IQVIA Holdings Inc. (IQV) Earnings Call Transcript & Summary

Hello, and welcome to today's IQVIA webinar where our speakers will be discussing capturing patient experience data or PED in clinical trials. My name is Kristin Mabry, and I will be your moderator for today's webinar. Before we begin, I'd like to review some key audience steps to give you the best webcast experience. [Operator Instructions]. We will try to answer these during the webcast but if a fuller answer is needed or if we run out of time, it will be answered later via e-mail. If you experience technical difficulties, try refreshing your browser by selecting F5 on your keyboard. You can also use the Help Widget for online troubleshooting, which covers common technical issues. You can expand your slide area by clicking on the maximize icon on the top right slide area or by dragging the bottom right corner. Please provide your feedback about today's webinar via the survey widget. Your feedback is important to us as it helps us to further improve future webinars. An on-demand version of the webcast will be sent to you within 1 business day from the conclusion of the webcast and can be accessed using the same audience link that was sent to you earlier. Now I'll turn it over to Stella to introduce our speakers and begin the presentation.

Thank you, Kristin. Hello, everyone. I'm Dr. Stella Karantzoulis, Senior Principal and Global Central Nervous System, or CNS Practice Lead. Today, I'll be joined by my colleagues from IQVIA's Patient Centered Solutions team. Joy Whitsett, she's Regulatory Director; and our Regulatory Center of Excellence Lead and Dr. Anna Krasnow, Associate Principal and our EMEA CNS lead. I'm very pleased to be part of a webinar that focuses on generating patient experience data in CNS disorders. I've spent close to 20 years now working directly with patients living with a variety of CNS disorders, spanning both neurology and psychiatric indications in both clinical and research settings. Our Patient Centered Solutions team are experts in understanding the patient experience and in developing strategies for how to best capture the patient's voice in clinical trials to support successful outcomes. And I'm pleased to say that we have a dedicated team in the U.S. and in EMEA that focuses almost exclusively on doing this work in CNS disorders. For today's webinar, Joy will provide us with a brief introduction into patient experience data or PED, what that is and how to collect it. And she's also going to talk to us about self-reporting factors to consider and she'll explore strategies in capturing PED when there are challenges to self-reporting. Anna will then walk us through select case studies focused on managing barriers to self-reporting and optimizing the quality of PED involving 4 different indications with CNS involvement. We'll then conclude this webinar with a Q&A session. So please don't forget to submit your questions via the Q&A widget. Okay. So let's go ahead and jump in. Joy the stage is yours, take it away.

