(T-023) Cost-effectiveness of paliperidone palmitate long-acting injectables for the treatment of schizophrenic patients: a joint pharmacometric/pharmacoeconomic approach

Haini Wen, Master's degree – PhD student, Department of Pharmacy, Uppsala University; Qingqing Chai, Master‘s degree in pharmacoeconomics – Clinical pharmacist, Department of Pharmacy, Shanghai Ninth People's Hospital; Han Liu, Phd student – Phd student, Department of Pharmacy, Uppsala University; Xiao Zhu, PhD in pharmacometrics – Associate Professor, Department of Clinical Pharmacy and Pharmacy Administration, Fudan University

Introduction: Schizophrenia, a chronic mental illness with high relapse rate. The conventional treatment relies on paliperidone palmitate extended release (PPER), which requires daily oral dosing. Poor adherence to treatment is common among patients and is associated with poorer outcomes including relapse and hospitalization. To improve patient adherence, two long-acting injectables (LAIs) have been successively approved that allow dosing monthly (1-month paliperidone palmitate, PP1M) or three-monthly (3-month paliperidone palmitate, PP3M). However, little is known about the relative cost and effectiveness of PP3M and PP1M when compared to PPER due to the lack of direct comparisons.

We developed a joint pharmacometric/pharmacoeconomic (PMPE) model to evaluate the treatments’ cost-effectiveness based on the Positive and Negative Syndrome Scale (PANSS) score via simulation.

Objectives:
• To quantify the relationship between PK and PD of PP1M and PP3M in patients of schizophrenia.
• To analyze the cost-effectiveness of PP1M and PP3M from a perspective of the Chinese healthcare system.

Methods: A PKPD model was developed based on published models[1]. Three independent population PK models of PPER, PP1M and PP3M were linked to a common PD model of paliperidone, respectively[2]. Information such as demographic characteristics and baseline PANSS score was extracted from previous clinical studies. Through stochastics simulation, individual time courses of PANSS total score and relapse rates were generated to calculate the transition probability in a Markov model.

The Markov model was developed for cost-effectiveness analysis which consisted 8 health states that reflected disease progress from stable, relapse to death. A time horizon for 5 years with schizophrenia transitioning on a monthly basis was used while considering the different levels of adherence and drug switch during the treatment. The utility parameters were extracted from previous studies.

Results: For each formulation, 1000 patients were simulated with a 12-month duration. The plasma concentration and PANSS score was predicted well by the PKPD model. The simulation showed that the relapse rate in each group is 14.5% (PPER), 9.7% (PP1M) and 9.8% (PP3M), respectively.
The PE analysis suggested that PP1M and PP3M were both cost-effective. Compared with PPER, PP3M exhibited absolute cost-effectiveness advantages with a lower discounted cost of $19,911 and a gain of 0.008 quality-adjusted life years (QALYs). When the willingness-to-pay (WTP) was set at $12,733, the probability of PP1M and PP3M being more cost-effective was 30.2% and 64.3% respectively.

Conclusion: This study quantified the relationship between plasma concentration and PANSS score of PP1M and PP3M, and highlighted the superior cost-effectiveness of PP3M. We had also demonstrated the added value of the PMPE models that could inform decision-making in healthcare reimbursement policy.

Citations: [1] Pilla Reddy, V., et al., Pharmacokinetic-pharmacodynamic modeling of antipsychotic drugs in patients with schizophrenia Part I: the use of PANSS total score and clinical utility. Schizophr Res, 2013. 146(1-3): p. 144-52.
[2] Magnusson, M.O., et al., Population Pharmacokinetics of a Novel Once-Every 3 Months Intramuscular Formulation of Paliperidone Palmitate in Patients with Schizophrenia. Clin Pharmacokinet, 2017. 56(4): p. 421-433.