public health decisions on PMTCT need to be part of broader country HIV programming (WHO Progress Report 2011). To optimally decrease MTCT and reduce maternal mortality, the highest priority should be to ensure that women who currently require ART for their own health receive it. Countries making decisions about increasing use of ART will need to consider how best to increase access for those needing ART for their own health while at the same time assuring sustainable access for those started on ART early.

Analyses of budgetary impact and cost-effectiveness that compare Options A and B (and B+) and address short- and long-term benefits and costs for both mothers and infants will be an important subjects for future research. Operational research will need to validate the cost-effectiveness of specific approaches in their settings, particularly in countries implementing Option B+.

11

Operational aspects of Option B+ Constraints on health systems and human resources will need to be addressed to accommodate an Option B+ approach. Task-shifting and possibly expansion of health staff to allow provision of ART in primary health care facilities, where the great majority of women receive AMNC care, and integration of ART and PMTCT programmes (for initiation, maintenance and retention on treatment) will be needed. Maintaining the drug supply chain and uninterrupted provision of ART to women during pregnancy and breastfeeding will be crucial. Additionally, strong support to women for continuous adherence to the regimen and retention in care and treatment after the perinatal and breastfeeding period needs to be assured.

Key questions and monitoring and evaluation needs for Option B+

While there are important potential programmatic advantages as well as health benefits to the Option B+ approach, there are also many important unanswered questions and an urgent need for careful monitoring and evaluation. Key questions include:

• Acceptability: How acceptable to women who would not otherwise be taking ART is early treatment for the purposes of prevention of HIV transmission to partners and infants when the clinical benefit to the woman herself is possible but unconfirmed?

• Adherence: To what extent are women adherent to the regimen throughout the MTCT risk period, and do they continue to be adherent after the MTCT risk period?

• Retention: How many women remain in care and treatment after the PMTCT period? • Service delivery: Can non-physician providers (e.g. nurses) provide ART on a large

scale in MCH settings? • Logistics: What is required to assure a reliable supply chain of ART and to avoid

stock-outs? • Monitoring: What are the appropriate strategies for clinical monitoring of response to

treatment (including CD4 testing and, potentially, measurement of viral load)? • Safety: What new evidence can be provided on the safety of EFV and TDF during

pregnancy and breastfeeding? The most reliable data will come from well-supported, prospective pharmacovigilance registries.

• Effectiveness for PMTCT: What is the impact of this intervention on MTCT rates (both early transmission, at six weeks, and overall rates, at the end of breastfeeding, and on HIV-free survival?

• Effect on the mother's health: What are the benefits to maternal health in women with high CD4 counts?

• Effectiveness for decreasing sexual transmission: What is the amount of benefit for serodiscordant couples in terms of increased prevention of sexual transmission of HIV and at the population level if  a  ‘treatment  as  prevention’  strategy  is  implemented?

• Cost and cost-benefit: What are the costs and cost-benefit of this approach? • Equity: What is the impact of this approach on the ability of the country programme

to serve other adults in need of ART?

12

PMTCT Implementation

After the release of the Rapid Advice and the 2010 PMTCT ARV guidelines, regions and countries moved quickly to adapt and update their guidelines. In general, in lower prevalence settings and in settings with more developed infrastructure, Option B was chosen. High-burden countries, particularly in sub-Saharan Africa, have favoured Option A, driven in large part by concerns about the increased drug cost of Option B and the lack of capacity to delivery triple ARVs in the MNCH care setting, and a desire for continuity with existing approaches. However, many of these countries are reassessing their initial choice based on programme experience and the rapidly changing context. Both low-prevalence and high-prevalence countries are moving towards Option B, and a number of countries are considering Option B+. More mature programmes (e.g. Botswana) and lower prevalence countries in West Africa have tended to choose Option B. Table 3 provides a list of 22 priority countries and national PMTCT policies as of end of 2011.

Table 3. The 22 priority countries for the Global Elimination Plan, the PMTCT national policy chosen as of the end 2011, and current considerations

Country

ANC HIV Prevalence

high (>15%) medium (5-15%)

low (<5%))

PMTCT option chosen at the End of

2011 Updates on PMTCT option, March 2012

(if changed from end of 2011) Angola Low B Botswana* High B Burundi Low B Cameroon Low A Chad Low ? Côte d'Ivoire Low B DRC Low A Ethiopia Low A Ghana Low B India* Low A Moving to B Kenya Medium A Considering B+ Lesotho High A Malawi* Medium B+ Implementing B+ Mozambique High A Namibia High A Nigeria Low A and B South Africa* High A Considering B+ Swaziland High A Tanzania* Medium A Uganda* Medium A Considering B+ Zambia* Medium A Considering B+ but funding/human

resource issues Zimbabwe* High A

* Countries with sites enrolling into PROMISE. Botswana is conducting P1077HS protocol, all other countries are conducting P1077BF and/or FF protocols.

13

Summary Option B and Option B+ have a number of benefits in addition to PMTCT that might make either of them a preferred option in some settings. These additional benefits include the following:

Implementation may be easier and more effective in areas where the availability of CD4 assays is limited.

These options assure that women who present in ANC in need of treatment for their own health receive appropriate ART without delay.

The same ARV regimen is administered to all pregnant women and continued during pregnancy, labour and postpartum.

The regimen may be able to be harmonized with those recommended for ART in non-pregnant individuals.

A  triple  ARV  regimen  may  improve  a  woman’s  health  (for  women  with  higher  CD4  counts) while she is taking it.

With good adherence and maintenance of viral suppression, the triple ARV regimens may reduce sexual transmission of HIV to sexual partners.

Despite these important advantages, which are now motivating some countries to move towards Option B or B+, there are still complex public health considerations for resource-limited countries. These include the limited availability of antiretroviral drugs; costs and health system burden; weighing the benefits of starting pregnant women on ART early against the risks of ARV drug toxicity to the mother and fetus/infant; issues related to adherence, resistance, ART failure and the availability of future treatment options; as well as assuring treatment for all individuals who meet current treatment guidelines.

Countries are in various stages of national PMTCT scale-up, have varying burdens of HIV infection and have different MCH infrastructure, resources and support from external partners. Countries with limited resources will ultimately have to determine how to best use ARVs and allocate resources for those who need treatment for their own health and to maximize the effectiveness of their PMTCT programmes.

WHO is not changing its 2010 PMTCT recommendations at this time, but with their interim program update (WHO Program Update April 2012) note that the Option B and the new Option B+ strategy may be able to link treatment and prevention on a large scale. Countries that are successfully implementing Option A and achieving their goals towards elimination were not told to plan an immediate change to Option B. However, WHO encouraged all countries, including those implementing Option A, to reassess their own situations in light of the changing understanding of benefits and uncertainties of the different options and approaches. They also noted that further data on feasibility, impact and cost-effectiveness, as well as better pharmacovigilance data to determine the safety of first-line regimens during pregnancy and breastfeeding, are needed to better inform country decision-makers.

References

14

World Health Organization. Antiretroviral drugs for treating pregnant women and preventing HIV infections in infants: recommendations for a public health approach: 2010 version. Geneva, Switzerland: World Health Organization.

UNAIDS. Countdown to zero: global plan for the elimination of new HIV infections among children by 2015 and keeping their mothers alive. 2011. World Health Organization, Geneva, Switzerland. URL:

http://www.unaids.org/en/media/unaids/contentassets/documents/unaidspublication/2011/20110609_JC2137_Global-Plan-Elimination-HIV-Children_en.pdf

Kitahata MM, Gange SJ, Abraham AG et al. Effect of early vs deferred therapy for HIV on survival. N Eng J Med 2009;360:1815-26.

When to Start Consortium (Sterne JA et al). Timing of initiation of antiretroviral therapy in AIDS-free HIV-1-infected patients: a collaborative analysis of 18 HIV cohort studies. Lancet 2009;373:1352-63.

Wright ST, Carr A, Woolley I, et al. CD4 cell responses to combination antiretroviral therapy in patients starting therapy at high CD4 cell counts. JAIDS 2011;58:72-9.

Writing Committee for the CASCADE Collaboration. Timing of HAART initiation and clinical outcomes in human immunodeficiency virus type 1 seroconverters. Arch Intern Med. 2011 Sep 26;171(17):1560-9.

Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011 Aug 11;365(6):493-505.

Kuhn L, Aldrovandi GM, Sinkala M, et al. Potential impact of new WHO criteria for antiretroviral treatment for prevention of mother-to- child HIV transmission. AIDS. 2010 Jun 1;24(9):1374-7.

