Putting Theory into Practice by Developing a Novel Digital Health Technology-Derived Endpoint in Sleep Quality

Abstract

Introduction

Sleep disturbances associated with menopause (SDM) are common and bothersome, but there are currently no specifically licensed treatments, and studies thus far have used different methodologies to measure sleep quality. Among those, digital health technologies (DHTs) present an innovative approach that supports patient-centric drug development by providing insights into how a patient responds to treatment in real-world settings. DHTs therefore may offer a solution to provide unobtrusive objective measurement of SDM. Here we describe the joint development of a novel DHT-derived endpoint for assessing sleep quality in menopausal women through a collaborative approach from evidence generation to analytical, clinical, and usability validation based on regulatory guidance.

Methods

To demonstrate the fit-for-purpose of the novel DHT-derived endpoint, Bayer (drug developer), Sleepiz AG (DHT provider), and DEEP Measures (collaboration platform provider) partnered and applied established frameworks to leverage prior work while compiling comprehensive data, conducting a gap analysis, and curating evidence in the DEEP Measures collaboration platform based on and in preparation for discussions with health authorities. Initial regulatory feedback from health authorities provided useful input and supported the study design on the incorporation of the DHT-derived endpoint into the clinical development program of elinzanetant. Through collaborative efforts between the drug developer and the DHT provider, the novel DHT-derived endpoint (Sleepiz One+ for continuous, home-based measurement of wake after sleep onset in SDM and other sleep parameters) was implemented as an exploratory endpoint in a phase 2 pilot study where data to demonstrate fit-for-purpose were generated and validated against polysomnography, the gold-standard objective measure for sleep. The study outcomes alongside the results of the gap analyses and leveraging prior work were then structured systematically in the DEEP Measures platform. Data were organized according to the DEEP Stack model (which included information on the measurement definition, target solution profile, and instrumentation), and these facilitated the integration of our outputs directly into the regulatory package used for following health authority interactions to drive the acceptance of the novel endpoint.

Conclusion

We outline how various stakeholders collaborated to leverage prior evidence, interacted with regulatory authority, and incorporated a novel DHT-derived endpoint into clinical development programs. Evidence and data generated in the present project have the potential to build the basis for further endpoint and DHT development and validation.

Introduction

Digital health technologies (DHTs) enable remote and near real-time data collection to objectively measure clinically meaningful aspects of health. The use of DHT in clinical research supports a patient-centric approach to drug development by providing insights into how a patient feels, functions, and responds to treatment in real-world settings, thereby removing the reliance on artificial environments. Understanding these advantages, remote data collection with DHTs has become increasingly prevalent [1, 2]. However, despite their increasing use and the existence of theoretical frameworks and health authority guidance [3–7], knowledge on their real-life application in clinical development programs remains limited.

DHT-derived endpoints need to be developed and validated prior to use in clinical research. In practice, this means that the DHT needs to show its fit-for-purpose and ability to measure the endpoint of interest and that the measurements captured with the technology are accurate and reliable. The clinical relevance and suitability of the endpoint for assessing the outcome of interest also need to be demonstrated. Established frameworks such as the V3+ (verification, analytical validation, clinical validation, usability) [3, 4] and leveraging prior work [5] can be used to demonstrate the fit-for-purpose of a DHT in clinical development programs. Health authorities, including the Food and Drug Administration (FDA) [6] and European Medicines Agency (EMA) [7], have also published guidance on the use of DHTs in clinical investigations. We therefore aimed to implement these existing frameworks and guidance and explore the development of a novel DHT-derived endpoint for sleep disturbances associated with menopause (SDM).

Sleep disturbances are among the most disruptive symptoms experienced by roughly 60% of menopausal women [8, 9]. Currently, no therapies are licensed specifically for the treatment of SDM [10], representing an unmet medical need. Elinzanetant (BAY3427080) is a dual neurokinin (NK)-targeted therapy (NK-1 and NK-3 receptor antagonist) approved in several countries for the treatment of menopausal vasomotor symptoms [11]. In addition to its efficacy in vasomotor symptoms, elinzanetant has demonstrated improvements in sleep disturbances captured with patient-reported outcome (PRO) tools in phase 2/3 studies [11, 12], which prompted the initiation of the phase 2 NIRVANA study to further explore its impact on SDM [13].

