D05: Antipsoriatics
ATC D05 antipsoriatics for modelers: a bounded score needs a bounded likelihood.
The disease these drugs treat, and where PASI sits among its biomarkers, is covered in the psoriasis primer.
A bounded score needs a bounded likelihood. PASI has been fitted with a bounded integer model, to respect the discrete and bounded nature of the data [1], and with a zero-inflated beta distribution, for the bounded nature and distributional characteristics of the score [2]. In one psoriasis analysis, fitting the 6-point physician’s global assessment alone carried a substantial structural bias that joint modeling with PASI removed [3].
References
[1]
Ooi QX, Kristoffersson A, Korell J, Flack M, L. Plan E, Weber B. Bounded integer model‐based analysis of psoriasis area and severity index in patients with moderate‐to‐severe plaque psoriasis receiving BI 730357. CPT: Pharmacometrics & Systems Pharmacology 2023;12:758–69. https://doi.org/10.1002/psp4.12948.
[2]
Tsamandouras N, Qiu R, Hughes JH, Sweeney K, Prybylski JP, Banfield C, et al. Employing zero-inflated beta distribution in an exposure-response analysis of TYK2/JAK1 inhibitor brepocitinib in patients with plaque psoriasis. Journal of Pharmacokinetics and Pharmacodynamics 2024;51:265–77. https://doi.org/10.1007/s10928-024-09901-2.
[3]
Hu C, Zhou H. Improving categorical endpoint longitudinal exposure–response modeling through the joint modeling with a related endpoint. Journal of Pharmacokinetics and Pharmacodynamics 2022;49:283–91. https://doi.org/10.1007/s10928-021-09796-3.