Stakeholder perspectives on potential open-label bias in patient-reported outcome results from cancer clinical trials: evidence from the IMPROvE project

Objectives: To explore stakeholder perspectives related to open-label bias in patient-reported outcomes (PROs) from cancer randomized controlled trials (RCTs) and identify possible risk factors for such bias. This study represents the first phase of the IMPROvE project, an international initiative investigating the impact of trial design (i.e., open-label vs blinded trials) on PRO results. Methods: Semi-structured interviews were conducted with 20 stakeholders, including PRO experts (researchers, oncologists, statisticians) and patients from seven countries. An interview guide addressed perceptions, experiences, and views on potential contributors of open-label bias in PROs. Thematic content analysis was used to identify recurrent themes and patterns. Findings were complemented with a targeted narrative literature review. Results: Stakeholders expressed divergent views regarding the presence and magnitude of open-label bias in PROs. Patient expectations, treatment type, disease stage, and clinician-patient interactions were frequently mentioned as potential contributors to such bias. While experts often framed open-label bias as a methodological concern, patients highlighted experiential and relational aspects. Both groups acknowledged that robust trial methodology and validated PRO instruments may mitigate risks. Literature findings were consistent with these perspectives, highlighting careful methodological planning and transparent reporting practices as critical to mitigate risk of bias. Conclusions: This study offers insights into stakeholder perspectives on potential open-label bias in PROs from cancer RCTs, highlighting both methodological and experiential dimensions. These exploratory findings may inform future evidence generation and guide the development of an "open-label bias PRO checklist."