To assess the efficacy of elinzanetant in sleep, a robust and consistent endpoint is needed; yet there is no commonly accepted endpoint for SDM in existing literature, and studies to date have used varying methodologies and captured different parameters. Although there are established and accepted primary efficacy endpoints, e.g., wake after sleep onset (WASO), in the space of insomnia [14–16], they have not been used as primary endpoints to measure SDM, thus highlighting the need to translate these endpoints into a new context of use (CoU) or potentially develop a novel primary endpoint for SDM.

PROs and polysomnography (PSG) are generally considered the gold standard for the assessment of sleep parameters in drug trials [17, 18]. Previous elinzanetant trials have used PROs in the form of questionnaires and/or electronic dairies, including the Pittsburgh Sleep Quality Index, Insomnia Severity Index, and Patient-Reported Outcomes Measurement Information System Sleep Disturbance Short Form 8b [11, 12]. Other studies have also applied similar PROs, albeit with varying durations or intervals of measurement [19]. PROs are subjective measures that collect information directly from patients on how they feel and function. Due to their ease of administration, PROs have the advantages of high patient compliance and reduced demand on specialists’ time [20]. They also help enrich the understanding on unique patient experiences and enable assessment of symptom bother from the patient’s perspective [21]. However, their subjective nature inherently has the potential to lead to inter-individual variability and high placebo effects [20].

By contrast, PSG has been used in clinical trials of other treatment modalities for SDM to objectively assess sleep parameters [22, 23]. PSG is performed at a sleep laboratory in a clinical center and utilizes multiple sensors worn across the body to collect data on biomarkers of sleep quality, such as total sleep time, sleep onset latency, and WASO. While the measurements are accurate, PSG is limited by its reliance on an artificial sleep environment that does not mimic the actual home environment, making it uncomfortable and not reflective of patients’ everyday sleep. It is also laborious and costly to administer. These factors limit its use in large multicenter phase 2/3 clinical studies that last over a prolonged period [20, 24].

Although the tools mentioned above are all common and well validated to evaluate sleep, the heterogeneity in their use and the lack of a standardized endpoint in SDM mean that it is often difficult to compare efficacy between treatments across studies and determine the true value of medication. Besides, there is currently limited regulatory guidance on efficacy parameters suitable for clinical trials involving women with SDM. While the FDA has issued recommendations for endpoints to be assessed in general sleep trials [25], work is needed to determine parameters specifically applicable in the context of SDM. Clinical studies using objective measures have identified WASO and nighttime awakenings as key aspects of sleep disturbances in peri-/postmenopausal women, while parameters like total sleep time appear less relevant [26–28]. More importantly, women with SDM tend to have high night-to-night variability in these parameters due to nightly fluctuations in sleep quality and characteristics [10, 29]. Hence, objective, continuous, and long-term monitoring of sleep in a natural environment is required to assess the efficacy of a novel treatment in this population.

We propose that a DHT in the form of a home-based sleep monitor may serve as a promising solution to this unmet need due to its convenient at-home use and ability to consistently measure sleep parameters over extended time. Considering this, Bayer (drug developer), Sleepiz AG (DHT provider), and DEEP Measures (collaboration platform provider) partnered to develop a novel digitally derived endpoint to measure sleep quality in menopausal women, with the Sleepiz One+ medical device being the DHT of choice. To demonstrate its fit-for-purpose, the drug developer and DHT provider have worked together to compile comprehensive V3+ data, conduct a gap analysis, leverage prior work, curate data/evidence in the DEEP platform, and hold discussions with health authorities. This article presents the joint development and the process to seek acceptance of this novel DHT-derived endpoint to support future trials in this CoU.

Methods

Choosing the Appropriate DHT for a Novel Digital Endpoint

The drug developer chose Sleepiz One+ (Sleepiz AG, Zürich, CH) among other available DHTs after conducting two rounds of landscape assessment. First, a global landscape analysis identified candidate devices, taking into account all technology types and forms including home PSG equipment, polygraphy equipment, wrist bands and other wearables, touchless mattress sensors, and sensors to be placed in the sleep environment. The second round assessed how well each device met the defined requirements according to consensus among the clinical study team. Key requirements were the ability for continuous home-based measurement of multiple sleep parameters such as total sleep time, WASO, and sleep efficiency; ease of use by patients; touchless operation; and evidence from past trials. Additionally, the device needed to be medical grade and approved by health authorities in at least one indication to facilitate efficient validation in the newly chosen CoU (SDM). For this same reason, a partner experienced in device validation and working with health authorities was preferred.

