Value of Liquid Biopsy for Genomic Profiling in Non-Small Cell Lung Cancer in France: A Pharmacoeconomic Model
Author(s)
Paul Hofman, Prof MD1, Yishan Chuang, PhD2, Elizabeth Sheppard, MBA3, Mehmet Berktas, MD MSc MICR4, Cyril Addi, PhD5, Hélène Cawston, MSc6, Wen Su, MSc7, Tam De Lacey, MSc7, Natasha B Leighl, Prof MD8;
1IHU RespirERA, Nice, France, 2AstraZeneca OBU Global Medical Diagnostic, Gaithersburg, MD, USA, 3AstraZeneca Oncology Global Global Pricing and Market Access, Cambridge, United Kingdom, 4AstraZeneca OBU Oncology Outcome Research, Cambridge, United Kingdom, 5AstraZeneca France Business Unit Oncology, Paris, France, 6Amaris Health Economics and Market Access, Paris, France, 7Amaris Health Economics and Market Access, London, United Kingdom, 8University of Toronto Princess Margaret Cancer Centre, IHPME, Dalla Lana School of Public Health, Toronto, ON, Canada
1IHU RespirERA, Nice, France, 2AstraZeneca OBU Global Medical Diagnostic, Gaithersburg, MD, USA, 3AstraZeneca Oncology Global Global Pricing and Market Access, Cambridge, United Kingdom, 4AstraZeneca OBU Oncology Outcome Research, Cambridge, United Kingdom, 5AstraZeneca France Business Unit Oncology, Paris, France, 6Amaris Health Economics and Market Access, Paris, France, 7Amaris Health Economics and Market Access, London, United Kingdom, 8University of Toronto Princess Margaret Cancer Centre, IHPME, Dalla Lana School of Public Health, Toronto, ON, Canada
OBJECTIVES: Liquid biopsy (LB) is a genetic testing method that illuminates disease prognosis, progression, and therapy responses, primarily using circulating tumour DNA (ctDNA). Our objective was to assess the cost-effectiveness of LB plus tumour tissue biopsy (TT), including concurrent and sequential TT and LB, LB with/without sequential TT versus TT alone in diagnosing non-small cell lung cancer (NSCLC) in France.
METHODS: We developed a decision tree followed by a partitioned survival model (PSM) to calculate costs and quality-adjusted life years (QALYs) over a 2-year time horizon. The decision tree included four strategies: TT alone, LB followed by TT, TT followed by LB, and concurrent TT and LB. The PSM had three health states: progression-free (PF), progressed disease (PD), and death. Patients entered the PSM in the PF state and received targeted therapy or chemotherapy based on the presence or lack-thereof of a genomic alteration. The VALUE trial and published literature informed sensitivity and specificity in the decision tree and published clinical trials informed survival in the PSM. French sources were used for the prevalence of genomic alterations and costs.
RESULTS: The model estimated that concurrent LB and TT led to more patients receiving targeted therapy compared to TT alone, with 4% more patients identified with a genomic alteration and greater total costs (58,713 Euro vs 56,875 Euro). Concurrent LB and TT increased testing costs (62%), time to treatment was shortened (4 days earlier, on average) and total QALYs increased compared to TT alone (0.892 vs 0.872). Sequential LB then TT incurred the longest time to treatment and the lowest costs.
CONCLUSIONS: LB concurrent with TT for diagnosis in NSCLC patients could increase mutation identification and shorten the time to treatment while slightly increasing total costs in France.
METHODS: We developed a decision tree followed by a partitioned survival model (PSM) to calculate costs and quality-adjusted life years (QALYs) over a 2-year time horizon. The decision tree included four strategies: TT alone, LB followed by TT, TT followed by LB, and concurrent TT and LB. The PSM had three health states: progression-free (PF), progressed disease (PD), and death. Patients entered the PSM in the PF state and received targeted therapy or chemotherapy based on the presence or lack-thereof of a genomic alteration. The VALUE trial and published literature informed sensitivity and specificity in the decision tree and published clinical trials informed survival in the PSM. French sources were used for the prevalence of genomic alterations and costs.
RESULTS: The model estimated that concurrent LB and TT led to more patients receiving targeted therapy compared to TT alone, with 4% more patients identified with a genomic alteration and greater total costs (58,713 Euro vs 56,875 Euro). Concurrent LB and TT increased testing costs (62%), time to treatment was shortened (4 days earlier, on average) and total QALYs increased compared to TT alone (0.892 vs 0.872). Sequential LB then TT incurred the longest time to treatment and the lowest costs.
CONCLUSIONS: LB concurrent with TT for diagnosis in NSCLC patients could increase mutation identification and shorten the time to treatment while slightly increasing total costs in France.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
EE239
Topic
Economic Evaluation
Disease
SDC: Oncology