To Treat or Not to Treat in
Clinically Isolated Syndrome
The decision whether to treat or not to treat patients presenting with a clinically isolated syndrome suggestive
of multiple sclerosis (MS) can be a difficult one. If the results of magnetic resonance imaging (MRI) are normal,
most clinicians would not treat, as the risk of development of MS is low. If the MRI is abnormal, however, the
decision is considerably more difficult. At present, we do not have any good markers that would allow us to deter-
mine which individuals will go on to develop MS and which individuals will remain stable.
An advantage of early treatment is a 15% to 20% further reduction in attack frequency over and above the
approximately 30% decrease that occurs with later treatment. This has been a consistent finding with early treat-
ment with the interferons and glatiramer acetate. Nevertheless, with early treatment we are subjecting many patients
to a regimen that will not help them in order to improve the outcome in those who will develop clinically definite
MS. This “medicalizes” those patients, who will suffer the same side effects as those who need the medication.
On the other hand, the placebo patients from the phase 3 trials who were treated with their second attack sub-
sequently have more attacks than those treated early, and probably more disability. The long-term subcutaneous
interferon beta-1b (Betaseron) results showed a much higher 17-year mortality rate in those receiving placebo than
in those treated with either low-dose or high-dose Betaseron in the phase 3 trial.
The following two articles by Richard A. Rudick (Mellen Center for Multiple Sclerosis Treatment and Research,
Cleveland Clinic, Cleveland, OH) and David A. Cottrell (Frenchay Hospital, Bristol, UK), respectively, present
both sides of this ongoing controversy.
—Robert M. Herndon, MD
Editor Emeritus and Controversies Editor
The Case for Early Treatment of Clinically Isolated Syndrome
with Disease-Modifying Drugs
Richard A. Rudick, MD
arly and continuous therapy with disease-mod- ment because they have mild MS. 2) Delaying treat-
ifying drugs (DMDs) is advisable for patients ment a few years to monitor and select patients for
with clinically isolated syndrome (CIS) at high DMD therapy is safe. 3) There is no evidence that early
risk of multiple sclerosis (MS), and for patients with and continuous treatment with DMDs translates into
established early relapsing-remitting MS (RRMS). long-term benefits. 4) Early and continuous treatment
is too expensive and carries unnecessary patient burden
Objections to this position include the following: 1)
and risk. Because health-care resources are finite and
Some patients do not need early and continuous treat-
MS drugs are expensive, the last of these objections
From the Mellen Center for Multiple Sclerosis Treatment and may have merit. Estimating the financial benefits of
Research, Neurological Institute, Cleveland Clinic, Cleveland, OH, DMDs is difficult, because financial costs are front-
USA. Correspondence: Richard A. Rudick, MD, Mellen Center
loaded, while financial benefits are delayed and are sub-
U10, Cleveland Clinic Foundation, 9500 Euclid Ave., Cleveland,
OH 44195; e-mail: firstname.lastname@example.org. ject to imprecise estimates of preventable health-care
International Journal of MS Care
costs. Also, support, education, advocacy, rehabilita- inflammation, which has been linked to axonal transec-
tion, and symptomatic therapy have many benefits, tion. Transected axons are abundant in acute MS
and some of these approaches are relatively inexpensive plaques and at the inflammatory edges of chronic
compared with DMDs. Cost-effectiveness of DMDs active lesions, even in patients with short disease dura-
relative to other approaches to MS therapy is an impor- tion.5 Transected axons are terminally injured and will
tant debate, but in this article, I will assume that degenerate, contributing to abnormalities in the nor-
mal-appearing white matter observed on MRI studies.6
DMDs are accessible and affordable and will focus on
Some gadolinium-enhancing lesions leave persistent
the scientific and clinical rationale for early treatment.