A consensus was reached among the study team that Sleepiz One+ satisfied all of the key requirements. Sleepiz One+ is a touchless home sleep monitor that enables remote data capture using radar technology to measure and analyze body movements for an unobtrusive assessment of vital signs and sleep. The hardware emits low-power electromagnetic waves that reflect from the human body to detect changes in the frequency and phase with sub-millimeter accuracy [30]. The device-associated software includes algorithms that use signal processing and machine learning to translate this information into sleep parameters (e.g., WASO, number of awakenings, latency to persistent sleep, sleep efficiency, total sleep time). Sleepiz One+ complies with regulatory requirements for medical devices in the USA and European Union (EU): in the USA, it is FDA 510(k) cleared as a vital signs monitor; in the EU, it is CE-marked (class IIa) under the Medical Device Regulation for vital signs monitoring and home sleep testing. As per its labeled CoU, Sleepiz One+ has been utilized in the past for sleep apnea screening [30, 31] and vital signs monitoring [32].

Leveraging Prior Work for the Use of the Novel DHT-Derived Endpoint in a Clinical Development Program

An initial development plan for the digitally derived endpoint was jointly formulated by the drug developer and DHT provider. To determine potential gaps in verification, analytical validation, clinical validation, and usability, we applied the concepts from the FDA Guidance for Industry: Digital Health Technologies for Remote Data Acquisition in Clinical Investigations, which provides recommendations for ensuring a DHT is fit for use in a clinical development program, including the level of validation required and the interpretability of data based on the DHT’s intended purpose and CoU within an investigation [6]; and from the framework for leveraging prior work, which informs on best practices to develop and validate DHT-derived endpoints using existing verification, analytical, and clinical data, thereby accelerating evidence generation and facilitating consistency across development programs [5] (Table 1).

Table 1.

Overview of gap analysis: leveraging prior data to support the use of a touchless home sleep monitor to assess sleep quality in menopausal women with sleep disturbances

Scenario	Verification	Analytical validation	Usability assessment	Clinical validation
Considerations: medical device status, intended use scope, endpoint status	Is the DHT accurate, precise, consistent across time and uniform across different environmental bench testing conditions?	Does the DHT accurately, reliably, and precisely generate the intended technical output from the input data? Is the data flow defined and validated?	Can the intent-to-treat population of the clinical trial use the DHT? What is the patient burden? Are usability studies needed?	What is the context of use (CoU)?	Does the measure identify or predict a meaningful clinical, biological, physical, functional state, or experience?
Best practice based on a theoretical and fictional case (insomnia) presented in the concept of leveraging prior work [4]
Measuring a novel endpoint within an authorized device label	Prior verification data can be leveraged since the device is CE-marked and FDA-cleared. The verification data that supported the marketing authorization should provide the information needed for verification	Prior analytical validation can be leveraged since the device is being used within its labeled intended use	Usability can be implied, and additional testing is not needed because the labeled intended use covers the intent-to-treat population of the clinical trial	The sponsor needs to confirm that the CoU in the clinical investigation matches the labeled intended use of the authorized device	If available, prior clinical validation may be leveraged since the device is designed to measure the same sleep parameters in the same setting as the intended use of the authorized device
Use of a cleared/approved medical device within its labeled intended use to measure a novel endpoint
				If the CoU for the device is patients with insomnia disorder, any existing clinical validation data that supported the device marketing authorization could provide some of the information needed for clinical validation of the DHT for use in the clinical investigation	However, depending on the analysis plans (e.g., more frequent data sampling compared to polysomnography or lower sensitivity parameters for wakefulness detection), the sponsor needs to generate additional data to justify that the measure predicts a meaningful clinical impact in the stated CoU
Practical case – developing a novel DHT-derived endpoint in sleep quality using a touchless home sleep monitor (case presented in this manuscript)
Description of novel endpoint and authorized device label Using Sleepiz One+, which is 510(k) cleared for the at-home assessment of vital signs and EU MDR Class IIa certified for vital signs and home sleep testing, to measure parameters of sleep quality (e.g., WASO, SE, TST) in subjects experiencing sleep disturbances, including menopausal women (CoU)	Verification data were generated by the DHT provider and can be leveraged. The following data sets are available: Hardware and software verification and validation Risk management file Electrical safety and electromagnetic compatibility compliance with international standards Declaration of conformity	Analytical validation data of the device manufacturer will be leveraged	Human factor/usability testing data are available from pre- and post-market surveillance studies in a general population	The clinical validation conducted by the manufacturer to achieve 510k clearance and EU CE-mark in the general adult population and patients suffering from sleep apnea will be complemented by data generated during the head-to-head comparison of the Sleepiz One+ device with polysomnography for sleep quality parameters in the SDM population during the NIRVANA phase 2 PoC study (NCT06112756)	Data on longitudinal response to therapy will be generated during the controlled, randomized NIRVANA phase 2 PoC study (NCT06112756) comparing the device with standard polysomnography, alongside the assessment of the drug candidate's efficacy and safety against the standard of care
			However, the symptoms experienced by the population of interest might potentially reduce their willingness to use the device
			To confirm device acceptance, supplementing existing tests in adults and sleep apnea patients, additional usability data could be collected from menopausal women with sleep disturbances

