Acceptance of Population-Adjusted Indirect Treatment Comparison Methods In Nice Assessments

• Published in: Value in health 2017-10, Vol.20 (9), p.A695
• Main Authors: Ivanescu, C, Skaltsa, K, Kráľ, P
• Format: Article
• Language: English
• Subjects: Clinical medicine; Comparative studies; Decision making; Feedback; Justification; Matching; Medical treatment; Meta-analysis; Oncology; Scrutiny; Technology
• ISSN: 1098-3015; 1524-4733
• DOI: 10.1016/j.jval.2017.08.1785

OBJECTIVES: Guidelines for applying methods for population-adjusted ITC, e.g. matching adjusted indirect comparisons (MAIC) and simulated treatment comparisons (STC), in submissions to NICE were published in December 2016. These methods overcome issues like disjointed evidence networks and heterogeneity in network meta-analyses (NMA), and can produce comparative evidence where it may otherwise be impossible. Previous research found 4 NICE appraisals using MAIC published between 2010 and 2016 [Thorn et.al 2016]. The objective of this study was to review recent submissions to NICE to determine the use and acceptability by the payer of the population-adjusted ITC methodology. METHODS: All manufacturer submissions, evidence review group reports and NICE committee feedback documents published between 01JAN2017 and 25JUN2017 were reviewed and information on methods for performing ITCs was extracted. RESULTS: In total, 37 single technology appraisals (STA) were identified. Ten did not include any ITC, 19 included standard NMA, four included a naive comparison and seven included MAIC; some appraisals included more than one method. The STC method was not applied in any of the appraisals. All seven STAs using MAIC were in oncology. Four were recommended, two were restricted and one received a preliminary negative recommendation. The method was used to address a disconnected network in three STAs and to adjust for trial population heterogeneity in the remaining four. In all cases, the method was accepted for decision-making although several limitations were highlighted, such as lack of rationale for the choice of method, limited justification for the choice of matching variables, limited possibility of matching outcome definitions or inclusion/exclusion criteria between studies. CONCLUSIONS: Despite the short timeframe, our research demonstrates that the use of population-adjusted ITCs has been increasing in NICE STAs, with 19% of STAs including MAIC. The method has been generally accepted, although considered equivalent to observational evidence and its underlying assumptions subjected to close scrutiny.