Thank you, Stella, and hello. I'm Joy Whitsett. I serve as a Regulatory Director in the IQVIA Patient Centered Solutions team and lead the PCS Regulatory Center of Excellence. I'm a quality and regulatory professional and I've been working in drug development for the past 20 years and have always had a passion for the integration of the patient voice in what we do. I started my career in cGMP drug product manufacturing and then moved to clinical quality and regulatory after about 10 years in the industry. And whether it be ensuring that the products we released for clinical trials met the highest quality standards to protect patient safety, or the clinical data we collect, informed decisions about meaningful benefit, I strongly believe that the concerns and safety of the patient come first. I joined the IQVIA PCS team a little over 4 years ago, where I've had the opportunity to work directly with outstanding scientists and statisticians that are truly experts in patient experience research. And my role is to provide leadership for our team and clients on the regulatory aspects and utility of patient experience data. I'm excited to be here today to discuss our thoughts and some pragmatic approaches to managing challenges in capturing patient experience data in CNS clinical trials. Today, also at the stage with the regulatory framework for patient experience data and my colleagues, Dr. Krasnow and Karantzoulis will follow with practical information and examples of how we've overcome challenges to collect patient experience data in CNS treatment areas. While the use of patient experience data to evaluate the efficacy of new medical products is not new and measurement science is a well-established field, the term patient experience data was formerly defined in the U.S. with the 21st Century Cures Act. And FDA has led the way in providing the series of patient-focused drug development guidances that lay out a framework for how to best collect patient experience data for regulatory decision-making. There are 4 PFDD guidances, but for our discussion today, I'll briefly lay out some information from PFDD-1, 2 and 3 that apply to these questions. What exactly is patient experience data? And how does it differ from other data that are collected in clinical research? How do you collect patient experience data? Where does patient experience data originate from? Is it always the patient? And what self-report factors we can consider and what barriers exist for patients to self-report? And how can we work to overcome those barriers and still represent the patient voice? So starting with the beginning, what is patient experience data? The formal definition of patient experience data are data that, 1, are collected by any persons, including patients, family members and caregivers of patients, patient advocacy organizations, disease research foundations, researchers and drug manufacturers; and 2, are intended to provide information about patients' experiences with a disease or condition, including, A, the impact, including physical and psychosocial impacts of such a disease or condition or a related therapy or clinical investigation; and B, patient preferences with respect to treatment of the disease or their condition. I know it's easy to get lost in the formality of this definition, and it can seem aspirational, but stay with me. Patient experience data is different from other data that are collected in clinical trials like blood tests, imaging or AE data because it's information on the experience with the disease and treatment symptoms, and it can't be measured by something like an MRI. It Includes the patient's impressions and relative importance of symptoms and impacts, and it extends to their treatment preferences. We'll cover this more later, but an example in CNS is agitation in patients that have traumatic brain injury. Patient experience data provide insight that this symptom is prevalent and important. The symptom may also be important to measure in clinical trials and patient experience data on frequency and severity of agitation can inform the development of new treatments that could provide a benefit that truly matters to the patient. So how do you collect patient experience data? When you're developing a strategy to collect patient experience data, start with your research question. Are you interested in understanding the patient journey? Or are you interested in understanding treatment effects. For research questions that are focused on understanding the patient journey, the methods outlined in patient-focused drug development guidance 2 are useful to understand what is important to the patient. Effective methods include qualitative, quantitative and mix methods. For a qualitative method, these are typically one-on-one interviews or focus groups and are used to understand in-depth information about patients' experiences, perspectives, priorities and their feelings. Quantitative methods are typically a research survey or questionnaire that can collect quantifiable or like numerical data that can be statistically analyzed to understand the patient experience. And mix method uses both. It's a combination of qualitative and quantitative data that are coming from the same study or a same set of related studies and that are integrated in a kind of predefined approach. So this might be something like a survey with open and fixed response questions or interviews that also include a survey with fixed responses. So again, when deciding what method to use to obtain patient experience data, it's important to think about your research question. And I've added some example scientific questions here for each methodology. For a qualitative method in CNS would be, what are the most important disease-related symptoms and impacts of age-related macular degeneration? And for quantitative methods, maybe it's [Technical Difficulty] treatment-related symptoms in Parkinson's disease. And for a mixed method, maybe we think about what is the patient experience with ALS, where we use interviews with patients to understand qualitatively what their experiences and then follow with the symptoms checklist, where both are analyzed separately and compared. Patient experience data can also be collected as outcome data using clinical outcome assessments. These data are incredibly useful to understand the treatment benefits, risks and tolerability and are often built into clinical trial endpoint strategies. The COA types that are outlined in patient-focused drug development guidance 3 are patient reported outcomes, clinician-reported outcomes, observer-reported outcomes and performance-reported outcomes. PROs are data that come directly from the patient. So these data would be coming from a questionnaire. It's a question that is intended to be only responded to by a patient. A clinician reported outcome is performed by a medically trained professional. And it's really important to incorporate rater training when you're using clinician-reported outcomes to ensure consistency between raters. An observer-reported outcome is something that's completed by a teacher or a parent that's not the patient, but that somebody that has frequent contact with the patient that can report on the things that they observe. And performance-rated outcomes are really interesting. This is an emerging field. We've often -- we've had the 6-minute walk test for -- we're measuring the task that the patients are performing, but this is also an emerging area where wearables and digital health technologies are being incorporated into endpoint strategy. COA by far, the most common form of patient experience data submitted to support regulatory decision-making on treatment effects. At IQVIA, a large part of our work is to help our clients select, modify or develop COAs that are appropriate for their clinical development programs. Moving now to the question of where patient experience data comes from and selecting a reporter? If at all possible, we look to the patient themselves to provide their perspective as they are the true experts in their disease. The questions to consider, especially in CNS, however, is the extent to which the patients can reliably and validly self-report. And some factors that are outlined in PFDD-1 from the FDA guidance series to take into account with your decision are listed here. What is the level of cognitive development function or mental status of the patients in this population. Are there any concerns that should be taken into account for language skills and numeracy skills, especially if you have a quantitative scale like selecting value between 0 and 10. What is the health literacy in the patient population? And what is their basic literacy? What are the health state and comorbidities that should also be taken into account when you're designing these studies. Also very important to engage experts to help with the determination of a reporter as they have experience in the treatment area and have seen lots of patients with different severities. This is especially important when the potential for patient report may be limited or compromised. A few populations where the ability to self-report needs to be examined closely are pediatric, those that may have full-time caregivers to assist with daily tasks, folks that have different capabilities such as vision or communication and language and cultural differences. Alternate methods that may be considered to overcome barriers to self-report might include interviews with shorter segments in pediatric populations to manage limited attention spans, or maybe you ensure that you conduct some usability assessments for materials that might have adjustable font or computerized tasks to overcome vision impacts. Caregivers can actually be really helpful to provide perspectives or [Technical Difficulty]. But it's important to have a good plan documented ahead of time for how they will be incorporated. And of course, language differences need to be addressed, but not just with translation, also it's important to incorporate cultural adaptation to ensure the reliability of the data. Let's shift now to those situations where we've looked for overcoming barriers to self-report but determined that the patients are unable to reliably self-report. In these cases, patient partners and clinicians can be incorporated into a strategy to collect patient experience data on both the patient journey and outcomes. They can inform the understanding of the journey using the qualitative and quantitative methods that we talked about before, and they can also inform outcomes with other types of COAs, like ClinROs and ObsROs. Also, performance outcomes can provide helpful data like we've discussed before. One caution when collecting patient experience data from reporters that are not the patient is to avoid proxy reporting. People that are not the patient are reliable reporters of behaviors or results of symptoms that they can observe but they cannot reliably report on behalf of the patient on concepts that can only be known to the patient. Such as maybe the level of itch or pain. We can and should incorporate data from patient partners to inform an understanding of the signs and impacts of disease and treatment, but not by having caregivers complete a PRO with what they think the patient would say, but instead by gathering information with an appropriately designed ObsROs. I'd like to now hand our presentation over to my colleague, Dr. Anna Krasnow, who will be presenting some case studies with our practical experience in collecting patient experience data in CNS clinical trials.