CE, Conformité Européenne; CoU, context of use; DHT, digital health technology; EU, European Union; FDA, Food & Drug Administration; MDR, Medical Device Regulation; PoC, proof of concept; SDM, sleep disturbance associated with menopause; SE, sleep efficiency; TST, total sleep time; WASO, wake after sleep onset.

The framework for leveraging prior work comprises six scenarios which consider the medical device status, intended scope of use, and endpoint status [5]. Briefly, the scenarios are (1) measuring a validated endpoint within an authorized device label, (2) measuring a validated endpoint outside an authorized device label, (3) measuring a novel endpoint within an authorized device label, (4) measuring a novel endpoint outside an authorized device label, (5) measuring a validated endpoint with a new DHT, (6) measuring a novel endpoint with a new DHT. In this case, we aimed to use the Sleepiz One+ within its label and expand on its CoU by developing a novel DHT-derived sleep quality endpoint applicable in trials evaluating SDM; specifically, it was incorporated into the phase 2 NIRVANA trial of elinzanetant to be used in parallel with PSG and continuously at home to measure sleep parameters [13]. We therefore determined that scenario 3 of the framework was applicable. A summary of this assessment is shown in Table 1, where the practical case section highlights the specific development needs for our novel DHT-derived sleep quality endpoint.

Verification and Analytical Validation

The verification phase ensures that the device is built to its design specifications. As safety aspects are defined as design requirements for medical devices, verifying their compliance with standards confirms the design specifications have been met. Analytical validation ensures that the signals collected by the device are appropriately handled by the algorithm to derive physiological metrics (i.e., WASO), and that these are verified against well-defined reference standards in a representative sample. As previously identified, Sleepiz One+ was chosen based on its existing approval by health authorities for use in sleep apnea screening and vital signs monitoring. This means that verification and validation data in these CoUs were already available from the device manufacturer and could be leveraged to demonstrate the performance of the device ahead of use in the current case example [30–32].

Usability Assessment

Similarly, robust usability assessment data from pre- and post-market studies in general and sleep apnea populations were available and could be leveraged. However, while these populations included women of menopausal age, a potential gap was identified since postmenopausal women with or without SDM may differ in their symptom experiences, which in turn influence their willingness to use Sleepiz One+. Additional usability testing in a small subset of women with SDM would further support the usability of the device in this CoU; this would be collected following initial successful validation of Sleepiz One+ in the NIRVANA trial.

Clinical Validation

During initial planning of the elinzanetant clinical development program, the drug developer conducted desk research and structured interviews with both women and healthcare professionals to identify the most relevant outcomes and preferred modalities for measuring SDM in women. Results revealed that nighttime awakenings, WASO, and insufficient sleep, among others, were important outcomes related to SDM [10] and therefore justified the use of WASO as our DHT-derived endpoint. This was followed by clinical validation where the study team compared sleep parameters measured by Sleepiz One+ with those by PSG – the current gold standard to capture objective outcomes in sleep studies [18]. Key aspects of clinical validation in this new CoU included assessing the device’s ability to detect wake and sleep epochs correctly (i.e., the sensitivity and specificity of Sleepiz One+), test-retest quality, the ability to detect changes in sleep parameters induced by treatment, as well as the bias of measurements on a parameter level (parameters as compared to PSG). Clinical validation data were planned to be generated in the NIRVANA trial after discussion with regulatory health authorities. Additional work is also ongoing to determine minimum clinically important differences using the DHT-derived endpoint. Outputs from interviews with women and healthcare professionals as well as data from the NIRVANA trial would be analyzed to identify appropriate thresholds for the interpretation of change [33].

