Efficacy and safety of sacubitril/valsartan in the treatment of heart failure: A systematic review incorporating unpublished clinical study reports

Problem

Sacubitril/valsartan is a first in class angiotensin-receptor neprilysin inhibitor used in the treatment of chronic heart failure. Several limitations have been highlighted with this novel medication’s pivotal phase III trial, PARADIGM-HF. The primary outcome in PARADIGM-HF was a composite of death from cardiovascular cause or first hospitalisation for heart failure. Our study systematically reviews and synthesizes all available RCT evidence on the efficacy and safety of sacubitril/valsartan in chronic heart failure, including unpublished sources of clinical trial data from Clinical Study Reports (CSRs).

Approach

We conducted a systematic search of literature databases, grey literature, regulatory drug documents and clinical trial registries, and we requested CSRs from the European Medicines Agency (EMA) and Clinical Study Data Request (CSDR). Two authors (DB and FM) independently screened all citations and screened titles and abstracts of remaining studies for full text suitability. Both authors then extracted data in duplicate from publications, clinical trial registries and CSRs separately. We used CSR data in preference as the presumed most reliable source, followed by journal publication and then trial registry data. We conducted a meta-analysis using CSR-informed estimates, using generic inverse variance (GIV) with random effects statistical models. Two authors (DB and TF) completed a risk of bias (RoB) assessment of included studies, using the Cochrane RoB tool. We used the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) methodology to report the certainty of evidence from included studies.

Findings

We identified 3,744 records in total. From these we identified 15 relevant clinical trials, of which 9 had results on clinical trial registries, and we obtained CSRs for three studies. The hazard ratio for the primary outcome of composite of cardiovascular death or first hospitalisation for heart failure was 0.99 (95% CI 0.70-1.4; p=0.959), for first hospitalisation for heart failure was 1.00 (0.69-1.44; p=0.997) and for cardiovascular death was 0.87 (0.62-1.22; p=0.410). These results showed a reduction in efficacy compared with the original pivotal trial.

For common safety outcomes, there was a greater risk of symptomatic hypotension with sacubitril/valsartan (RR 1.44; 1.21-1.71, p<0.001) and of hyperkalaemia (RR 1.95; 1.10-3.48, p=0.023). The risk ratio (RR) for myocardial infarction was 1.02 (95% CI 0.84-1.23, p=0.84) and for atrial fibrillation was 1.05 (0.97-1.14, p=0.639), which is in contrast to the results from PARADIGM-HF which reported significantly reduced risk with sacubitril/valsartan for these two outcomes.

Consequences

Our meta-analysis found no evidence of a significant difference between sacubitril/valsartan and comparator for the original primary efficacy endpoint and a greater risk with sacubitril/valsartan for selected safety outcomes. Our results showed a reduction in the effect estimate for sacubitril/valsartan for 3 main efficacy outcomes compared with the original PARADIGM-HF trial, with a change in statistical significance to these outcomes now being non-significant.