Presentation (CTI)

OBJECTIVES: Network meta-analyses (NMAs) are widely used in health technology assessment (HTA) to allow indirect comparisons of clinical trial data. Such data are often immature at the time of submission, leading to an incomplete understanding of relative efficacy. This study replicates an NMA from a past HTA and explores how including more recent data could affect results and influence decision-making.
METHODS: A Bayesian NMA of overall survival in non-small cell lung cancer (NSCLC) was replicated, encompassing eight treatments, using publicly available documentation from the National Institute for Health and Care Excellence (NICE) Technology Appraisal 557 (TA557). Where necessary, the analysis was supplemented through estimating redacted values from published data. The NMA was then iteratively updated with the latest data cuts from relevant clinical trials to evaluate the impact of new evidence that was unavailable at the time of the original appraisal.
RESULTS: Using information from the literature, results were produced that were comparable to TA557. After incorporating results from the most recent published data cuts, some results were materially different; the hazard ratio (HR) and 95% credible intervals for pembrolizumab plus pemetrexed-platinum versus placebo plus pemetrexed-platinum was 0.49 (0.35, 0.68) in TA557, which increased to 0.60 (0.48, 0.77) with newly published data.
CONCLUSIONS: The results of the current study demonstrate that incorporating more recent data can lead to meaningful changes in results, highlighting the dynamic nature of NMAs in HTAs. Seemingly small changes in point estimates of HRs can have a profound impact on cost-effectiveness results, and thus, decision-making. Additionally, publicly available information can be limited, hindering reproducibility. These findings emphasise the need for updating NMAs when new data is made available, greater transparency in reporting, and improved access to complete data to ensure health policy recommendations reflect the most current evidence.