Table 1.
Model input parameters.
| Model input | Value | PSA distribution | Sources |
|---|---|---|---|
| Patient characteristics | |||
| Starting age, years | 55 | Lee et al. (9) | |
| Proportion of curable recurrence, % | Lee et al. a (10) | ||
| Treatment | 56.6 | ||
| Control | 47.8 | ||
| Incidence rate of curative treatments (/person-year), mean (95% CI) | Lee et al. a (10) | ||
| Resection | 0.0888 (0.0385, 0.1390) | Normal | |
| Radiofrequency ablation | 0.3033 (0.2105, 0.3962) | Normal | |
| Percutaneous ethanol injection | 0.2071 (0.1304, 0.2839) | Normal | |
| Liver transplantation | 0.0148 (0.0000, 0.0353) | Normal | |
| Utilities | |||
| Health state utility (95% CI) | Pollom et al. (16) | ||
| Recurrence-free state | 0.88 (0.85, 0.92) | Beta | |
| Curable recurrence | 0.88 (0.85, 0.92) | Beta | |
| Incurable recurrence | 0.40 (0.32, 0.48) | Beta | |
| Treatment-related disutility | Ock et al. b (17) | ||
| Resection | - 0.26 | ||
| Radiofrequency ablation | - 0.33 | ||
| Percutaneous ethanol injection | - 0.33 | ||
| Transplantation | - 0.21 | ||
| Costs (USD), mean (SD) c | |||
| Treatment cost (per injection) | 3,807 | ||
| Health state cost (per cycle) | |||
| Recurrence-free state d | 211 (21) | Gamma | Micro-costing |
| Curable recurrence d | 211 (21) | Gamma | Micro-costing |
| Incurable recurrence | 2,505 (711) | Gamma | HIRA data |
| Event cost for curative treatments | HIRA data | ||
| Resection | 8,082 (3,015) | Gamma | |
| Radiofrequency ablation | 2,085 (1,039) | Gamma | |
| Percutaneous ethanol injection | 1,640 (1,282) | Gamma | |
| Liver transplantation | 67,142 (23,888) | Gamma | |
| End-of-life cost d | 6,798 (679) | Gamma | Yang, (18) |
CI, confidence interval; HIRA, Health Insurance Review & Assessment Service of Korea; PSA, probabilistic sensitivity analysis; SD, standard deviation.
Derived from post-hoc analysis of recurrence data from phase III trial (Lee et al., 2015 & Lee et al., 2018).
Converted to disutility from Ock et al., 2017.
1 USD = 1,166.51 KRW.
Standard deviation is assumed to be 10% of the mean value.