The effect of disease-modifying therapies on brain atrophy in patients with clinically isolated syndrome: a systematic review and meta-analysis

Abstract

Objectives:

Brain atrophy is associated with cognitive deficits in patients with clinically isolated syndrome (CIS) and can predict conversion to clinical definite multiple sclerosis. The aim of the present meta-analysis was to evaluate the effect of disease-modifying drugs (DMDs) on brain atrophy in patients with CIS.

Methods:

Eligible placebo-control randomized clinical trials of patients with CIS that had reported changes in brain volume during the study period were identified by searching the MEDLINE, SCOPUS, and Cochrane Central Register of Controlled Trials (CENTRAL) databases. This meta-analysis adopted the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for systematic reviews and meta-analyses.

Results:

Three eligible studies were identified, comprising 1362 patients. The mean percentage change in brain volume was found to be significantly lower in DMD-treated patients versus placebo-treated subgroups (standardized mean difference [SMD]: = −0.13, 95% confidence interval [CI]: −0.25, 0.01; p = 0.04). In the subgroup analysis of the two studies that provided data on brain-volume changes for the first (0–12 months) and second (13–24 months) year of treatment, DMD attenuated brain-volume loss in comparison with placebo during the second year (SMD = −0.25; 95% CI: −0.43, −0.07; p < 0.001), but not during the first year of treatment (SMD = −0.01; 95% CI: −0.27, 0.24; p = 0.93). No evidence of heterogeneity was found between estimates, while funnel-plot inspection revealed no evidence of publication bias.

Conclusions:

DMDs appear to attenuate brain atrophy over time in patients with CIS. The effect of DMDs on brain-volume loss is evident after the first year of treatment.

Introduction

Brain atrophy has recently emerged as an independent predictor of long-term neurological deterioration, impaired quality of life, and sustained disability progression in patients with multiple sclerosis (MS) [Filippi et al. 2010]. Brain atrophy has been shown to be present right after the first MS clinical event [Audoin et al. 2010], and to progress more rapidly throughout the course of MS than in healthy individuals [Chard et al. 2002; De Stefano et al. 2010].

Results from longitudinal magnetic resonance imaging (MRI) studies of patients with clinically isolated syndrome (CIS) indicate that whole-brain atrophy and gray matter-volume loss can predict conversion to clinically definite MS [Pérez-Miralles et al. 2013; Bergsland et al. 2012; Uher et al. 2014], better than measurements such as lesion load or brain volume on baseline MRI alone [Di Filippi et al. 2010]. Brain atrophy was also found to be related to cognitive deficits (e.g. mood disturbances, sexual dysfunction, personality disorders), with the cognitive impairment in CIS resembling that found in the later stages of MS [Štecková et al. 2014; Giorgio and De Stefano, 2010; Benedict et al. 2004].

To the best of our knowledge the efficacy of disease-modifying drugs (DMDs) in attenuating brain atrophy in patients with CIS has not been investigated systematically using a meta-analytical approach. In view of the former considerations we conducted a systematic review and meta-analysis to evaluate the effect of available DMDs on brain atrophy in patients with CIS using available data from randomized controlled trials (RCTs).

Methods

Trial identification and data abstraction

This meta-analysis adopted the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [Liberati et al. 2009]. Eligible placebo-control RCTs of patients with CIS that reported changes in brain volume during the study period were identified by searching the MEDLINE, SCOPUS, and Cochrane Central Register of Controlled Trials (CENTRAL) databases. The combination of search strings that was used in all database searches included the terms: ‘clinically isolated syndrome’, ‘CIS’, ‘brain atrophy’, and ‘brain volume’. The complete search algorithm that was used in the MEDLINE search is available in the supplementary file. No language or other restrictions were imposed. The last literature search was conducted on 27 October 2014. Reference lists of all articles that met the criteria and of relevant review articles were examined to identify studies that may have been missed by the database search.

All retrieved studies were scanned independently by two reviewers (GT and AHK) to include only placebo-control RCTs of CIS patients that had reported changes in brain volume during the study period. We excluded from the final analysis: (a) observational studies; (b) case series; (c) case reports; (d) RCTs without placebo subgroups. In the case of any disagreement regarding the literature search results between the two coauthors, the senior coauthor (KV) was consulted and disagreement was resolved with consensus.

In each study that met the inclusion criteria for the quantitative analysis a predefined seven-point quality control was used to address for biases. For each quality item the corresponding risk of bias was categorized as low, high, or unclear according to the suggestions made by Higgins and colleagues [Higgins et al. 2011]. Quality control and bias identification were performed by three independent reviewers (GT, AHK, and KV) and all emerging conflicts were resolved with consensus.

