Use of RWE to Inform Survival Extrapolation in Health Economic Appraisals of Cancer Drugs in Sweden: A Review of National Health Economic Reports 2010-2024
Author(s)
Emma Söreskog, MSc1, Mariam Kasrashvili, MSc2, Sophie Langenskiöld, PhD2, Sigrid Klaar, PhD2, Niklas Zethraeus, PhD1.
1Department of Learning, Informatics, Management and Ethics, Karolinska Institutet, Stockholm, Sweden, 2Department of Medical Sciences, Uppsala University, Uppsala, Sweden.
1Department of Learning, Informatics, Management and Ethics, Karolinska Institutet, Stockholm, Sweden, 2Department of Medical Sciences, Uppsala University, Uppsala, Sweden.
OBJECTIVES: Long-term survival predictions in health economic assessments of cancer drugs, typically based on clinical trials, are often associated with considerable uncertainty. To reduce uncertainty, several health technology assessment (HTA) agencies recommend incorporating real-world evidence (RWE) to inform or validate survival extrapolations. Within the SURPASS project, we examine the accuracy of survival predictions and how they can be improved in cost-effectiveness analyses. This sub-study specifically reviews whether, and how, RWE has been used to inform survival extrapolations in the Swedish Dental and Pharmaceutical Benefits Agency’s (TLV) cancer drug assessments.
METHODS: We reviewed all TLV-published health economic assessments of cancer drugs submitted for national reimbursement or regional hospital recommendations between 2010 and 2024. For each report, we extracted several parameters, including the evaluated indication, comparators, decision analytic model, time horizon, survival extrapolation method, trial characteristics, external RWE sources (study design, setting, patient population), predicted life years (LYs) and LYs gained, identified uncertainties, reimbursement decisions, and recommendations for use.
RESULTS: In total, 238 reports for 137 cancer drugs were reviewed. Commonly assessed indications included lung, prostate, lymphoma, blood, and breast cancer (68% of all reports). In 40 cases (17%), RWE, including, administrative registry or claims data, were used to inform survival extrapolation. RWE was used in 32 cases (13%) to externally validate survival models or inform long-term survival beyond the initial trial phase. In 9 cases (4%), RWE was used to inform comparator extrapolated survival where the trial lacked comparator arm. Additional results will be presented.
CONCLUSIONS: Explicit testing of external or internal validity of fitted survival regression models was rare. Despite its potential to enhance the credibility of survival extrapolations, RWE remains underutilised in TLV’s health economic appraisals of cancer drugs. Increased and systematic use of RWE could potentially reduce uncertainty in survival projections and strengthen the evidence base for healthcare decision-making.
METHODS: We reviewed all TLV-published health economic assessments of cancer drugs submitted for national reimbursement or regional hospital recommendations between 2010 and 2024. For each report, we extracted several parameters, including the evaluated indication, comparators, decision analytic model, time horizon, survival extrapolation method, trial characteristics, external RWE sources (study design, setting, patient population), predicted life years (LYs) and LYs gained, identified uncertainties, reimbursement decisions, and recommendations for use.
RESULTS: In total, 238 reports for 137 cancer drugs were reviewed. Commonly assessed indications included lung, prostate, lymphoma, blood, and breast cancer (68% of all reports). In 40 cases (17%), RWE, including, administrative registry or claims data, were used to inform survival extrapolation. RWE was used in 32 cases (13%) to externally validate survival models or inform long-term survival beyond the initial trial phase. In 9 cases (4%), RWE was used to inform comparator extrapolated survival where the trial lacked comparator arm. Additional results will be presented.
CONCLUSIONS: Explicit testing of external or internal validity of fitted survival regression models was rare. Despite its potential to enhance the credibility of survival extrapolations, RWE remains underutilised in TLV’s health economic appraisals of cancer drugs. Increased and systematic use of RWE could potentially reduce uncertainty in survival projections and strengthen the evidence base for healthcare decision-making.
Conference/Value in Health Info
2025-09, ISPOR Real-World Evidence Summit 2025, Tokyo, Japan
Value in Health Regional, Volume 49S (September 2025)
Code
RWD74
Topic Subcategory
Health & Insurance Records Systems
Disease
SDC: Oncology