Interactions with Regulatory Health Authority

A pre-investigational new drug application type B meeting with the FDA was held to discuss the clinical development program evaluating the use of elinzanetant in women with SDM. For the purposes of this article, we focus on discussions surrounding the novel digitally derived sleep endpoint. In the briefing package to the FDA, we included the rationale for this novel endpoint generation, namely, to allow us to maintain a natural sleep environment and so eliminate the influence of artificial environment on sleep patterns, as well as to address night-to-night variability in sleep parameters through longitudinal measurements with low participant burden. With Sleepiz One+ meeting these requirements, we also provided a summary of the regulatory status of the DHT and the advantages of touchless home sleep monitoring. The FDA provided encouraging and insightful feedback on incorporating this DHT-derived endpoint into the clinical development of elinzanetant.

The FDA initially suggested an alternative of using type II at-home PSG devices to measure sleep endpoints since they may already have the validation data needed to demonstrate fit-for-purpose. They further added that if Sleepiz One+ were to be used in future elinzanetant trials, there needs to be sufficient data to support its use and validity, along with the validity of all endpoints derived from it. Ultimately, we preferred a DHT-based solution since the continuous and contactless nature of Sleepiz One+ better aligns with our objective of measuring sleep in a natural environment over a longer period of time, and this was accepted by the FDA as a way forward. They also agreed that the necessary data to demonstrate its utility, accuracy, and fit-for-purpose in women with SDM could be collected through the NIRVANA trial. Specifically, the primary endpoint of the trial, WASO, was assessed using PSG in a certified sleep laboratory. In parallel with the onsite PSG, Sleepiz One+ was used in the sleep laboratory and at home for continuous measurement of WASO, with its data evaluated as an exploratory endpoint. Provided that the data demonstrate fit-for-purpose for the DHT-derived endpoint, the drug developer plans to further engage with the FDA to discuss the use of Sleepiz One+ instead of PSG in potential future studies. Preliminary data from the trial indicated that epoch-level comparison and parameter-level results showed good agreement between Sleepiz One+ and PSG in postmenopausal women with SDM [34].

Utilizing DEEP to Collate Evidence and Facilitate Discussions

Role of DEEP Measures in DHT-Derived Endpoint Development

DEEP Measures is a collaborative platform used to support the end-to-end development of the DHT-derived endpoint. During initial planning, the drug developer identified the need to seek a third party with expertise in DHT and endpoint qualification that could offer a platform to collate data from both the drug developer and device manufacturer and assess data gaps. DEEP Measures was identified as a partner with such capabilities while having the additional experience in navigating regulatory approval of DHT-derived endpoints with health authorities.

Figure 1 provides an overview of the workflow between the drug developer, DHT provider, and DEEP to achieve endpoint acceptance. The DEEP Measures platform provided a common space for stakeholders to build a development strategy and identify key characteristics of the proposed DHT, work through the validation process (including pulling in existing evidence from other indications available on the platform), and prepare the necessary documentation to establish regulatory acceptance of the DHT.

Fig. 1.

Workflow between stakeholders in this use case utilizing the DEEP platform for collaboration. CE, Conformité Européenne; CoI, concept of interest; CoU, context of use; EU, European Union; FDA, Food and Drug Administration; MAH, meaningful aspect of health; MDR, Medical Device Regulation.

Developing a Target Solution Profile and Curation of Available Evidence for Validation

The first step of the process involved translating the initial conceptual framework into a target solution profile (TSP). A TSP outlines the desirable and essential characteristics for a specific modality (e.g., actigraphy-based wearables, radar- or voice-based touchless devices) to measure a specific concept of interest (CoI) within a CoU. It describes details of the measurement method profile, raw data profile, algorithm profile, performance requirements, and supported health data variables, as well as any available regulatory intelligence and acceptance details [35]. TSPs with sufficient evidence and validation can be used as standards, which in turn enable reusability of evidence for different CoUs.