Data on brain-volume changes in all subgroups between time points, or during time intervals within studies, were extracted independently by the two authors who performed the literature search (GT and AHK).

Statistical analyses

Unadjusted mean differences of percentage changes in brain volumes between treatment and placebo subgroups were pooled as standardized mean differences (SMDs). SMD estimates were calculated as the mean differences divided by the corresponding pooled standard deviations and were subsequently interpreted using a general rule of thumb reported by Cohen, in which an SMD of 0.2 represents a small effect, an SMD of 0.5 represents a medium effect, and an SMD of 0.8 or larger represents a large effect [Cohen, 1988]. A random-effects model (DerSimonian–Laird) was used to calculate the pooled SMDs as previously described [Tsivgoulis et al. 2014].

Heterogeneity between studies was assessed with the Cochran’s Q and I² statistics. For the qualitative interpretation of heterogeneity, I² values of at least 50% were considered to represent substantial heterogeneity, while values of at least 75% indicated considerable heterogeneity, as stated in the Cochrane Handbook for Systematic Reviews of Interventions [Deeks et al. 2011]. Publication bias (i.e. assessment of bias across studies) was graphically evaluated using a funnel plot, given that the Cochrane handbook dictates as a rule of thumb that tests for funnel-plot asymmetry should be used only when there are at least 10 studies included in the meta-analysis [Sterne et al. 2011].

We subsequently conducted subgroup analyses according to the duration of disease-modifying therapies using data from studies that provided information on changes in brain volume for both time periods (i.e. 0–12 months and 13–24 months). The mixed-effects model was used to calculate both the pooled point estimate in each subgroup and the overall estimates. According to the mixed-effects model, we used a random-effects model (DerSimonian–Laird) to combine studies within each subgroup and a fixed-effect model (Mantel–Haenszel) to combine subgroups and estimate the overall effect. We assumed the study-to-study variance (tau-squared) to be the same for all subgroups. Tau-squared was first computed within subgroups and then pooled across subgroups.

All statistical analyses were conducted using Review Manager (RevMan) Version 5.2 software (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark).

Results

Study selection and study characteristics

Systematic searches of the MEDLINE and SCOPUS databases yielded 117 and 146 results, respectively. A subsequent search in the CENTRAL database retrieved no additional RCTs. After removing duplicates, the titles and abstracts from the remaining 166 studies were screened and four potentially eligible studies for the meta-analysis were retained. After retrieving the full-text versions of the aforementioned four studies, one retrieved full text was excluded because only the baseline brain-volume measurements were reported [Arnold et al. 2013]. In the final presentation of the literature search results, there was no conflict or disagreement between the two reviewers and the three studies that met the study protocol’s inclusion criteria, that is, the Early Treatment of Multiple Sclerosis (ETOMS) trial [Filippi et al. 2004], the PRECISE trial [Comi et al. 2013], and the TOPIC [Miller et al. 2014], were included both in the qualitative and quantitative synthesis (Figure 1). We included only the data from the randomized part of the PRECISE trial [Comi et al. 2013]. The characteristics of the included studies, comprising 1362 patients (66.6% women, mean age 30.7 years, mean baseline Expanded Disability Status Scale score: 1.4), are summarized in Table 1.

Figure 1.

Flowchart presenting the selection of eligible studies.

Table 1.

Baseline characteristics of patients with clinically isolated syndrome included in the meta-analysis of randomized control trials.

Reference	Study name	Subgroup	Dose	Number of patients	Time points (months)	Females	Age (years)	Baseline Expanded Disability Status Scale
Filippi et al. [2004]	ETOMS	Inteferon beta-1a (22 µg)	1/ week (sc)	131	12/24	60%	28.8 ± 5.8	1.0 (0.0–7.0)
		Placebo	1/ week (sc)	132		67%	27.9 ± 6.1	1.0 (0.0–5.0)
Comi et al. [2013]	PRECISE	Glatiramer acetate (20 mg)	1/ daily (sc)	243	12/24	65.4%	31.5 ± 6.9	1.1 ± 1.0
		Placebo	1/ daily (sc)	238		68.5%	30.8 ± 7.0	1.0 ± 1.0
Miller et al. [2014]	TOPIC	Teriflunomide (14 mg)	1/daily (pos)	216	27	71%	32.8 ± 8.1	1.8 ± 0.97
		Teriflunomide (7 mg)	1/daily (pos)	205		63%	31.6 ± 9.0	1.5 ± 1.02
		Placebo	1/daily (pos)	197		69%	32.0 ± 8.4	1.71 ± 1.00

pos, per os; sc, subcutaneously.

