Real-World Evidence: Can it Really Be Used For Drug Approvals?
- Posted by Jennifer King
- On February 1, 2017
With the signing of the 21st Century Cures Act, the US Congress tasked the FDA with developing a framework to evaluate how the use of data from sources other than traditional clinical trials may be used to support drug approvals. This framework will apply to post-market commitments and to new indications for approved drugs. This brings a further advantage to sponsors utilizing the 505(b)(2) pathway, on top of the already recognized benefit of streamlined development programs. So how and when will this option be available to sponsors, and has anything really changed?
What’s in the 21st Century Cures Act?
The 21st Century Cures Act that was enacted into law in December 2016 describes the amendment of the Federal Food, Drug, and Cosmetic Act by inserting, among other things, a provision to assess the potential for use of real-world evidence (RWE) in drug approval programs. (We have previously blogged about earlier versions of this legislation here).
The Act defines real-world evidence as ‘data regarding the usage, or the potential benefits or risks, of a drug derived from sources other than randomized clinical trials.’ The sources of data include ‘ongoing safety surveillance, observational studies, registries, claims, and patient-centered outcomes research activities.’
The Act states that the framework that FDA must develop will cover gaps in data collection activities, standards and methodologies for collecting the data, and priority areas that the program will address.
The framework must be developed in consultation with industry, academia, medical professional organizations, patient advocacy organizations, consumer organizations, and disease research foundations, and must be implemented within 2 years of the enactment date. A draft guidance for industry must be issued within 5 years followed by a final document 18 months after public comment on the draft document.
The section further notes that real-world evidence may be used for other purposes when there is a sufficient basis for doing so, and that the Act does not alter standards of evidence under which studies are evaluated.
The Problems with Using Real-World Evidence
RWE, or real-world data (RWD), are gathered from large datasets such as electronic health records (EHR), insurance claims, or even personal devices. However, the datasets are often not designed to provide the level of evidence required for approval of a product, or for addressing regulatory questions. Instead, the data may be collected with the primary goal of determining treatment for an individual patient, insurance reimbursement, or performing financial analyses. Or the data may be collected by patients and/or their personal devices, with varying degrees of accuracy and compliance.
Not surprisingly, RWD is frequently not of the quality required to meet current regulatory standards. Problems may include missing data and its retrospective interpretation or potential remedy at the data collection stage, accuracy, lack of adverse event data, and unintended bias. This is in contrast to clinical trials in which extensive measures are taken to reduce variability, to ensure the quality of the data collected, and to obtain detailed data on every adverse event that occurs. After the RWD has been generated, retrospective determination of appropriate methods of analysis can introduce further bias. Patient privacy must be considered, not just from a patient’s perspective, but also for the potential to further reduce data integrity. This list is not exhaustive but gives an idea of how incorrect inferences can be drawn from analyzing RWD.
What Does This Mean for the FDA?
The use of RWE to support regulatory approvals challenges the traditional paradigm in which the only authoritative evidence is generated through prospective randomized clinical trials (RCTs).
The enactment of the 21st Century Cures Act puts the FDA in a difficult situation in which it must establish the framework for evaluating real-world evidence that often does not contain the controls and methodology of the standards of a well-conducted clinical trial. Yet the standard of evidence, of course, cannot be lowered.
The FDA has begun to address the challenges by publishing scholarly articles defining RWE and the factors which improve its utility for regulatory purposes. The FDA and the National Institutes of Health (NIH) are working on ways to harmonize data collected from EHRs, claims data, and registries, and to facilitate the provision of actual data that does not require external review or interpretation. To its credit, the FDA has embraced the positive uses for RWE, particularly its potential for informing hypotheses and study design, and for increasing our knowledge of the effects of a product on more diverse populations than those studied in clinical trials.
The FDA Perspective
In a journal article published by 15 senior members of the FDA, including directors of the 3 main centers (drugs, devices, and biologics) and of multiple offices and divisions, real-world evidence is interpreted specifically to mean ‘information on health care that is derived from multiple sources outside typical clinical research settings, including EHRs, claims and billing data, product and disease registries, and data gathered through personal devices and health applications.’ The FDA does not include the word ‘randomization’ in its definition, and makes clear that the distinction between RWD and data collected from research intensive/clinical trials is not to be based on the presence or absence of a planned intervention or the use of randomization, as both are compatible with RWE. In fact, the FDA notes that for interventional studies, prospectively planned interventions and randomization can improve the validity of RWE.
The FDA also includes some clinical trials in an earlier definition of RWE in a medical devices guidance for industry, including ‘large simple trials, or pragmatic clinical trials.’
The FDA has also raised the issue of the increasing costs of conducting clinical trials, without a corresponding increase in the quantity of evidence produced to support decisions about healthcare. For these reasons, FDA notes that many researchers and sponsors are becoming interested in the expanding RWE data available from sources such as EHR, registries, and even social media. But once again, as these data were often not collected with the aim of being used for product approvals, there will likely be issues with bias, quality, accuracy, and reliability.