World Health Organization. Antiretroviral therapy for HIV infection in adults and adolescents: recommendations for a public health approach: 2010 revision. Geneva, Switzerland, World Health Organization.

Hoffman RM, Black V, Technau K, et al. Effects of highly active antiretroviral therapy duration and regimen on risk for mother-to-child transmission of HIV in Johannesburg, South Africa. JAIDS. 2010;54:35-41.

Chasela CS, Hudgens MG, Jamieson DJ, et al. Maternal or infant antiretroviral drugs to reduce HIV-1 transmission. N Engl J Med 2010;362:2271-2281.

The Kesho Bora Study Group (de Vincenzi I et al). Triple antiretroviral compared with zidovudine and single-dose nevirapine prophylaxis during pregnancy and breastfeeding for prevention of mother-to-child transmission of HIV-1 (Kesho Bora study): a randomised controlled trial. Lancet Infect Dis.2011;11:171-180.

Severe P, Juste MA, Ambroise A, et al. Early versus standard antiretroviral therapy for HIV-infected adults in Haiti. N Engl J Med. 2010 Jul 15;363(3):257-65.

Carter RJ, Dugan K, El-Sadr WM, et al. CD4+ cell count testing is more effective than HIV disease clinical staging in identifying pregnant and postpartum women eligible for antiretroviral therapy in resource-limited settings. JAIDS 2010;55:404-10.

Ciaranello AL, Perez F, Keatinge J, et al. What Will It Take to Eliminate Pediatric HIV? Reaching WHO Target Rates of Mother-to-Child HIV Transmission in Zimbabwe: A Model-Based Analysis. PLoS Med. 2012 Jan;9(1):e1001156.

Nachega J, Uthman O, Mills E, et al. Adherence to ART during and after pregnancy in low-, middle-, and high-income countries: a systematic review and meta-analysis. 19th Conference on Retroviruses and Opportunistic Infections, Seattle, WA, March 5-8 2012 (Abstract 1006).

Myer L, Cornell M, Fox M, et al. Loss to –follow-up and mortality among pregnant and non-pregnant women initiating ART: South Africa. 19th Conference on Retroviruses and Opportunistic Infections, Seattle, WA, March 5-8 2012 (Abstract 22).

Clouse K, Maskew M, Bassett J, et al. Delayed diagnosis of HIV and high rates of loss-to-follow-up among pregnant women attending antenatal services at a primary health clinic: Johannesburg, South Africa. 19th Conference on Retroviruses and Opportunistic Infections, Seattle, WA, March 5-8 2012 (Abstract 1004).

Schouten EJ, Jahn A, Midiani D et al. Prevention of mother-to-child transmission of HIV and the health-related Millennnium Development Goals: time for a public health approach. Lancet 2011;378:282-4.

World Health Organization. Programmatic update: use of antiretroviral drugs for treating pregnant women and preventing HIV infection in infants – executive summary. April 2012. World Health Organization, Geneva, Switzerland, 2012.

15

World Health Organization. Guidance on couples HIV testing and counseling, including antiretroviral therapy for treatment and prevention in serodiscordant couples – recommendations for a public health approach. April 2012. World Health Organization, Geneva, Switzerland, 2012.

Nachega J, Uthman C, Mills E, Muessig K et al. Adherence to antiretroviral therapy (ART) during and after pregnancy in low, middle and high income countries: a systematic review and meta-analysis. 19th Conference on Retroviruses and Opportunistic Infections, Seattle, WA, March 5-8, 2012 (Abstract 1006).

Wang B, Losina E, Stark R, et al. Loss to follow-up in a community clinic in South Africa--roles of gender, pregnancy and CD4 count. S Afr Med J. 2011;101:253-7.

Kaplan R, Orrell C, Zwane E, et al. Loss to follow-up and mortality among pregnant women referred to a community clinic for antiretroviral treatment. AIDS. 2008;22:1679-81.

Rhame FS. When to start antiretroviral therapy. Curr Infect Dis Rep 2011;13:60-7 The Kesho Bora Study Group. Impact of triple-ARV prophylaxis during pregnancy and breastfeeding compared

with short-ARV prophylaxis for MTCT prevention on maternal Disease Progression. XVIII International AIDS Conference, Vienna, July 2010 (Abstract no. ThLBB105).

Fowler MG, Brown E, Maldonado Y, et al. Maternal disease progression in the first year after delivery among HIV+ African women: HPTN 046 clinical trial. 19th Conference on Retroviruses and Opportunistic Infections, Seattle, WA, March 5-8 2012 (Abstract 1015).

Strategies for Management of Antiretroviral Therapy (SMART) Study Group, El-Sadr WM, Lundgren JD, Neaton JD, et al. CD4+ count-guided interruption of antiretroviral treatment. N Engl J Med. 2006;355:2283-96.

Tungsiripat M, Drechsler H, Aberg JA. Discontinuation of antiretroviral therapy postpartum: no evidence for altered viral set point. JAIDS 2006;44:16-17.

Powis KM, Kitch D, Ogwu A, et al. Increased risk of preterm delivery among HIV-infected women randomized to protease versus nucleoside reverse transcriptase inhibitor-based HAART during pregnancy. J Infect Dis. 2011 Aug 15;204(4):506-14.

Cohen D, Young S, Murray K, et al. Maternal nutritional status predicts adverse birth outcomes among HIV+ rural women receiving cART: Uganda. 19th Conference on Retroviruses and Opportunistic Infections, Seattle, WA, March 5-8, 2012 (Abstract 1027).

Chen J, Ribaudo H, Parekh N, et al. HAART and adverse birth outcomes among HIV+ women: Botswana. 19th Conference on Retroviruses and Opportunistic Infections, Seattle, WA, March 5-8, 2012 (Abstract 1028).

Shapiro R, Souda S, Parekh N, et al. High prevalence of chronic placental hypertensive damage but no in utero HIV transmission among women on HAART who deliver stillbirths: Botswana. 19th Conference on Retroviruses and Opportunistic Infections, Seattle, WA, March 5-8, 2012 (Abstract 1031).

Homsy J, Brunnell R, Moore D, et al. Reproductive intentions and outcomes among women on antiretroviral therapy in rural Uganda: a prospective cohort study. PLosOne 2009;4:e4149

King R, Khana K, Nakayiwa S et al. 'Pregnancy comes accidentally--like it did with me': reproductive decisions among women on ART and their partners in rural Uganda. BMC Public Health. 2011 Jul 5;11:530.

Jhangri GS, Heys J, Alibhai A, Rubaale T, Kipp W.Unmet need for effective family planning in HIV-infected individuals: results from a survey in rural Uganda. Fam Plann Reprod Health Care. 2012;38:23-9.

Mahy M, Stover J, Kiragu K, et al. What will it take to achieve virtual elimination of mother-to-child transmission of HIV? An assessment of current progress and future needs. Sex Transm Infect 2010;86 (Suppl 2):ii48-55.

WHO/UNAIDS. The Treatment 2.0 framework for action: Catalyzing the next phase of treatment, care and support. 2011. World Health Organization, Geneva, Switzerland. URL: http://www.unaids.org/en/media/unaids/contentassets/documents/unaidspublication/2011/20110824_JC2208_outlook_treatment2.0_en.pdf

Bristol-Myers Squibb Company. Efavirenz FDA-approved drug label – December 19, 2011. http://www.accessdata.fda.gov/drugsatfda_docs/label/2011/020972s040lbl.pdf

European Medicines Agency. Efavirenz product information – December 20, 2011. http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000797/WC500028102.pdf

Ouattara EN, ANglaret X, Wong AY, et al. Projecting the clinical benefits and risks of using efavirenz-containing ART regimens in women of childbearing age in Sub-Saharan Africa. AIDS 2012;26:625-34.

16

Ford N, Calmy A, Mofenson L. Safety of efavirenz in the first trimester of pregnancy: an updated systematic review and meta-analysis. AIDS. 2011 Nov 28;25(18):2301-4.

World Health Organization. Progress report 2011: Global HIV/AIDS response. Epidemic update and health sector progress towards universal access. WHO, UNICEF, UNAIDS. Geneva, Switzerland. URL: http://whqlibdoc.who.int/publications/2011/9789241502986_eng.pdf

! 26!

Appendix 3

Scientific Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment

Science Roadmap through 2016 (See Appendix for timeline for trial results)

For this review, ClinicalTrials.gov was searched for any active trials on PMTCT, as well as information was gathered through personal contact with investigators who are active in the PMTCT field. There are 7 trials related to PMTCT (including PROMISE) that will have results available between 2011 and 2016 to inform guidelines.