Thank you, Joy. Hello, everyone. My name is Anna Krasnow. I'm an Associate Principal and EMEA CNS lead in the Patient Centered Solutions team. I'm a neuro scientist by training and my current role, I oversee delivery of projects relating to clinical outcome assessments in neurology and psychiatry. In this section of the webinar, I will walk you for 4 case studies. Each of these provides different examples of best practices for overcoming barriers to self-report that Joy covered earlier. We will start with a case study on Fragile X syndrome were we encounter cognitive and communication challenges in the pediatric population with caregivers. Here, we'll learn how to leverage publicly available information to reduce burden on patients and caregiver communities. The next 3 case studies are based on research projects conducted by PCS team at IQVIA. In Case Study 2, we will discuss another pediatric population with caregivers. A rare form of epilepsy that results in cognitive and intellectual challenges. This case study is an example of collecting input from reporters other than patients. Case Study 3 involves participants with physical, sensory, intellectual and communication challenges that result from traumatic brain injury. Here, we learn some best practices for conducting patients and caregiver interviews in the context of generating evidence for an existing co-instruments. The final case study enrolls participants with a visual impairment. Here we show how barriers to self-reports can be overcome by adapting methodology. And let's dive in. The first step for understanding patient experience is the reviewed available [ infrastructure ]. The [indiscernible] reviews can be very effective for generating foundational evidence, especially in common conditions that have a rich body of evidence. However, it can be difficult to even know where to start when you are working with rare disease population. Here, evidence is limited, especially research using qualitative methods and it can be difficult to recruit patients for studies to understand their experience and priorities for treatment. Here, we use the example of Fragile X syndrome, which is a rare genetic condition that causes intellectual disability, behavior and learning difficulties and specific physical characteristics. The severity and manifestation of symptoms vary and tend to change throughout the lifespan. The condition is more common and more severe in men. In recent years, several Fragile X trails have failed and the patient and family community have become increasingly frustrated by the slowed progress of clinical development. One source of frustration is the feeling that they have not had a strong enough voice in providing a direction for new drug developments. For example, what outcome measures are used to determine their efficacy of treatments. To better understand the perspective and priorities of the FXS community, the National Fragile X Foundation hosted PFDD meeting in 2020, which was conducted in collaboration with the FDA. This meeting is an example of a larger initiative by the FDA called patient listening sessions. These sessions are requested by the agency or [ APAC ] to generate awareness and evidence of patient experience and specific diseases. They are a way of establishing a dialogue between the FDA and the patient community for the following objectives. To inform regulatory decision-making, to make patients and their advocates understand the work done by the agency to provide a starting point to inform early stage research and also to educate FDA staff about the unmet needs of patients and their extended communities. Prior to the meeting, the foundation implemented a large survey designed to investigate the main symptoms, daily living challenges, family impacts and treatment priorities for individuals with FXS and their families. Of the 467 survey participants, only 8 were patients. They were all females over the age of 13. And while their report is extremely valuable, it's unlikely to be representative of the full spectrum of patient experience. Overall, the survey indicated that generating comprehensive picture of the patient experience in FXS would require integrating information from multiple sources. This survey also identified some primary concepts of interest that were used to anchor the subsequent PFDD meeting. The meeting itself was attended by 226 participants, including members of the FXS community, representatives of biotech and pharma companies, especially those with relevant treatments in development, and members of government entities and professional community, including researchers, clinicians and patient advocacy groups. The meeting was structured as a combination of presentations from experts and guided focus groups discussions. The 2 main topics covered in the meeting included: Understanding the symptoms and burden of living with FXS from the perspective of patients; and what goals and priorities patients and families have for treatments, including what level of risk and side effects were willing to tolerate in exchange for potential benefits. A few dominant themes emerged from the meeting. Anxiety was a symptom reporting to be the most important to target with intervention according to parents. Anxiety is a difficult concept to measure in a population where a self-report is limited, because it is an internal state and severity of anxiety cannot be inferenced by an observer. Instead, the experts describe a need to develop an observer reported outcome measure, an ObsRO, that can capture anxiety behaviors, such as social avoidance, eye aversion or over anticipation and preparation for upcoming events with repeated questioning. The second symptom of interest for targeting with treatment was cognition. Experts describe promising efforts to modify existing cognitive performance outcome measures to be appropriate for individuals with significant intellectual visibility. Many topics explored were related to available treatment. All parents express concerns about the lack of available treatment for Fragile X and their dissatisfaction was the off-label treatments that they were using. Some parents also express frustration that some experimental treatments that they felt worked for their child were subsequently not approved. And there was an overall sense that one size fits all approach may not be appropriate and that drugs that benefit some individuals with FXS may be worthwhile, even if not effective for everyone. Overall, in this case study, we see an example of how publicly available information may serve as a useful starting point for understanding the patient experience and guiding the design of future trials. Next, we turn to a case study where IQVIA conducted interviews with caregivers and clinicians to assess novel outcome measures in pediatric epilepsy. Let's start with some context. The sponsor for this research was developing a treatment for a very rare form of epilepsy, which affects only 0.5% of children diagnosed with epilepsy. The typical age of onset is between the ages of 4 and 5, although symptoms can emerge as late as in early adolescence. Children can be developing normally before the onset of seizures. After seizures begin, a slow cognitive decline follows. It is so gradual that caregivers often don't even notice at the first. Symptoms are heterogeneous and depend on the location of the brain abnormality. They can include cognitive problems such as difficulty with completing tasks and memory hyperactive impulsive or autistic-like behavior, emotional symptoms, predominantly anger and anxiety, learning difficulties, physical and motor symptoms, for example, disturbance or impaired fine motor skills and lastly, social difficulties. The client wanted to collect clinical data for this particular patient population using validated instruments. IQVIA conducted initial research that included a targeted literature review, social media intelligence review and clinician interviews. Based on this, the study team determined that no existing COA instruments were suitable for this context of use. The next step was to conduct an item generation meeting, which resulted in development of 4 new instruments, including 2 observer-reported outcomes, ObsRO and 2 clinician-reported outcomes, ClinROs. Those draft instruments were debriefed in caregiver and clinician interviews and modified based on feedback to ensure that they were understandable, relevant and clear, both to caregivers and clinicians in this specific rare form of epilepsy. The resulting instruments were implemented in sponsor's clinical trial. This provided an opportunity to conduct exit interviews and explore what level of change would be meaningful for each instrument. These interviews were conducted inside and outside of the United States with a small sample of caregivers and clinicians. Next, I'll share some details of the methods employed in these 2 interview studies. The stand-alone cognitive debriefing interviews and exit interviews. The cognitive debriefing interviews for the ObsROs were conducted