Once the TSP was developed, all available evidence and literature supporting this profile were curated according to the Stack model in the cloud via a secure shared workspace. In our case, evidence from the validation of Sleepiz One+ in vital signs monitoring [32] and sleep apnea assessment [30, 31] was collated next to technical documentation of the DHT including verification, validation, and usability data. The DEEP catalog, structured according to the DEEP Stack model [35], supported the implementation of V3+ and allowed us to conduct a systematic gap analysis identifying prior data that could be leveraged and potential evidence gaps. Moving through the validation process, we used the DEEP platform’s functionality to integrate evidence directly into documents such as regulatory meeting packages and briefing books. Specifically, the DEEP Stack model organizes information into three blocks: measurement definition block, TSP block, and instrumentation block. Figure 2 illustrates a general schematic of the elements included in each block of the DEEP Stack model that supported our development of a novel DHT-derived endpoint for SDM.

Fig. 2.

Overview of the DEEP Stack model concept and elements included. CoI, concept of interest; CoU, context of use; DHT, digital health technology; MAH, meaningful aspect of health; TSP, target solution profile.

Outcomes Generated from DEEP

In line with the workflow shown in Figure 1, collaboration partners first conducted a profiling workshop to understand the needs of the drug developer for this specific use case. Results from this workshop were then translated into a TSP for the modality of touchless home sleep monitoring. Details of the TSP, including the determined essential and desirable requirements of the measurement modality, are shown in online supplementary Table 1 (for all online suppl. material, see https://doi.org/10.1159/000549984). Importantly, the platform also facilitated the initial landscape analysis of all available DHTs, which contributed to the subsequent identification of Sleepiz One+ as our DHT of interest. Throughout this process, there was no particular preference toward any DHT, as the most important factor was to have a contactless device with sufficient accuracy that could assess sleep patterns with minimal disruption to clinical trial participants’ normal sleeping habits. Approaching profiling in this manner provides the opportunity to evaluate a variety of sensors that could be used against this TSP. In the end, Sleepiz One+ was confirmed as the solution which best fulfilled our identified requirements.

After developing the TSP and identifying existing evidence for the DHT, the curation of the Stack model began with DEEP entering the agreed CoI, meaningful aspect of health, and CoU in the measurement definition block; followed by the drug developer and DHT provider entering information into the TSP and instrumentation blocks (refer to the elements shown in Fig. 2). Examples of information included in each block of the Stack model are provided in online supplementary information.

For our case, the key discussions in the measurement definition block were the positioning of the meaningful aspect of health (i.e., nightly sleep quality) and CoI (i.e., WASO) based on the available evidence and literature, potential gaps in clinical interpretation of the measure after the completion of the NIRVANA trial, and considerations for cultural adaptation (e.g., differences in sleep habits among different cultural and ethnic groups). These discussions were also supported by industry expertise and publicly available knowledge that had been added to DEEP from the development of other digital measures by other stakeholders [36, 37]. In the instrumentation block, information about the algorithm (e.g., flowchart of the data processing steps), data pipeline (how data are transferred from the DHT and where they are handled throughout their lifecycle), satisfactory sleep/wake detection, and performance within different sleep arrangements (e.g., partner in bed, pets in bed) were identified as critical discussion points at the next regulatory health authority engagement.

Discussion

We acknowledged the current limitations in assessing SDM in women and worked to develop a novel DHT-derived endpoint using the Sleepiz One+ to enable convenient and continuous at-home monitoring of sleep quality over several weeks. The present case illustrates how drug developers, DHT manufacturers, and collaboration platform providers can effectively work together across industry boundaries to put theory, guidance, and frameworks into practice. We describe the joint development of a novel DHT-derived endpoint, utilizing the home-based sleep monitor Sleepiz One+, to assess sleep quality in women with SDM. In particular, we highlight the processes of leveraging prior evidence, engaging with health authority (FDA), and using the collaborative DEEP platform to facilitate discussions. The development of this endpoint puts existing frameworks in practice and demonstrates the potential of incorporating DHTs into clinical investigations. In the context of SDM, the novel endpoint holds promise in enabling continuous, natural assessment of sleep parameters over extended time. A home-based sleep monitor may also be advantageous over PSG, particularly in this population, as it removes the need to rely on artificial environments which may be disruptive to women’s sleep. Findings from the NIRVANA trial will therefore serve as an important milestone to validate the utility and fit-for-purpose of the DHT, allowing it to be incorporated into not only the elinzanetant development program but also potentially other sleep trials in the future.

