COST-EFFECTIVENESS OF A DYNAMIC CTDNA-GUIDED ADJUVANT STRATEGY IN STAGE II-III COLON CANCER. A SAGITTARIUSMODEL-BASED ANALYSIS
Author(s)
CARLO FEDERICI, PhD1, Francesco De Pretis, PhD2, Luca Lazzari, MD3, Ylenia Silvestri, MD3, Aleksandra Torbica, PhD4, Anna Testa, MSc5, Clara Montagut, MD6, Silvia Marsoni, MD3.
1Centre for Research on Health and Social Care Management, SDA Bocconi University, MILANO, Italy, 2Centre for Research on Health and Social Care Management, SDA Bocconi University, milano, Italy, 3Precision Oncology Unit, IFOM ETS, milano, Italy, 4Bocconi University, Milan, Italy, 5IFOM ETS, Milan, Italy, 6Department of Medical Oncology, Hospital del Mar, Barcelona, Spain.
1Centre for Research on Health and Social Care Management, SDA Bocconi University, MILANO, Italy, 2Centre for Research on Health and Social Care Management, SDA Bocconi University, milano, Italy, 3Precision Oncology Unit, IFOM ETS, milano, Italy, 4Bocconi University, Milan, Italy, 5IFOM ETS, Milan, Italy, 6Department of Medical Oncology, Hospital del Mar, Barcelona, Spain.
OBJECTIVES: Circulating tumor DNA (ctDNA)-guided strategies may enable dynamic risk stratification and treatment adaptation after resection for stage III and high-risk stage II colorectal cancer. We evaluated the cost-effectiveness of liquid biopsy (LB)-guided management, based on the SAGITTARIUS trial design, versus standard 6-month CAPOX from the Italian healthcare perspective.
METHODS: An individual-level discrete-event simulation with a lifetime horizon compared standard of care with two LB-guided strategies, analysed separately by post-operative ctDNA status. In ctDNA-positive patients, standard care was compared with adaptive adjuvant treatment guided by three sequential LBs and biomarker-tailored treatment switching. In ctDNA-negative patients, standard care was compared with LB-enhanced surveillance and treatment de-escalation. Efficacy and toxicity inputs for standard adjuvant regimens were derived from published trials; for biomarker-guided treatments without adjuvant evidence, metastatic efficacy was conservatively extrapolated. Costs were based on Italian NHS tariffs. Cohorts of 30,000 patients per arm were simulated. Outcomes included costs, relapses, life-years (LYs), quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios (ICERs), assessed against a €21,000/QALY threshold.
RESULTS: In ctDNA-positive patients, LB-guided management reduced relapses by 24% and increased mean survival by 5.8 LYs and 3.2 QALYs, with incremental costs of €11,699 and an ICER of €3,172/QALY. In ctDNA-negative patients, relapse rates were similar (12.3% vs 13.0%), while LB-guided surveillance generated 0.26 additional QALYs, mainly through reduced treatment-related toxicity, with incremental costs of €3,427 and an ICER of €13,268/QALY. Results were most sensitive to treatment-effect assumptions in ctDNA-positive patients and follow-up utility assumptions in ctDNA-negative patients.
CONCLUSIONS: Dynamic LB-guided management was cost-effective in both ctDNA-defined populations, with distinct value drivers: relapse reduction through adaptive escalation in ctDNA-positive patients and toxicity reduction through de-escalation in ctDNA-negative patients. These findings support further refinement and validation using mature SAGITTARIUS data.
METHODS: An individual-level discrete-event simulation with a lifetime horizon compared standard of care with two LB-guided strategies, analysed separately by post-operative ctDNA status. In ctDNA-positive patients, standard care was compared with adaptive adjuvant treatment guided by three sequential LBs and biomarker-tailored treatment switching. In ctDNA-negative patients, standard care was compared with LB-enhanced surveillance and treatment de-escalation. Efficacy and toxicity inputs for standard adjuvant regimens were derived from published trials; for biomarker-guided treatments without adjuvant evidence, metastatic efficacy was conservatively extrapolated. Costs were based on Italian NHS tariffs. Cohorts of 30,000 patients per arm were simulated. Outcomes included costs, relapses, life-years (LYs), quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios (ICERs), assessed against a €21,000/QALY threshold.
RESULTS: In ctDNA-positive patients, LB-guided management reduced relapses by 24% and increased mean survival by 5.8 LYs and 3.2 QALYs, with incremental costs of €11,699 and an ICER of €3,172/QALY. In ctDNA-negative patients, relapse rates were similar (12.3% vs 13.0%), while LB-guided surveillance generated 0.26 additional QALYs, mainly through reduced treatment-related toxicity, with incremental costs of €3,427 and an ICER of €13,268/QALY. Results were most sensitive to treatment-effect assumptions in ctDNA-positive patients and follow-up utility assumptions in ctDNA-negative patients.
CONCLUSIONS: Dynamic LB-guided management was cost-effective in both ctDNA-defined populations, with distinct value drivers: relapse reduction through adaptive escalation in ctDNA-positive patients and toxicity reduction through de-escalation in ctDNA-negative patients. These findings support further refinement and validation using mature SAGITTARIUS data.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
EE678
Topic
Clinical Outcomes, Economic Evaluation, Health Technology Assessment
Topic Subcategory
Trial-Based Economic Evaluation
Disease
Oncology