in ways of 3 to 10 caregivers. Caregivers were recruited by social media and online forms. The eligibility criteria included 1, minimum age of 18; 2, caregiver status of a child age 4 to 12 with a confirmed diagnosis of the pediatric epilepsy of interest; and 3, their ability to participate in one-on-one 90-minute telephone interview. In short, we are looking to speak to parents or legal guardians that provided daily care for a child diagnosed with epilepsy and could share their insights about their experience with this condition. The cognitive debriefing interviews for the ClinROs were conducted with 4 clinicians that had experienced treating and/or conducting research with patients with this pediatric epilepsy of interest. All cognitive debriefing interviews were conducted with participants living in the U.S. The sample for the exit interviews recruited from the sponsors clinical trial was smaller but more varied geographically. It included one caregiver in the U.S. and 1 in Switzerland and 2 clinicians in the U.S., 1 in Switzerland and 2 in Spain. This case study is a good example where a conceptual model of a disease was properly developed with input from reporters other than patients. Here, caregivers and clinicians. It showcases development of the de novo COAs to serve as fit-for-purpose instruments in the sponsor's trial. And it also is an example of conducting qualitative research in and outside of U.S. to help accounting for any language and cultural differences and generalizing the findings. The next case study is about research conducted by IQVIA to understand the patient experience with traumatic brain injury and the use of neuropsychiatric inventory or clinicians or NPI-C in this particular context of use. IQVIA conducted a qualitative interview study with the aim to understand the patient experience with symptoms related to TBI and provide evidence of content validity for NPI-C. NPI-C is a ClinRO that uses caregivers and patients as informants for the clinician rating. It was originally developed for use in dementia and there was no existing evidence that was fit for purpose in TBI. Also similarly to many ClinROs designed for use in CNS indications, development of NPI-C predated FDA's PFDD guidance and the process did not generate evidence that meets the current regulatory standards for outcome measures in clinical trials. Before we dive into the methods, let's consider some of the characteristics of this patient population. People with TBI can experience a wide range of symptoms that can impact their ability to self-report. The overall severity is a combination of the symptom areas that include: Cognitive symptoms related to memory, concentration and attention; sensory symptoms that may include impact on vision or sensitivity to lights or sounds; and lastly, psychological impacts, such as mood changes and agitation. People who experienced a moderate severe TBI often needs supervision and help from caregivers. Many who left alone before the injury must move in with their family members for support. Some people with severe TBI may be able to share insights on their experience in an interview setting, which was the case for this particular case study. Let's have a look. The key to developing a successful methodology in studies, including challenging populations is to carefully determine the study sample and eligibility criteria. Here, we interviewed a total of 21 patients and caregivers. The sample included 9 patient caregiver diets, 2 solo patients and 1 solo caregiver. In total, we've captured 12 unique patient experiences. Trained IQVIA moderators conducted one-on-one interviews with 11 patients, aged 18 to 75, for the confirmed diagnosis of TBI and severity spanning from mild to severe, as confirmed by their health care provider and also 10 caregivers aged 18 or more. We're screening patients. The study team looked for a medical history of TBI that occurred at least 6 months prior to the screening, any mental health diagnosis that could confirms the research, hence they check that the patients would be able to complete the interview, either in a single 75-minute session or 2 1-hour sessions, depending on the participants' ability and preference. Caregivers were screened for how much time they spend with the patients with the minimum requirement of 2 hours per day, 3 days per week. In cases where the patient had a legally authorized representative, the caregiver had to be that person. Lastly, caregivers were also checked for their ability to complete the interview. In addition to speaking to patients and caregivers, the study team completed interviews with 5 clinicians, including 3 psychiatrists and 2 neurologists with extensive experience in treating TBI. The moderators who conducted the patient interviews shared some of their strategies for interviewing people with TBI. For example, formulate clear and direct questions, use simple and correct language, allow sufficient time for the patient to process, repeat and summarize information whenever this is needed and also provide positive feedback when this is appropriate. The study team found some differences in reported behavior between patients and caregivers. For example, when discussing refusing medication, the patients would say that this never happens, while caregivers will report on some instances of such behavior. Another example is reporting on the nature and the extent of aggression. Some example quotes are displayed on the screen. Last but not least, interviews with the 3 types of participants, patients, caregivers and clinicians confirms that NPI-C items are relevant to people who suffered a traumatic brain injury and that the instructions, items and response options are understandable and appropriate in this context of use. The final case study spanned CNS and ophthalmology. Here, I will cover a qualitative patient interviews that IQVIA conducted to assess a novel PRO in dry age-related macular degeneration or dry AMD for short. It's a progressive disease characterized by the formation of insoluble retinal deposits and vision loss. Geographic atrophy that is secondary to dry AMD occurs in the later stages of the disease and impairs visual acuity. It severely impacts the quality of life and may leads to permanent blindness in some patients. We included a couple of quotes from patients describing some of their typical symptoms, having restrict their visual field and poor lighting adaptation. Here this sponsor was developing a treatment for this condition and wanted to collect clinical trial data using validated instruments. First, they conducted a literature review and interviews with clinicians and patients to understand the patient experience with dry AMD. They found that there were no suitable existing PROs to measure visual impairments associated with geographic atrophy that is secondary to dry AMD. Hence, the sponsor is set out in the path to develop a new one. They called it 10 item visual impairment symptom severity assessment or VISSA-10 for short. As the next step in the development process, IQVIA conducted combined [ quantitative validation ] and cognitive debriefing interviews. First, the moderator explored symptoms and impacts of geographic atrophy. Next, they debriefed the VISSA-10 with patients to assess if the items are relevant, [ no matter ] the instructions, items and response options are understandable and appropriate. The study sample included 19 patients diagnosed with GA secondary to dry AMD, age 50 or older. Patients had a known visual acuity score and moderate or severe visual impairment in the affected eye. All interviews were conducted over the phone. The VISSA-10 was verbally administered by the moderator with the aim to assess the instrument as comprehensive and easy to understand. Typically, in an interview setting, some visual stimuli are used allowing the moderator to screen share the PRO items and response options. Here, the moderator adapted the methodology to read out the items and frequently repeat their response options to the participant. The study being budgeted for additional time to allow for this fully verbal administration of the instrument. What they found is that patients frequently reported poor light adaptation, blurred vision and difficulty reading and driving as part of their experience. Moderators noted that the interviews were highly emotional to the patients who are reflecting on their visual impairments. The instrument itself was found to be suitable for assessing GA secondary to dry AMD, both in terms of coverage of the salient symptoms and the clarity and ease of use. The next steps for this instrument will involve some form of psychometric work and considerations for how to best implement it in the clinical setting, perhaps in clinic or eCOA implementation where someone verbally administering VISSA-10 to the participants. This case study nicely illustrates how barriers to self-report can be overcome by adapting methodology. It was also our last case study. I'll now hand over to my colleague, Stella, to close today's session.