Evidence and data generated in the present project have the potential to build the basis for further endpoint and DHT development and validations. Supported by a legal framework agreed to by all stakeholders, the drug developer, DHT provider, and collaboration platform provider have the ability to extrapolate from this case to support potential future projects. For example, pending outcomes of the NIRVANA trial, further interactions with regulatory authorities regarding the use of the novel digitally derived endpoint in future elinzanetant trials may be planned. DEEP would be used to create the meeting briefing package. Afterward, feedback from regulatory authorities would be added to the Stack model to continue strengthening the evidentiary base for the use of the DHT in the elinzanetant development program. Based on data generated in the trial, the drug developer and the DHT provider plan to seek acceptance of the novel DHT-derived sleep quality endpoint to support future trials in this CoU. In addition, learnings from developing this novel DHT-derived endpoint could be translated into other CoUs where sleep parameters may serve as meaningful endpoints, e.g., to monitor diseases and therapeutic interventions with the DHT in other drug trials. Further, findings could also be leveraged to broaden the applications of the DHT within its intended use for vital signs monitoring in clinical development programs where the CoIs involve continuous measurements of vital parameters.

Other potential next steps could include the use of the DEEP platform, with agreed-upon legal frameworks between stakeholders, to support a public qualification of the novel endpoint, whereby validation data and regulatory opinion are made public, meaning that any company can use the validated endpoint in their own studies. For example, in the past, DEEP and the EMA piloted an innovation task force meeting which focused on qualifying a novel digitally derived endpoint to assess nocturnal scratch in atopic dermatitis treatment and involved stakeholders across the pharmaceutical industry (Leyens, L. et al., 2026, manuscript in preparation). Further potential next steps could include Sleepiz AG working with various drug developers in different indications to qualify Sleepiz One+ for other novel sleep and vital signs endpoints and facilitate its use in measuring established endpoints in other drug development programs. The generated data added to the DEEP Stack model by the DHT provider may support other drug developers to accelerate their access to the sleep and vital signs endpoints digitally derived by the DHT. Finally, DEEP could launch projects (supported by agreed upon legal frameworks) in other therapeutic areas where additional or different stakeholders, from industry and academia, could contribute to and build on the existing sleep data, device knowledge, and other evidence.

Acknowledgments

The authors thank the elinzanetant clinical development team and the DEEP team for their support. Editing and medical writing support was provided by Gavin Chiu and Jessica Sampson-Taylor of Highfield Communication, Oxford, UK.

Statement of Ethics

An ethics statement was not required for this work since no human or animal subjects or materials were used.

Conflict of Interest Statement

The authors declare no nonfinancial competing interests but the following financial competing interests. F.K., J.P., K.O.’S., and A.B. are employees and shareholders of Bayer. J.K. and M.S. are employees of Sleepiz AG. J.B. and P.Y. are employees of DEEP Measures. The views on the topics discussed in this publication are those of the authors and do not express the positions of the respective companies.

Funding Sources

The NIRVANA study was funded by Bayer Pharmaceuticals AG, Germany, who was responsible for the study design, execution, and analysis. The Bayer employees named above were involved in manuscript conception, planning, writing, and the decision to publish the present article.

Author Contributions

F.K., J.P., K.O.’S., and A.B. provided the drug developer perspective; J.B. and P.Y. provided the DEEP perspective and developed Figure 2; J.K. and M.S. provided the device manufacturer’s perspective. F.K. and J.K. developed Table 1. F.K., J.K., J.B., and P.Y. developed Figure 1. All authors contributed to the conceptualization and writing of the manuscript and online supplementary information, revised the manuscript critically, approved the final version to be published, and are accountable for all aspects of the work.

Funding Statement

The NIRVANA study was funded by Bayer Pharmaceuticals AG, Germany, who was responsible for the study design, execution, and analysis. The Bayer employees named above were involved in manuscript conception, planning, writing, and the decision to publish the present article.

Data Availability Statement

All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.