Two of the studies [Filippi et al. 2004; Comi et al. 2013] evaluated brain atrophy by using structural image evaluation in the normalization of atrophy method, which provides an automated brain-volume change analysis from baseline [Smith et al. 2002]. In the third study protocol the method of brain-atrophy measurement in MRI scans was not mentioned in the paper of the included study [Miller et al. 2014], but has been extensively described in other publications from the same authors [Rovaris et al. 2000, 2001].

Risk of bias for independent studies

The risk of bias in the included studies is summarized in Figure 2. Overall, a low risk of bias was found within individual studies, except for the uncertainty of bias related to allocation concealment [Filippi et al. 2004; Miller et al. 2014], and blinding of outcome assessment [Comi et al. 2013]. All study protocols were supported partly [Filippi et al. 2004] or solely [Comi et al. 2013; Miller et al. 2014] by the pharmaceutical companies that produce and market the drug under consideration in each study. In one of them [Miller et al. 2014], it was clearly stated that the study data were collected by the investigators and analyzed by the sponsor pharmaceutical company, which also developed the study protocol, increasing thus the susceptibility to a possible bias [Bero, 2013].

Figure 2.

(a) Risk of bias summary: review authors’ judgments about each risk of bias item for each included study. (b) Risk of bias graph: review authors’ judgments about each risk of bias item presented as percentages across all included studies.

Overall analysis and subgroup analyses

The mean percentage change in brain volume was found to be significantly lower in DMD-treated patients versus placebo-treated subgroups (SMD = −0.13, 95% confidence interval [CI]: −0.25 to −0.01; p = 0.04) (Figure 3). No evidence of heterogeneity was found between estimates (I² = 19%; p = 0.30 by Cochran’s Q statistics). In view of the small number (n = 3) of included studies, inspection of the funnel plot revealed no evidence of publication bias (Figure 4).

Figure 3.

Percentage changes in mean brain volume in patients with clinically isolated syndrome receiving disease-modifying therapy compared with those receiving placebo.

Figures 4.

Funnel plot of the included studies. SE, standard error, SMD, standardized mean difference.

In the subgroup analysis of the two studies [Filippi et al. 2004; Comi et al. 2013] that provided data on brain volume changes for the first (0–12 months) and second (13–24 months) year of treatment, DMDs (glatiramer acetate or interferon beta-1a 22 µg) attenuated brain-volume loss compared with placebo during the second year (SMD = −0.25; 95% CI: −0.43 to −0.07; p < 0.001), but not during the first year of treatment (SMD = −0.01; 95% CI: −0.27 to 0.24; p = 0.93) (Figure 5). Moreover, the difference in percentage brain-volume change between CIS patients under treatment with DMDs and CIS patients randomized to placebo tended to be significant over the treatment years (p = 0.08). Finally, we detected no evidence of heterogeneity between estimates during the second year of treatment (I² = 0%; p = 0.93 by Cochran’s Q statistics), while moderate heterogeneity was identified between estimates during the first year of treatment (I² = 61%; p = 0.11 by Cochran’s Q statistics)

Figure 5.

Subgroup analysis of the randomized control trials on percentage brain-volume changes between the first, second, and third year of included studies.

Discussion

The findings from the current systematic review and meta-analysis indicate that DMDs appear to attenuate brain atrophy over time when compared with placebo in patients with CIS. More specifically, our analyses documented a modest beneficial effect of DMDs on brain atrophy, with no evidence of heterogeneity across trials using different immunomodulatory treatments such as, subcutaneous interferon beta-1a, glatiramer acetate, and teriflunomide, all currently considered first-line treatments for MS. Moreover, we documented a low risk of selection, performance, detection, attrition, and reporting biases using the validated, quality-control methodology of the Cochrane collaboration for the assessment and quantification of biases in individual studies included in comprehensive meta-analyses [Higgins et al. 2011]. Finally, our subgroup analyses indicated that the efficacy of DMDs in terms of brain volume-loss attenuation appears to be evident following the first year of treatment. No firm implications could be made for potential differences in the degree of brain-atrophy attenuation between the DMDs under consideration using data from the present meta-analysis.

