New model could assist in guiding RRMS treatment decisions
Researchers relied on measurable factors to predict long-term disease course
A statistical model that takes clinical and demographic factors into account could help guide treatment decisions in people with relapsing-remitting multiple sclerosis (RRMS) who haven’t yet started on a multiple sclerosis (MS) therapy, a study reports.
“Our study offers a predictive tool that fulfills an unmet need for a validated, robust model for RRMS prognosis,” researchers wrote in the study, “Development and validation of a scoring system for predicting disease activity in treatment-naïve patients with relapsing-remitting multiple sclerosis,” which was published in Multiple Sclerosis and Related Disorders.
There are more than 20 approved MS treatments and most are indicated for relapsing forms of MS such as RRMS, meaning these patients have a wide variety of treatments to choose from.
Disease-modifying therapies for MS have their own pros and cons. In general, some medicines are less powerful at easing MS activity and slowing disability progression, but are generally safer, whereas other therapies are highly potent, but carry more substantial safety risks.
In theory, giving less powerful therapies to people with less active disease could be beneficial, while more potent therapies could be reserved for those with very aggressive disease. This is an especially important consideration in low- and middle-income countries, where it may be difficult to access high-efficacy therapies for all patients.
Actually doing this type of stratification is challenging because it’s hard to predict long-term outcomes in people with MS who are about to start treatment.
Factors that predict long-term outcomes
Here, a team led by scientists in Egypt analyzed data from 518 RRMS patients in an Egyptian database who hadn’t started any MS treatment. Their goal was to use readily measurable factors — such as age at disease onset, sex, disability progression, and disease activity — to predict long-term disease course, thus offering a tool to help make decisions about treatment.
“Adopting an algorithm based on the level of disease activity can help prioritize patients for high-efficacy therapies while reserving less intensive treatment for those with lower disease activity,” the scientists wrote.
Each of the patients in the analysis was rated by five independent MS experts. The patients were grouped into four different levels of MS activity, including “active” (10.8%), “highly active” (40%), “very highly active” (35.5%). and “aggressive” (13.9%). The researchers then constructed statistical models to see how well different factors could predict MS activity.
The final model showed good accuracy and was able to accurately predict MS severity in 92% of the patients. In that final model, the most important factors were disability scores, having a high relapse rate and the initial relapse recovery, and the presence of motor weakness and bladder symptoms. Other factors like sex, age, and lesion load also played a role, but were less crucial mathematically.
The researchers used the final model to create a mathematical tool called a nomogram that allows clinicians to quickly predict disease activity in RRMS patients. Further work will be needed to validate the model and explore its applications in clinical practice, the researchers said.
“We have developed a validated model that leverages an ordinal logistic regression to accurately predict the disease activity class at the early stages of RRMS,” the scientists wrote. “The resulting scoring system, coupled with a user-friendly nomogram, offers clinicians the ability to make informed decisions on patient treatment for treatment naïve RRMS patients.”
About the Author
