Abstract
Objective: The authors examined the effect of 6.0 MIU interferon beta-1a (IFNβ-1a) administered IM each week on the evolution of monthly magnetization transfer ratio (MTR) within new gadolinium-enhancing (Gd+) lesions in patients with very early relapsing-remitting (RR) MS.
Background: IFNβ is an effective disease-modifying treatment for patients with RRMS. Among other effects, it has been shown to decrease the number of new Gd+ and T2-weighted lesions. MTR is a putative marker for irreversible tissue damage and evolution of MTR within a lesion may reflect recovery of tissue damage. It is not known whether IFNβ-1a affects the recovery phase of lesions.
Methods: Eight untreated patients with RRMS who completed up to 14 monthly brain MRI sessions elected to initiate treatment with IFNβ-1a. Four out of eight patients developed new Gd+ lesions during treatment. MTR of lesions at the time of appearance and subsequent rate of change of monthly MTR were compared before and after treatment (stratified Mann-Whitney test).
Results: The difference between MTR at appearance of 47 new Gd+ lesions before treatment versus 23 new Gd+ lesions during treatment was not significant. Twenty-two of 47 new Gd+ lesions before treatment and 11 of 23 new Gd+ lesions after treatment were monitored for up to 6 months. After appearance of new Gd+ lesions, the rate of increase in MTR was faster during therapy (p = 0.037).
Conclusion: MTR abnormalities within new Gd+ lesions evolve at a faster rate during treatment with IFNβ-1a than before initiating therapy. This is consistent with the hypothesis that IFNβ-1a promotes resolution of new Gd+ lesions.
MRI has helped to clarify the natural history of MS and is currently accepted as a primary outcome measure for screening promising disease-modifying therapies.1 A focal disturbance of vascular permeability, as detected by leakage of IV gadolinium contrast-enhancing (Gd+) agents into the brain, is an early event in the formation of new lesions.1 Disease activity, as measured by the appearance of new or enlarging focal lesions, can be detected 5 to 10 times more frequently by monthly brain MRI scans than by monthly change in standardized measures of neurologic impairment and disability.1 When compared with white matter from healthy control subjects, diffuse and subtle MRI abnormalities also can be detected in MS normal-appearing white matter (NAWM). These abnormalities include prolonged T1 and T2 relaxation time constants,2 reduced magnetization transfer ratios (MTR),3 elevated apparent diffusion coefficients,4,5 decreased spectral intensity for N-acetylaspartate,6-8 and increased spectral intensity for creatine.9 These changes are consistent with the notion that diffuse microscopic edema, demyelination, gliosis, and axonal loss are far more extensive than observed on conventional MRI.
We previously reported that quantitative monthly MRI measures are, on average, more abnormal in NAWM regions in which new Gd+ lesions become visible.10 We have also reported a pattern of quantitative monthly MR measures that precedes, accompanies, and follows the appearance of many new Gd+ lesions. This pattern consists of the following sequence of events. First, low MTR is apparent for at least several months before new Gd+ lesions later become visible. Second, MTR abruptly declines in concert with the appearance of a new Gd+ lesion. Third, MTR increases monthly for 2 to 6 months after the appearance of new Gd+ lesions but generally does not recover to prelesion levels. These observations have been confirmed by other investigators.11 We speculated this pattern of change in MTR first reflects appearance and resolution of focal edema and demyelination and the residual reduction in MTR 3 or more months following the appearance of new Gd+ lesions provides a surrogate measure of irreversible demyelination and axonal loss.10
Interferon beta (IFNβ) is an effective disease modifying treatment for patients with relapsing-remitting (RR) and secondary progressive (SP) MS.12-15 Perhaps the most impressive therapeutic benefit of this treatment is the dramatic reduction in number of new Gd+ and T2-weighted (T2W) lesions.12-17 However, lesions imaged with T2W sequences do not reliably distinguish reversible edema and inflammation from irreversible demyelination and axonal loss. Thus, it was unknown whether the effects of IFNβ reflected a benefit on reversible or irreversible tissue damage. In a preliminary effort to determine whether IFNβ could reduce change in a putative marker of irreversible tissue damage, we studied MTR in new Gd+ lesions that became visible before and after initiating treatment with IFNβ-1a, 6.0 MIU administered IM each week.18 We reported MTR was similar in new lesions that became visible before and during treatment. Richert et al. independently reported serial whole brain MTR was similar before and after initiating IFNβ-1b.19
The overall goal of the current study was to define the effect of IFNβ-1a, 6.0 MIU administered IM each week, on the evolution of monthly MTR within new Gd+ lesions. We hypothesized the slope of recovery of monthly MTR within new Gd+ lesions that became visible during therapy would be more positive than the slope of recovery of monthly MTR within Gd+ lesions that became visible before initiating therapy. Therefore, our study was designed to answer the following question: Do monthly quantitative measures of MTR in new Gd+ lesions evolve differentially before and after initiating therapy with IFNβ-1a?