Aluvia (Lopinavir-ritonavir) for PMTCT in Africa: Results from this observational study were presented at October 2011 ICAAC (Ngoma M et al. ICAAC 2011 Abs. H-1-1153). AZT/3TC/LPV-r was provided to HIV-infected pregnant women through 12 months of breastfeeding in Zambia. The study confirmed low transmission rates with triple drug prophylaxis: overall MTCT was 3%, with 1.6% occurring between ages 6-12 months (despite continued maternal prophylaxis). Adherence and safety data were not provided (although mean gestational age at birth was 37.9 months so may have some preterm infants), and there was no control group. These are first data on prolonged maternal triple drugs through 12 months breastfeeding.

Study

Antepartum

Intrapartum

Postpartum Mother

Postpartum Infant

Comments

LPV/r for PMTCT in Africa (NCT 01088516) Zambia Observational

study, breastfeeding (BF)

N=279 enrolled btn 12/2008-12/2009; 225 delivered 230 infants, 9 stillbirths

CD4 <350 60%

14-30 weeks gestation AZT/3TC/LPV/r

AZT/3TC/LPV/r + sdNVP

During breastfeeding AZT/3TC/LPV/r

sdNVP plus AZT x 5 d

Results ICAAC Sept 2011 Overall MTCT at 12 mos

3% - Birth, 1/215 (0.5%) - 6 wks, 2/214 (0.9%) - 6 mos (exclusive BF)

0/191 (0%) - 6-12 mos (BF +

complementary food) 3/186 (1.6%)

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment 2

Zambia Pilot Triple ARV PMTCT Program: At March 2012 CROI, results from a community-

based evaluation of a triple ARV PMTCT pilot program in 4 community clinics in Zambia were presented (Gartland M et al. 19th CROI 2012 Abs 23LB). HIV incidence and HIV-free survival were evaluated by community-based testing of infants age <2 years in 4 communities before/after implementation of a pilot triple ARV PMTCT program given to all HIV-infected women through 12 months of breastfeeding, compared to 4 communities with standard of care, in which women with CD4 <350 cells/mm3 were referred to ART clinic and women with CD4 >350 cells/mm3 received AZT/sdNVP without extended infant prophylaxis (the antepartum component of Option A). MTCT at birth was not significantly different between groups (0 infections in the triple ARV group and 1 infection in the AZT/sdNVP group). However, not surprisingly, postpartum infections through breastfeeding were significantly different, with 1 infection in the triple ARV group compared to 13 in the AZT/sdNVP group in which no extended prophylaxis of mother or infant was received. Loss to follow-up by 12 months had a trend toward being higher in the triple ARV communities than the control AZT/sdNVP communities, primarily after 6 months postpartum, although this was not statistically significant (p=0.11).

Study

Antepartum

Intrapartum

Postpartum Mother

Postpartum Infant

Comments

Zambia Pilot Program (NCT 00753324) Zambia Observational, BF 4 clinics will

provide triple ARV for PMTCT, N=160

“Before-after”  community survey to evaluate HIV-free survival based on maternal-infant HIV testing

28 weeks gestation Pilot clinics AZT/3TC/NVP or LPV/r (depending on CD4 Control clinics (“active  comparator”): AZT

Pilot clinics AZT/3TC/NVP or LPV/r Control clinics AZT/sdNVP

Breastfeeding to 12 mos Pilot clinics AZT/3TC/NVP or LPV/r Control clinics AZT/3TC tail x1 wk?

Pilot clinics sdNVP + AZT x 1 wk Control clinics sdNVP plus AZT x 1wk – no extended prophylaxis

The “active  comparator”  communities where AZT/ sdNVP are standard will be compared but there is no extended infant prophylaxis

In utero infection not different: - 0 with triple ARV - 1 (0.7%) with

AZT/sdNVP Postpartum infections

occurred primarily in the control group without extended prophylaxis; by 12 months: - 1 with triple ARV - 13 with

AZT/sdNVP without extended prophylaxis\

There was a trend toward higher loss-to-FU in the triple ARV communities. By 12 months: - 25 in triple ARV

lost FU - 18 in AZT/sdNVP

lost FU

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment 3

ANRS 12174: In 2013, results from the infant post-exposure prophylaxis (PEP) ANRS 12174

randomized trial will be available. In this 4 country African study, mothers who are not eligible for ART receive the national program PMTCT prophylaxis regimen, their infants receive sdNVP, and then infants are randomized to daily 3TC or LPV/r for up to 12 months of breastfeeding. This will be the first study to provide data on the efficacy and safety of prolonged infant prophylaxis for 12 months as currently recommended by WHO (only 6 month regimens have been studied to date).

Study

Antepartum

Intrapartum

Postpartum Mother

Postpartum Infant

Comments

ANRS 1217`4-PEP (NCT00640263) Burkina Faso,

South Africa, Uganda, Zambia

Randomized trial, BF

Comparing infant prophylaxis with LPV/r vs 3TC during BF for 50 wks

N=1,500

Mother not eligible for ART - receives national PMTCT program

National PMTCT program

National PMTCT program

sdNVP at birth to all infants, then randomized: Arm 1: 3TC from day 7 to 50 wks (1 wk post cessation BF) Arm 2: LPV/r from day 7 to 50 wks (1 wk post cessation BF)

Endpoint: MTCT between age 7 d and 50 wks (in infants uninfected at birth)

2/3 enrolled; enrollment will close 4/30/2012

Last child completes 12 mo follow-up 4/2013

Results available by mid-2013

Interim analysis planned in Jan-Feb 2012

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment 4

PHPT-5: In 2013, results from PHPT-5 will be available. PHPT-5 originally was a 3 arm study

comparing AZT plus maternal/infant sdNVP, AZT plus infant sdNVP only, and AZT/LPV/r. When Thailand changed PMTCT guidelines to triple ARV prophylaxis, these arms were stopped – MTCT rates were not significantly different: AZT + mom/baby sdNVP, 3.6% (95% CI 1-8%); AZT+ infant sdNVP, 1.6% (95% CI 0.2-6%); AZT/LPV/r, 1.4% (955 CI 0.2-5%). Shorter duration of antepartum ARV was associated with increased risk of MTCT; the study was redesigned to a single arm to evaluate an intensification regimen when women receive <8 weeks of antepartum triple ARV; in addition to the triple ARV, mothers receive intrapartum sdNVP, and infants receive intensified prophylaxis. This will address optimizing PMTCT in late presenters in formula feeding population.

Study

Antepartum

Intrapartum

Postpartum Mother

Postpartum Infant

Comments

PHPT-5 (NCT00409591) Thailand Observational,

single arm, formula feeding

Evaluating “intensification”  if receive <8 wk anetpartum ART to reduce intrapartum MTCT

N=410

AZT/3TC/LPV/r (current standard of care)

If <8 wks ART before delivery: AZT/3TC/LPV/r plus sdNVP If >8 wks ART before delivery (observational) AZT/3TC/LPV/r

If <8 wks ART before delivery: AZT/3TC/LPV/r x 1 mo; CD4 >350 stop, <350 continue If >8 wks ART before delivery (observational) Postpartum, CD4 >350 stop, <350 continue

If <8 wks ART before delivery: AZT/3TC/NVP x 2 wk, then AZT/3TC for 2 additional weeks If >8 wks ART before delivery (observational) AZT x 4 wk

3 interim analyses Stopping rules: – Futility: >80%

probability no decrease intrapartum MTCT;

– Efficacy: >80% probability >50% efficacy (hypothesized 0.8% with intensified triple ARV vs 2.6% with historical triple ARV alone)

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment 5

PROMOTE (Prevention of Malaria and HIV Disease in Tororo): Also in 2013, results from the

PROMOTE trial will be available. This study randomizes HIV-infected pregnant women living in a malaria endemic area to a triple ARV regimen that is either PI-based (AZT/3TC/LPV/r) or NNRTI-based (AZT/3TC/EFV) during pregnancy and breastfeeding up to 12 months. The primary objective is to evaluate malaria incidence in the women by study regimen. While not primary endpoints, the study will also evaluate safety of the two triple ARV regimens and MTCT rates.

