Real-world evidence across a medicine's lifecycle
Real-world evidence serves a medicine at every stage of its life. In early development it informs how a product is positioned and how its trials are designed, within a single study it could serve design, enrolment and analysis, and after launch it addresses long-term effectiveness and safety, cost and value, comparison with competitors and the effects of switching to generics.
Its role across the lifecycle
Real-world evidence has emerged as a vital complement to randomised controlled trials, addressing questions of effectiveness, safety and cost-effectiveness across a medicinal product's lifecycle. Regulators use real-world data for post-marketing safety assessment and surveillance of medicinal products, and payers and health technology assessment bodies use them to inform cost-effective coverage decisions.
Real-world evidence historically served mainly to assess disease statistics and to fulfil post-authorisation safety monitoring obligations. Real-world data and evidence have long served many purposes, such as characterising population health and disease trends and studying the risk associated with different exposures, and they have been an important part of drug safety surveillance, especially once a newly approved medicine starts to be used in clinical practice. Real-world data have also long supported continued benefit-risk evaluations and label extension applications, for example in rare diseases with limited patient populations or where traditional clinical trials would be unfeasible or unethical.
More recently, real-world evidence has taken on a strategic role in all phases of the product lifecycle, from early development to post-launch. It has become an essential component of evidence generation across the lifecycle. Real-world data and evidence are increasingly used in clinical development and in regulatory decision-making on new drug applications, and real-world data are becoming an important supplementary source of evidence in the clinical development of new medicines and in regulatory approval decisions.
The evidence required differs along the lifecycle:
- Development: the standard of care and natural history, unmet need and disease burden, patient recruitment and trial design.
- Launch and conditional review: budget impact and post-marketing commitments.
- Growth phase: comparative safety and effectiveness, patient-reported outcomes, adherence, differentiation in subpopulations, and utilisation and prescribing patterns.
- Mature phase, as competitors go generic: head-to-head comparative safety and effectiveness, the effects of switching to generics on therapy outcomes, and patients' preferences.
Early development
Real-world evidence is best planned early, with a forward-looking approach that anticipates future needs. In early development its generation focuses on supporting decisions on positioning investigational products and on informing trial design and start-up, work that includes characterising patient populations, understanding disease burden and gathering insights into clinical practice.
The European Scleroderma Trials and Research (EUSTAR) registry has been used to generate real-world evidence on systemic sclerosis-associated interstitial lung disease. A prospective analysis of five years of lung function data from 800 patients in the registry identified risk factors for the disease and patterns of its progression. A retrospective analysis of 6,000 patients in the registry used stepwise cohort enrichment to identify patients at risk of progression and to determine how many would be eligible for a trial, with the aim of informing the selection of inclusion criteria and target populations.
Claims databases have similarly been used to generate real-world evidence on unmet needs in generalised pustular psoriasis. A Japanese study compared patients with generalised pustular psoriasis and patients with plaque psoriasis against the general population for comorbidities, medication use, healthcare resource use and costs over one year of follow-up. Patients with generalised pustular psoriasis had a higher disease burden, greater reliance on systemic treatment and more healthcare use than those with plaque psoriasis. A US claims study showed the significant economic burden of generalised pustular psoriasis and palmoplantar pustulosis, with higher healthcare costs and more frequent inpatient visits than in the general population.
Clinical development
Big data could be applied at any stage of a study, from design through enrolment to data analysis. Regulatory medical writers need to consider how best to bring it into their work outputs. Before a study starts, existing data from registries and electronic health records could help identify an appropriate target population, define more targeted eligibility criteria, show whether enough patients are likely to be available, and even supply baseline data. Study efficiency could thus be greatly improved by cutting the cost of and time for recruitment, reducing attrition and minimising later changes to the protocol.
Throughout the study, data from digital sources such as mobile applications and wearables may be available faster than conventionally collected data, allowing prompt futility analyses in a study or benefit-risk assessment in post-marketing surveillance, and so more rapid decision-making. Registries also help track patients during study follow-up under their usual care, minimising loss to follow-up and reducing missing data.
After launch
After launch, real-world evidence generation focuses on long-term effectiveness and safety in routine clinical practice, to support clinical trial findings, payer discussions and the clinical adoption of a new product. Post-marketing surveillance data and real-world data from medical and administrative claims and electronic health records may be used to validate safety and efficacy claims from randomised controlled trials. Real-world evidence also evaluates cost-effectiveness, complementing the findings of randomised trials. Its insights into cost-effectiveness and healthcare resource utilisation enhance understanding of a medicine's value, supporting discussions with payers and informing treatment planning from a pharmacoeconomic perspective. Real-world evidence might also be needed to support pricing discussions, the development of treatment guidelines and clinical adoption.
