BARCELONA—Monitoring for relapses and MRI activity is the best means of predicting treatment response in patients with multiple sclerosis (MS), according to research presented at the 31st Congress of the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS). Relapses are less frequent than MRI activity, however, especially among newly diagnosed patients.
Isolated MRI activity may help assess treatment response, but technical issues such as variability in image acquisition and scan comparison may make this approach difficult. “There is not a magic threshold for the number of new T2 lesions to be tolerated,” said Luca Prosperini, MD, PhD, Researcher in the Department of Neurology and Psychiatry at Sapienza University in Rome. The location of the lesions may be more relevant for defining treatment response than their number, he added.
Soon after disease onset, neurologists should monitor the patient for signs of aggressive disease, although this MS phenotype is less frequent. Characterizing clinical and MRI features is important because patients with aggressive disease may require aggressive treatment such as the induction approach, said Dr. Prosperini.
Examining MRI Activity
“MRI activity is probably the most sensitive tool to detect biological disease activity, even a few months after the treatment starts,” said Dr. Prosperini. Various published studies, in which patients underwent MRI at six to 12 months after treatment initiation, support early on-treatment MRI assessment as a way to predict future response to treatment, future relapses, and future disability in the medium and long terms.
In a 2004 post hoc analysis of the pivotal trial of interferon beta-1a, the investigators found that on-treatment MRI activity, especially new T2 lesions, was associated with increased disability and increased brain atrophy at the end of the two-year follow-up in the active group, but not in the placebo group. In a 2013 post hoc analysis that included data from the pivotal trial’s extension phase, patients who accumulated the highest level of disability at 15 years after treatment initiation had had gadolinium-enhancing lesions after one year of treatment.
In a 2009 study of 394 patients with relapsing-remitting MS who started different formulations of interferon beta and were reassessed after one year, Dr. Prosperini and colleagues found a clear dose-effect relationship between new T2 lesions and the risk of future disability. “The greater the number of new T2 lesions, the higher the risk of disability progression over a follow-up time of four or five years,” said Dr. Prosperini. “It’s even more relevant that this statistical figure can be replicated even after removing patients who relapsed in the first year. In other words, the dose-effect phenomenon of new T2 lesions assessed at one year is independent of early clinical disease activity.”
In 2013, Dr. Prosperini and colleagues compared two criteria for assessing nonresponse to treatment. They used the European Medicines Agency criteria for escalating to a second-line therapy, as well as the isolated MRI activity criteria (ie, patients with one or more gadolinium-enhancing lesion or two or more T2 lesions). Both sets of criteria predicted patients’ risk of future disability worsening with similarly high accuracy.
Monitoring Clinical and MRI Activity
In 2015, Boster et al presented a post hoc analysis of data from the FREEDOMS I and II trials, which followed 1,693 patients with relapsing-remitting MS who received fingolimod. The study’s follow-up lasted four years. On-treatment relapses, MRI activity, and a combination of both factors effectively predicted future relapses and future disability. In addition, focal MRI activity, defined as one or more gadolinium-enhancing lesion or two or more new T2 lesions, was the strongest predictor of failure to achieve no evidence of disease activity.
Some neurologists believe that isolated MRI activity is insufficient reason to change treatment, and a 2011 post hoc analysis by Sormani et al supports this idea. The researchers examined 560 patients with relapsing-remitting MS in the PRISMS trial of interferon beta-1a. The combination of relapses and new T2 lesions fully estimated the treatment effect at the end of two years, but each of these measures alone estimated only about 63% of the treatment effect.
Next, Sormani and colleagues developed a statistically optimized version of the Rio score in 2013 and analyzed how accurately it could classify patients’ risk of disability worsening after one year of treatment with interferon beta. Their data indicated that patients with a modified Rio score of 2 or greater have a high risk (ie, 50%) of disability worsening. A score of 2 corresponds to having at least two relapses in the first year or one relapse and at least five new T2 lesions, said Dr. Prosperini. Because patients at medium risk (ie, those with a modified Rio score of 1) are the most difficult to classify, in terms of treatment response, Sormani suggested that they be reassessed at 18 months of treatment. If these patients have one or more relapse, or two or more new T2 lesions, they should be considered nonresponders, the researchers concluded.
The amount of MRI activity to be tolerated remains a matter of debate. To investigate this question, Sormani and colleagues examined a large amount of data from the MRI in MS network about patients treated with interferon beta. The investigators determined that patients with two or more relapses or one relapse and three or more new T2 lesions were at highest risk of disability worsening. The investigators initially classified 130 patients as being at high risk, but the final number of patients who had sustained disability worsening was more than 300. “So, this score suffered from a low sensitivity,” said Dr. Prosperini.
Lesion Location May Predict Outcomes
“Another important question is the topography of the new lesions,” he continued. Galassi et al presented a study at ECTRIMS of 390 patients with relapsing-remitting MS who initiated treatment with interferon beta and were assessed at one year. Follow-up lasted for four years. When the researchers chose disability worsening as their final outcome and performed multivariable Cox regression analysis, they found that new infratentorial lesions and new spinal cord lesions were independent predictors of disability worsening. Lesion count, however, did not contribute to model fit.
For patients with a good prognosis, early and safe treatment should be the goals, said Dr. Prosperini. Early and effective treatment should be the goals for patients with a bad prognosis, he added. “Consider induction strategy in aggressive MS, even if it’s the less frequent phenotype. After treatment starts, we should assess patients correctly and regularly.”
—Erik Greb