Given the fact that brain-volume loss can be noninvasively and reproducibly detected and quantified by MRI [Grassiot et al. 2009], whole-brain atrophy has recently emerged as an attractive measure of long-term tissue loss and as a substrate for clinical disability and therapy effectiveness [Bermel and Bakshi, 2006; Barkhof et al. 2009]. In particular, brain-volume loss has been correlated with disability progression and cognitive impairment in MS, with the loss of gray-matter volume having greater clinical relevance in comparison with the loss of white-matter volume [Barkhof et al. 2009]. The development of brain atrophy in CIS patients is influenced by previously detected inflammatory activity in the brain, as atrophy appears only after a delay of months following inflammation [De Stefano et al. 2014]. There is accumulating evidence indicating that brain atrophy is evident even in the earliest clinical stages of MS (newly diagnosed CIS) compared with healthy controls [Geurts et al. 2012; Lukas et al. 2015]. Moreover, MS patients exhibiting disease progression were found to have higher rates of spinal cord-atrophy progression [Bakshi et al. 2005]. Finally, the investigators of a recent study evaluating brain metabolic changes suggestive of axonal damage in patients with radiologically isolated syndrome (RIS) have reported lower normalized brain volumes in patients with RIS than in healthy individuals [Stromillo et al. 2013]. However, it should be noted that for the time being brain-volume loss can hardly be used as a decision-making biological marker in everyday clinical practice. On the other hand, its potential use as an outcome measure in RCTs evaluating possible ‘neuroprotective’ effects of DMDs in different stages of MS is currently under consideration [Geurts et al. 2012].

Our sensitivity analyses indicated that the pooled beneficial effect of DMDs on brain atrophy was significantly greater during the second year (SMD: −0.25) compared with the first (SMD: −0.01). This observation may be attributed to an increase in nontissue-related brain-volume loss during treatment with disease-modifying therapies (termed ‘pseudoatrophy’) during the first year of treatment, which is thought to be due to the resolution of inflammation and brain edema [Paolillo et al. 2004; Khan et al. 2012]. It is still unknown whether pseudoatrophy is a temporary phenomenon or if its extent depends on the type, dose, and administration route of DMDs [Zivadinov et al. 2008]. Thus, our observations suggest that the development of pseudoatrophy may confound MRI measurements of brain-volumeloss during the first months of treatment and an observation period of more than 1 year may be optimal to evaluate the potential beneficial effect of DMDs on brain atrophy in future RCTs. Interestingly enough, the impact of fingolimod on brain-volume loss was observed at early time-point comparisons (6 and 12 months) in patients with relapsing-remitting MS, thereby indicating the lack of pseudoatrophy in MS patients taking fingolimod during the first year [Barkhof et al. 2014]. However, whether such an effect of the same treatment might be valid in patients with CIS remains to be clarified.

Certain limitations of this report need to be acknowledged. First, the number of included studies was relatively small, as only three of the RCTs of patients with CIS provided atrophy measures on baseline and follow up. Brain-volume loss was not reported in the other three large CIS trials with interferon beta-1a, that is, the Controlled High Risk Avonex Multiple Sclerosis Study (CHAMPS) [Jacobs et al. 2000], the REbif FLEXible dosing in early MS (REFLEX) trial [Comi et al. 2012], and the Betaseron in Newly Emerging Multiple Sclerosis for Initial Treatment (BENEFIT) trial [Kappos et al. 2006], and thus they could not be included in the present analysis. Despite evaluating independent data from the different subgroups and/or time points of each study, the findings of the present meta-analysis need to be reproduced in future using a larger number of RCTs. It should also be noted that neither the relationship between brain atrophy and spinal cord-atrophy progression, nor a subgroup analysis between the type of CIS (e.g. optic neuritis, sensible signs/symptoms, motor deficits) and progression of brain atrophy after disease-modifying therapies were reported in any of the three trials that were included in the meta-analysis [Filippi et al. 2004; Comi et al. 2013; Miller et al. 2014]. Second, even though quality control of the included studies suggests an overall low risk of bias, we cannot exclude bias related to funding source, as all study protocols had financial and/or other support from pharmaceutical industries with a clear conflict of interest on the study outcomes. Third, variable DMDs (interferon beta-1a, glatiramer acetate, teriflunomide), in different dosages (teriflunomide 7 mg/14 mg), in different routes of administration (oral, subcutaneously), and in different administration schemes (once daily, once weekly) were investigated in different RCTs. Thus, it is possible that the aforementioned differences in the treatment subgroups could be, at least partially, responsible for the identified correlations. Fourth, the imaging methods that were used for the measurement of brain-volume change were identical in two out of three included studies and this may have confounded the reported associations. However, it should be kept in mind that no significant heterogeneity was detected among trials.

In conclusion, DMDs appear to attenuate brain atrophy over time in patients with CIS and this effect becomes evident following the first year of treatment. Future RCTs with longer follow-up periods and standardized imaging methodology are required to evaluate the potential protective effect of DMDs on brain and spinal-volume loss in CIS patients beyond the first 2 years of treatment, and to remove potential confounding in MRI measurements due to pseudoatrophy.