Patient Centric Trials
Chair: Colleen Russell (Parexel)
Vice Chair: Shuyen Ho (Consultant)
Speaker: Laura Lee Johnson (FDA)
Title: You Can Have Patient-Centric Simulation Models, and Just Because There is a PRO Included in a Study Does Not Make It Patient Centric
Abstract: Patients centric does not stop at endpoints that matter to patients. Patient centric means including patients early in the planning process to help inform multiple aspects of the trial design and conduct, troubleshooting logistics during a trial, and interpreting results. It means continuing to listen to patients and seeking their input when determining how to conduct a trial that will answer questions relevant to society and patients. Statisticians have a responsibility to solicit, engage, and collaborate with patients when planning trials and analyses. Statistical models are not just for study designs and data patterns. They need to answer the patients’ questions. Listening to patients and caregivers to understand their questions and insights can support many estimand and statistical analysis plan choices integrated into trial design. Why might missing data be missing, and what logistics and design choices could lower the rates of missing data? What elements might encourage enrollment and retention? How frequently can and should data be collected, where, and how? How will all this impact the statistical model, variance of the measurements, and sample size? If you are doing simulations, how did you choose your parameter space? We will discuss examples where statisticians designed patient centric analysis and simulation plans, and how diverse sources of information were key in the decision making for development programs.
Speaker: Stephen L. Yates (UCB)
Title: Amplifying the Voice of the Patient in Clinical Research: Development of Toolkits for Use in Designing and Conducting Patient-Centered Clinical Studies
Abstract: Incorporating patient perspectives into clinical studies is recognized as important to the development of high-quality, safe, and effective fit-for-patient medicines. However, no widely accepted methodology to help design more patient-centered studies has been established systematically. TransCelerate Biopharma Inc., a non-profit organization promoting collaboration across biopharmaceutical companies, organized a Patient Experience (PE) Initiative to create tools to intentionally include the patient perspective into the design and implementation of clinical studies. The resulting tools include the Patient Protocol Engagement Toolkit (P-PET), to engage patients early in protocol development, and the Study Participant Feedback Questionnaire (SPFQ), to assess patient experiences during clinical studies. To develop these toolkits, TransCelerate conducted a literature review and identified aspects of clinical studies that patients find either valuable or burdensome, or that affect participation, adherence, and engagement in a clinical study. The concepts identified were refined through elicitation of feedback from patient advisors, clinical study site advisors, and subject matter experts from member companies (MCs) of TransCelerate. This feedback was considered in identifying gaps, defining scientific methodology to understand how to evaluate patients’ needs, and developing and refining the P-PET and the SPFQ. As part of the development process, descriptions/drafts of the tools were shared with patients, clinical site advisory groups, MCs, and the US Food and Drug Administration, and then revised. MCs simulated use of the tools, and feedback was incorporated into the final versions of the P-PET and SPFQ prior to public release. The P-PET and SPFQ are available free on the TransCelerate website.
(Previously published in: Ther Innov Regul Sci (TIRS) 2020 Nov;54(6):1489-1500, https://pubmed.ncbi.nlm.nih.gov/32617912/)
Speaker: Kevin Weinfurt (Duke)
Title: Clarifying the Role of Within-Patient Change for Interpreting the Meaningfulness of Treatment Effects in Parallel Groups Designs
Abstract: Clinical trials are often designed to evaluate how an intervention affects how patients feel, function, or survive. Feeling or functioning is typically measured using clinical outcomes assessments (COAs) such as patient-reported, observer-reported, clinician-reported, or performance outcome measures. Because the metrics of such measures are often unfamiliar and difficult to interpret, researchers will attempt to define a minimally important change and use this to either define some type of responder endpoint or to help interpret endpoints based on COA scores at a fixed follow-up time or change-from-baseline. In this talk I will argue that, in the context of parallel groups designs, there has been undue emphasis on “meaningful within-patient change” and that this has distracted from more straightforward ways of assessing the meaningfulness of an observed treatment effect.
Discussant: Ceciel T. Rooker (International Foundation for Gastrointestinal Disorders)
