Note from the National Guideline Clearinghouse (NGC): The National Institute for Health and Clinical Excellence (NICE) commissioned an independent academic centre to perform an assessment of the manufacturer's submission on the technology considered in this appraisal and prepare an Evidence Review Group (ERG) report. The assessment report for this technology appraisal was prepared by the Southampton Health Technology Assessment Centre (see the "Availability of Companion Documents" field).
Clinical Effectiveness
Critique of Manufacturer's Approach
Description of Manufacturer's Search Strategy and Comment on whether the Search Strategy Was Appropriate
The sources used by the manufacturer for the search (Embase, Medline, Medline in Process, NICE, Cochrane, NCCHTA, American Society of Clinical Oncology [ASCO], National Health Service (NHS) CRD, Internal databases, internet), are appropriate and comprehensive. Additional databases that could have been used to obtain the clinical evidence are Biosis and Web of Science, although it is unlikely that they would have yielded any additional key results. The manufacturer has documented the use of ASCO, which is the key source of information for sourcing ongoing cancer trials. The search documentation could have been widened or clarified to include mention of sources such as the national research register, controlled clinical trials, clinicaltrials.gov, in order to track any ongoing trials.
The search strategies in the manufacturer's submission (MS) are transparent, fully documented, and reproducible. The ERG reproduced components of the search on 23rd May 2006. The main search (Search 1, MS) yielded similar results, but the ERG identified 457 citations with the paclitaxel search (after amending to take account of extra references since November 2005), compared with 84 in the manufacturer's search. The manufacturer's Embase search was from 1988, whereas the ERG's was from 1980, but searches were otherwise as similar as was feasible. A brief scan of the identified references suggested that none of the 'extra' references were relevant to the systematic review.
The MS states that the search included data up until the 28th November 2005. For the sake of completeness, the ERG considers that an update search should have been re-run for all the study drugs.
Statement of the Inclusion/Exclusion Criteria Used in the Study Selection and Comment on whether They Were Appropriate
The MS describes an appropriate method of identifying and screening references for inclusion in the systematic review. Three independent reviewers applied pre-specified inclusion/exclusion criteria to citations identified by the searches, and discussed any unclear references until agreement was reached.
The MS specified the following inclusion criteria for the systematic review of the literature:
- Study design - original studies reporting final results of phase III clinical trials
- Interventions - gemcitabine/paclitaxel, docetaxel/capecitabine, paclitaxel monotherapy or docetaxel monotherapy
- Population - patients with metastatic breast cancer (MBC) who have been treated and failed on prior anthracycline treatment in an adjuvant or neoadjuvant setting
- Outcome measures - no outcome measures were specified in the inclusion/exclusion criteria
Phase I and II trials, observational studies, letters to the editor and editorials were excluded from the systematic review. The manufacturer did not state whether published systematic reviews would be considered in the review, and did not state clearly whether conference abstracts would be included or excluded. The specified inclusion/exclusion criteria were appropriate and reflect the information given in the decision problem.
Economic Evaluation
Cost Effectiveness Searches
The searches for cost-effectiveness studies are not clearly described in the MS. The searches described in the clinical effectiveness section of MS appear to have covered cost-effectiveness, since the reviewers identified studies from these which were only applicable to the economic model. However, the cost-effectiveness section then describes a separate search (dated 8th September 2005) of all the key databases. This search is not well documented, and only basic keywords are included in table 19 of the MS. The citations identified by this search are different from those identified in the earlier stage of the review, and appear to have been used to inform the design of the economic model.