Expert consensus on novel medicines to prevent preterm birth and manage preterm labour: Target product profiles

To develop target product profiles (TPPs) for new medicines for preterm birth prevention and preterm labour management that address the real‐world need of women and healthcare providers, informed by views and agreement amongst globally diverse stakeholders.

Preterm birth (i.e.birth before 37 completed weeks of gestation) is the leading cause of death in children aged under 5 years globally; 35% of neonatal deaths are caused by complications of preterm birth. 1 Nearly 15 million babies are born preterm annually, approximately 80% of which occur in Africa and South Asia, in predominantly limited-resource settings. 2Preterm babies are at increased risk of short-and long-term adverse health outcomes, including chronic lung disease, infections, and neurological, visual and auditory disabilities. 3,4Up to 50% of preterm births are the result of spontaneous preterm labour, with an additional 25%-30% resulting from preterm prelabour rupture of membranes (PPROM). 5The aetiology of spontaneous preterm birth, although not well understood, is multifactorial, and can involve infection, inflammation, uteroplacental ischaemia or haemorrhage, uterine overdistension or stress. 6urrently, there are very few effective medicines recommended for the prevention of preterm birth. 7Effective preventive agents are available for women with specific risk factors, such as progesterone for women with previous preterm birth and/or short cervix. 8In women experiencing spontaneous preterm labour, tocolytic agents can slow or stop contractions, providing time to administer corticosteroids for fetal lung maturation or to allow the transport of women to higher-level health facilities. 9Several tocolytic agents are used internationally, including calcium channel blockers, betamimetics, nitric oxide donors, cyclooxygenase (COX) inhibitors, magnesium sulphate and oxytocin receptor antagonists, which are likely to prolong pregnancy for 2-7 days, with some also improving newborn health outcomes. 10espite 2.4 million neonatal deaths, 2 million stillbirths and 295 000 maternal deaths occurring globally each year, [11][12][13] drug development for pregnancy-related conditions remains severely underfunded and inadequate. 14In 2020, the Accelerating Innovation for Mothers (AIM) project was created with the goal of invigorating investment in research and development (R&D) to address the 'drug drought' in medicines for pregnancy-specific conditions.AIM is a collaboration between the Concept Foundation, Policy Cures Research and the Burnet Institute, which envisions a world where medicines and technologies are developed and made accessible to women for pregnancy-specific conditions.Across all AIM activities, data have been collected from over 20 countries.
Target product profiles (TPPs) are a well-recognised strategy to promote the development of innovative medical products, such as devices, diagnostic tests and therapeutics, [15][16][17] and are an important resource for multiple stakeholders in the R&D pathway, including funders, researchers, product developers, manufacturers and regulators. 17TPPs are documents that specify the minimum and preferred characteristics that a new product should take, to fulfil a clinical need.TPPs can also be used to identify, prioritise and evaluate new candidate products.Developing a TPP is a consensusgenerating process, requiring input from diverse key stakeholders to align around a clear set of product goals and characteristics. 159][20][21] However, a review of the literature and the World Health Organization (WHO) Health Product Profile Directory revealed that there are no publicly available TPPs for medicines for maternal conditions, including preterm birth.The WHO and United Nations International Children's Emergency Fund (UNICEF) have developed TPPs for neonatology, including for antibacterial agents for neonatal sepsis and devices used for newborn care. 21,22As part of the AIM project, we developed two new TPPs for medicines to prevent and treat pre-eclampsia. 23In this study, we used quantitative and qualitative methods to develop new TPPs describing medicines for the prevention of preterm birth and the management of preterm labour.

| M ET HODS
We prepared a study protocol on TPP development informed by methods used in recent TPPs for HIV cures and diagnostic tests for sexually transmitted infections, and adopted the five-step process used by Lewin et al. (Figure 1). 15,18,19,24he protocol was approved by the Alfred Ethics Committee for Human Research (project no.108/21), and informed consent was obtained from all participants prior to their participation.

