Bringing Real-World Data to Multiple Sclerosis Treatment Decisions

Multiple sclerosis (MS) damages the insulating cover of nerve cells, throwing off the central nervous system’s efforts to communicate. Caused by a mix of genetic and environmental factors, MS is characterized by 4 distinct phenotypes that affect individuals differently; thus, the data that emerge from randomized controlled trials (RCTs) may only go so far in telling physicians how to treat the patient in front of them. By their nature, RCTs typically stop before gathering information about how a drug affects patients over the long term.

Regulators, pharmaceutical companies, and payers want answers they can’t get from RCTs, and increasingly they are looking to real-world data (RWD) for answers, according to the keynote speaker who opened ECTRIMS 2019, the 35th Annual Congress of the European Committee for Treatment and Research in Multiple Sclerosis, taking place in Stockholm, Sweden.

Use of registries, surveys, patient medical records, and other data sources presents can open new doors, said Maria Trojano, MD, professor of neurology at the University of Bari, Italy, where she is head of the Department of Basic Medical Sciences, Neurosciences and Sense Organs and director of the neurophysiopathology unit.

Using these sources can present challenges, but most of these can be overcome, she said, and RWD has already provided some important answers to MS researchers, especially about the long-term effectiveness of therapies. The strength of this evidence is that it represents thousands of observations, sometimes over an extended period.

“High-quality real-world studies can be used to verify the transferability of expert opinion in the clinical practical context,” she said.

Trojano explained what questions are answered by RCTs versus observational studies or pragmatic trials, why RWD studies are needed in MS treatment, how the rise of new data sources and tools have made these studies easier, and what RWD has brought to the field so far.

“What do we mean when we talk about real-world data?” she asked, before going to the list of available data sources that can be mined for answers about how therapies work or how a disease behaves in certain populations. RCTs and studies that use RWD both have strengths and weaknesses—a key one being the risk of bias outside the clinical trial—but the 2 can coexist.

The fundamental difference, Trojano explained, is that observational studies and pragmatic trials reflect the choices made by doctors and patients, and thus, what happens in the clinic every day.

“The pragmatic clinical trial represents the best method for real-world comparisons of alternative therapies, but unfortunately, its use is still limited in MS,” she said.

Trojano noted that both the FDA and the European Medicines Agency are working to promote the use of RWD in the regulatory process; the FDA is doing so through a framework document developed in response to the 21st Century Cures Act. Even without this push, Trojano said the use of RWD studies has soared in MS, she pointed to data that show the number of studies has climbed from 17 to more than 1100 in the past 15 years.

The reason? “Observational real-world studies can fill crucial gaps in evidence,” she said. Indeed, Trojano is a leader in a European MS registry consortium that includes Italy, Denmark, Sweden, France, and the international neuro-immunology registry MSBase.

RWD has come to the fore due to the increased availability of data and the rise of advanced statistical methods, which allow researchers to calculate to overcome selection biases that exist simply because patient records reflect those who chose to seek treatment (in the United States, they may reflect who has insurance).

But Trojano said ECTRIMS is holding events like summer seminars to train researchers in advanced statistical methods to work through these issues. And, she said, use of RWD has already yielded answers, including 20-year follow-up studies of injectable therapies, which demonstrate the long-term benefits in relapsing-remitting MS. Another question that data can help answer is whether early, intensive therapy works better than an escalation approach.

“What are the major challenges?” Trojano asked. Data quality remains an issue. Sometimes data are not complete, and sometimes biomarkers or patient-reported outcomes (PROs) are not captured.

“Missing data is a common issue,” she said. There is also a problem of heterogeneity in outcome assessment across different countries, which can hamper comparison effectiveness research. International meetings seek to set standards to resolve these issues.

Finally, Trojano said, “We need to better integrate data into clinical practice.” This point was echoed at a satellite symposium, “The Evolution of MS Study Outcomes: Relevance to Clinical Practice Today,” where speaker Paul Matthews, MD, DPhil, FRCP, of Imperial College, London, spoke about the importance of gathering PROs to assess a patient’s cognitive status. He called PROs “the filter through which all [the] biology and pharmacology acts.”

Fellow symposium speaker Tjalf Ziemssen, MD, professor of Clinical Neuroscience, University Clinic Carl-Gustav Carus, discussed the emergence of pragmatic trials and said in a disease that is extremely heterogeneous, “they will help us individualize treatment.”

Trojano agreed. RWD, she said, “is no longer a ‘nice to have’ in MS treatment.”