OBJECTIVES: A recent phase II clinical trial demonstrated the efficacy of xanomeline with trospium (XT) in decreasing positive and negative symptoms of schizophrenia. With a phase III trial currently underway, it is important to understand the potential cost-effectiveness of this new treatment option. We aimed to determine the list price at which XT could be cost-effective, compared to lumateperone, in people with markedly ill schizophrenia.

METHODS: We developed a Markov model from a US healthcare payer perspective over a lifetime horizon for a cohort of 18-year-olds diagnosed with markedly ill schizophrenia. Effectiveness data for XT and lumateperone were based on Positive and Negative Syndrome Scale (PANSS) scores in two clinical trials (NCT03697252 and NCT0228276). We mapped PANSS scores to EQ-5D utilities. Varying the price of XT, we calculated the ICER to determine the price at which XT would be cost-effective at willingness-to-pay (WTP) thresholds of $50,000, $100,000, and $150,000 per QALY. We conducted one-way and probabilistic sensitivity analyses. Costs and outcomes were discounted at 3%; costs are presented in 2021 USD.

RESULTS: XT generated 15.12 discounted QALYs, compared to 15.94 for lumateperone (incremental QALY loss of 0.82). Total discounted lifetime costs for lumateperone were $318,201. Imposing WTP thresholds of $50,000, $100,000, and $150,000, XT was cost-effective at $15,300, $12,650, and $9,950 per year, respectively. One-way sensitivity analyses identified the drug costs and health state utilities as key drivers of the model. Probabilistic sensitivity analysis showed that at a threshold of $50,000/QALY, there is 76.8% likelihood that XT is cost-effective when priced at $15,300.

CONCLUSIONS: XT was less effective than lumateperone. With an ICER in the south-west quadrant of the cost-effectiveness plane, the reduction in costs from the reimbursement of XT would allow spending to be dedicated to healthcare activities that could generate more QALYs than lumateperone.