black holes, which represent areas of reduced axonal
Four fundamental considerations support my rec-
ommendation for early diagnosis and continuous dis- These observations demonstrate that inflamma-
ease treatment. tion within acute MS lesions irreversibly damages
1. MS evolves to a disabling disease in a significant central nervous system (CNS) tissue, suggesting an
proportion of affected individuals; benign MS exists but ongoing destructive pathologic process during the
cannot be accurately predicted early in the disease. Wein- course of RRMS. Not surprisingly, longitudinal
shenker and colleagues 1 reported that MS patients MRI studies demonstrated accelerated brain atrophy
required assistance with walking a median of 15 years in MS patients.8,9 Brain volumes decline five to eight
from their diagnosis. Confavreux and colleagues 2 times faster than in healthy, age-matched controls,
reported that once MS patients experienced limitations reflecting increasing loss of myelin and axons over the
in ambulation, they could expect progressively increas- course of the disease. This process begins very early in
ing disability over time. Clinical experience confirms MS. Rapid rates of brain atrophy have been observed
these observations. Middle age brings progressive walk- in CIS patients who convert to MS,10,11 and rapid brain
ing problems for many MS patients. Once walking atrophy continues throughout the disease. Like all dis-
problems begin, gradual deterioration is the rule, not ease-severity measures, atrophy rates vary across MS
the exception. Undoubtedly some patients have mild patients, and the rate of atrophy during RRMS is only
MS, and some—perhaps 20% of the total—never expe- weakly correlated with relapses or Expanded Disability
rience disability from their disease. However, some Status Scale (EDSS) score.12 This suggests that progres-
patients classified as benign after 10 years progress to sive loss of brain tissue may be clinically inapparent
higher levels of disability after another 10 years.3 Also, during early stages of the disease. Brain atrophy during
some patients classified as having “benign MS” have RRMS is highly relevant, however, as it has been shown
significant cognitive impairment. Thus the definition to predict future ambulatory13 and cognitive9 disability.
and diagnosis of benign MS are problematic. The real Accelerated gray matter atrophy occurs in patients at all
problem, however, is that benign MS can be diagnosed stages of MS as well14 and correlates better with disabili-
accurately only in retrospect. No precise markers of ty than other MRI measures.11
benign MS exist that can be applied at the beginning of The picture emerges of a disease that is pathologically
the disease, or within a few years. Patients are not well active from onset, in which CNS tissue may be dam-
served by an imprecise prediction that they might have aged or destroyed without obvious clinical manifesta-
benign MS. Delaying treatment in patients who may tions during early disease stages. Over time, with cumu-
have benign MS will inevitably result in lost opportuni- lative brain tissue injury and the additive effects of
ties for many individuals who need treatment. aging, a threshold is surpassed beyond which neurologic
2. Irreversible axon and myelin injury occurs early in function fails and progressive disability ensues. This is
the course of MS and is mostly subclinical but leads to sec- the likely basis for SPMS. If this is the correct sequence
ondary progressive MS (SPMS). Longitudinal magnetic of events, then waiting for progressive disability before
resonance imaging (MRI) studies have demonstrated a treating is inappropriate. This model supports the need
high frequency of new subclinical brain lesions, indi- for early preventive therapy.
