Restoring invisible and abandoned trials: a call for people to publish the findings

We write to issue a call to action to restore the reporting of a randomized clinical trial on evolocumab in patients with cardiovascular disease sponsored by AMGEN (FOURIER trial):

Sabatine MS, Giugliano RP, Keech AC, Honarpour N, Wiviott SD, Murphy SA, Kuder JF, Wang H, Liu T, Wasserman SM, Sever PS, Pedersen TR, for the FOURIER Steering Committee and Investigatorset al. Evolocumab and Clinical Outcomes in Patients with Cardiovascular Disease. N Engl J Med 2017;376(18):1713-1722. doi: 10.1056/NEJMoa1615664. ClinicalTrials.gov Identifier: NCT01764633.

We understand this trial was misreported since cardiovascular events were not published in full and also relevant limitations of the trial were disregarded.

We intend to restore the written record for this trial in accordance with the principles of the Restoring Invisible and Abandoned Trials (RIAT) initiative (1).

Our rationale for correcting the record
The protocol of the FOURIER clinical trial reads as follows: “Given that evolocumab is expected to reduce LDL-C by even more than 1.0 mmol (39 mg/dL) in patients already on an intense dose statin therapy (based on phase 1b results), it is expected that the FOURIER study will demonstrate that evolocumab administration will produce a significant reduction of cardiovascular events, compared to placebo, in high risk patients on maximally tolerated atorvastatin background therapy” (protocol synopsis, page 39).

The published trial reports the results on a pre-defined composite cardiovascular primary end point (cardiovascular death, myocardial infarction, stroke, hospitalization for unstable angina, or coronary revascularization) and on the following secondary outcomes (2):
- cardiovascular death, myocardial infarction, or stroke (key secondary endpoint)
- death from any cause
- hospital admissions for worsening heart failure

However, in the NEJM publication there is no information on the total cardiovascular events and serious cardiovascular events in the evolocumab group vs placebo.

According to the study protocol, the “Safety Objective” is to evaluate the safety and tolerability of evolocumab, compared with placebo, in subjects with clinically evident cardiovascular disease. The approved version of the protocol (pages 78-79) defines Serious Adverse Effects (SAE) in the same way as the International Conference for Harmonization (ICH) does. At the same time, the protocol indicates that for grade 4 “life threatening” adverse events it was left to the investigator’s judgement to also report these abnormalities as SAE (page 175), which could have led to underestimating this safety outcome.

After patient recruitment had already started, the protocol was amended in order to clarify that all cause death, myocardial infarction, stroke, revascularization, hospitalization for unstable angina, hospitalization for heart failure, and transient ischemic attack will not be considered serious adverse events in this study, but will be handled as efficacy endpoints that the data monitoring committee will be requested to follow. This means that CV events classified as endpoints were not considered as SAEs even though they also met the ICH definition for SAEs. In addition, since CV adverse events were not reported in the NEJM publication, the incidence of overall and serious CV events remains unknown.

The protocol was also amended to have some secondary outcomes switched after the trial had already got started. In the published article there was no mention about this outcome switching carried out during the implementation of the trial, and also relevant potential conflicts of interest of the main author were not reported. According to Amgen transparency disclosures and ProPublica website, the main author of the NEJM publication received payments during 2014-2016 worth over 130,000 US $ for consulting activities directly related to Repatha (evolocumab). However the main author declares having no conflicts of interest in relation to the FOURIER trial.

According to the information in the Clinical Study Report (CSR), the trial failed to secure blinding and this should also be stated among the limitations in the published article.

Scope of our restoration
We aim at restoring the information on cardiovascular events, serious CV adverse events and also show some relevant limitations of the trial that were not reported. If appropriate, events re-adjudication will be carried out to correctly adjudicate and analyze events.

Our sources
On March 27, 2018, the EMA granted Juan Erviti access to the CSR of the FOURIER trial. Over the last year most of the information requested was sent to the researcher and the rest is expected to be received within the next few months. The information already obtained is enough to proceed to the restoration of the trial. Other publicly available sources like Amgen transparency disclosures and ProPublica website will be used to restore the information on potential conflicts of interest.

Our objectives
- To report serious and total CV adverse events according to the ICH definition.
- To report serious and total adverse events according to the ICH definition.
- To re-adjudicate events and re-calculate the results for the primary and secondary outcomes if appropriate.
- To state the real conflicts of interest of the authors in relation to the FOURIER trial, failure to secure blinding and report about the outcome switching carried out during the implementation of the trial.
- Other possible inconsistencies between the results in the CSR and the published article that may arise.

References
1. Doshi P, Dickersin K, Healy D, Vedula SS, Jefferson T. Restoring invisible and abandoned trials: a call for people to publish the findings. BMJ. 2013 Jun 13;346:f2865.
2. Sabatine MS, Giugliano RP, Keech AC, Honarpour N, Wiviott SD, Murphy SA, et al. Evolocumab and Clinical Outcomes in Patients with Cardiovascular Disease. N Engl J Med 2017;376(18):1713-1722. doi: 10.1056/NEJMoa1615664