Study

Antepartum

Intrapartum

Postpartum Mother

Postpartum Infant

Comments

PROMOTE (NCT00993031) Uganda Randomized trial,

BF PI vs NNRTI

triple ARV and malaria incidence in pregnant/ postpartum women

N=500

12-28 weeks gestation Arm 1: AZT/3TC/LPV/r Arm 2: AZT/3TC/EFV

Arm 1: AZT/3TC/LPV/r Arm 2: AZT/3TC/EFV

Through 12 mos BF Arm 1: AZT/3TC/LPV/r Arm 2: AZT/3TC/EFV

All infants receive daily NVP from birth to 6 weeks

Primary endpoints are malaria but will evaluate HIV MTCT, mom CD4 and RNA, and toxicity/ pregnancy outcome

Enrolling: 270 as of 9/2011

First interim analysis 2012; final analysis 2013

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment 6

Safety Study of TDF-Containing Triple Regimen for PMTCT of HIV and Hepatitis B (HBV)

(TiP): In 2014, the TiP trial results from China will be available. This study will enroll 80 pregnant HIV/HBV co-infected women, who are randomized to 2 different triple ARV regimens: AZT/3TC/LPV/r or TDF/FTC/LPV/r. The primary endpoint of the study will be safety of TDF in the 40 pregnant women who receive it and in their infants; secondary endpoints includes MTCT of HIV and HBV and safety of discontinuation of triple ARV in HIV/HBV co-infected women postpartum.

Study

Antepartum

Intrapartum

Postpartum Mother

Postpartum Infant

Comments

TiP (NCT 01125696) China Randomized trial,

formula feeding Comparing

AZT/3TC vs TDF/3TC dual NRTI backbone in triple ARV PMTCT regimen in HIV/HBV coinfected women

N= 80 women with HIV/HBV coinfection,

14-28 weeks gestation Arm 1: TDF/3TC/LPV/r Arm 2: AZT/3TC/LPV/r

Arm 1: TDF/3TC/LPV/r Arm 2: AZT/3TC/LPV/r

If CD4 <350 continue ARV regimen; if >350, stop ARV regimen

All infants receive AZT x 6 wks

Primary endpoint safety (DXA, stopping postpartum); PK TDF

Secondary: 12 mo HIV and HBV MTCT; HBV viral load; ARV drug resistance (HIV, HBV)

Opened to enrollment 9/2011

F/U through 12 mos postpartum

Estimated completion 3/2014

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment 7

PROMISE/IMPAACT P1077 (Promoting Maternal Infant Survival Everywhere):

PROMISE/IMPAACT P1077 is designed to address in an integrated fashion 3 critical questions facing HIV-infected pregnant and postpartum women who do not yet require treatment for their own health, and their infants: 1) Will an antenatal maternal triple ARV regimen be more effective than AZT/sdNVP in reducing the risk of in utero

and intrapartum MTCT in women with higher CD4 count (>350) and will it be safe? (Antepartum Component) 2) Will a postpartum maternal triple ARV regimen given for the duration of breastfeeding (to 18 months) be more

effective than infant NVP prophylaxis in reducing the risk of postnatal MTCT in infants of women with higher CD4 count (>350) and will the interventions be safe? (Postpartum Component)

3) Will stopping a maternal triple ARV regimen given solely for PMTCT in HIV-infected women with higher CD4 counts either at delivery or at the cessation of breastfeeding be  deleterious  to  the  mothers’  health  and  disease  progression? (Maternal Health Component)

The overall PROMISE protocol has 3 separate sequential interventional components to address each of these four questions (there was a 4th component looking at co-trimoxazole in uninfected infants after weaning, but this component has been discontinued by the team; no patients had enrolled). Due to variations in the standard of care for HIV-infected pregnant and postpartum women and their infants at different IMPAACT trials sites globally, not all of these questions are relevant at all sites of the network. Therefore, three versions of the PROMISE protocol have been developed, each containing only those components relevant to the different settings of the IMPAACT network. One version, P1077BF, is designed for areas of the world where HIV-infected women are advised to breastfeed, and addresses all 3 questions with 3 sequential mother/infant randomizations. Another version, P1077FF, is designed for low-resource settings where HIV-infected women formula-feed, and addresses questions 1 and 3 with 2 sequential randomizations. The third version, P1077HS, is designed for countries where triple combination drug regimen use during pregnancy for PMTCT as well as formula feeding is standard; this protocol addresses question 3 only.

In addition to these main protocols, PROMISE also includes modules for intensive study of potential bone and renal toxicity of TDF in women/infants; evaluation of ARV resistance in women and infants who become infected; sub-study of HBV/HIV co-infected women; and a cost effectiveness analysis. PROMISE will follow all women and infants for at least 96 weeks. The study provides first-line regimens to all participants randomized to triple ARV or who require treatment during the course of follow-up, and has several second-line regimens available for those who need them.

Study

Antepartum

Intrapartum Postpartum Mother

Postpartum Infant

Comments

PROMISE/P1077 (NCT01061151) Multi-component

sequential randomized trial, BF and formula

P1077BF/FF: India, Malawi, Uganda, S Africa, Tanzania, Zambia, Zimbabwe; N=4,650

P1077HS: US, Argentina, Brazil, Peru, Haiti, China, Thailand, Botswana; N=2,000

>14-28 weeks gestation

1. Antepartum Component*: AZT/3TC/LPV/r vs AZT/sdNVP + 1 week TDF/FTC  “tail” for reducing in utero and intrapartum MTCT

(*in version 1, HBV+ women randomized to triple ARV arm are randomized to AZT/ 3TC/LPV/r vs TDF/ FTC/LPV/r; in version 2.0 of protocol, all women randomized to triple ARV will be randomized between the two regimens

2. Postpartum Component: Maternal TDF/FTC/LPV/r vs infant NVP for the duration of BF (to 18 mos) for reducing postnatal MTCT

3. Maternal Health Component: Stop vs continue maternal triple ARVs in women with high CD4 counts after MTCT risk ceases - effect on maternal progression

P1077BF/FF enrolling ~115/mo; as of 5/3/12, 1160 (of 4400) enrolled - 1st efficacy analysis for Antepartum/ Postpartum in 2012; final in 2014 - 1st efficacy analysis for Maternal Health in 2013, final in 2015 (Infant)/ 2016 (Maternal)

P1077HS enrolling ~40/mo; as of 5/3/12, 661 (of 2000) enrolled - 1st efficacy analysis in 2014, final in 2016

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment 8

Appendix: Timeline for Results of Ongoing Studies on Prevention of Mother to Child HIV Transmission

PMTCT study 2011 2012 2013 2014 2015 2016

Triple ARV with LPV, Zambia (observational)

X

Zambia 4-clinic pilot (observational community survey)

X

ANRS 12174 (PEP) Infant prophylaxis options ( 3TC vs LPV/r x 50 wk)

X

PHPT-5 (single  arm  “intensification”) X

TiP (HIV/HBV infected pregnant women randomized AZT/3TC vs TDF/FTC triple ARV)

X

PROMOTE (PI vs NNRTI triple ARV for malaria in pregnant women) (analyses based on malaria hypotheses)

* ** X

Final analysis

PROMISE 1077BF/FF (randomized) (assumptions based on accrual increase as planned and events rates as projected in protocol or higher).

Antepartum AZT /sdNVP vs triple drugs

* ** X Final analysis

Postpartum Infant NVP vs maternal triple drugs through 18 mos of BF

* ** *** X

Final analysis

Maternal health Stop vs continue after breastfeeding (or at delivery if FF)

* ** *** X Final analysis

Infant health CTX vs placebo to 18 mos if uninfected and weaned <12 mo

* ** X

Final analysis

PROMISE 1077HS (randomized to stop vs continue triple ARV regimen after delivery)

* ** X

Final analysis

*First interim efficacy DSMB ** Second interim efficacy DSMB *** Third interim efficacy DSMB

Science Roadmap on Antiretroviral Drugs for PMTCT and Maternal Treatment 9

! 27!

Appendix 4

PROMISE protocol modifications

Draft Summary of Changes to Page 1 of 3 15 May 2012 IMPAACT 1077BF/FF/HS

SUMMARY OF CHANGES INCLUDED IN THE FULL PROTOCOL AMENDMENTS OF:

IMPAACT 1077BF/FF/HS

PROMISE Study (Promoting Maternal and Infant Survival Everywhere)

Summary of Major Revisions and Rationale for Changes made across all versions of PROMISE (BF, FF, and HS): 1) Reference to the results of the HPTN 052 study, which demonstrated the benefit of HAART in

reducing transmission of HIV in discordant couples, is added to the body of the protocol and the sample informed consent form for screening and enrollment. Also added is reference to current US DHHS guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents (updated in March 2012) and new WHO guidance on couples HIV testing and counseling including antiretroviral therapy for treatment and prevention in serodiscordant couples (issued in April 2012).