| Step 1: initial drafting phase
On 9 March 2021, the AIM project convened a multidisciplinary advisory group of 11 experts from diverse backgrounds (drug development research, maternity clinical practice, consumer advocacy, maternity health programme implementation, social enterprise, funders and international agencies), with diversity in gender and geographical location.In consultation with this advisory group, preterm birth was selected for the development of two TPPs for: (1) agents for preventing preterm birth in women at increased risk; and (2) agents for the management of preterm labour (tocolysis; Figure 1).
Initially, we sought to define the intended use-case scenario for prophylactic medicines for preterm birth and therapeutic medicines for preterm labour, the key variables for the TPPs, and the acceptable minimum and preferred targets for each of these variables.Through literature review and AIM team consultations, intended use-case scenarios were developed, which were revised in subsequent phases (Box 1 presents the final versions).Given the considerable burden of preterm birth affecting women in low-and middle-income countries (LMICs), we specified that a primary focus of the TPPs were to drive the development of medicines that could be used safely and effectively for women living in limitedresource settings.We collated TPP templates and guidance produced by several reputable organisations (United States Food and Drug Administration (USFDA), Bill & Melinda Gates Foundation, WHO, PATH and others) and developed a TPP template with 21 domains.Through literature reviews we developed draft minimum and preferred targets for each domain, along with additional annotations, including the rationale and supporting evidence (original draft TPPs available as Appendixes S1 and S2).

| Step 2: international stakeholder survey
We conducted an international online stakeholder survey to seek participants input on use-case scenarios and the minimum and preferred targets for each variable.We used F I G U R E 1 Overview of target product profile (TPP) development process the survey approach of Pelle et al. in their development of TPPs for clinical decision support tools, 19 using Qualtrics (Provo, UT, USA), and pre-tested it on two individuals prior to launch.Respondents were asked to rate their agreement with the minimum and preferred targets for each domain using a Likert scale (where 1 equals strongly disagree and 5 equals strongly agree).Optional comments were invited for each domain.
Professionals working in the field of maternal and perinatal health (including clinicians, researchers, funding agency staff, international public organisation staff, programme implementers, policymakers, representatives of consumer advocacy organisations and other relevant maternal health systems stakeholders) were invited to participate in the survey.We sought diverse representation from participants in high-, middle-and low-income countries.In total, 273 individuals were invited to participate using the following databases: (1) AIM project database of relevant maternal health R&D experts; (2) a database of all individuals who had participated in WHO maternal and perinatal health guideline development groups in the past 12 years; 25 and (3) members of the WHO Multi-Country Survey on Maternal and Newborn Health research network. 26n addition, the survey was distributed through other clinician-researcher networks and listservs, including the Cochrane Pregnancy and Childbirth network and the Perinatal Society of Australia and New Zealand; as such, we cannot calculate an exact response rate.We pre-specified a minimum of 50 responses per domain to evaluate the degree of consensus.We defined agreement as ≥75% of respondents selecting agree or strongly agree for a specific variable.The survey was active for 31 days (Figure 1).Demographics of the final survey respondents are presented in the results.

| Step 3: stakeholder interview phase
In parallel with the online survey, we conducted in-depth, one-on-one interviews with stakeholders with a particular interest in preterm birth, maternal health drug development and implementation.The goal of these interviews was to seek more detailed feedback on the use-case scenarios and requirements for each variable.Stakeholders were identified from the same databases used for the survey.In addition to appropriate expertise on the topic of interest, diversity of gender, geographical location and technical expertise was sought (Figure 1).Interview participants were not identified from the survey participants.
Stakeholders were initially invited to participate via email and were provided drafts of the TPPs to read prior to the interviews.Interviews were conducted over Zoom by an AIM project researcher, using a semi-structured, pre-tested interview guide.Informed consent was obtained before the interview began and each interview lasted approximately 60-90 minutes.Interviews were conducted in English.The interview guide involved discussing the TPP variables sequentially, with particular emphasis on variables relevant to their area of expertise or interest.We explicitly sought participants' views on applicability across different countries and resource levels.Participant responses were captured via interview recordings and field notes.Interview recordings were only referred to if field notes were unclear on a particular variable.

