Bayesian semiparametric meta‐analytic‐predictive prior for historical control borrowing in clinical trials

2021 ◽  
Author(s):  
Bradley Hupf ◽  
Veronica Bunn ◽  
Jianchang Lin ◽  
Cheng Dong
2013 ◽  
Vol 13 (1) ◽  
pp. 41-54 ◽  
Author(s):  
Kert Viele ◽  
Scott Berry ◽  
Beat Neuenschwander ◽  
Billy Amzal ◽  
Fang Chen ◽  
...  

2013 ◽  
Vol 32 (21) ◽  
pp. 3609-3622 ◽  
Author(s):  
Sandro Gsteiger ◽  
Beat Neuenschwander ◽  
Francois Mercier ◽  
Heinz Schmidli

Haematologica ◽  
2017 ◽  
Vol 102 (3) ◽  
pp. e118-e119
Author(s):  
Nicola Gökbuget ◽  
Herve Dombret ◽  
Renato Bassan ◽  
Martha Wadleigh ◽  
Michael Doubek ◽  
...  

Biometrics ◽  
2014 ◽  
Vol 70 (4) ◽  
pp. 1023-1032 ◽  
Author(s):  
Heinz Schmidli ◽  
Sandro Gsteiger ◽  
Satrajit Roychoudhury ◽  
Anthony O'Hagan ◽  
David Spiegelhalter ◽  
...  

2019 ◽  
pp. 216847901986253
Author(s):  
Naoki Isogawa ◽  
Kentaro Takeda ◽  
Kazushi Maruo ◽  
Takashi Daimon

Background: In the process of research and development of a new treatment, clinical trials are conducted to evaluate its safety and efficacy. Key to streamlining the process is to utilize appropriate historical information on an outcome of a control treatment when designing and analyzing a clinical trial. Methods: For the use of such historical control information, there exist a meta-analytic approach and power prior approach. In this article, we evaluate their performance with regard to the type I error (TIE) rate and power through a simulation study where we analyze the data on a binary outcome of an experimental treatment and a control treatment from a new small-scale trial, along with the corresponding data of the control treatment from multiple historical trials. The reason is that the difference in the performance between the 2 approaches has not been clear. Results: When historical trials were homogeneous, the power was higher in the power prior approach and the meta-analytic approach using a beta-binomial model with a less noninformative prior than the other approaches. However, when heterogeneous historical trials were mixed, the power was lower, or the TIE rates got inflated. Conclusions: To make use of historical control data, if importance is attached to control of the TIE rate, the meta-analytic approach using a normal-normal hierarchical model may be preferable to the power prior approach, whereas if attached to improvement of the power, this preference be reversed. Anyway, the best approach should be chosen by comparing the operational characteristics of the approaches.


2021 ◽  
Vol 5 (Supplement_1) ◽  
pp. A257-A257
Author(s):  
Sanjiv Kaul ◽  
Olulade Ayodele ◽  
Nicole Sherry ◽  
Elyse Swallow ◽  
Fan Mu ◽  
...  

Abstract Chronic hypoparathyroidism is associated with an increased risk of cardiovascular (CV) complications. This study investigated CV events over a period of up to 5 years in adult patients with chronic hypoparathyroidism treated with recombinant human parathyroid hormone (1–84), rhPTH(1–84), compared with a historical control cohort of patients who did not receive rhPTH(1–84). The rhPTH(1–84)-treated patient cohort was derived from the NCT00732615 (REPLACE), NCT01268098 (RELAY), and NCT01297309 (RACE) clinical trials. A control cohort of adult patients who did not receive rhPTH(1–84) or rhPTH(1–34) was selected from the US Explorys electronic medical record database (Jan 2007−Aug 2019) using selection criteria similar to the enrollment criteria used in the trials. Index date was the day after initiation of treatment for the rhPTH(1–84) cohort and the day after the first calcitriol prescription for the control cohort. The primary outcome was the risk of a composite CV event (defined as any event of cerebrovascular disease, coronary artery disease, heart failure, or peripheral vascular disease) in the rhPTH(1–84) cohort compared with the control cohort through 5 years post-index. Patients with a CV event at baseline were excluded from the analysis. Risk of a CV event was assessed in a Kaplan-Meier analysis and a Cox proportional hazards model adjusted for demographic characteristics, baseline clinical conditions, and baseline serum calcium levels. The analysis included 113 patients in the rhPTH(1–84) cohort and 618 patients in the control cohort. Patients in the rhPTH(1–84) cohort, compared with the control cohort, were younger (mean ± SD age, 47.8±12.0 vs 51.0±16.8 years; P<0.05), a higher proportion were White (94.7% vs 81.9%; P<0.01), and fewer had acute manifestations of hypoparathyroidism before the index date (22.1% vs 69.6%; P<0.001). In a Kaplan-Meier analysis, rhPTH(1–84)-treated patients had a significantly reduced risk of developing a CV event compared with patients in the control cohort (P<0.01); 3.5% of rhPTH(1–84)-treated patients and 16.3% of control cohort patients developed a CV event over the 5-year follow-up period. The adjusted hazard ratio for developing a CV event associated with rhPTH(1–84) treatment vs no rhPTH(1–84) treatment was 0.23 (95% CI, 0.07−0.74; P<0.05). This analysis is limited by differences in patient management under predefined clinical trial protocols for the rhPTH(1–84) cohort vs real-world clinical practice for the control cohort. Over 5 years, patients with chronic hypoparathyroidism treated with rhPTH(1–84) in clinical trials had a significantly reduced risk of CV events compared with a control cohort of patients who did not receive rhPTH(1–84). Further research is needed to better understand the mechanism underlying the association between chronic hypoparathyroidism and risk of developing a CV event.


Sign in / Sign up

Export Citation Format

Share Document