2) Schedules of Evaluation: To reduce participant burden and operational complexity, the overall

frequency of study visits, the administration of questionnaires, and the collection, testing and storage of laboratory specimens specified in the Schedules of Evaluation (SoE) are decreased in all three of the study components. These changes do not impact the study objectives in any way. If a site clinician determines that additional evaluations (e.g., liver function tests) are needed for clinical management, additional testing may be performed. In addition, minor clarifications regarding assessment timing have been incorporated in the footnotes where relevant.

3) The sample informed consent form for women who become pregnant while on study drug is updated

in its entirety to make the form more specific to the decision to continue study-supplied study drugs during pregnancies that may occur during study follow-up. Information on individual antiretrovirals that may not be applicable to all participants at all sites is removed and information on WHO recommendations for use of anti-retrovirals during pregnancy is added to complement information on US Public Health Service recommendations.

Important BF and FF specific changes are: 4) Because use of TDF-containing triple ARV regimens during pregnancy is becoming increasingly

common in many countries, including those in which the PROMISE studies are being conducted, the Antepartum Component (1077BA and 1077FA) has been modified to include an additional antepartum triple ARV regimen for women without Hepatitis B (HBV) co-infection. ALL women will now to be randomized in a 1:1:1 ratio to receive either the study’s single ARV antepartum regimen (ZDV + sdNVP + TRV tail) or one of 2 triple ARV regimen receiving either 3TC-ZDV/LPV-RTV, or TDF-FTC/LPV-RTV. Both 3TC-ZDV and TDF-FTC are recommended for use in pregnant women per current WHO guidelines, therefore this modification will ensure greater scientific and programmatic relevance of the study findings by providing much needed data on safety, tolerability, and adherence for both regimens, in addition to meeting the (unchanged) primary objectives of comparing the efficacy and safety of single versus triple ARV antepartum regimens for PMTCT. Reflective of this change, IMPAACT P1084s (the Tenofovir Safety Sub study described in protocol Section 1.4) will expand antepartum enrollment to HBV-uninfected women at sites with DXA scan capabilities, further maximizing the study’s assessment of safety and broadening the generalizability of findings to HBV-uninfected women as well as HBV-infected women.

Draft Summary of Changes to Page 2 of 3 15 May 2012 IMPAACT 1077BF/FF/HS

5) The Infant Health Component (1077BI) has been eliminated from the study. After careful consideration, the protocol team determined that conduct of this study component of the study is no longer feasible, due to the lengthening duration of breastfeeding at many study sites. The team also determined that this component is unlikely to yield sufficiently compelling data, beyond that expected from other ongoing studies with similar objectives, to justify randomization of participants and conduct of additional research procedures. As such, this component, as currently described in the protocol, will not be conducted. Descriptions of and references to this component and the planned analyses have been deleted throughout the protocol as have the associated schedule of evaluations and informed consent form.

6) The instructions in Section 4.4 regarding counseling on HIV and infant feeding were updated for

consistency with the World Health Organization’s 2010 Guidelines on HIV and Infant Feeding.

7) The Antepartum, Postpartum and Maternal Health Components all include a section on management of women who are HBV co-infected (Sections 2.6136, 4.726, and 5.526, respectively); in each of these, the stated requirement for plasma storage has been modified for consistency with the relevant schedules of evaluation.

8) The estimated prevalence of HBV co-infected women enrolled in the study has been updated from

approximately 10% to 3-5%. This estimate reflects data from the first 900 women enrolled in the Antepartum Component of the study. Consequently, the total number of HBV co-infected mothers estimated to be included in the Hepatitis B sub study has been reduced to 130–220 mothers and their infants (from the original estimate of 440 mothers and their infants). The reduced number of mother-infant pairs will provide sufficient statistical power for the primary study comparisons. This change has been incorporated where relevant throughout the protocol.

9) The specifications for use of study-supplied ARVs for infant prophylaxis have been expanded to

permit use of study-supplied NVP (or 3TC) after six weeks of age among breastfeeding infants who are ineligible for the Postpartum Component and whose mothers are not otherwise on a triple ARV regimen, if prophylaxis is not otherwise available locally as standard of care (outside of the study) (Section 2.612).

10) An investigation of the use of hair as a biomarker of maternal and infant adherence and exposure to

ARVs has been added within the Postpartum Component. With the known limitations of self-reported drug adherence data, objective biomarkers are critical for monitoring ARV use, particularly as prophylaxis. However, collection, processing and testing of the typically necessary specimens is expensive and burdensome in many settings. Adherence and exposure to lopinavir/ritonavir (LPV/r) for women and their infants randomized to 1077BP Step 1 Arm A and to nevirapine (NVP) for infants randomized to 1077BP Step 1 Arm B will be assessed by measuring hair drug concentrations of the relevant antiretrovirals in a longitudinal fashion on selected samples. At the end of the study, the relationship between hair drug concentration levels and infant HIV acquisition, secondary outcomes (maternal or infant toxicities), and plasma drug concentration levels will also be assessed. Participation in this investigation is optional for sites and for participants and requires collection of hair samples from mothers and their infants while receiving ARVs. No additional study visits associated with this investigation are required. Protocol Sections 1.0, 4.0, and 7.0, Postpartum Enrollment Consent Form and Appendix IA have been modified to allow for its inclusion and for collection of hair samples.

Draft Summary of Changes to Page 3 of 3 15 May 2012 IMPAACT 1077BF/FF/HS

11) An investigation of the maternal immune correlates of protection against breast milk transmission of HIV has been added, as these have not been previously defined and yet would critically inform development of immunologic interventions to further reduce breast milk transmission of HIV. Such interventions are important, as breast milk transmission continues to occur in the setting of antiretroviral prophylaxis and because large-scale implementation of maternal antiretroviral therapy during lactation may be hindered by toxicity, non-compliance, development of drug resistant mutations as well as operational challenges. The primary objective of this investigation is to compare the HIV-specific heterologous and autologous neutralizing and binding antibody responses in plasma and breast milk of postnatal transmitting and non-transmitting mothers at the time of postnatal HIV transmission. Approximately 40 mother-infant pairs with breast milk transmission and 80 non-transmitting mother-infant pairs are targeted for these analyses.

HS-specific changes are: 12) The timeframes for participant recruitment, screening, and enrollment are clarified and expanded for

consistency and operational flexibility: • Potential study participants may be recruited during pregnancy or after pregnancy outcome

(modified from during pregnancy or after delivery/postpartum). • Informed consent may be obtained during pregnancy or after pregnancy outcome (modified from

during pregnancy or after delivery/postpartum). • Study-specific screening procedures may be initiated, after informed consent has been obtained, in

the third trimester or after pregnancy outcome (modified from 36 weeks gestation or after delivery/postpartum).

• Enrollment may occur within 42 days after pregnancy outcome (modified from within 28 days postpartum).

13) The eligibility criteria for entry into the study (Step 1) are modified to permit broader representation of the target study population: • At all sites, women who have reached 18 years of age or the minimum age of independent

informed consent will be considered eligible for inclusion; additionally, at sites that obtain institutional review board approval to enroll younger participants, women age 16-17 years will be considered eligible for inclusion with the consent of their parent or legal guardian (criterion 4.1.1; modified from previous requirement to have reached 18 years of age or minimum age of independent consent, whichever was greater).

• Women with a CD4+ cell count of at least 400 cells/mm3 or at least 50 cells above the country threshold for initiation of therapy within 120 days prior to initiation of HAART in the current pregnancy will be considered eligible for inclusion (criterion 4.1.7; modified from at least 400 cells/ mm3 and from 60 days to 120 days).

14) To expand the antiretroviral options available to enrolled women, tenofovir disoproxil

fumarate/emtricitabine/rilpivirine (Complera®) has been added to the list of study drugs, in the body of the protocol (for all three study components) and in the sample informed consent forms. Ritonavir capsule formulation has been removed.

! 28!

Appendix 5

WHO Programmatic Update 2012

hiv/aids Programme

Use of AntiretrovirAl DrUgs for treAting PregnAnt Women AnD Preventing Hiv infection in infAnts

EXECUTivE sUMMaRY

April 2012

PRogRaMMaTiC UPdaTE

1

EXECUTIVE SUMMARY

Recent developments suggest that substantial clinical and programmatic advantages can come from adopting a single, universal regimen both to treat HIV-infected pregnant women and to prevent mother-to-child transmission of HIV. This streamlining should maximize PMTCT programme performance through better alignment and linkages with antiretroviral therapy (ART) programmes at every level of service delivery. One of WHO’s two currently recommended PMTCT antiretroviral (ARV) programme options, Option B, takes this unified approach.