| Step 4: public consultation
The draft TPPs were made available online for public comment via the Burnet Institute and Concept Foundation websites.The consultation period lasted approximately 4 weeks (concurrent with the international online survey) and was distributed via social media (Figure 1).

| Step 5: synthesis and finalisation
The results of the international online survey were analysed, and domains where consensus was not reached were identified.Consensus was defined as 75% or more of respondents who agreed or strongly agreed with the minimum and preferred criteria.The outputs of expert interviews were summarised, with key themes and 'major' or 'minor' concerns identified.Variables where consensus was not achieved, or where strong disagreement was identified via interviews, BOX 1 Use-case scenarios for TPPs in the prevention of preterm birth and treatment of preterm labour

• Prevention of preterm birth
A prophylactic agent that can be administered to pregnant women at increased risk of spontaneous preterm birth.The medicine can be administered in any healthcare setting where pregnant women receive care, would have an excellent safety profile during pregnancy, can be commenced early in pregnancy and can be continued throughout pregnancy, as required.

• Management of preterm labour
A therapeutic agent that can be administered by skilled health personnel to pregnant women experiencing spontaneous preterm labour, accompanied by the monitoring of maternal and fetal well-being in antenatal care settings.The therapeutic agent will inhibit uterine contractions, facilitate the prolongation of pregnancy, be safe for women and babies, and ideally improve perinatal health outcomes.
were modified based on stakeholder feedback.The final drafts were shared with the AIM expert advisory group for comments before finalisation and publication (Figure 1).

| Survey and public consultation results
We received 46 responses to the survey.The respondents were from across all WHO geographical regions (Figure 2A).Participants represented diverse expertise, selecting the best description of their current role as clinicians (doctor, midwife or nurse, 39.1%), researchers (37.0%), epidemiologists (4.3%), staff of not-for-profit organisations (6.5%), staff of international public organisations (2.2%), staff of health programmes (4.3%), staff of funding agencies (2.2%) and other (including participants who identified as more than one option, 4.3%).
The survey results showed high agreement (consensus; ≥75%) across almost all variables of both TPPs (Figure 3).For preterm birth prevention, agreement was less than 75% for the minimum requirements regarding the acceptable level of adherence to the medicine regimen (20% of respondents disagreed or strongly disagreed; Figure 3A).For treatment of preterm labour, agreement was less than 75% for the preferred requirements regarding the WHO prequalification of the medicines (29% neutral, 0% disagree or strongly disagree; Figure 3D).Consensus was achieved for all other variables in both TPPs.No comments were received via the public consultation website.

