Decision Making for a Companion Diagnostic in an Oncology Clinical Development Program
- PDF / 340,994 Bytes
- 9 Pages / 602.986 x 782.986 pts Page_size
- 4 Downloads / 190 Views
Decision Making for a Companion Diagnostic in an Oncology Clinical Development Program
Drug Information Journal 46(3) 294-302 ª The Author(s) 2012 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0092861512438748 http://dij.sagepub.com
Lee Kaiser, PhD1, Claus Becker, PhD, MBA2,*, Sharad Kukreti, PhD, MBA2, and Bernard Fine, MD, PhD3
Abstract The decision to incorporate the specific evaluation of a candidate companion diagnostic (CDx) in a clinical development plan (CDP) is often difficult and is exacerbated by the lack of relevant decision tools. In this article, we discuss a novel method to assess the probability of technical success (PTS) of a CDP that adequately evaluates a CDx compared with a CDP that doesn’t. We propose splitting the PTS into subjective (biological uncertainty) and quantitative (clinical uncertainty) components, assessing each separately, and then combining them in a decision theoretical manner to obtain an overall success probability of a CDP with and without a CDx. Keywords personalized medicine, expected net present value, portfolio management, probability of technical success, R&D management
Introduction Personalized medicine frequently is described as central to the commercial success of novel medicines and to the societal goal of managing the cost and improving the quality of health care.1,2 Despite this, and perhaps because of the long cycle times in drug development, there are few examples of successful personalized medicines, and these are concentrated in oncology, where companion predictive diagnostics (CDx) help match patients to a specific therapy.3 Although oncology therapeutics are still typically developed in patients based on the histological characteristics of their tumor, it is becoming more common for these therapeutics to be developed in subsets of patients whose tumors have a certain molecular characteristic. Examples of this include her-2 positivity for trastuzumab in breast cancer4 and activating BRAF mutations for a RAF inhibitor in melanoma.5 The decision to design the clinical development plan (CDP) to adequately evaluate a patient subset is typically based on the strength of hypotheses about biological characteristic of the disease, the mechanism of action of the new molecular entity (NME), and the candidate CDx, as well as on commercial considerations. Because these hypotheses were strong for trastuzumab, the clinical trials supporting its approval were limited to patients testing positive with the corresponding assay. In other cases, there may be greater uncertainty that a specific assay will identify patients who differentially benefit from the NME, with
consequent uncertainty whether the CDP should include a robust evaluation of this candidate predictive diagnostic through specific study designs and primary objectives. We describe an approach to this issue that uses the portfolio management concept of the probability of technical success (PTS), which quantifies a development program’s risks. Portfolio management is commonly applied i
Data Loading...