Methods.
Clinical methods.
This study was approved by the University of California, San Francisco (UCSF) Committee on Human Research. Participation was offered to all patients who were evaluated at the UCSF/Mt. Zion MS Center between September 21, 1995 and November 23, 1996, and who met the following entry criteria: 1) clinically definite20 RRMS,21 2) age between 18 and 55 years, 3) Expanded Disability Status Scale score (EDSS)22 < 6.0, 4) the presence of one or more focal Gd+ lesions on T1-weighted MRI, and 5) no prior treatment with IFNβ, glatiramer acetate, or immunosuppressant drug. All MS patients were evaluated with monthly neurologic examinations and brain MRI scanning sessions for as long as 14 months. Those patients who exhibited three or more new Gd+ lesions during the last six scanning sessions were offered treatment with IFNβ-1a administered IM each week. Those patients who elected to initiate therapy underwent monthly neurologic examinations and brain MRI scanning sessions for up to 6 months.
Unscheduled neurologic examinations were performed before and during treatment, as soon as practical after report of new or worsening symptoms. Exacerbations were treated with 500 mg IV methylprednisolone per day for 3 consecutive days when symptoms were accompanied by loss of ability to perform a routine activity of daily living. To minimize the effect of corticosteroids on MRI measures of interest,23 imaging procedures were delayed 14 or more days after methylprednisolone therapy was completed.
MRI methods.
MRI methods for image acquisition, selection of regions of interest (ROIs), and definition of lesion ROIs have been previously described in detail.10 In summary, all MR data were acquired on a Siemens Vision 1.5-T system (Erlangen, Germany) using a quadrature head coil. Before MR acquisition, the B0 homogeneity over the entire head was optimized using an automated routine supplied by the manufacturer. Two-dimensional (2D) gradient echo scout images, acquired in three planes, were used for positioning of subsequent MRI scans. All subsequent MRI scans were collected with full brain coverage using a 192 × 256 matrix collected over a 180 mm × 240 mm field of view and 3-mm contiguous slices that were angulated parallel to an imaginary line connecting the anterior commissure to the posterior commissure. The nominal in-plane resolution was 1 mm2. During each monthly MRI session six full brain image sets were acquired in the order listed below. Proton density (PD; echo time [TE] 20/repetition time [TR] 2500) and T2W (TE 80/TR 2500) scans were acquired using a double spin echo sequence. A 3D gradient echo (TE 6/TR 50/flip angle [FA] 15) scan was acquired without (GE) and with (GEMT) an off-resonance saturation pulse, but with the same receiver gain and processing factors. The scans were acquired with the same receiver gain. The saturation pulse was of Gaussian shape, 8 msec in duration, positioned 1.6 kHz off-resonance from the narrow proton water frequency, and had a mean amplitude of 3.7 μT. A T1-weighted MRI (TE 17/TR 600) scan was acquired before (T1W) and 10 minutes after (T1Gd+) IV injection of gadodiamide (Omniscan, Nycomed Inc., Princeton, NJ; 0.1 mmol/kg). All MRI were coregistered using the Woods’ algorithm24,25 to the PD MRI collected during the first MRI session. New lesions were identified by visual inspection from T2W and T1Gd+ MRI using software developed in-house. This software allowed the user to compare the visual appearance and anatomic coordinates of new lesions, and to extract raw pixel intensities for any of the imaging measures of interest for any ROI and across all of the MRI sessions.