Thank you, Anna, and thanks to both of you for the great presentations today. As we close out this webinar, I can't help but reflect as a clinician who has worked so closely with patients living with CNS disorders for so many years now on how far we've come as a field with the guidance of the agency over the years, towards amplifying and measuring the patient experience, especially in these very complicated CNS disorders. I think the case study that Anna presented on Fragile X was a very nice example of how we can right away leverage what is available to us to move forward our understanding of the patient experience in CNS disorders, especially in those rare CNS disorders and start the work that is needed towards developing novel outcomes assessments that more meaningfully reflect the patient experience in what currently exists. Likewise, the third case study that Anna presented, involving patients with traumatic brain injury, showed us that to generate a truly comprehensive picture of a patient experience wherein there is known or suspected brain injury resulting in cognitive and behavioral sequelae, careful attention must be paid from the start to understand the point at which patients can and cannot report for themselves and when additional sources of information are needed to ensure the patient experience, the data that is being collected is truly reliable and valid. In CNS, we know that there are so many clinical outcome assessments that were developed long ago, developed mostly by clinicians, for clinicians without direct patient input. And therefore, just don't map on [ relative to ] the patient experience. Nevertheless, what we heard today, what Anna showed us in these case studies, against the regulatory backdrop that Joy so nicely outlined, is that if we're willing to do the work and apply well-designed, practical approaches to generate patient experience data, we can, in fact, collect meaningful patient experience data in these complicated and challenging CNS disorders, even when there are barriers to self-report.

