Manas et al. (2020)25 UK Funding source: A consultancy agreement between York Health Economics Consortium and Boston Scientific | Analysis: Cost-utility analysis Time horizon: 20 years Perspective: UK National Health Service | A hypothetical cohort of patients (≥ 65 years of age) with hepatocellular carcinoma who were unresectable at presentation and were eligible for transarterial embolization, cTACE or DEB-TACE; the cohort was 75% male | Intervention: TARE with TheraSphere Comparators: transarterial embolization, cTACE, and DEB-TACE | A cohort-based Markov model with a cycle length of 4 weeks. The 8 health states included: watch and wait, resection, pharmacological management, pre-transplant, post-transplant, no hepatocellular carcinoma post-transplant, no hepatocellular carcinoma (other), and dead | Effectiveness inputs and utility values were derived from National Liver Offering Scheme data, clinical trials, and assumptions informed by clinical experts; cost inputs were taken from publicly available sources, including the NICE formulary and National Schedule of Reference Costs 2018–201929 |
There was no hepatocellular carcinoma recurrence after a successful liver transplantation Patients who had hepatocellular carcinoma recurrence after a resection did not receive further first-line treatment or curative interventions (e.g., transarterial embolization, cTACE, SIRT, resection, or liver transplant) Transplants were considered curative (i.e., there was no transition from post-transplant to pharmacological management) Several inputs related to transition rates and mortality rates used assumptions based on expert opinion
|
Muszbek et al. (2020)26 UK Funding source: Sirtex Medical Ltd. (a manufacturer of SIR-Spheres) | Analysis: Cost-utility analysis Time horizon: a lifetime horizon (scenario analyses considered a 5-year horizon) Perspective: UK National Health Service | Patients with unresectable intermediate (BCLC stage B) or advanced (BCLC stage C) hepatocellular carcinoma for whom any transarterial embolization therapies (e.g., TACE or transcatheter arterial embolization) were inappropriate, who did not have extrahepatic disease, had a low tumour burden (≤ 25%), and who had preserved liver function (ALBI grade 1) | Intervention: SIRT with SIR-Spheres (90Y resin microspheres) Comparator: Sorafenib | A lifetime partitioned survival analysis model that incorporated 4 health states: progression-free, post-progression, received curative therapy, and dead | Effectiveness inputs and utility values for SIRT and sorafenib were extrapolated from the SARAH trial,33 effectiveness inputs for treatments with curative intent were selected from a targeted literature review, and resource uses were from registries and clinical surveys; utility values were EQ-5D scores mapped from EORTC QLQ-C30 within the SARAH trial33 |
Several assumptions were made to extrapolate values for overall survival and progression-free survival curves to a lifetime horizon Utility consequences of adverse events were assumed to be captured in trial values used to inform the model When a published study or other source for model inputs did not provide standard errors or confidence intervals, a 20%v variation of the mean was assumed
|
Walton et al. (2020)20 UK Funding source: The NIHR Health Technology Assessment program | Analysis: Cost-utility analysis conducted as part of an HTA Time horizon: 10 years Perspective: UK National Health Service and Personal Social Services | Patients with unresectable intermediate (BCLC stage B) or advanced (BCLC stage C) hepatocellular carcinoma and without extrahepatic disease for whom any conventional transarterial embolization therapies (e.g., TACE) were inappropriate | Intervention: SIRTs, including TheraSphere, SIR-Spheres, and QuiremSpheres Comparator: Established clinical management without SIRT using sorafenib or lenvatinib | A decision tree representing the outcome of the work-up procedure transitioning into a 3-state (i.e., progression-free survival, post-progression, and dead) partitioned survival model | Treatment-effectiveness parameters and other inputs calculated from the SARAH,33 SIRveNIB38 and REFLECT trials39 were used; additional effectiveness inputs were drawn from the NMA conducted as part of the HTA; costs of SIRTs and health-state costs were obtained from manufacture’s submissions and the National Schedule of Reference Costs 2017–201830 | The model incorporated several assumptions due to limited availability of clinical literature in some instances:
QuiremSpheres and TheraSphere efficacy are equal to SIR-Spheres No downstaging to curative therapy was permitted Bilobar treatments are performed in 2 separate procedures Proportions of patients who would receive various systemic therapies (these were validated with clinical experts)
|
Rognoni et al. (2017)27 Italy Funding source: ASBM Srl through an unrestricted grant to CERGAS, Bocconi University | Analysis: Cost-utility analysis Time horizon: a lifetime horizon (scenario analyses considered the following time horizons: 1 year, 2 years, 4 years, 6 years, 8 years, 10 years, 15 years, 20 years) Perspective: Italian health care service | Patients with intermediate or advanced hepatocellular carcinoma at 3 oncology centres in Italy; patients with metastatic disease or early or terminal hepatocellular carcinoma were excluded | Intervention: TARE Comparator: Sorafenib (target dosage; 800 mg/day) | A cohort-based multistate Markov model that had a cycle length of 1 month. The model consisted of 5 possible health states (i.e., post-transplantation, stable disease, disease progression, death from the disease, and death from other causes) | Transition probabilities were obtained from prospectively collected real-world data from patients receiving care at 3 centres in Italy; health utility values were retrieved from studies identified in a literature search or came from the Cost-Effectiveness Analysis Registry; health care resource consumptions and costs were from the Regional Health Care Service price list and the from diagnosis-related group-reimbursement rates |
|
