Feb 22, 2023
Clopidogrel, prasugrel and ticagrelor: lack of data for network meta-analysis
One drug manufacturer did not provide the requested additional data. However, exploratory analyses show that these would have been important.
Is clopidogrel, prasugrel or ticagrelor – each in combination with acetylsalicylic acid (ASA) – better for preventing atherothrombotic events in adults with acute coronary syndrome undergoing primary or delayed percutaneous coronary intervention (PCI)? According to the less than satisfactory results of a report published by the German Institute for Quality and Efficiency in Health Care (IQWiG), this remains an open question, even though the 3 drugs have been in use for many years. This is because the informative value of the network meta-analysis depends to a large extent on study data that the Institute requested from a drug manufacturer, but did not receive.
An English translation of the rapid report is now available.
Comparison of 3 antiplatelet drugs
The P2Y12 receptor antagonists clopidogrel, prasugrel and ticagrelor are used in combination with ASA to inhibit platelet aggregation in acute coronary syndrome and thus prevent atherothrombotic events. All 3 drugs have been subject to benefit assessments by IQWiG, starting with clopidogrel in 2004. Ticagrelor was assessed in 2011 in the first early benefit assessment after the Act on the Reform of the Market for Medicinal Products (AMNOG) came into force in Germany.
What was still missing was a comparison of the 3 drugs with each other. The Federal Joint Committee (G-BA) commissioned IQWiG to carry out a comparative benefit assessment in 2021, specifically in the therapeutic indication for prasugrel, i.e. for the prevention of atherothrombotic events in adults with acute coronary syndrome undergoing primary or delayed PCI.
Lack of data for an informative network meta-analysis
Such a triangular comparison is performed using a network meta-analysis. However, this can only provide informative results if the study data considered are sufficiently similar. If studies differ in design, it may be necessary to use subpopulations in order to compare sufficiently similar populations.
Although IQWiG identified studies for each of the three legs of the triangle, the published data lacked information on how to identify subpopulations suitable for the comparison. The Institute therefore requested additional data from 2 drug manufacturers and an academic research group. While the research group and one of the manufacturers provided all the necessary information, the other manufacturer, who was responsible for a particularly large study that was therefore essential to the analysis, did not provide any information. In the case of this report, the G-BA and IQWiG were dependent on the good will of the manufacturers because, unlike with AMNOG, there is no legal obligation to provide data for comparisons of several drugs that have already been subject to regular benefit assessments.
Exploratory comparison shows differences
To assess whether the evidence was relevant for the results of the comparison of the drugs, IQWiG assessed the available data from the 3 largest studies, each of which compared 2 drugs and thus covered the entire triangle. One of the 3 studies included exactly the population that was considered in the regulatory approval process (“approval population”), which was relevant for the benefit assessment. In a second study, the approval population could be separated from the larger study population. In the third study, however, this was not possible because the manufacturer did not provide the requested data.
The calculated network meta-analyses were sensitive to this difference. For example, all-cause mortality was lower for one drug in the overall study population, but higher for another in the approval population.
Volker Vervölgyi, Division Head in IQWiG's Drug Assessment Department, noted: "Both the separate consideration of patient populations and the consideration of the drug approvals appear to be important to answer the G-BA's research question."
Procedure of report production
In April 2021, the G-BA commissioned IQWiG to prepare the comparative benefit assessment of clopidogrel, prasugrel and ticagrelor in acute coronary syndrome as a rapid report in an accelerated procedure. The interim reports were therefore not published and were not subjected to a commenting procedure. IQWiG sent the (German-language) rapid report to the commissioning agency, the G-BA, in January 2023 and published it online in February 2023. The English translation of the report was published in September 2023.