Thank you all for that great discussion. Before we begin the Q&A, we have 2 poll questions for the audience. Please take a look at your screen and answer the following questions. What types of PED are you most familiar with? COA data, such as, PRO, ClinRO, ObsRO, PerfO; qualitative, such as interviews or focus groups; quantitatives such as surveys or mixed methods. Thank you all for your input. Our next question for our audience is, what decision do you think could best be informed by PED? Regulatory, reimbursement, treatment options, clinical development or all of the above. Thank you all for submitting your answers. I'll now turn it over to Stella to discuss the results and start the Q&A.

Wonderful. Thank you very much, Kristin. Wow, I have to say that I am sort of pleasantly surprised with that second poll questions around the decisions that we think would be best informed by PED data. Our team was actually expecting that most would say that the answer was, A, regulatory. So very impressed. And then for the first one, what types of PED data we're most familiar with? I think we also thought that the majority of you would respond, A, COA data. So nice to see that a lot of you are familiar with qualitative data. Okay. So now what I want to do is get the Q&A speakers on board. I'm going to introduce one new person to the panel here. So in addition to Joy and Anna, we have Dr. Meagan Farrell. Meagan is an Associate Director in our PCS team. She has 10 years of industry experience in clinical outcomes research, and she is a particular expertise in the assessment of cognitive function across diverse and challenging populations. Welcome, Meagan to the speaker panel.

Thank you. Good to be here.

So we have quite a lot of questions coming through. So I'm going to go ahead and do my best to orient to the panel here. So the first question I'm going to go with -- has to do with real-world data, something that our team engages in quite often with our real-world colleagues at IQVIA. The question is, how can patient experience be captured prospectively in the real world, i.e., outside clinical trials? Joy, I'm going to ask you to try and answer that question, please.