cating that MS is frequently active without clinical dis- 3. Early treatment reduces brain inflammation, is neu-
ease manifestations.4 New MRI lesions reflect tissue roprotective, and delays SPMS. Phase 3 trials of subcuta-
International Journal of MS Care
neous interferon beta-1b (IFNβ-1b) (Betaseron), 15 (Rebif ) persisted at 4 years. Patients with continuous
intramuscular interferon beta-1a (IFNβ-1a) (Avonex),16 treatment for 4 years were significantly better than
subcutaneous IFNβ-1a (Rebif ),17 and subcutaneous patients with an initial 2-year treatment with place-
glatiramer acetate (Copaxone)18 have demonstrated a bo.28,29 Assessment of patients from the placebo-con-
reduction in relapse rate of approximately 30% in trolled study of intramuscular IFNβ-1a (Avonex)
RRMS. The beneficial effect on new MRI lesions is showed that treatment with IFNβ-1a rather than place-
greater. In some trials, reduced disability progression bo for 2 years led to better patient outcomes at 8 years,
measured using the EDSS was also observed, 16,17 even though all patients were treated about the same
although some argue that permanent disability is diffi- between 2 and 8 years.30 A 5-year extension of the
cult to measure in early MS because most patients do CHAMPS study showed a persistent benefit for the
not become significantly disabled, the EDSS may patients originally treated with IFNβ-1a (Avonex), even
measure impairment rather than disability at the low though placebo patients were switched to IFNβ-1a
end of the scale, and some of the change may be noise after converting to clinically definite MS (CDMS).31
in the measure or reversible change.19 However, some Third, studies of IFNβ and glatiramer acetate have
studies12,20,21 demonstrated reduced rates of brain atro- been largely negative in patients with SPMS; thus, no
phy, clearly indicating that therapy during the RRMS drugs are approved for treatment of SPMS, because no
stage can inhibit irreversible brain tissue destruction. drug has established efficacy in slowing disability pro-
This was also seen in the AFFIRM study, in which gression once patients enter this stage of the disease.
natalizumab (Tysabri) significantly reduced brain atro- Brain atrophy studies are consistent with the clinical
phy in the second year of the clinical trial.22 Because findings. In contrast to RRMS, there are no examples
the rate of brain atrophy has been linked to future dis-
of treatment interventions that slow the rate of atrophy
ability, it is reasonable to speculate that reduced brain
in SPMS.32,33 It seems that patients become refractory
atrophy during RRMS will translate into delayed tran-
to treatment at later stages of MS.
sition to SPMS. This was recently reported in a longi-
Collectively, these studies suggest a “window of
opportunity” for treating MS. Some prominent thought
4. Delayed treatment is less effective than early treat- leaders have proposed very early, aggressive immune
ment, and no treatment is effective once the patient transi- suppression in patients with relapsing MS, with favor-
tions to SPMS. What is the harm in waiting to decide able early results.34 Although the jury is still out on the
about DMDs in MS patients? The problem with this benefits and risks of very aggressive, early treatment,
strategy is that early treatment appears to be more continuous treatment of MS patients early in the disease
effective than delayed treatment. First, studies using the with currently available, safe DMDs is clearly beneficial
same drug at different disease stages have shown consis- and advisable.
tently higher therapeutic effects in earlier stages of dis-
ease. Studies of IFNβ and glatiramir acetate have Financial Disclosures: Dr. Rudick has received consulting fees
shown that these agents reduce relapse rates by about from Millennium, Novartis, Teva Neuroscience, Wyeth Pharmaceuti-
30% in established RRMS but by about 50% in cals, Biogen Idec, and Genzyme-Bayer.
patients with CIS.24-26 Interferon beta-1a reduces sus- References
tained EDSS worsening by about 40% in RRMS, but 1. Weinshenker BG, Bass B, Rice GP, et al. The natural history of multi-
by only 20% at most in SPMS.27 The data suggest that ple sclerosis: a geographically based study, I: clinical course and dis-
ability. Brain. 1989;112:133–146.
MS becomes progressively less responsive to existing 2. Confavreux C, Vukusic S, Adeleine P. Early clinical predictors and
therapy. progression of irreversible disability in multiple sclerosis: an amnesic
Second, follow-up studies from randomized place- process. Brain. 2003;126:770–782.
3. Sayao AL, Devonshire V, Tremlett H. Longitudinal follow-up of
bo-controlled studies show persistent benefits from “benign“ multiple sclerosis at 20 years. Neurology. 2007;68:
early treatment; patients with delayed treatment do not 496–500.
“catch up” with the therapeutic benefits of early treat- 4. McFarland HF, Stone LA, Calabresi PA, Maloni H, Bash CN, Frank
JA. MRI studies of multiple sclerosis: implications for the natural histo-
ment. The PRISMS-4 study showed that the benefits ry of the disease and for monitoring effectiveness of experimental
of initial randomization to subcutaneous IFNβ-1a therapies. Mult Scler. 1996;2:198–205.