| Findings from stakeholder interviews
A total of 40 stakeholders were invited to participate in an interview, of which 21 stakeholders (13 females and eight males) responded; all interviews were conducted between August and October 2021.Like the survey respondents, the interview participants represented diverse gender, geographical and expertise.Participants were from Africa, Asia/Pacific, Europe, USA and South and Central America (Figure 2B), and included nine obstetrician/researchers, two drug development experts, two consumer representatives, two midwives, one neonatologist, two maternal medicines procurement experts, one basic scientist, one programmes implementation expert, and one WHO staff member.Compared with survey respondents, there was a slightly higher proportion of interview participants from the Western Pacific region, and a slightly lower proportion from the African region.There were no interview participants from the Eastern Mediterranean region, although they were represented in the survey.
Although survey respondents strongly agreed with the target population for both preterm birth prevention (minimum, 89%; preferred, 92%) and preterm labour management (minimum, 88%; preferred, 88%), several interviewees disagreed with the definition of the target population in both TPPs.Numerous interviewees raised concerns with the broad definition of the target population for the prevention of preterm birth and felt that a more precise definition was required.Specifically, there was confusion as to whether all forms of preterm birth (i.e.spontaneous, PPROM and provider initiated) were being targeted, and whether one drug would be able to prevent preterm birth in all of these situations.
In the TPP for medicines to treat preterm labour, many interviewees did not agree with the proposed target population of women at <34 weeks of gestation in spontaneous preterm labour.The initial draft of the TPP used <34 weeks of gestation because of the WHO's 2015 recommendation that women at risk of imminent preterm birth be treated with antenatal corticosteroids (ACSs) for fetal lung maturation in this situation. 9However, interviewees suggested that women at <37 weeks of gestation would be a more appropriate target population, considering that: tocolytics can be useful after 34 weeks of gestation to support the transport of women in spontaneous preterm labour to a higher level of care; that in many low-resource settings it is not possible to determine with confidence the gestational age of a pregnancy; and that in some settings women at risk of preterm birth between 34 and <37 weeks of gestation are given ACSs. 9,27-29Some interviewees also suggested that the minimum acceptable level of treatment adherence (i.e. a tolerable discontinuation rate of <35%) was too high, which was reflected in a high level of disagreement on this variable in the survey.One interviewee remarked that any preventive medicine with a discontinuation rate greater than 20% would be unlikely to have meaningful public health impact.
There was strong agreement amongst all interviewees that new medicines should not require cold chain transport or storage, that any new medicines should not require additional clinical monitoring of mother or baby and that medicines should be affordable in LMICs.These findings were reflected in the survey results.All but two interviewees strongly agreed that any new medicine should not require the use of a companion diagnostic test as a requirement for its routine use.Two interviewees suggested that the use of F I G U R E 3 International survey responses.Results from international stakeholder survey.Percentage of respondents that strongly agreed (dark green), agreed (light green), were (grey), disagreed (orange) or strongly disagreed (red) in response to the minimum and preferred requirements for each variable in the TPPs for new medicines to prevent preterm birth (A, B) and to manage preterm labour (C, D).The vertical dashed lines indicate the consensus level a companion diagnostic test for bacterial vaginosis could be beneficial for targeting the use of preventive agents but agreed that such a requirement may limit widespread implementation in limited-resource settings.
The question of whether evidence of improvement to perinatal health outcomes should be included as a minimum (rather than a preferred) requirement in both TPPs was raised by numerous interviewees.Some felt that for any new agent to be used widely, the demonstration of improved perinatal outcomes (i.e.reduced neonatal mortality) was essential.However, other interviewees felt that it was acceptable as a preferred requirement, given the difficulty in conducting large trials that could demonstrate benefits.These interviewees considered that evidence of a reduced incidence of preterm birth (prevention TPP) and evidence of a clinically important difference in extending pregnancy duration (treatment TPP) were themselves compelling evidence of benefit and were thus appropriate as a minimum requirement.
Although the survey showed a high level of agreement for the safety requirements in both TPPs, a small number of interviewees disagreed with the inclusion of lactating women, suggesting that this was not relevant as these medicines would not be used postpartum.However, many interviewees felt that lactating women should be specified, as medicines with a long half-life may have implications for breastfeeding immediately after birth.Several participants felt that protecting breastfeeding was particularly critical for women living in LMICs, because of the considerable barriers associated with formula feeding, such as cost and access.One interviewee with experience as a healthcare professional in LMICs stated that pregnant women may also be breastfeeding other infants while pregnant.Another minor issue identified by some interviewees was that in both TPPs, vaginal administration should be included as a non-invasive administration route.