Thank you, Stella, and thank you for the question. It's a good consideration. The challenge, right, when we're putting studies together that collect real-world data are the quality of the COAs that are being collected in clinical practice. A lot of times, we work very hard to make sure that our COAs that are put into clinical trials are well defined. They're taken at specific time points. It's very well controlled. But in the real world, the COAs might be not as consistently used, the frequency of assessments is a little challenging. But there are some good examples of COAs that are commonly used like the Kansas City Cardiac questionnaire in clinical practice. And so you would want to make sure that you select COA or a piece of outcome data that's sensitive and is well defined and also commonly used in clinical practice when you're designing a real-world study, collecting outcomes data. There's also this question about how would you understand the patient journey. And there are a set of real-world evidence guidances that have come out. And yes, they're often considering databases that are massive and collecting biomarker data. But there is a mention of patient-generated real-world data and trying to understand what that is. IQVIA has worked as a part of the real-world evidence alliance. And we've spent some time considering this question of what is patient-generated real-world data. And an example would be something that is not in a clinical trial protocol. So it's not putting a wearable into a clinical trial protocol. It's something that people do in their daily life. So it might be collecting from a database that -- that includes Apple Watches, right, where people are routinely monitoring their heart rate data, that would be patient-generated real-world data. And again, those are just examples, but the design and the research question and ensuring that you're capturing the concept of interest is really the primary consideration in capturing patient experience data in the real world.

Fantastic. Thank you, Joy. I'm going to ask you to stay on because I see another question related to regulatory. So let's just go with that. And the question is, do you have any updates regarding regulatory expectation of PED format? How should it be presented to regulators? There is massive and heterogeneous datasets...

Yes, agreed. And there are different approaches, right? So we definitely have the same format that has been adopted for several years, which is submitting a stand-alone dossier for COA evidence. So if you have a COA or a PRO that's incorporated into your clinical endpoint strategy, you want to ensure that you have an evidence dossier that has all of the sort of qualitative and quantitative evidence supporting that COA as fit for purpose. But what about other patient experience data, right? Like what about data that we've collected that help us understand the patient experience with treatment or the patient experience with symptoms and impacts. There have been some approaches to just compiling all of that data and submitting it with your new drug application as sort of a stand-alone patient experience dossier. These can be data dumps and they aren't really helpful for regulators when they're trying to inform a regulatory decision. So we advise our clients against putting together an additional dossier with just patient experience data. But instead to incorporate a narrative throughout your communications with FDA, where it's important for them to understand what the patient experience is with the disease. So if you think about it, like in terms of boilerplate language or clinical overview language, where we consistently see descriptions of symptomology or biomarkers, but no mention of what the patient experience with the disease is. What we're trying to do is, help our sponsors integrate what we know about the patient experience more throughout instead of siloed in their communications with FDA. So I know it's not an easy, here's the format question, but it's important to think about these things and to present them in a way that's interpretable and useful for regulatory decisions.

Fantastic. Thank you, Joy. Okay. Next question that I'm going to bring there. There is a question that speaks to more of the quantitative statistical analyses with PED. We don't have a statistician as part of our panel today, although we have many statistician as part of our PCS team. But I'm going to read the question. I'm going to ask Meagan if she can contribute something to this question. Meagan, the question is, how could we conduct a quantitative statistical analysis if the PED collected from surveys and questionnaires are generally in ordinal scales. What kind of approach do we use it? Can you provide an example?

You give me the hard questions. First, let me just say, I'm not a statistician. But actually, most of the PROs and ObsROs that we use utilize ordinal scales in some form, because they often ask patients and caregivers to use Likert scales 3 things like how much does a given symptom bother you. And we don't necessarily know that the distance between these response scales are equally distributed. So if you just want to describe differences between groups or between individuals over time, you can do something like a nonparametric version of a T test, like the Mann-Whitney U test is something that I've used before. In many cases, researchers actually try to develop more continuous summary scores using various Likert-type scales. So they sum up the responses on the Likert items. It's important that when you're creating a summary or a composite score to represent this data that you weight more heavily, items that we know to be more important than others, if that information is known at the time. So once you have that continuous summary score, then you can use any type of method like [ ANOVA ] or a linear mixed methods to look at change over time on that continuous summary score. Perhaps the person is asking too -- I mean there are a lot of preference studies that use a rank order component, so they ask patients to rank how important various dimensions of their disease are or preferences in terms of treatment. And that's -- those types of analysis do get a little bit more complicated if you want to maintain that rank ordering. And there are various methods out there. Of course, you can just do kind of descriptive statistics to talk about the frequency with which certain choices are ranked more highly than others. But there's various options out there. And I believe there is some guidance, the PFDD guidance to that, that speaks about the different types of scales.