The lesion ROI was defined at appearance as all pixels in which MTR differed by three or more SDs from the mean MTR in 20 pixels of NAWM located 5 mm distant from the PD-defined edge of the new lesion.10 MTR for each lesion ROI was defined as the mean MTR for all pixels in the lesion ROI. The pixels for each lesion ROI at appearance were held constant every month thereafter.
Statistical analysis.
MTR was measured in new Gd+ lesions detected before and during treatment. Lesions were monitored monthly for up to 6 months. The rate of change of monthly MTR was expressed as a slope estimated by a least squares regression for each lesion. MTR of lesions at the time of appearance and individual slopes of MTR evolution after enhancement were compared before and after treatment using Mann-Whitney tests stratified by patient. This approach avoids the assumptions of normally distributed data and independence of different lesions within the same patient, both of which were not valid. For each patient who had at least one lesion both before and after treatment, it compares the before lesions to the after lesions, looking for within-patient change. It then combines evidence across patients to obtain an overall p value for whether observed changes within patients might have arisen by chance in the absence of any influence of treatment on lesion MTR values.
Results.
Accounting of patients and MRI activity.
Twenty-two untreated patients completed up to 14 monthly MRI scans. The mean age was 38 years, 50% were women, and the mean disease duration from the first symptom was 12 months with mean EDSS score at entry of 1.4. Eight of these patients exhibited three or more new Gd+ lesions during the last six scanning sessions and elected to initiate weekly therapy with IFNβ-1a. Of these eight patients, four developed one or more new Gd+ lesions during treatment. Lesions from three of these four patients could be monitored for 2 to 6 months after becoming visible: 22 before treatment and 11 during treatment. Lesions that appeared within 2 months before treatment initiation were excluded because the rate of change (recovery) that would have occurred without treatment could not be estimated. Similarly, lesions appearing at the last or second to last month of the study were also excluded. One patient required one steroid treatment while under therapy with IFNβ-1a. To eliminate potential confounding of steroid effects, any lesion that developed in this patient following steroid therapy was excluded from the analysis. Similarly, no lesion was followed into the poststeroid treatment period.
Cross-sectional comparisons of MTR within new Gd+ lesions before and during treatment.
Differences between the MTR of 47 new Gd+ lesions before treatment and 23 new Gd+ lesions during treatment in the four patients at the time of appearance did not reach statistical significance. The range of MTR values at appearance for these lesions is given in table 1.
Range of magnetization transfer ratio (MTR) values of lesions at time of appearance
Longitudinal comparisons of MTR within new Gd+ lesions before and during treatment.
The rate of recovery as measured by the slope in monthly MTR following lesion appearance was faster during treatment than before treatment (p = 0.037) (table 2). Meaningful multivariate analyses to directly evaluate whether any of this association could be explained by confounding factors, such as lesion size, highest gadolinium intensity, or lowest MTR, are not possible with this small sample. As a crude assessment of the potential for such confounding, we examined rank correlations with recovery slope over all lesions from the three patients who contributed data to this analysis. These rank correlations were −0.36 for lowest MTR (95% CI −0.63 to −0.02, p = 0.04), −0.25 for lesion size (95% CI −0.57 to +0.10, p = 0.16), and −0.14 for highest gadolinium intensity (95% CI −0.46 to + 0.21, p = 0.43).
Median rates of recovery of magnetization transfer ratio (MTR) values of lesions before and during treatment*
Discussion.
Converging lines of evidence support the notion that MS is an autoimmune disease in which T cells mediate destruction of myelin within the CNS. This process may be triggered by exposure to non-self antigens that are presented to T cells in the context of major histocompatibility complex (MHC)-II molecules. T cells, upon recognition of these antigens, express vascular endothelial adhesion molecules and liberate matrix metalloproteinases that increase vascular permeability and facilitate migration of T cells into the CNS. T cells then migrate through CNS parenchyma along chemokine concentration gradients and may encounter myelin antigens that molecularly mimic non-self antigens initially leading to T cell activation. The inability to distinguish self from non-self antigens may partially explain T-cell–mediated destruction of myelin.
