Real-World Evidence & Biosimilars: Weighing The Potential Impact
By Anna Rose Welch, Editorial & Community Director, Advancing RNA
A few months ago, I chaired a panel on real-world evidence (RWE) at the DIA Biosimilar Conference. Almost daily, we get word of the long-term safety and efficacy of biosimilars in naïve patients or following a transition from an originator. But there’s a lot of work sponsors need to do behind-the-scenes to get the answers stakeholders need today. During the conference, four epidemiologist experts presented on a variety of RWE-related topics, including how to evaluate the appropriate data sources, design fit-for-purpose studies, and use RWE to answer some of our most pressing questions (article forthcoming!).
Like all great panels, however, there was hardly enough time to engage the experts in a Q&A following their presentations. So, I reached out to each expert after the conference and asked them to answer a few (more) questions. Here, IQVIA’s Jaclyn Bosco, Sandoz’s Edward Li, and the University of Messina’s Dr. Gianluca Trifirò discuss how RWE can impact regulator and payer decisions, shape treatment pathways, and encourage greater stakeholder comfort and action toward biosimilars.
Anna Rose Welch: How is RWE best used to inform biosimilar development? Do you see real-world evidence impacting or advancing the biosimilar regulatory pathway or payer decisions in the future, and how?
Jaclyn Bosco: Stakeholders are seeking evidence demonstrating that, in routine clinical practice, a biosimilar performs similarly in terms of effectiveness and safety when compared with the biologic. As biosimilars can be used in indications for which the approval was received through extrapolation, external comparator data of originator biologics will help inform which safety events and effectiveness outcomes we would expect to see for the biosimilars.
Gianluca Trifirò: From the European perspective, I think RWE generation may assist regulators and mostly payers in areas which cannot be properly addressed by premarketing randomized clinical trials (RCTs). RWE can be used to further demonstrate the feasibility of extrapolation and, perhaps even more importantly, the interchangeability of originators and their biosimilars and of biosimilar-to-biosimilar transitions. Though interchangeability is of great interest to payers, there is currently uncertainty on the best way to demonstrate interchangeability and on the most suitable approach to generate evidence on the clinical effects of switching from the originator to biosimilars. This issue can hardly be addressed fully in the premarketing setting — for example, through the transition studies requested by FDA — as the scenario is much more complex in the real world.
Welch: Dr. Trifiro, can you share an example in which you’ve seen the use of real-world evidence evolve regulatory and/or clinician thinking about biosimilars?
Trifirò: I think that RWE generation (assuming that the quality of observational studies is high!) may support regulators for specific controversial issues like interchangeability, as I mentioned previously. More importantly, RWE has been essential in evolving clinical thinking about the value of biosimilars. As an example, the first European Crohn's and Colitis Organisation (ECCO) position statement was not supportive of the use of biosimilars and came out against switching from an originator to biosimilar. This position statement has recently been substantially updated, now supporting the use of biosimilars in naïve patients as well as patients transitioning from the originator. This shift in opinion was based on cumulated evidence coming from real world data.
Welch: What best practices can be employed to encourage stakeholder action using RWE?
Trifirò: In general, the confidence of clinicians and patients towards biosimilars is much higher if evidence is generated from real world data coming from their clinical practice. My suggestion is to identify the main doubts from clinicians/patients and then put forth efforts to generate RWE targeting doubtful aspects of biosimilars (e.g. dosing penalty, etc.). Involvement of clinicians and patients in RWE generation is really helpful.
Welch: Each therapeutic indication will have questions that can be answered by real-world evidence, which, in turn, can advance the overall treatment pathways. In this age of innovative and biosimilar oncology treatments, how do you see real-world evidence impacting the overall oncology treatment pathway?
Edward Li: Our understanding of the best oncology treatments for patients continues to evolve as new therapies and new data for existing treatments become available. Clinicians make the best decisions for their patients based on the most current evidence at hand. For newly approved medicines, the only available evidence will be in the form of randomized controlled trials conducted to obtain regulatory approval. After market uptake, RWE will emerge to complement and help validate the safety and efficacy data generated by the registration trials. Decision-makers can then use this newly acquired RWE to make changes, if necessary, to the treatment algorithm. This will be especially important for cancer-agnostic targeted and immuno-oncology therapies, where the pre-approval trials included relatively smaller patient numbers per tumor type or more narrow populations compared to patients typically seen in an oncology practice.
Welch: What questions will be the most important to ask in the oncology space when it comes to RWE — specifically for oncology biosimilars?
Li: Within oncology, the next major question to answer via RWE is whether biosimilars will improve access to therapies, and therefore improve population-based outcomes because more people are able to receive therapy. If a reference biologic is not currently being used because of cost concerns, a less-costly biosimilar may be a cost-effective treatment option. Use of a biosimilar in such settings would expand access to biologic therapies for more patients. RWE could then be used to more precisely measure outcome effects and total costs of care attributed to the biologic (reference or biosimilar).
Bosco: I think the questions for RWE are the same whether it is an oncology or non-oncology biosimilars (e.g., is the biosimilar as safe and effective as the originator biologic). RWE oncology studies may be complex due to considerations around the multi-modality of treatment regimens patients receive. For example, the dose of the biosimilar received may depend on the combination of other therapies a patient is receiving. There are also two types of oncology biosimilars, those that are administered to help reduce the incidence of certain side effects of other therapies and those that are used to treat the cancer itself. Being able to discern whether any observed safety event (for example) is due to the biosimilar of interest versus another treatment a patient is receiving will be an important consideration to the design of a RW study. It’s also critical to understand which oncology data sources can be used to generate real-world evidence when conducting real-world oncology studies.
Welch: What one other therapeutic area do you feel is particularly ripe for biosimilar RWE, and how might the questions for this differ from those of oncology?
Li: The other therapeutic area where generating RWE for biosimilars will be important is within the inflammatory disease spectrum – particularly for rheumatic diseases (e.g., rheumatoid arthritis) and inflammatory bowel diseases. Seeing as patients with chronic diseases receive therapy for a relatively long time (compared to oncology), RWE can help to complement the evidence from biosimilar registration trials about transitioning a patient to a biosimilar from the reference.
Welch: What is one piece of advice you would give sponsors who want to explore the use of RWE?
Bosco: Keep in mind that real-world data are typically collected for reasons other than a specific research question. So, it is important to assess and understand what data are included, how the data are collected, recorded, and curated, which definitions are used, and what level of missingness may exist. This will be essential for understanding whether RWD can be used to generate valid RWE to answer a given stakeholder’s needs.