Positron emission tomography: establishing priorities for health technology assessment.

Abstract

Positron emission tomography (PET) is an expensive diagnostic imaging technology. Despite the long history of PET development, the costs and effectiveness of its use in routine clinical practice remain unknown. Against this background of uncertainty regarding the clinical role of PET, the UK Standing Group on Health Technology requested a review of its current and potential role which would enable research priorities in this area to be established. This 3-month project had two explicit objectives: (1) to review the state of knowledge regarding the clinical applications of PET; (2) to determine the key health technology assessment (HTA) research questions relating to the use of PET in the UK. A literature review to ascertain the state of knowledge regarding the clinical applications of PET and a three-round Delphi study to inform the key HTA research questions relating to the use of PET in the UK were undertaken. The results of an earlier systematic review, published by the Veteran's Health Administration (VHA) in the USA in 1996, were used as the starting point for the literature review. The VHA review was updated and extended by means of MEDLINE and Cochrane Library database searches. Participants in the Delphi study were selected by discussion with five individuals in the UK with an interest in, and awareness of, developments in PET. As a result of their suggestions, 43 individuals were initially invited to participate, of whom two did not feel appropriately qualified. Questionnaires were sent by facsimile to all invited participants, who were asked to return the completed forms by facsimile within a week. The content and structure of the Delphi study was informed by the results of the literature review. The responses and comments of the participants were a major source of information for this report. Clinical applications for PET have been advocated in three broad disease groups: oncology, cardiology and neuropsychiatric disorders. There are currently four PET modalities that need to be considered when assessing its potential clinical role in the UK: full ring PET scanners operating in two or three dimensions (available at five sites); partial ring rotating PET scanners (one currently operating in the UK); coincidence imaging with modified gamma camera technology; and high-energy collimator imaging of 511 keV photons with modified gamma camera technology. There is a paucity of available evidence relating to the cost-effectiveness of the various PET modalities in all of the clinical indications for which the technology is currently being advocated. In addition, many existing reports on the diagnostic accuracy of PET are limited because they are liable to bias and often relate only to very small patient numbers. The results of the Delphi study indicated that the four most important research priorities for the NHS, in descending order of their importance, are: (1) the relative cost-effectiveness of (a) full ring PET, (b) gamma camera PET using coincidence imaging and (c) existing diagnostic strategies to determine staging prior to operative intervention for lung cancer; (2) partial ring PET compared with full ring PET in oncology (3) the relative cost-effectiveness of (a) full ring PET, (b) gamma camera PET using coincidence imaging and (c) existing diagnostic strategies to stage and monitor treatment response in breast cancer; (4) the relative cost-effectiveness of (a) gamma camera PET using coincidence imaging and (b) 511 keV collimated positron imaging for assessing myocardial viability when selecting patients for revascularisation surgery. Vignettes describing each of the research priorities are provided in the main report. The findings of this project, which was undertaken rapidly in order to inform HTA research prioritization in the UK, provide a contemporary overview of the potential clinical role for PET in the NHS. Evidence is needed that using PET as a diagnostic