IFNβ dramatically reduces the number of new Gd+ lesions in patients with RRMS.17,27 This effect is generally observed within 4 weeks of initiating therapy and is believed to reflect the ability of IFNβ to reduce expression of vascular endothelial adhesion molecules28 and liberation of matrix metalloproteinases29,30 and to modulate the actions of IFNγ and other proinflammatory cytokines.31,32 However, the effect of IFNβ on the rate of resolution of imaging metrics in new Gd+ lesions has never been studied. For this reason, we compared the rate of improvement in MTR, a putative imaging marker of demyelination and edema, in new focal Gd+ lesions that became visible before and during treatment with IFNβ-1a. The rate of improvement in MTR was more rapid in lesions that became visible during treatment than before treatment. Thus our observations provide the first evidence that IFNβ hastens recovery of MTR in some new Gd+ MRI lesions, and this effect was independent of lesion size or gadolinium intensity.
Although monthly MTR in new focal Gd+ MRI lesions improved more rapidly during treatment with IFNβ, it is unclear to what extent this reflects resolution of edema, remyelination, or both. We speculate some degree of remyelination is probable if acute reduction in MTR predominantly reflects demyelination. This appears to be the case, as MTR is only slightly reduced in acute experimental allergic encephalitis, a condition in which focal lesions are largely edematous with little or no demyelination.33 In contrast, MTR is reduced to a much greater extent in acute MS lesions that are characterized histopathologically by demyelination and reduced axonal density.34
There are several limitations to our study, none of which invalidate our findings. First, the observation period was brief and not uniform across patients. Lesions were only followed for 2 to 6 months. Thus, we cannot compare final level of recovery for MTR in lesions appearing before and during therapy within patients. We also cannot exclude the possibility that more robust treatment effects would be evident with longer follow-up. Second, our cohort was highly selected for very early RRMS. Thus, the treatment effect observed in this cohort might differ or not be evident in patients with RRMS whose disease duration is longer or in patients with SPMS or primary progressive MS. Third, the numbers of lesions and patients in this study were small. Thus, the treatment effect observed in this cohort should be considered as a preliminary finding and may not be evident in other patients with early RRMS or in all lesions from individuals with early RRMS. Small sample size also prevented us from directly assessing whether faster recovery during treatment might be due to confounding factors such as lesion size, highest gadolinium intensity, or lowest MTR. Fourth, we administered single-dose gadolinium to enhance the T1-weighted images. Higher doses of gadolinium might detect more numerous and less severe lesions. Our study was not designed to address this question and we do not know whether treatment effects observed with single-dose gadolinium are evident with triple-dose gadolinium. Fifth, there is no histopathologic correlation for the treatment effect observed during our study. Thus, we cannot determine to what extent improvement in the rate of recovery of MTR reflects resolving edema, remyelination, or both processes.
Our observations indicate that treatment with IFNβ-1a favorably influences the rate of recovery of low MTR following the appearance of some new Gd+ lesions. This is the first indication that IFNβ-1a not only reduces the number of new focal lesions but also affects the evolution of MTR within those lesions. To what extent this treatment effect represents more rapid resolution of microscopic edema or earlier remyelination is uncertain. However, either benefit could help explain confirmed observations that IFNβ-1a reduces clinical manifestations of disease activity in patients with relapsing forms of MS.12,16,17 Our preliminary findings provide support for the notion that MTR and quantitative MRI in MS can be used not only to monitor new disease activity but to monitor the resolution of new MRI lesions as influenced by disease-modifying therapies.
Acknowledgments
Supported in part by the National Multiple Sclerosis Society, RG2655-A-4 (D.E.G.).
Footnotes
-
None of the authors has any financial interest in Biogen or its products. D.E.G. and M.K. have received honoraria for presenting educational lectures at scientific meetings sponsored by Biogen.
- Received August 10, 1999.
- Accepted January 27, 2000.
References
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Alert Me
Recommended articles
A serial study of new MS lesions and the white matter from which they arise
A magnetization transfer histogram study of normal-appearing brain tissue in MS
A conventional and magnetization transfer MRI study of the cervical cord in patients with MS
Comparison of MS clinical phenotypes using conventional and magnetization transfer MRI