ARE CURRENT EVIDENCE SYNTHESIS STANDARDS FIT FOR THE EU JOINT CLINICAL ASSESSMENT ERA? GAPS ACROSS PRISMA, COCHRANE, ISPOR, AND HTA GUIDANCE
Author(s)
Rozee Liu, MSc1, Stacy Grieve, PhD1, Sharada Harricharan, PharmD2, Anna Forsythe, MBA, MSc, PharmD1.
1Oncoscope, Miami, FL, USA, 2Frontier HEOR, Toronto, ON, Canada.
1Oncoscope, Miami, FL, USA, 2Frontier HEOR, Toronto, ON, Canada.
OBJECTIVES: The EU Joint Clinical Assessment (JCA) requires timely, population/intervention/comparator/study design (PICOS)-specific, and continuously-updated evidence submissions. However, traditional systematic literature reviews (SLRs) remain largely static and resource-intensive, while guidance for living systematic reviews (LSRs) and artificial intelligence (AI)-assisted evidence synthesis continues to evolve. This study evaluated whether existing methodological guidance adequately supports JCA evidence requirements and identified gaps relevant to HTA practice.
METHODS: A structured review of seven major guidance sources was conducted, including PRISMA, Cochrane guidance, Cochrane RAISE guidance, The Professional Society for Health Economics and Outcomes Research (ISPOR) recommendations, National Institute of Clinical Excellence (NICE) guidance, Canada’s Drug Agency (CDA) guidelines, and EU JCA methodological documents. All guidance were qualitatively assessed against requirements relevant to JCA-ready evidence generation, including continuous updating, AI-assisted review processes, transparency, governance, reproducibility, auditability, and support for evolving PICOS frameworks.
RESULTS: Three major gaps were identified. First, LSRs are recognized within methodological guidance but remain insufficiently operationalized for routine HTA use. Second, AI-assisted evidence synthesis is increasingly acknowledged, yet governance, validation, and auditability requirements remain inconsistently defined. Third, no reviewed framework fully supports continuously updated, PICOS-specific evidence generation aligned with JCA timelines. These identified gaps have informed the development of a Real-time AI-assisted Living Systematic Literature Review (REAL-SLR) framework integrating continuous evidence surveillance, AI-assisted study identification and extraction, expert validation, version-controlled updates, and transparent audit trails. The framework further enables dynamic evidence stratification across key clinical dimensions including disease stage, biomarker status, line of therapy, comparator class, trial design, guideline status, and HTA relevance.
CONCLUSIONS: Current guidance treats LSRs and AI as separate methodological developments rather than components of a continuous evidence ecosystem. A framework integrating living evidence methods, AI, and HTA governance has been proposed to bridge this gap, and better align evidence generation with JCA requirements to support more timely, transparent, and decision-ready submissions.
METHODS: A structured review of seven major guidance sources was conducted, including PRISMA, Cochrane guidance, Cochrane RAISE guidance, The Professional Society for Health Economics and Outcomes Research (ISPOR) recommendations, National Institute of Clinical Excellence (NICE) guidance, Canada’s Drug Agency (CDA) guidelines, and EU JCA methodological documents. All guidance were qualitatively assessed against requirements relevant to JCA-ready evidence generation, including continuous updating, AI-assisted review processes, transparency, governance, reproducibility, auditability, and support for evolving PICOS frameworks.
RESULTS: Three major gaps were identified. First, LSRs are recognized within methodological guidance but remain insufficiently operationalized for routine HTA use. Second, AI-assisted evidence synthesis is increasingly acknowledged, yet governance, validation, and auditability requirements remain inconsistently defined. Third, no reviewed framework fully supports continuously updated, PICOS-specific evidence generation aligned with JCA timelines. These identified gaps have informed the development of a Real-time AI-assisted Living Systematic Literature Review (REAL-SLR) framework integrating continuous evidence surveillance, AI-assisted study identification and extraction, expert validation, version-controlled updates, and transparent audit trails. The framework further enables dynamic evidence stratification across key clinical dimensions including disease stage, biomarker status, line of therapy, comparator class, trial design, guideline status, and HTA relevance.
CONCLUSIONS: Current guidance treats LSRs and AI as separate methodological developments rather than components of a continuous evidence ecosystem. A framework integrating living evidence methods, AI, and HTA governance has been proposed to bridge this gap, and better align evidence generation with JCA requirements to support more timely, transparent, and decision-ready submissions.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
HTA272
Topic
Health Technology Assessment, Methodological & Statistical Research, Study Approaches
Topic Subcategory
Systems & Structure
Disease
Oncology