ISSUE: Progress in understanding of the underlying biological mechanisms in cancer has led to a paradigm shift in drug development and to many products receiving approval based on single arm trials (SATs). SATs have been favoured over randomized controlled trials (RCTs) to avoid allocating patients to potentially less effective control therapies and address difficulties in enrolling rare subsets of disease. A recent review reported that the FDA granted 153 new oncology indications based on SATs between 2001 and 2020 (102 accelerated and 51 regular approvals).
Absence of RCT evidence presents challenges for health technology assessment (HTA) in evaluating comparative effectiveness of new interventions versus current treatments. Although statistical methods exist, there remains a substantial potential risk of bias from differences in unknown or unmeasured prognostic factors and the definition or ascertainment of endpoints. These issues, as well as small sample sizes, may result in substantial uncertainty in comparative effectiveness estimates. Progress towards higher quality evidence reflecting effectiveness in routine practice (e.g., pragmatic trials) has been very limited. Rather, with increased use of SATs, are we going backwards?
OVERVIEW: Vishal Bhatnagar will discuss circumstances where an SAT may be appropriate, concerns about and experience with regulatory approvals based on SAT evidence. Adrian Vickers will describe alternative comparative effectiveness methods based on patient-level and aggregate data; their strengths, limitations, and potential for bias and increased uncertainty that may present challenges for HTA. Anne-Pierre Pickaert will discuss the importance of efficient research, regulatory, and HTA processes in making new treatments available rapidly whilst ensuring that they truly offer improved outcomes. The panel will debate learnings from regulatory approvals and HTAs and how HTA processes may accommodate uncertainty. Audience polling on key questions and audience questions will be discussed (10 min). Stakeholders interested in issues surrounding SATs including industry, researchers, regulators, and payers may be interested.
