Valuations as Mediators Between Science and the Market: How Economic Assumptions Shape Pharmaceutical Trial Designs

Abstract

How can economic assumptions be present in the heart of commercially driven drug development research? Such assumptions underpin industry-based bio-statistical discussions around a new pharmaceutical trial design, the ‘compound finder’. This example illustrates several ways in which trials might be designed and situated in the larger setting of interlinked valuation practices central to the development, distribution, and use of pharmaceuticals. It shows how economic assumptions and considerations can be differently entwined with endeavors to produce knowledge. Different trial designs may further differ in what knowledge they produce. Adaptive design trials (ADTs), of which the compound finder is one kind, share the feature that they might be the object of thousands of simulations to specify the design taking many different kinds of considerations into account. These considerations include several economic aspects such as trial costs and assumptions about the future market. ADTs will likely continue to become more common in the years to come, even if the future for the specific compound finder trial design is uncertain. Yet, the continued rise in importance of ADTs means a further intimate entwining of economic assumptions into the specification of trial designs. This will be consequential for what knowledge is produced as well as where and how treatments are assessed.

Keywords:

• drug development
• economic assumptions
• trial design
• medical research
• valuation

Acknowledgements

The overall project takes an interest in the experimental design of biomedical experiments as a site of valuations. The fieldwork for this sub-study was done by Helgesson. This version has benefited from comments by participants at a CSI seminar at Mines ParisTech, and in particular comments from Madeleine Akrich, Liliana Doganova, Alexandre Mallard, and Vololona Rabeharisoa. We would finally like to thank the editors Les Levidow and Kean Birch as well two anonymous reviewers for helpful comments on earlier versions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Claes-Fredrik Helgesson is a professor in Technology and Social Change at Linkoping University, Sweden. His research interest concerns in broad terms the intertwining of economic organizing, science and technology. He works in the fields of economic sociology and science and technology studies (STS), currently often in the emerging field of valuation studies. Helgesson is co-founder and co-Editor-in-Chief of Valuation Studies, and is a co-editor with Isabelle Dussauge and Francis Lee of Value Practices in the Life Sciences and Medicine (Oxford University Press, 2015). His webpage is available at cfhelgesson.net.

Francis Lee is associate professor of Technology and social change at Uppsala University, Sweden. His research deals with valuations and politics of knowledge production. Of particular concern is how technological infrastructures—such as standards, algorithms, or big data—become part of knowledge production. He has published widely in science and technology studies (STS), and is a coeditor with Isabelle Dussauge and Claes-Fredrik Helgesson of Value Practices in the Life Sciences and Medicine (Oxford University Press, 2015). His webpage is available at francislee.org.

ORCID

Claes-Fredrik Helgesson http://orcid.org/0000-0002-6074-6088

Francis Lee http://orcid.org/0000-0002-7206-2046

Notes

1 For a discussion on how the study valuations can be used to avoid separating the epistemic and economic as belonging to autonomous spheres, see Dussauge et al. (2015).

2 21st Century Cures Act, H.R. 34, 114th Cong. (2015), section 3021. FDA published a draft guidance on ADTs already in 2010 (Food and Drug Administration, 2010).

3 The approach used for statistical analysis is another area where there might be differences. Whereas traditional RCTs rely on so-called frequentist statistical analysis, ADTs might rely on so-called Bayesian inference. Yet, ADTs might also rely on traditional frequentist approaches (Chevret, 2012). An extensive discussion of such differences between RCTs and ADTs is beyond the scope of this paper.

4 The specific details of the design actually involved two doses for each compound, which gave the trial a total of seven arms (two for each compound plus placebo). We have chosen to present the case in a simplified form as if it only had four arms, since the added complication of two doses per compound is unimportant for our overall analysis.

5 One example: The rules for how different subjects are to be allocated to different treatments can have large consequences for how many patients end up with what might emerge to be considered a sub-par treatment.

Additional information

Funding

This paper reports a sub-study within the ‘Trials of Value: On the valuation practices in designing medical experiments’ project funded by Riksbankens Jubileumsfond [grant number P11-0034:1].