GENAI FOR NEXT-GENERATION JCA WORKFLOWS: AUTOMATING EVIDENCE GENERATION WHILE ENSURING TRANSPARENT, TRACEABLE EU-HTA SUBMISSIONS
Author(s)
Inderpreet S. Marwaha, MSc, RPh1, Rajdeep Kaur, PhD1, Barinder Singh, RPh2, Gagandeep Kaur, M.Pharm1, Shubhram Pandey, MSc1;
1Pharmacoevidence Pvt. Ltd., SAS Nagar, Mohali, India, 2Pharmacoevidence Pvt. Ltd., SAS Nagar Mohali, India
1Pharmacoevidence Pvt. Ltd., SAS Nagar, Mohali, India, 2Pharmacoevidence Pvt. Ltd., SAS Nagar Mohali, India
OBJECTIVES: The EU Joint Clinical Assessment (JCA) introduces swift and rigid timelines for evidence scoping, dossier submission, and report finalization, creating unprecedented pressure on HEOR teams to rapidly integrate diverse clinical evidence with significant risks of delays and cost of rework. The objective of this paper is to propose a pragmatic framework for integrating Generative AI (GenAI) into JCA workflows to overcome resourcing bottlenecks and ensure the delivery of timely, aligned, and traceable evidence packages.
METHODS: We developed a RAG-enabled GenAI evidence pipeline to support the JCA lifecycle from readiness assessment to submission. The proposed framework integrates structured trial data and unstructured documents and enables automated literature screening and PICO extraction, with configurable logic to adapt PICO elements to country- and assessment-specific requirements. The resulting structured evidence was used to support feasibility assessment, Indirect treatment comparisons, and dossier preparation. System components were evaluated by cross-functional experts to assess accuracy, completeness, traceability to source evidence, and alignment with predefined PICO scopes.
RESULTS: The GenAI-enabled system was tested across multiple use cases using both structured and unstructured evidence. The system enabled rapid screening and identification of evidence aligned with predefined PICO scopes. Automated PICO extraction adapted outputs to country- and assessment-specific requirements, supporting evidence reuse across HTA contexts. Generated evidence supported feasibility assessments, ITC inputs, and dossier drafting, with outputs remaining fully traceable to source documents. Expert review confirmed that the outputs met predefined quality criteria, while reducing manual effort and supporting faster evidence generation.
CONCLUSIONS: Success in the JCA era demands a shift from purely manual drafting to cost and time-efficient intelligent automation. Prioritizing the development of GenAI-ready evidence structures is critical to facilitating the timely completion of JCA documents. By adopting the proposed GenAI framework, manufacturers can accelerate evidence readiness, reduce inefficiencies, and improve the likelihood of timely market access across Europe.
METHODS: We developed a RAG-enabled GenAI evidence pipeline to support the JCA lifecycle from readiness assessment to submission. The proposed framework integrates structured trial data and unstructured documents and enables automated literature screening and PICO extraction, with configurable logic to adapt PICO elements to country- and assessment-specific requirements. The resulting structured evidence was used to support feasibility assessment, Indirect treatment comparisons, and dossier preparation. System components were evaluated by cross-functional experts to assess accuracy, completeness, traceability to source evidence, and alignment with predefined PICO scopes.
RESULTS: The GenAI-enabled system was tested across multiple use cases using both structured and unstructured evidence. The system enabled rapid screening and identification of evidence aligned with predefined PICO scopes. Automated PICO extraction adapted outputs to country- and assessment-specific requirements, supporting evidence reuse across HTA contexts. Generated evidence supported feasibility assessments, ITC inputs, and dossier drafting, with outputs remaining fully traceable to source documents. Expert review confirmed that the outputs met predefined quality criteria, while reducing manual effort and supporting faster evidence generation.
CONCLUSIONS: Success in the JCA era demands a shift from purely manual drafting to cost and time-efficient intelligent automation. Prioritizing the development of GenAI-ready evidence structures is critical to facilitating the timely completion of JCA documents. By adopting the proposed GenAI framework, manufacturers can accelerate evidence readiness, reduce inefficiencies, and improve the likelihood of timely market access across Europe.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
HTA84
Topic
Health Technology Assessment
Disease
No Additional Disease & Conditions/Specialized Treatment Areas