International Journal of MS Care
5. Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, Bo L. Axonal 20. Hardmeier M, Wagenpfeil S, Fisher E, et al. Predictors of atrophy
transection in the lesions of multiple sclerosis. N Engl J Med. during treatment with rIFN-beta 1a IM in relapsing multiple sclerosis:
1998;338:278–285. a three year follow-up. Neurology. 2003;60:A301.
6. Filippi M, Tortorella C, Bozzali M. Normal-appearing white matter 21. Rovaris M, Comi G, Rocca MA, Wolinsky JS, Filippi M. Short-term
changes in multiple sclerosis: the contribution of magnetic resonance brain volume change in relapsing-remitting multiple sclerosis: effect of
techniques. Mult Scler. 1999;5:273–282. glatiramer acetate and implications. Brain. 2001;124:1803–1812.
7. van Walderveen MA, Kamphorst W, Scheltens P, et al. Histopatholog- 22. Polman CH, O‘Connor PW, Havrdova E, et al. A randomized, place-
ic correlate of hypointense lesions on T1-weighted spin-echo MRI in bo-controlled trial of natalizumab for relapsing multiple sclerosis. N
multiple sclerosis. Neurology. 1998;50:1282–1288. Engl J Med. 2006;354:899–910.
8. Fisher E. Measurement of central nervous system atrophy. In: Cohen J,
23. Trojano M, Pellegrini F, Fuiani A, et al. New natural history of interfer-
Rudick R, eds. Multiple Sclerosis Therapeutics. London: Informa
on-beta-treated relapsing multiple sclerosis. Ann Neurol. 2007;61:
9. Bermel RA, Bakshi R. The measurement and clinical relevance of
24. Jacobs LD, Beck RW, Simon JH, et al.; CHAMPS Study Group. Intra-
brain atrophy in multiple sclerosis. Lancet Neurol. 2006;5:158–170.
muscular interferon beta-1a therapy initiated during a first demyelinat-
10. Dalton CM, Brex PA, Jenkins R, et al. Progressive ventricular enlarge-
ing event in multiple sclerosis. N Engl J Med. 2000;343:898–904.
ment in patients with clinically isolated syndromes is associated with
the early development of multiple sclerosis. J Neurol Neurosurg Psy- 25. Comi G, Filippi M, Barkhof F, et al. Effect of early interferon treatment
chiatry. 2002;73:141–147. on conversion to definite multiple sclerosis: a randomised study.
11. Fisniku LK, Chard DT, Jackson JS, et al. Gray matter atrophy is related Lancet. 2001;357:1576–1582.
to long-term disability in multiple sclerosis. Ann Neurol. 2008;64: 26. Kappos L, Polman CH, Freedman MS, et al. Treatment with interferon
247–254. beta-1b delays conversion to clinically definite and McDonald MS in
12. Rudick RA, Fisher E, Lee JC, Simon J, Jacobs L; Multiple Sclerosis Col- patients with clinically isolated syndromes. Neurology. 2006;67:
laborative Research Group. Use of the brain parenchymal fraction to 1242–1249.
measure whole brain atrophy in relapsing-remitting MS. Neurology. 27. European Study Group on interferon beta-1b in secondary progres-
1999;53:1698–1704. sive MS. Placebo-controlled multicentre randomised trial of interferon
13. Fisher E, Rudick RA, Simon JH, et al. Eight-year follow-up study of beta-1b in treatment of secondary progressive multiple sclerosis.
brain atrophy in patients with MS. Neurology. 2002;59:1412– Lancet. 1998;352:1491–1497.
1420. 28. PRISMS Study Group and the University of British Columbia MS/MRI
14. Fisher E, Lee JC, Nakamura K, Rudick RA. Gray matter atrophy in Analysis Group. PRISMS-4: Long-term efficacy of interferon-beta-1a in
multiple sclerosis: a longitudinal study. Ann Neurol. 2008;64:255– relapsing MS. Neurology. 2001;56:1628–1636.
