May Issue - Reflections from the real world – benefits and limitations of phase IV studies and registries
Emeritus Professor Sylvia Haas, MD
Sylvia Haas is an Emeritus Professor of Medicine and was previously Director of the Haemostasis and Thrombosis Research Group at the Institute for Experimental Oncology and Therapy Research, Technical University of Munich, Germany. Her scientific focus is the development of new antithrombotic therapies, laboratory monitoring of anticoagulants, biomarkers, and tumour-associated thrombosis. She is a member of several international professional societies and is on the editorial boards for several peer-reviewed journals. Click here for a full biography.
Welcome to the May 2015 edition of the Thrombosis Adviser newsletter.
Dear Ladies and Gentlemen,
Differences between randomized controlled trials and phase IV studies/registries Randomized controlled trials (RCTs) are well established as the gold standard for assessing drug efficacy and safety. Although vital in drug development and approval, RCTs have limitations because the trial design and study outcomes may not readily translate into routine clinical practice. The need to generate clinical evidence by controlled RCTs, including the enforcement of strict inclusion/exclusion criteria, may lead to under-representation of subpopulations within a target patient group or treatment conditions that are not fully reflective of the ‘real world’. For example, patients with co-morbidities or those taking multiple medications are often not included in clinical trials.
By contrast, phase IV studies and registries capture real-world epidemiological data, treatment patterns and drug use in a wider cross-section of patients. They are the first ‘real’ test of a drug in a clinical practice setting across a large number of patients, i.e. ‘where the rubber meets the road’. The main objective for phase IV studies and registries is to identify or confirm the safety landscape seen in phase III studies, with additional input on rare or long-term adverse events – often only observed when several thousand patients are studied. Cost-effectiveness outcomes and quality of life observations may also be developed from these data. Observational studies and registries also have limitations, such as the potential for over-reporting of adverse events when a drug is first approved (the so-called Weber effect) and the potential for vague or missing data due to less rigorous reporting standards than RCTs.
The key differences between RCTs and real world studies are listed below:
The terms ‘phase IV studies’ and ‘registries’ should not be used interchangeably. Although the two are similar, non-interventional phase IV studies are normally limited to the observation of the behaviour of a specific drug or medical device in the real world with a set number of patients and defined study duration. Registries have a much broader scope and often do not have a predefined time frame. A patient registry uses observational study methods to collect a range of data (clinical and non-clinical) and form a comprehensive database. Registries have prespecified efficacy and safety outcomes for a population with a particular disease, condition or exposure for a specific purpose, but may also provide epidemiological or other data that were not necessarily anticipated at the outset. Data are captured in an inclusive way, disqualifying few patients and producing supplementary evidence that may be generalized to more diverse patient populations.
Real-world studies of anticoagulants
Registries are especially important when evaluating new therapies such as novel anticoagulants for stroke prevention in patients with atrial fibrillation and management of venous thromboembolism and acute coronary syndrome. Through phase IV studies and registries, a more thorough understanding of the impact of new treatments on healthcare systems can be acquired. Anticoagulation registries can also obtain real-world data about how bleeding events are managed and provide more information on benefit–risk profiles in specific patient groups. In the anticoagulation setting, there are multiple ongoing phase IV studies/patient registries and many have published data. These include:
|Study||Patient type||Start date||Number of patients||Duration of patient follow-up||Key observations||References*|
|XAMOS phase IV study||Elective hip or knee replacement surgery or hip-fracture surgery||Early 2009||17,701||Enrolment to 3 months post-surgery||Confirmed favourable benefit–risk profile of rivaroxaban seen in phase III studies||Turpie et al 2012|
|Turpie et al 2014|
|XALIA phase IV study||Objectively confirmed acute DVT treated with rivaroxaban||June 2012||~4800 planned (ongoing)||1 year from the final date of patient enrolment||Results expected in 2015||Ageno et al 2014|
|RIETE registry||Acute DVT or PE||March 2001||>40,000 to date (ongoing)||Minimum of 3 months||Both fatal PE and fatal bleeding are more common in cancer patients with VTE than in patients without cancer||Monreal et al 2006|
|In outpatients with DVT, home treatment was associated with a better outcome than hospital treatment||Lozano et al 2014|
|XANTUS phase IV study||Non-valvular AF prescribed rivaroxaban, irrespective of stroke risk||June 2012||~6000||1 year||Findings not yet reported||Camm et al 2014|
|GARFIELD-AF registry||Newly diagnosed non-valvular AF and ≥1 investigator-defined stroke risk factor||December 2009||55,000 planned (ongoing)||At least 2 years||OACs frequently not used according to guidelines||Kakkar et al 2012|
|Kakkar et al 2013|
|PREFER in AF registry||Patients with a diagnosis of AF in Europe||January 2012||7243||1 year||Frequent and possibly inappropriate use of combined oral anticoagulant and antiplatelet agents||Kirchhof et al 2014|
|De Caterina et al 2014|
|GLORIA-AF registry||Newly diagnosed with AF at risk of stroke||May 2011||Up to 56,000 (ongoing)||3 years||Findings not yet reported||Huisman et al 2014|
|ORBIT-AF registry||Community-based outpatients with AF||June 2010||10,098||³2 years||OAC use is high and driven by bleeding risk as well as stroke risk||Cullen et al 2013|
|RE-LY AF registry||Patients presenting to emergency department with AF||Sep 08||15,4||1.5 years||Large variations in demographics and risk factors exist between global regions||Oldgren et al 2014|
|Dresden NOAC registry||Patients treated with novel OACs||October 2011||1776 treated with rivaroxaban||2 years||Rates of major bleeding with rivaroxaban may be lower and the outcome similar to that seen with VKA||Beyer-Westendorf et al 2014|
*Not an exhaustive list.
AF, atrial fibrillation; DVT, deep vein thrombosis; OAC, oral anticoagulant; PE, pulmonary embolism; VKA, vitamin K antagonist; VTE, venous thromboembolism.
Registries can tell us how innovations may be applied for the greatest benefit to each patient on the individual level, and are, therefore, as important as RCTs. Phase IV observational studies and registries may also provide data for additional or enhanced value post-approval, to include quality of life and cost-effectiveness outcomes.
The journey from drug discovery to clinical use is a long one. RCTs are obviously essential steps, and drug development is generally thought to end with the phase III trials that lead to marketing authorization. However, phase IV studies are just as important, being the first time that a drug is assessed under real-world conditions. In addition, registries give us the clearest picture of how patients are treated in clinical practice and may generate hypotheses for future RCTs. In conclusion, real-world data and RCTs should be seen as having complementary roles in evaluating patient outcomes.