Statistical Considerations on Clinical Efficacy Studies of Biosimilar for PMDA Submission
- PDF / 703,323 Bytes
- 4 Pages / 595.276 x 790.866 pts Page_size
- 38 Downloads / 148 Views
ANALYTICAL REPORT
Statistical Considerations on Clinical Efficacy Studies of Biosimilar for PMDA Submission Yumiko Asami, MS1 · Jean Pan, PhD2 · MyungShin Oh, PhD2 · Akira Sato, MS3 Received: 14 December 2019 / Accepted: 19 February 2020 © The Drug Information Association, Inc 2020
Abstract In general, similar to FDA and other health authorities, the PMDA requires clinical efficacy study(ies) to evaluate equivalence between a reference biological product and a Biosimilar product for new drug applications. Even if an identical clinical efficacy study is included in both of PMDA and FDA submissions, the coefficients of confidence interval (CI) used for comparison with the equivalence margins could be different between the two submissions (e.g., 95% CI vs. 90% CI). In this article, we will focus on clinical efficacy studies of Biosimilar products and provide an overview of the two one-sided tests (TOST) and the type I error rate for equivalence design. Then, we summarize published PMDA review reports of Biosimilar products in terms of the coefficients of CI and other elements of the primary endpoints, and explain some Japanese guidelines of Biosimilar and Statistics behind the difference between PMDA and FDA submissions. In addition, we discuss how to use statistical methods correctly and efficiently for PMDA submissions. Keywords Biosimilar · New drug application (NDA) · Equivalence margin · Two one-sided tests (TOST) · Synthesis method
Background Biosimilar (i.e., follow-on biological medicinal product in Japanese guidelines) is defined as a new biotechnological medicinal product developed to be similar in terms of quality, safety, and efficacy to an already licensed, biotechnology medicinal product (hereinafter referred to as reference biological product or RBP) developed by a different marketer–manufacturer in Japan [1, 2]. In general, clinical pharmacokinetic (PK) study(s) (or pharmacodynamic (PD) study(s), PK/PD study(s)) and clinical efficacy study(ies) are required to evaluate equivalence (EQ) between a RBP and a Biosimilar product for new drug applications (NDAs) to Pharmaceuticals and Medical Devices Agency (PMDA), similarly to the Food and Drug
Administration (FDA) and the European Medicines Agency (EMA).
Two One‑Sided Tests (TOST) for EQ Designs Let Δ present the true difference between the test and reference groups, and ΔL (≤ 0) and ΔU (≥ 0) present pre-specified EQ margins, the hypothesis to evaluate EQ is
H0 ∶ Δ ≤ ΔL or Δ ≥ ΔU vs. HA ∶ ΔL < Δ < ΔU
(1)
In another expression, the above hypothesis can be tested using two one-sided tests (TOST)
H0L ∶ Δ ≤ ΔL vs. HAL ∶ Δ > ΔL and
(2)
H0U ∶ Δ ≥ ΔU vs. HAU ∶ Δ < ΔU * Yumiko Asami [email protected] 1
Biostatistics, CSL Behring, Aoyama Building, 1‑2‑3 Kita‑Aoyama, Minato‑ku, Tokyo 107‑0061, Japan
2
Biostatistics, Biosimilars, Amgen, Thousand Oaks, CA, USA
3
Global Project Management, Daiichi Sankyo, Tokyo, Japan
A TOST at significance level 𝛼 for (2) is identical to comparing the two-sided 100 (1–2𝛼 )% CI to the pre-specifie
Data Loading...