Now a new, third option (Option B+) proposes further evolution—not only providing the same triple ARV drugs to all HIV-infected pregnant women beginning in the antenatal clinic setting but also continuing this therapy for all of these women for life. Important advantages of Option B+ include: further simplification of regimen and service delivery and harmonization with ART programmes, protection against mother-to-child transmission in future pregnancies, a continuing prevention benefit against sexual transmission to serodiscordant partners, and avoiding stopping and starting of ARV drugs. While these benefits need to be evaluated in programme settings, and systems and support requirements need careful consideration, this is an appropriate time for countries to start assessing their situation and experience to make optimal programmatic choices.

This programmatic update is meant to provide a current perspective for countries on the impor-tant changes and new considerations arising since publication of WHO’s PMTCT ARV guide-lines, 2010 version, especially as a number of countries are now preparing to adopt Option B+. WHO has begun a comprehensive revision of all ARV guidelines, including guidance on ARVs for pregnant women, planned for release in early 2013.

Prevention of mother-to-child transmission of HIV (PMTCT) is a dynamic and rapidly changing field. Current World Health Organization (WHO) PMTCT antiretroviral (ARV) guidelines on treating pregnant women and preventing infection in infants (1), issued in 2010, were a major step towards more effica-cious regimens. The WHO guidelines emphasize the impor-tance of providing lifelong antiretroviral therapy (ART) to all HIV-infected pregnant women eligible for such treatment and recommend two short-term antiretroviral prophylaxis options (Option A and Option B) for women not eligible under current criteria, as determined by CD4 count, for treatment for their own health (Table 1). Recently, a third option, to provide life-long ART to all HIV-infected pregnant women, regardless of CD4 cell count, has emerged (Option B+), and a number of countries are already adopting or considering this approach.

Although many low- and middle-income countries are still in early stages of implementing the 2010 guidance, new evidence and recent experience warrant a programmatic update to reassess preferences between Options A and B for prophylaxis in HIV-infected pregnant women who do not need treatment for their own health and to weigh the poten-tial advantages and considerations of the new Option B+ approach in a public health perspective.

Current WHO guidance on ARV use in HIV-infected pregnant women

The 2010 WHO PMTCT ARV guidelines are based on the need to distinguish between treatment and prophylaxis. Consistent with the 2010 WHO adult ART guidelines (2), they recommend and prioritize starting all women with CD4 counts d350 cells/mm3 or WHO Stage 3 or 4 disease (approximately 40–50% of all HIV-infected pregnant women) on ART for life for their own health as well as for the prevention of infant HIV infection. For women with CD4 counts >350 cells/mm3, who are not eligible for treatment according to current criteria, the PMTCT ARV guidelines recommend starting ARV prophylaxis early in preg-nancy and, in breastfeeding settings, providing extended ARVs to either the mother or child during the postpartum risk period.

The two recommended prophylaxis options, A and B, are quite different programmatically but were judged to be equally effi-cacious, if implemented appropriately, in reducing the risk of infant infections for women with CD4 counts >350 cells/mm3. Because of the difference in the prophylaxis options, it is sometimes not well understood that Options A and B include both treatment and prophylaxis components, as shown in Table 1. The overall effectiveness, both for the mother’s health

2

and for preventing new infant infections, of implementing either of the options depends on providing both ARV treat-ment to those with low CD4 counts and prophylaxis to those with higher CD4 counts. Countries were asked to weigh the benefits and uncertainties of the two approaches, particularly the operational issues, in order to determine the best approach for their national programme.

Rationale for this update

In the short time since the 2010 PMTCT ARV guidelines were developed, the context and expectations for PMTCT programmes have changed considerably. Major changes include:

� the ambitious goals for eliminating paediatric HIV infec-tion of the new Global Plan Towards the Elimination of New HIV Infections Among Children by 2015 and Keeping Their Mothers Alive (3), together with substantial progress in the global scale-up of PMTCT and ART coverage (4);

� new evidence to support ARV treatment as HIV preven-tion—notably that provision of ART to HIV-infected indi-viduals with higher CD4 cell counts, who are not eligible for treatment, significantly reduces sexual transmission to a serodiscordant (uninfected) partner (5); this evidence has led to new WHO recommendations on couples

counselling and treatment for serodiscordant couples regardless of CD4 count (6);

� increasing country experience with operational and programme implementation challenges with both Option A and Option B;

� the proposal by some countries to move to the new Option B+ approach of lifelong ART for PMTCT for all HIV-infected pregnant women, rather than stopping ARVs for women not eligible for treatment, as in both Option A and Option B (7);

� the launch of the Treatment 2.0 Initiative to simplify and optimize the use of ARVs and standardize the first-line treatment regimen (8,9);

� reassuring data on the safety of efavirenz in pregnancy (10); and

� the decreasing cost of ARV drugs (11,12).

In addition, concerns have been raised that WHO’s recommen-dation of two different options for PMTCT prophylaxis for HIV-infected women who do not require treatment for their own health might be confusing and should be reconsidered in light of newly recognized potential benefits, operational experiences and the programme requirements of the various options.

Table 1. Three options for PMTCT programmes

Woman receives:

Infant receives:

Treatment (for CD4 count d350 cells/mm3)

Prophylaxis (for CD4 count

>350 cells/mm3)

Option Aa Triple ARVs starting as soon as diagnosed, continued for life

Antepartum: AZT starting as early as 14 weeks gestation

Intrapartum: at onset of labour, sdNVP and first dose of AZT/3TC

Postpartum: daily AZT/3TC through 7 days postpartum

Daily NVP from birth through 1 week beyond complete cessation of breastfeeding; or, if not breastfeeding or if mother is on treatment, through age 4–6 weeks

Option Ba Same initial ARVs for bothb: Daily NVP or AZT from birth through age 4–6 weeks regardless of infant feeding method

Triple ARVs starting as soon as diagnosed, continued for life

Triple ARVs starting as early as 14 weeks gestation and continued intrapartum and through childbirth if not breastfeeding or until 1!week after cessation of all breastfeeding

Option B+ Same for treatment and prophylaxisb: Daily NVP or AZT from birth through age 4–6 weeks regardless of infant feeding method

Regardless of CD4 count, triple ARVs starting as soon as diagnosed,c continued for life

Note: “Triple ARVs” refers to the use of one of the recommended 3-drug fully suppressive treatment options. a Recommended in WHO 2010 PMTCT guidelinesb True only for EFV-based first-line ART; NVP-based ART not recommended for prophylaxis (CD4 >350)c Formal recommendations for Option B+ have not been made, but presumably ART would start at diagnosis.

3

This programmatic update, while not presenting new guide-lines, reviews the currently recommended Options A and B, discusses the rationale for Option B+, and provides an update from WHO indicating and weighing preferences as much as possible among the range of options. This update summarizes key issues that need to be addressed in field settings and in national programmes. It also highlights evidence gaps that need to be addressed to build a base for future revision of guidelines.

Key findings

This programmatic update indicates that Options B and specifically B+ are likely to prove preferable to Option A for operational, programmatic and strategic reasons. While Option A has been successfully implemented in a number of high-burden countries, generally it has been difficult to implement in many low-resource settings due to the changes in drugs delivered across the care continuum (antenatal, delivery and postpartum) and the requirement for timely CD4 testing to determine which women should initiate ART for their own health. In contrast, Option B and Option B+ start all HIV-infected pregnant women on triple ARV regimens without need for an initial CD4 cell count (although CD4 testing is still needed in Option B and desirable in Option B+). Thus, Options B and B+ provide greater assurance that women in need of treatment receive a fully suppressive triple ARV regimen, to minimize the risks of infant infection and maximize the benefit to their own health, and avoid inadvertently receiving a subop-timal ARV prophylaxis intervention, particularly in settings with limited access to CD4 testing. Limited access to timely, reli-able CD4 testing, and thus the inability to identify women in need of treatment and to initiate treatment, is a major concern in many resource-constrained settings, especially at the primary care level, where most women obtain maternal and child health (MCH) care.

