BACKGROUND: When clinically diagnosed as hepatocellular carcinoma in China, 70%-80% of patients are late-stage along with high treatment costs. Several diagnostic strategies are employed but the detection of early-stage HCC is still limited. The emerging GAAD algorithm has demonstrated better performance on early-stage HCC detection, whereas its cost-effectiveness in China is unclear.

OBJECTIVES: To evaluate the cost-effectiveness of different diagnostic strategies for HCC patients in China.

METHODS: A discrete event simulation model was developed combining a decision tree and Markov structure with over 40 years of patients at high risk of HCC on a lifetime horizon, to evaluate the costs and health outcomes (QALYs gained) of ultrasonography (US), alpha-fetoprotein (AFP), protein induced by vitamin K absence-II (PIVKA-II), GAAD, US+AFP, AFP+PIVKA-II, and GAAD+US. Patients detected as positive by the above strategies would receive the advanced diagnosis, and then be treated according to the HCC treatment regimen, while patients diagnosed as negative (either true or false) would be followed up or continue previous treatment. Epidemiologic, diagnostic performance, treatments’ utility, and cost data were obtained from literature, interviews, and real-world data. Uncertainties on key parameters were explored through deterministic and probabilistic sensitivity analyses.

RESULTS: The 3 strategies of US, GAAD, and GAAD+US formed the cost-effectiveness frontier, in which GAAD+US brought the best health outcomes (13.69 QALYs) with a lifetime cost of RMB 56,932. The ICER of GAAD over the US was RMB 37,448 per QALY and the ICER of GAAD+US over GAAD was RMB 145,535 per QALY. Both DSA and PSA proved the stability of the results.

CONCLUSIONS: With 3 times China GDP per capita as the threshold, the GAAD+US has been demonstrated the most cost-effective strategy for early HCC diagnosis among the diagnostic strategies in China.