| Finalisation of TPPs
Following the synthesis of findings from the survey and the stakeholder interviews, the authors refined the target population definition in both TPPs (Tables S1 and S2).For medicines to prevent preterm birth, the target population in the TPP specifies 'women with identified risk factors for spontaneous preterm labour' only (Table S1).For medicines to treat preterm labour, the upper gestational age limit of the target population in the TPP is <37 completed weeks of gestation (Table S2).In the TPP for the prevention of preterm birth, the minimum treatment adherence was changed to <30% discontinuation (Table S1).Vaginal administration was also included within the 'formulation, dosage and administration' variable in both TPPs.Improving neonatal outcomes remained as part of the preferred requirement for clinical efficacy, and 'not contraindicated in lactating women' remained a minimum safety requirement in both TPPs (Tables S1 and S2).

| Main findings
Using multiple methods, including quantitative surveys and qualitative in-depth interviews with diverse experts, we have conducted the first study to comprehensively evaluate the requirements for new maternal medicines to prevent preterm birth and manage preterm labour.Our findings highlight that across multiple stakeholder groups (including basic research, obstetrics, midwifery, neonatology, drug development, maternal medicines procurement, programme implementation, WHO and consumer advocacy) there was strong agreement on the characteristics that new medicines for preterm birth and labour must have to meet the real-world needs of women and providers.There was agreement from surveys and interviews that for any novel or repurposed medicine: (1) cold chain transport and storage should not be required; (2) that any medicine should require minimal to no additional drug-specific clinical monitoring of mother or baby; (3) non-invasive administration routes such as oral medicines are preferred; and (4) medicines would need to be affordable in LMICs.
In this study, the variables that provoked the highest level of disagreement or discussion were about the target population for both TPPs.Additional areas of discussion included the importance of safety during breastfeeding of any new medicines and the benefits versus implementation barrier of a required clinical diagnosis for the use of medicines.