The FDA has suggested the importance of understanding RWE such that expectations for its use can be set appropriately. While tightly controlled clinical trials provide evidence of efficacy for a product, RWE can complement this finding by reducing the ‘uncertainty about generalizability’ of the findings. RWE can also provide data on patient compliance outside of the rigors of a clinical trial.
For many of the reasons stated above, the FDA sees the potential for RWE as ‘complementing’ data from clinical trials and playing a significant role in supporting/strengthening the evidence base for safety and effectiveness across the life cycle of drug development (encompassing both pre- and post-market regulatory decisions). RWE can support regulatory decision-making associated with label extension or a new indication of an approved drug, support confirmatory studies for drugs approved under the FDA’s expedited programs, and improve the efficiency of Phase IV post-marketing monitoring and surveillance.
When Real-World Evidence Can Be Used
Typically, RWD will not be available for use with a new chemical entity, but will be applicable to products suitable for approval via the 505(b)(2) pathway and in fulfilling post-market commitments.
RWE clearly plays roles in guiding healthcare policy decisions, safety surveillance, and reimbursement schemes. Most experts agree that drug or device developers can use RWE to generate hypotheses, or to focus clinical trial design. There are also cases where RWE could support the approval of a product by increasing the diversity in age, gender, race, and concurrent medication use in the patient population studied in the clinical trials. But if not collected with prospective methodology, and given the potential for bias, it is difficult to imagine that RWE will be accepted in place of a pivotal study with the exception of historical control data being used to compare the efficacy of a proposed product for life-threatening conditions.
For those intent on using RWE prior to issuance of the guidance, the medical device RWE guidance provides detail on factors that would increases the regulatory relevance of RWD. As far as adapting the use for drugs or biologics, as with most things, Camargo advises early and through discussion with the FDA, and careful adherence to the feedback given.
Orphan Diseases and Real-World Evidence
In rare diseases where new drug development is especially challenging due to the small numbers of people affected by each disease and the lack of medical understanding of many disorders, RWE may confirm clinical benefit for drugs approved under the FDA’s expedited programs. FDA’s approval of Exondys 51 (eteplirsen; Sarepta Therapeutics) for Duchenne muscular dystrophy may be an example of this. Exondys 51 was controversially approved in September 2016 via the accelerated approval pathway despite an external advisory committee determination that production of the surrogate marker was NOT induced to a level that would predict clinical benefit.
Although the FDA has required Sarepta to conduct a post-approval randomized controlled trial to confirm clinical benefit, it is likely that RWE, complementary to the clinical trial data, may play a critical role in confirming clinical benefit. In this setting, with an orphan indication/sparsity of patients, potential study recruitment difficulties due to a placebo treatment arm, if feasible, the addition of RWE may further support confirmatory study findings. If Sarepta cannot complete a randomized, controlled trial, it may turn to RWE to demonstrate clinical benefit.
What Has Changed?
RWE in the form of prospective registries or single-arm trials has already been evaluated for the approval of medical devices. A draft FDA Guidance document describing the Use of RWE to Support Regulatory Decision-Making for Medical Devices was issued in July 2016. In this draft guidance, the FDA recognizes that additional valuable RWD can be collected to help inform or augment the FDA’s understanding of the benefit-risk profile, to support an expansion of indication, or a newer version of an approved device, and could potentially be used to aid in regulatory decision-making. However, the FDA explains that the situation is different for drugs and biologics, as devices are typically developed in an iterative fashion such that substantial knowledge of the effect of confounding factors is often available a priori.
Camargo has already been using RWE in support of product approvals. This includes drug distribution and adverse event data of marketed unapproved drugs, and historical control data for comparison with single-arm study data for life-threatening indications. In each of these situations, we have provided a strong justification of how the data was collected and why the RWE supports the proposed product.
As the FDA has already been receptive to RWE in situations where it is well justified, we believe the new statute legislates for a commonsense framework that is broadly explainable as to how RWE can be collected and justified for 505(b)(2) applications and post-market commitment studies. At this time, the best way to view RWE from a regulatory perspective is providing complementary knowledge to that gained from traditional clinical trials.
Updated Note: President Trump’s executive order, signed on Monday, January 30, 2017, may potentially delay the implementation of the RWE framework and/or issuance of the Guidance documents, or indeed other sections of the 21st Century Cures Act. As part of the executive order, new regulations must be accompanied by the identification of 2 regulations to be eliminated. This requirement is expected to cause delays for the FDA in issuing Guidance for Industry documents.
To learn more about how RWE could be used in the development of your drug, device, or combination product, contact us.
Stacey Ayres, PhD, Senior Director of Scientific and Regulatory Affairs, Camargo Pharmaceutical Services.