The postdoctoral scholar at the University of California, San Francisco, spoke about a prognostic marker that can be used to study the role of genetic, epidemiologic and immune variables on MS, and to measure the long-term impact of treatment in clinical trials.
Antje Bischof, MD
One of the major challenges multiple sclerosis (MS) research is understanding progressive disease. Spinal cord area shows the strongest correlations with MS disability and is able to differentiate progressive from relapsing-remitting disease subtypes.
Antje Bischof, MD, postdoctoral scholar at the University of California, San Francisco, and colleagues, conducted a longitudinal study to evaluate spinal cord atrophy measured in patients with MS through MRI over 12 years and found that upper cervical cord atrophy obtained from routine brain MRI is a strong indicator of impending conversion from relapsing-remitting MS to secondary progressive MS (SPMS).
Brain volumetric and spinal cord area at C1 level were analyzed from brain MRI in order to evaluate their potential to differentiate between the 2 matched groups prior to the conversion period. Those who developed SPMS demonstrated an accelerated rate (-2.15% per year) before conversion to secondary progressive compared to those with relapsing-remitting MS matches who did not convert to SPMS (-.74% per year). The data suggest that the difference exists at least 4 years before conversion to SPMS. Bischof concluded that the cervical atrophy rate at C1 level can be used as a prognostic marker to study MS and measure the long-term impact of treatment in trials.
To further highlight Bischof's research, NeurologyLive sat down with Bischof at the fourth annual Americans Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS) Forum 2019 held in Dallas, Texas.
Antje Bischof, MD: We presented work that we have been working on for probably the last 2 years. We are very much interested in finding biomarkers to look at the prognosis of MS patients that start out with a relapsing course but then at some point convert to secondary progressive disease, because we know that not all patients do eventually convert to progressive disease and we really would like to have biomarkers that guide our treatment decisions early on.
When we looked into the literature about the evidence on which biomarkers would be promising, the most promising is certainly the spinal cord atrophy and that corresponds very well with the clinical notion that patients who have a progressive disease have a predominant myelopathy. At UCSF we have a very long-term observational cohort study that has been studied now for almost 15 years, so we were basically able to go back in time knowing the outcome of the patients today and see how they were differentiating in terms of MRI markers at the baseline. We had a median disease duration of 10 years at baseline.
We looked very comprehensively at all metrics of MRI that have been suggested in the literature to be possible biomarkers of disease outcome and who had good correlations with clinical disability. We were able to look at spinal cord atrophy because we developed a novel method that allowed us to extract spinal cord areas from brain images which nobody has really succeeded on a prospectively collected longitudinal cohort before, so that was the main strength of our study.
What we found was that patients who went on to convert to secondary progressive disease during our study period had much higher rates of spinal cord atrophy than those patients who remained stable during the entire observation period until today, which is 2 decades of disease since disease onset. This was really surprising for us that the rates were so different between the 2 groups and it was really clear-cut. Statistically, for those who are a little bit firm in statistics, the confidence intervals of the slopes of the 2 patient groups were actually not overlapping at all which gave us a really highly significant p-value, so that was very interesting to see. This tells us that patients who convert to secondary progressive disease actually differentiate already during the relapsing phase from those who remain a more benign disease subtype and remain relapsing for more than 2 decades, and that is certainly the most promising biomarker that we currently found in our study.
We also looked at brain metrics; we did a very comprehensive study and tried to include brain lesion loads from T1- and T2-weighted images, we also looked at slowly-enlarging lesions which have just in the recent years become a focus of the research because they have been shown to represent probably chronic information that is not targeted by treatments so we wanted to know whether these lesions are also more prominent in patients who convert to progressive disease, but we did not find any difference.
We looked at all commonly used brain atrophy metrics and that's means global brain atrophy and also regional brain atrophy, so atrophy of the white matter, the gray matter, deep gray matter, structures including thalamus, putamen, and caudate, and we did not find significant differences between the 2 patient groups. It was really the spinal cord atrophy that stood out from all these markers to be prognostic of the conversion of patients to secondary progressive disease.
Currently this is a research study, so, unfortunately, we don't have these metrics available in the clinics yet, but we are working on methods to automize this and hopefully translate this into the clinics. I think we are already judging atrophy on MRIs when we look at the spinal cord clinically, and it seems that these patients who demonstrate a lot of atrophy are those who are at risk of converting to SPMS, whether this translates into a treatment algorithm we don't know yet because we don't know what actually underlies this atrophy, it might be that patients who develop a lot of atrophy also had a lot of inflammation and our data suggests that this might be the case.
We saw that the patients who converted to SPMS later had larger areas of atrophy, although this was not statistically significant, they had larger areas of the cervical cord than for example controls, and given the fact that these patients had the disease for more than 10 years when we started studying them, this suggests that they had larger areas than the controls, you would assume that after 10 years of having atrophy they should have smaller areas and this was not the case, so why this is we don't know, but it might be related to ongoing inflammation that kind of leads to a swelling of the cord, so it might be that this can be targeted with treatments and this is what we are going to study.
The interesting or important aspect of the fact that we can analyze spinal cord atrophy now from brain images is that we can go back into all the datasets that are available worldwide and have been collected in clinical trials and observational cohorts and we can actually study this retrospectively which will give us answers much faster than if we were to do prospective clinical trials. We can essentially look at all clinical trials for all the medications that are available to determine if there is an effect of this medication on the spinal cord atrophy or not, and I think this will probably push us into guiding our treatments much better with a targeted approach that is really on an individual patient level. This is our hope, of course, this has to be studied, and I hope we will have the results within the next years on this, at the moment we are positive that we will get these results.
I think something that was also surprising we found is that if we went back in time from the conversion time point and looked at a subset of patients where we had scans available consecutively from every year going back up to 4 years before the conversion, we still found that the rates did not change which probably means that the rates are already high quite early in the disease in those patients who convert progressive disease—this tells us that this is a very distinct subset of patients that have or might be at risk for a more severe disease course and might need to be addressed differently in terms of treatment. We don’t have enough data at the moment to be able to look at the dynamics of the spinal cord atrophy over time, whether this is changing in an individual patient or whether this is always happening at the same speed, but this is something we really want to look at as well in the future if we have more data.
Transcript edited for clarity.