| Interpretation, in light of other evidence
Development of new obstetric medicines -including those for preterm birth prevention and management -is underresourced and poorly coordinated. 30The mapping of maternal research funding demonstrates that very few donors place high priority on maternal health intervention studies. 31This is particularly true for the pharmaceutical industry -a 2014 analysis by Footman et al. assessed 2340 studies on maternal health interventions in LMICs, reporting that pharmaceutical companies were involved in only 7% of these studies. 31There is a clear market failure in the R&D for new maternal medicines.As of 2008, there were fewer drugs in active development for all maternal conditions than for the rare disease amyotrophic lateral sclerosis. 14This 'drug drought' in maternal medicines has led to only two new drugs, the tocolytic atosiban and carbetocin, for the prevention of postpartum haemorrhage, being licenced specifically for use in pregnant women in the last 30 years. 32Unlike the global efforts for HIV and malaria treatments, there has been no global systematic effort to improve maternal medicine R&D and identify candidate medicines that meet the real-world needs of women and providers, particularly for those living in LMICs where newborn and maternal morbidity and mortality are unacceptably high. 11,31,33The TPPs developed by the WHO for other areas of reproductive health -such as sexually transmitted infection diagnostics -are highly valued by R&D companies, who use them to guide the development of new tests. 34TPPs describing new medicines for obstetric conditions have not been prioritised prior to the AIM project, but are now being identified as a tool to promote innovative advances for women.In addition to the two TPPs we have developed for pre-eclampsia medicines, 23 researchers at the University of Birmingham are also developing a TPP for endometrial receptivity tests for recurrent miscarriage. 35The TPPs developed during this study will provide guidance to prioritise research towards candidate medicines that best meet the needs of women at risk of or experiencing preterm birth.
Underfunding and siloed research means that there are very few medicines available for women at risk of preterm birth or who are in preterm labour.Currently, calcium channel blockers, betamimetics, nitric oxide donors, COX inhibitors, magnesium sulphate and oxytocin receptor antagonists are used for tocolysis, although there is only low to very low GRADE (Grading of Recommendations Assessment, Development and Evaluation) evidence that these tocolytics themselves substantively improve perinatal outcomes. 10tosiban, an injectable oxytocin antagonist that requires cold chain transport, is the only recommended drug specifically developed as a tocolytic and is only available in some highincome settings, and not in the USA.All other medicines are repurposed tocolytics and have undesirable safety profiles (e.g.betamimetics) or accessibility challenges (e.g.nifedipine). 10,36Noting the lack of evidence of benefits for tocolytic agents, the WHO's 2015 guidelines recommend against tocolysis for women in spontaneous preterm labour, 9 although they are currently under review.However, they do specify that oral nifedipine can be considered in order to provide a treatment window for the administration of ACS and/or transport to an appropriate care setting. 9Nitric oxide donors such as nifedipine are the highest ranked tocolytic for delaying preterm birth, 10 and oral nifedipine meets many of the variables specified in the current TPP, including no need for cold chain transport, non-invasive administration and minimal additional monitoring of the mother.Many national guidelines recommend the use of tocolytics for women in preterm labour to delay birth by 48 hours, but the choice of tocolytic differs between guidelines. 379][40] We therefore believe that these TPPs will be helpful to guide the development of the new products that are needed.
There are few options for medicines to prevent preterm birth.2][43] Progesterone meets many of the criteria within the current TPP, such as no cold chain requirements and non-invasive administration; however, questions remain about the most effective administration and formulation of progesterone, and the indications for use (such as short cervical length, which requires the performance of an ultrasound) may not be easy to identify in all settings. 8In addition, the USFDA proposed that Makena® (hydroxyprogesterone caproate injection) be withdrawn from the market because of evidence for a lack of clinical benefit in preventing preterm birth. 44From a clinical perspective, new medicines for both the prevention of preterm birth and the management of preterm labour are needed.
This study has identified strong agreement from diverse stakeholders, on the requirements new preterm birth medicines should meet.TPPs, such as those developed from this research, provide guidance to involved stakeholders, researchers and companies that can help to drive the development of candidates that are most likely to meet real-world needs.For funders, these TPPs can be used to set targets for R&D and promote research investment in the most promising candidates that meet or mostly meet the parameters defined in this study.TPPs are also a tool that allow regulatory agencies to evaluate new candidates and their potential for implementation in a range of settings.Prioritising R&D funding for candidate medicines that meet or mostly meet the requirements of the TPPs can drive the development of medicines that will address the needs of women and providers, particularly in LMICs.

| Strengths and limitations
We have described the development of the first TPPs for medicines to prevent preterm birth and to manage preterm labour.Given the global diversity of expertise of the participating stakeholders, the results of this study can be considered a reflection of the needs of women, healthcare providers, researchers and drug developers, and will facilitate the development of new medicines for preterm birth that can be implemented globally.One limitation was that we performed only one round of the international survey to identify levels of agreement.In other TPP development activities (e.g. for HIV cures), high levels of disagreement for some TPP variables necessitated a second survey round with stakeholders to reach consensus. 15However, in this study we reached consensus across nearly all variables in a single round.Although the exact response rate of the survey cannot be determined (as it was disseminated through multiple channels and listservs), a larger number of responses may have yielded different levels of agreement; however, given the high level of agreement here, we believe this to be unlikely.Another limitation is that we were unable to define a specific target affordable price that new medicines should meet, which is often included in TPPs. 24However, given the variety of preterm medicines currently being evaluated, 45,46 as well as the different pricing structures used across global markets, we considered that defining a specific target price would be inappropriate. 24In future updates of the TPPs, and as more specific target medicines are identified, economic analyses could be incorporated as a key variable in these TPPs.
Distribution of stakeholders by WHO global region.The proportion of stakeholders who participated in the survey (A, n = 46) and interviews (B, n = 21) in each of the WHO global regions is presented.AFR, African region (yellow); AMR, region of the Americas (blue); EMR, Eastern Mediterranean region (dark green); EUR, European region (red); SEAR, South-East Asian region (light green); WPR, Western Pacific region (black)