Regimen efficiency and simplification. Another key advantage of Options B and B+ is greater efficiency, very much in accord with Treatment 2.0 principles. First, the same simplified, fixed-dose combination ARV regimen can be used throughout the PMTCT intervention. Further, it is possible, and highly desirable, to provide the same regimen both for PMTCT and as the first-line national ART regimen for non-pregnant individuals. The ability to use the same regimen for PMTCT and for first-line ART considerably simplifies drug forecasting, procurement, supply to facilities, and drug stock monitoring. The first-line regimen of tenofovir/lamivudine/efavirenz (TDF/3TC/EFV) is available as a single-pill fixed-dose combi-nation and has been recommended recently as the optimized regimen for first-line adult treatment, including for pregnant women (9). An important advantage of efavirenz in the first-line regimen is that it can be used in all women, regardless of CD4 count (unlike nevirapine, which cannot be used in women with high CD4 counts). Although concerns remain about the safety of efavirenz in early pregnancy, and enhanced phar-macovigilance monitoring is needed, review of recent data is reassuring, and benefits are likely to outweigh risks (10).

Many HIV high-burden countries initially chose Option A because of limited PMTCT programme support, challenges of

scale-up, lower drug costs, ease of adding on to prior PMTCT approaches and training, and limited capacity to provide triple ARVs in MCH settings. However, these factors are changing, and a number of high-burden countries are considering moving from Option A to Option B or B+.

Costs. The cost of ARV drugs was a major determinant in countries’ choice of a PMTCT option. In 2009 the average ARV drug cost of Option B was three to five times higher than the cost of Option A (depending on regimen and assuming the provision of both ART and prophylaxis). However, by the end of 2011, this differential had diminished to two times higher. The annual cost of two-pill formulations of TDF/3TC/EFV has decreased by 30% over the past three years and is now US$150; the newer TDF/3TC/EFV single-pill fixed-dose regimen costs approximately US$180 per year (11,12). Further declines are anticipated. With the differing initial cost of drugs now less of a factor, analyses of long-term costs, cost-benefit and cost-effectiveness will be more appropriate for guiding policy decisions than per person initial cost.

Option B+ advantages. The Option B+ approach of life-long ART for all HIV-infected pregnant women, regardless of CD4 count, has important advantages over both Options A and B (if viral suppression is maintained) but needs to be evaluated in programme and field settings. These advantages include:1. further simplification of PMTCT programme requirements—

no need for CD4 testing to determine ART eligibility (as required in Option A) or whether ART should be stopped or continued after the risk of mother-to-child transmission has ceased (as in Option B) (although CD4 counts or viral load assays are still desirable for determining baseline immuno-logical status and monitoring response to treatment);

2. extended protection from mother-to-child transmission in future pregnancies from conception;

3. a strong and continuing prevention benefit against sexual transmission in serodiscordant couples and partners;

4. likely benefit to the woman’s health of earlier treatment and avoiding the risks of stopping and starting triple ARVs, especially in settings with high fertility; and

5. a simple message to communities that, once ART is started, it is taken for life.

Challenges and questions. Still, there are important programmatic, operational and clinical challenges and ques-tions about Option B+ that need to be addressed, including service organization and service delivery of ART in MCH and primary care settings, cost and sustainability, ARV adher-ence and retention in care, referral mechanisms and transi-tions from the PMTCT programme to HIV care and treatment programmes, concerns about HIV drug resistance with long-term use of ART when initiated in early HIV disease, safety of increased ARV exposure for the fetus/infant, acceptability and equity. Thus, countries implementing Option B+ or plan-ning demonstration projects should be supported to monitor this approach closely to address these issues and assess the feasibility, cost-benefit and public health impact of Option B+.

4

WHO advice to countries

In light of global and country commitments to elimination of new paediatric infections and the changes outlined in this programmatic update, all countries should examine their own policy, goals and implementation experiences and assess how they can better simplify, optimize and integrate their PMTCT and ART programmes. Countries that are successfully imple-menting Option A and achieving their targets of decreasing mother-to-child transmission of HIV and treating mothers eligible for ART do not need to plan an immediate change to Option B or B+. Countries that are considering changing their PMTCT guidelines should anticipate and prepare adequately for the changes, to assure that clear policy, implementation strategy, proper messaging, training and an ARV demand fore-casting and supply system are in place.

Options B and specifically B+ seem to offer important programmatic and operational advantages and thus could accelerate progress towards eliminating new paediatric infec-tions. If Option B+ can be supported, funded, scaled up at the primary care level and sustained, it will also likely provide the

best protection for the mother’s health, and it offers a prom-ising new approach to preventing sexual transmission and new HIV infections in the general population.

There is an urgent need to assess country experiences and evidence that address the preferences among Options A, B and B+ outlined here. Evidence on the operational advantages of providing triple ARVs to all HIV-infected pregnant women (Options B and B+), on how to best meet the programme requirements of these approaches, and on the acceptability, effectiveness and prevention impact of providing lifelong ART to all HIV-infected pregnant women (Option B+) will help inform upcoming guidelines revision.

This programmatic update is meant to provide a current perspective for countries on the important changes and new considerations arising since the 2010 PMTCT ARV guide-lines, especially as a number of countries are now preparing to adopt Option B+. WHO has begun a comprehensive revision of all ARV guidelines, including guidance on ARVs for preg-nant women, planned for release in early 2013.

1Antiretroviral drugs for treating pregnant women and preventing HIV infec-tions in infants: recommendations for a public health approach, 2010 version. Geneva, World Health Organization, 2010. http://www.who.int/hiv/pub/mtct/guidelines/en/2Antiretroviral therapy for HIV infection in adults and adolescents: recommen-dations for a public health approach: 2010 revision. Geneva, World Health Organization, 2010. http://www.who.int/hiv/pub/arv/adult2010/en/3Joint United Nations Programme on HIV/AIDS (UNAIDS). Countdown to zero: global plan for the elimination of new HIV infections among children by 2015 and keeping their mothers alive, 2011–2015. Geneva, UNAIDS, 2011. http://www.unaids.org/en/media/unaids/contentassets/documents/unaid-spublication/2011/20110609_JC2137_Global-Plan-Elimination-HIV-Chil-dren_en.pdf 4World Health Organization (WHO), Joint United Nations Programme on HIV/AIDS, and United Nations Children’s Fund. Global HIV/AIDS response: epidemic update and health sector progress towards universal access: pro-gress report 2011. Geneva, WHO, 2011. http://whqlibdoc.who.int/publica-tions/2011/9789241502986_eng.pdf5Cohen MS et al. Prevention of HIV-1 infection with early antiretroviral thera-py. New England Journal of Medicine, 2011 Aug 11, 365(6):493–505. http://www.nejm.org/doi/full/10.1056/NEJMoa11052436Guidance on couples HIV testing and counselling and antiretroviral therapy for treatment and prevention in serodiscordant couples: recommendations for a public health approach. Geneva, World Health Organization, 2012 (in press,

REFERENCES

expected publication April 2012).7Schouten EJ et al. Prevention of mother-to-child transmission of HIV and the health-related Millennium Development Goals: time for a public health approach. The Lancet, 2011, 378:282–284. http://www.itg.be/itg/Uploads/Nieuws/2011-Lancet%20Viewpoint%20-%20PMTCT%20Public%20Health%20Approach.pdf8World Health Organization (WHO) and Joint United Nations Programme on HIV/AIDS. The Treatment 2.0 framework for action: catalyzing the next phase of treatment, care and support. Geneva, WHO, 2011. http://www.unaids.org/en/media/unaids/contentassets/documents/unaidspublica-tion/2011/20110824_JC2208_outlook_treatment2.0_en.pdf9Short-term priorities for antiretroviral drug optimization. Meeting report (18–19 April 2011, London, UK). Geneva, World Health Organization, 2011. http://www.who.int/hiv/pub/arv/short_term_priorities/en/index.html 10Technical update on treatment optimization: use of efavirenz during preg-nancy in a public health perspective. Geneva, World Health Organization (in preparation, expected publication May 2012).11Transaction prices for antiretroviral medicines and HIV diagnostics from 2008 to July 2011: global price reporting mechanism, GPRM, October 2011. Ge-neva, World Health Organization, 2011. http://www.who.int/hiv/amds/gprm_summary_report_oct11.pdf12World Health Organization, HIV/AIDS Department. Global price reporting monitoring, March 2012 (unpublished analysis).

5

� Easier implementation could expand services. Reported difficulties with implementing PMTCT pro-grammes, including the challenge of providing ARV treatment in MCH settings and at the primary care level, highlight the importance of simplifying drug regimens and operational delivery, as exemplified by Options B and B+. Easier implementation should facilitate expan-sion of services and more effective programmes. This will, however, require strengthened antenatal services, task-shifting, more effective ARV service delivery in MCH settings and direct linkages with ART programmes.