Awesome. Thank you very much, Meagan. There is another question here, Meagan, that I think is also -- I'm going to start with you, and then I think, Anna, I'm going to ask you to chime in here, too. Something that we didn't touch on in this webinar but very important and relates to the use of technology and collection of PED data. So Meagan, let me ask you first. Tell me how you might think of technology as being used to collect PED in CNS?

Yes. Thank you. So there's a lot of enthusiasm around this right now, because there's many new wearable devices that have come on to the scene and they're really designed to capture important aspects of the patient experience that may be especially relevant in individuals with CNS implications. So they may be able to characterize complex characteristics of speech or gait. And these technologies have really come a long way in recent years. I think there is an appeal to using them and incorporating these types of outcomes in clinical trials because they offer a more objective measure of patient performance when you compare them to a PRO or an ObsRO or even a ClinRO. And to go back to the question about real-world, I mean, they obviously offer some more real-world validity about these types of measures in the patient's actual daily life. That being said, there's a lot of work that still needs to be done to demonstrate that these wearable technologies can be used reliably by patients, especially those with cognitive impairments that the scoring algorithms and technologies work consistently. And most importantly, in light of this whole presentation today, that they actually characterize meaningful aspects of the patient experience. But I do know of at least 1 digital COA that has recently been approved by the [ EMEA ] as a secondary end point, and it's developed -- it was developed to measure aspects of gait and gait velocity in patients with Duchenne muscular dystrophy. And so this was a device that is intended to be worn around the ankle, it's a sensor. So it measures various aspects of gait. And the endpoint that they derived is a measure of peak performance or sort of peak gait velocity. And the researchers were able to demonstrate that it's sensitive to change and that it maps onto known measures -- or clinical measures of these types of constructs. So I think there's a lot of promise there, and it's only going to develop further as the technologies improve.

Great. Anna, I know we don't have that much time, but I know that you were just at an international conference on Alzheimer's disease and Parkinson's disease. Anything that you want to add to that around the tech base and collection of PD data?

Yes, absolutely. Thank you, Stella. Just listening to all the talks in the conference, there's definitely an increasing interest in using digital to measure cognitive performance. And 2 examples come to mind: One is a virtual research study that's sponsored by Biogen in collaboration with Apple. The study is called Intuition. And the goal is to measure changes in thinking and memory in adults and to study their longer-term changes in the brain health. It uses Apple Watch and iPhone and it's a non-interventional study, just observational. And the other example is about using speech tasks and also voice biomarkers to test language and memory skills. One such online test has been recently implemented by ADNI, so the Alzheimer's disease neuroimaging initiative. And it's worth mentioning that none of those 2 methods are yet validated in a way that would make them fit for purpose in the context of a registrational clinical trial, but this seems to be where that field is headed and that validation would have to happen along the way.

Great. I see that we only have 3 -- 2 more minutes now. There are some questions around publications on the case study, Anna, that are coming through. May I ask you to just sort of guide the audience here?

Yes, absolutely. So the first case study on Fragile X, that's publicly available. If you were to type up patient listening sessions, FDA in browser, they will direct to a website where you can find not just a Fragile X case study, but all the summaries that were conducted for like a variety of different indications. In terms of the 3 case studies that were conducted by IQVIA, one of them has been published. So the one that's about validation of VISSA-10, it's out. So if you were to pop it in the [ top match ], you should be able to find it and you can always e-mail us if you have any trouble, and we're very happy to direct you.

Wonderful. Thank you for that. There is 1 minute left, and there's quite a number of questions. One of them, can you quickly, can you provide an example of a time when you had to modify your methodology after initiating a study in a CNS therapeutic area? I can quickly take that since I often do the qualitative interviews with patients with sort of complex CNS disorders. Yes, it happens all the time. So we get into an interview format, the patient might be in an acutely psychotic state, and we cannot proceed with the interview. The patient may be lethargic, maybe not able to sustain their retention despite the referring physician having stated that they would be able to engage in this type of work. Having to pivot in that moment in a pediatric population I had where we weren't expecting cognitive impairment and there was clear intellectual disability and cognitive impairment that was prohibiting the kids from speaking with me. So yes, it does happen all the time. It's very important to have trained moderators, especially in CNS indications when you're conducting qualitative interviews in these patient populations. I realize we're on the hour right now. So we are going to have to thank you all. Happy to engage in continued conversations. Please reach out to any one of us and we'd be happy to continue these conversations. Thank you very much for attending, everyone.