Post-authorisation safety studies may be needed to fulfil post-marketing commitments to regulators, and they are often conducted in non-interventional settings to complement the efficacy and safety data available at the initial marketing authorisation. One such study addressed emerging safety concerns suggesting a link between several glucose-lowering therapies, sodium-glucose cotransporter-2 inhibitors among them, and acute pancreatitis, a relatively rare but potentially serious and occasionally fatal condition. Using two large US claims databases, it compared the incidence of acute pancreatitis between patients with type 2 diabetes on metformin starting empagliflozin and those starting a sulfonylurea, and its results supported existing evidence that empagliflozin is not associated with an increased risk.
Empagliflozin from trial to routine care
The EMPA-REG OUTCOME trial, prospective and placebo-controlled, showed lower relative risks of cardiovascular death, all-cause mortality and hospitalisation for heart failure with empagliflozin in patients with type 2 diabetes and established cardiovascular disease. On the basis of this trial the FDA expanded the indication for empagliflozin to reducing the risk of cardiovascular death in patients with type 2 diabetes and cardiovascular disease, and major clinical guidelines on diabetes treatment for patients with cardiovascular disease changed accordingly.
The EMPRISE programme followed the trial. It is a global, multi-year monitoring programme with a new-user, active-comparator cohort design, in which patients starting empagliflozin or a comparator are matched 1:1 by propensity score. It evaluates the effectiveness, safety and healthcare utilisation of empagliflozin in routine care across a broad spectrum of baseline cardiovascular risk. Using data from the US, Europe and Asia, the programme confirmed that, in a broader population than the trial's, the risks of cardiovascular events, heart failure hospitalisation and mortality are lower with empagliflozin than with a dipeptidyl peptidase-4 (DPP-4) inhibitor or a glucagon-like peptide-1 (GLP-1) receptor agonist.
The final analysis of EMPRISE US, with more than 115,000 matched pairs starting empagliflozin or a DPP-4 inhibitor, found lower risks with empagliflozin of myocardial infarction or stroke, heart failure hospitalisation, major adverse cardiovascular events, and cardiovascular and all-cause mortality. Patients starting empagliflozin also had a higher risk of diabetic ketoacidosis and lower risks of acute kidney injury, severe hypoglycaemia and retinopathy progression than those starting a DPP-4 inhibitor. Subgroup analyses in EMPRISE US showed benefits in subgroups such as older patients and patients with a history of atherosclerotic cardiovascular disease or heart failure, adding to the evidence from EMPA-REG OUTCOME. EMPRISE Europe and Asia, with more than 85,000 matched pairs in 11 countries, validated the US findings in international settings. The programme's real-world evidence thus complemented the results of randomised controlled trials with insight into empagliflozin's effectiveness in patients with type 2 diabetes with or without a history of cardiovascular disease or heart failure, across diverse routine care models.
EMPRISE also enabled a direct comparison of the cardiovascular effects of empagliflozin and GLP-1 receptor agonists, which have shown cardioprotective potential in clinical trials. Empagliflozin was associated with similar risks of myocardial infarction or stroke and lower risks of heart failure hospitalisation and cardiovascular mortality than GLP-1 receptor agonists.
In US routine practice, healthcare utilisation and costs of care were lower in patients starting empagliflozin than in those starting a DPP-4 inhibitor, irrespective of underlying cardiovascular disease. An analysis of EMPRISE data in Sweden showed that empagliflozin reduced rates of hospitalisation and of in- and outpatient visits in patients with type 2 diabetes.
Tiotropium/olodaterol in chronic obstructive pulmonary disease
Real-world studies of tiotropium/olodaterol in chronic obstructive pulmonary disease also show how real-world evidence can be used after launch. A study using US claims and commercial laboratory data compared the real-world effectiveness and safety of tiotropium/olodaterol with long-acting β2-agonist/inhaled corticosteroid combinations and found lower risks of exacerbations, pneumonia and escalation to triple therapy with tiotropium/olodaterol, highlighting its importance in avoiding inhaled corticosteroid overuse and reducing the risk of exacerbations. Two studies of treatment patterns in US and UK healthcare databases showed that inhaled corticosteroids are overprescribed in both countries, despite guidelines recommending them only for patients with severe disease who meet certain criteria, potentially putting patients at risk of side effects and increasing unnecessary healthcare costs.