EP-013 - DEVELOPMENT OF POPULATION PHARMACOKINETICS MODEL OF PYRAZINAMIDE AND OPTIMAL DOSE RECOMMENDATION FOR ELDERLY WITH DIABETES MELLITUS IN KOREAN TUBERCULOSIS PATIENTS.

R. Kim¹, R. Jayanti¹, T. Jang², J. Min³, J. Kim⁴, J. Oh⁵, H. Lee⁶, H. Lee⁷, K. Shin⁸, B. kang⁹, H. Kim¹⁰, Y. Ko¹¹, J. Kim¹², Y. Choi¹, P. Nguyen¹, Y. Cho¹, J. Shin¹; ¹Inje University, Busan, Republic of Korea, ²Kosin University, Busan, Republic of Korea, ³The Catholic University of Korea, Seoul St. Mary's Hospital, Seoul, Republic of Korea, ⁴The Catholic University of Korea, Incheon St. Mary's Hospital, Incheon, Republic of Korea, ⁵Korea University Guro Hospital, Seoul, Republic of Korea, ⁶Inje University Busan Paik Hospital, Busan, Republic of Korea, ⁷The Catholic University of Korea, Eunpyeong St. Mary's Hospital, Seoul, Republic of Korea, ⁸Yeungnam University Medical Center, Daegu, Republic of Korea, ⁹Dong-A University Hospital, Busan, Republic of Korea, ¹⁰Gyeongsang National University Changwon Hospital, Changwon, Republic of Korea, ¹¹Kangdong Sacred Heart Hospital, Seoul, Republic of Korea, ¹²The Catholic University of Korea, Uijeongbu St. Mary's Hospital, Seoul, Republic of Korea.

Background: Diabetes mellitus (DM) is common comorbidity that appeared among the elderly population, likely to complicate their clinical conditions. The effect of DM on the pharmacokinetics (PK) of anti-tuberculosis (TB) drugs remained arguable. However, DM potentially hampered successful TB treatment. Through the development of the population PK model, we aimed to provide optimal dosing of pyrazinamide (PZA) for the geriatric DM population with TB in regard to the probability of achieving the PK target.

Methods: PZA concentrations at random post-dose points, demographic characteristics, and clinical information were collected in a multicenter prospective cohort study from 13 hospitals in Korea. Population PK model was developed using a non-linear mixed effect method and dose simulation was carried out. Data from 610 TB patients were divided into a training and test dataset with a 4 to 1 ratio.

Results: A one-compartment model with allometric scaling for body size effect adequately described PK parameters of PZA. Geriatric patients (>70 years old) with DM were found as significant covariates, increasing the apparent clearance (CL/F) of PZA by 30% (CL/F: 4.49 L/h). Our model was externally evaluated using the test set and provided better predictive performance compared to the previously published model. Following our simulations to assess efficacy and toxicity, geriatric DM patients with TB required higher doses of PZA in order to achieve the area under concentration from 0 to 24 hours (AUC0-24) target of ≥ 363 mg.h/L.

Conclusion: The established population PK model sufficiently described the PK parameters of PZA in Korean TB patients. Our model can be useful in therapeutic drug monitoring to suggest the optimal dose of PZA, particularly for the geriatric DM population with TB.