Please use this identifier to cite or link to this item: http://hdl.handle.net/2381/42261
Title: Using Simulated Individual Patient Data (IPD) From Published Registry and Ipd From The Seer-Medicare Registry To Extrapolate Results From Randomised Clinical Trials (RCTS) in Metastatic Melanoma
Authors: Martina, R.
Jenkins, D.
Bujkiewicz, S.
Dequen, P.
Abrams, K. R.
Lees, M.
Davies, J.
Kalf, R.
Makady, A.
First Published: 31-Oct-2016
Publisher: Elsevier for International Society for Pharmacoeconomics and Outcomes Research
Citation: Value in Health, 2016, 19 (7), pp. A717-A717 (1)
Abstract: Objectives Pre-marketing authorisation estimates of survival are generally restricted to those observed directly in randomised clinical trials (RCTs). However, for regulatory and Health Technology Assessment (HTA) decision-making a longer time horizon is often required than is studied in RCTs. Therefore, extrapolation is required to provide long-term evidence of treatment effect. Registry data can provide evidence to support extrapolation of treatment effects. The aim of this work is to use real world data (RWD) and non-linear regression models to evaluate long term survival. Methods IPD was simulated from published survival curves of patients with MM in real world. Additionally, SEER-Medicare registry data were combined with RCT data to estimate long-term survival of patients with MM. Exponential, Weibull, Gompertz, log-logistic, log normal parametric survival models and a non-parametric model were fit to the RCT data, and used to extrapolate the data from 48 months to 72 months. A naive extrapolation was applied as well as extrapolation based on simulated IPD, summary data and SEER-Medicare data. Adequacy of the models was assessed through comparisons of the log-likelihood, whilst treatment effects were estimated using the restricted Area under the Curve (AUC). Reliability was assessed through visual inspection of the fit to the long-term data. Results Blending RCT and registry data allowed for reliable estimation of long term survival of patients with MM using a log-logistic, lognormal model and non-parametric model. The log-logistic and lognormal model estimated long term survival with reduced uncertainty when including the SEER-Medicare database compared to a naive extrapolation approach. Conclusions The results showed that the use of the SEER-Medicare registry decreased the uncertainty in long term prediction of overall survival in patients with MM. The use of registry data may be an acceptable approach for pharmaceutical companies, regulatory and HTA decision bodies for assessing long term survival of cancer treatments.
DOI Link: 10.1016/j.jval.2016.09.2121
ISSN: 1098-3015
eISSN: 1524-4733
Links: https://www.sciencedirect.com/science/article/pii/S1098301516334878?via%3Dihub
http://hdl.handle.net/2381/42261
Version: Post-print
Status: Peer-reviewed
Type: Conference Paper
Rights: Copyright © Elsevier 2016. This version of the paper is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
Description: Abstract only
Appears in Collections:Conference Papers & Presentations, Dept. of Health Sciences

Files in This Item:
File Description SizeFormat 
Martina+et+al+Abstract+ISPOR+2016+final.pdfPost-review (final submitted author manuscript)348.83 kBAdobe PDFView/Open


Items in LRA are protected by copyright, with all rights reserved, unless otherwise indicated.