� Unknowns need research. Concerns and unknowns with Options B and B+ include possible increased ARV multi-drug resistance in women due to poor adherence and in infants infected despite maternal ART, and the accept-ability and feasibility for women of remaining in care and on lifelong ART, especially for women starting treatment earlier than is currently recommended for adults generally. In particular, rapid scale-up of ARVs, including efavirenz, for pregnant women will greatly increase early fetal exposure, including exposure from conception in future pregnancies, and prolonged exposures during breastfeeding. Pharma-covigilance, drug resistance monitoring, implementation research and programme monitoring are necessary.

� No easy fix. Moving from current Option A or Option B to Option B+ will not, on its own, resolve the key challenges and problems of expanding coverage and successfully transitioning pregnant women from PMTCT programmes to HIV care and treatment programmes. Well-supported referral systems and strong MCH and ART programme linkages are essential.

� Adherence and retention crucial. Postpartum drop-out rates in PMTCT programmes are especially high, in part due to weak postpartum services. PMTCT interventions during breastfeeding have yet to be fully implemented successfully with any option. Maintenance of viral suppression with ARV treatment—achieved by supporting continued adherence to the ART regimen—is crucial to the additional benefits of the Option B and B+ interventions and to minimizing adverse consequences.

� And especially with Option B+. While programmes need to provide effective support for adherence and retention in care with all three PMTCT options, additional support will be required for Option B+. It is particularly important for programmes implementing Option B+ to develop strong systems to support adherence and reten-tion and to build evidence of successful practices through implementation science.

� Family planning still essential. Even in the context of expanded access to ART for HIV-infected pregnant women, family planning services still need to be strength-ened to avoid unintended pregnancies.

� Quality assurance needed for HIV testing. Reli-able HIV rapid testing in antenatal settings is important for all options, as the entry point to PMTCT interven-tions. Robust quality assurance systems and confirma-tory testing will be especially important in the context of Option B+, where every pregnant woman who tests HIV-positive is started on treatment for the rest of her life.

� Time to reassess. New developments warrant reas-sessment of current PMTCT and treatment options. WHO is not changing its guidance now but will review its PMTCT ARV guidelines as part of a comprehensive review and consolidation of all ARV-related guidance in 2013.

� Options B and B+ have advantages. WHO recog-nizes that in many settings there are likely to be impor-tant clinical and programmatic advantages to the currently recommended Option B (maternal triple ARVs for all HIV-infected pregnant women and continued lifelong for those eligible for treatment) and the emerging Option B+ (lifelong treatment for all HIV-infected pregnant women, regardless of CD4 count) over Option A (ART for pregnant women eligible for treatment; AZT antenatal single-drug prophylaxis and infant prophylaxis during breastfeeding).

� Options B and B+ better assure treatment. While current data do not indicate differences in the efficacy of Options A and B when used as prophylaxis for women not eligible for treatment, Options B and B+ provide greater assurance that women in need of treatment, especially in settings with limited access to CD4 testing, receive a fully suppressive triple ARV regimen to minimize the risk of infant infection and to benefit their own health.

� Benefits beyond PMTCT. Option B and particularly Option B+ offer women benefits beyond PMTCT, including likely additional benefit for women’s own health by starting treatment earlier and prevention of sexual HIV transmis-sion to uninfected partners, including the common situa-tion of HIV serodiscordant couples.

� Higher cost but more cost-effective? Initial drug costs are higher for Options B and B+ than for Option A, but the cost of the drugs is decreasing. The benefits gained for the costs expended are likely to be much greater.

� Options B and B+ simpler for programmes. These regimens are, in many aspects, simpler for programmes—the same regimen could be given to all HIV-infected pregnant women (available as a once-daily fixed-dose combination); there is no initial distinction between treat-ment and prophylaxis; CD4 counts are not needed for starting ARVs; there is no change in regimen during the pregnancy/postpartum period (as in Option A); and the regimen could be harmonized with adult ART regimens for easier logistics if an efavirenz-based regimen is used.

� Option B+ has further advantages. Compared with Option B, Option B+ would provide protection against sexual transmission of HIV that extends past the period of risk for mother-to-child transmission, protect the next pregnancy starting from conception, and avoid stopping and restarting ARVs with the next pregnancy or when CD4 count later drops below 350 cells/mm3.

� More countries moving toward Option B or B+. Many high-burden countries in sub-Saharan Africa initially favoured Option A, due to lower drug cost and continuity with prior PMTCT recommendations, but some are now reassessing this choice. Countries with lower prevalence or more developed infrastructure tended to choose Option B. Malawi was the first to adopt Option B+, for its ease of implementation and potential prevention benefit; addi-tional countries are now considering Option B+.

Programmatic update on ARVs for pregnant women and PMTCT: Key points

! 29!

Appendix 6

Letters received by the PROMISE study teams from the

Ministries of Health of Malawi and Uganda

Telephone: + 265 789 400 Facsimile: + 265 788 403

All Communications should be addressed to: The Secretary for Health

Prof Francis Martinson and Dr Newton Kumwenda IMPAACT 1077BF UNC and JHU Projects Malawi

Dear Francis and Newton,

In reply please quote No.

MINISTRY OF HEALTH P.O. BOX 30377

LILONGWE3 MALAWI

25TH AUGUST, 2011

IMPAACT 1077BF: PROMISE- PROMOTING MATERNAL AND INFANTS SURVIVAL EVERYWHERE

I would like to thank you very much for your recent letter and discussions we have had on the above project.

The Ministry of Health through HIV and AIDS Department supports the conduct of the PROMISE study. We feel the study will provide extra information for the current and future PMTCT policy development in our setting.

In view of the above, Ministry of Health supports the current version of the research protocol and wishes you the best in your tasks to further enlighten us on the newly approved PMTCT policy that is being implemented. Please keep us regularly informed as you pursue this important study.

Dr Frank M Chimbwandira Director, HIV and AIDS Dept. FOR: SECRETARY FOR HEALTH

Mini ster' s Office: 256-414-340871 Office of the Minister of Health Telephone: General Lines: 256-414-340874/

231563-9 P.O Box 7272,

Kampala,

Uganda.

Fax: 256-414-253842

IN ANY CORRESPONDENCE ON _ A. nM 1051269101 THE REPUBLIC OF UGANDA !HIS SUBJECT PLEASE QUOTE Nlr :-:-:': . . : .. . . . . . .

5 March 2012

Dr. Philippa Musoke and Dr. Maxensia Owor Makerere University- John Hopkins University Research Collaboration Upper Mulago Hill Rd- Mulago Hospital P.O Box 23491 KAMPALA

LETTER OF SUPPORT FOR THE CONTINUATION OF THE IMPAACT 1077BF-PROMOTING MATERNAL AND INFANTS SURVIVAL EVERYWHERE (PROMISE) STUDY BEING CONDUCTED BY MUJHU IN UGANDA

Following the meeting held in my office in the Ministry of Health on Friday 241h February 2012 to discuss and share the information regarding the PROMISE clinical trial being conducted at MU-JHU, in cooperation with the Department of Obstetrics and Gynaecology and Mulago Hospital, I have understood that the PROMISE study which is being carried out across many sites in Africa, is evaluating the optimal and safest regimens for use by HIV infected pregnant and post partum breastfeeding women and their HIV exposed infants . The study will also assess the adherence of pregnant and post partum women to more complex PMTCT regimens and whether there is early emergence of resistance.

All this information is critically important as Uganda moves ahead in a phased manner to implement PMTCT guidelines aimed at the virtual elimination of new paediatric HIV and the harmonization of adult treatment, prevention and PMTCT guidelines. Indeed, this information is needed in order to help inform Ministry of Health, International donors and policy makers in providing evidence based direction to the HIV I AIDS response in general and PMTCT in particular. In this regard, the Ministry supports the continuation of the PROMISE study in Uganda. I believe it will provide the much needed information on the safety and effectiveness of available PMTCT regimens, which will further inform PMTCT and treatment policies.

The purpose of this letter is to indicate the support of the Ministry to the ongoing PROMISE research protocol being conducted by MU-JHU Research Collaboration at Mulago Hospital.

I wish you well as you continue with the enrolment and move toward completion of this important study.

Hon Dr Ondoa J D Christine MINISTER OF HEALTH

Copy to: Hon Minister of State for Health (General Duties) Ag Permanent Secretary, Ministry of Health Director General Health Services, Ministry of Health Ag Director Health Services (Community and Clinical Services)