265. 29. Schwid SR, Bever CT Jr. The cost of delaying treatment in multiple
15. The IFN-b Multiple Sclerosis Study Group. Interferon beta-1b is effec- sclerosis: what is lost is not regained. Neurology. 2001;56:1620.
tive in relapsing-remitting multiple sclerosis, I: clinical results of a mul- 30. Rudick RA, Cutter GR, Baier M, et al. Estimating long-term effects of
ticenter, randomized, double-blind, placebo-controlled trial. Neurolo- disease-modifying drug therapy in multiple sclerosis patients. Mult
16. Jacobs LD, Cookfair DL, Rudick RA, et al.; Multiple Sclerosis Collabo-
31. Kinkel RP, Kollman C, O‘Connor P, et al. IM interferon beta-1a delays
rative Research Group (MSCRG). Intramuscular interferon beta-1a for
definite multiple sclerosis 5 years after a first demyelinating event.
disease progression in relapsing multiple sclerosis. Ann Neurol.
32. Molyneux PD, Kappos L, Polman C, et al.; European Study Group on
17. PRISMS (Prevention of Relapses and Disability by Interferon beta-1a
Interferon beta-1b in secondary progressive multiple sclerosis. The
Subcutaneously in Multiple Sclerosis) Study Group. Randomised dou-
ble-blind placebo-controlled study of interferon beta-1a in relapsing/ effect of interferon beta-1b treatment on MRI measures of cerebral
remitting multiple sclerosis. Lancet. 1998;352:1498–1504. atrophy in secondary progressive multiple sclerosis. Brain. 2000;
18. Johnson KP, Brooks BR, Cohen JA, et al. Copolymer 1 reduces 123(Pt 11):2256–2263.
relapse rate and improves disability in relapsing-remitting multiple 33. Filippi M, Rovaris M, Iannucci G, Mennea S, Sormani MP, Comi G.
sclerosis: results of a phase III multicenter, double-blind placebo-con- Whole brain volume changes in patients with progressive MS treated
trolled trial. Neurology. 1995;45:1268–1276. with cladribine. Neurology. 2000;55:1714–1718.
19. Ebers GC, Heigenhauser L, Daumer M, Lederer C, Noseworthy JH. 34. Coles AJ, Compston DA, Selmaj KW, et al. Alemtuzumab vs. interfer-
Disability as an outcome in MS clinical trials. Neurology. 2008;71: on beta-1a in early multiple sclerosis. N Engl J Med. 2008;359:
International Journal of MS Care
The Case for Delaying Treatment of
Clinically Isolated Syndrome
David A. Cottrell, MB ChB, MRCP, BSc, PhD
s a practicing neurologist in the United treatment group, compared with 5% of the delayed
Kingdom with a special interest in multiple treatment group, having entered the progressive phase
sclerosis (MS), I often prescribe interferon over the 10 years of follow-up. This was despite a
beta (IFNβ) and glatiramer acetate for patients with maintained 10-year risk reduction in the number of
relapsing-remitting MS (RRMS) and, if relapses are patients developing clinically definite MS (CDMS) in
prominent, in patients with secondary progressive MS. those receiving immediate treatment. The results from
The decision whether to start one of these agents is extension trials are, however, very difficult to interpret
made with the patient after we have discussed the avail- because of their significant dropout rates, the nature of
able evidence of efficacy and the potential side-effect crossover trials, and problems with validation using
profiles of these therapies. small changes in the EDSS as a measure of unremitting
An early, safe, tolerable treatment that, most impor- disability.4
tantly, has long-term efficacy is highly desirable for I remain unconvinced by the data presented from
newly presenting patients, their families, and MS spe- these trials that initiating either IFNβ or glatiramer
cialists. Hence I have followed the evolving industry- acetate after a patient’s first clinical episode of demyeli-
sponsored clinical trials of drug treatment for patients nation is in the patient’s best interests. I outline my
with clinically isolated syndrome (CIS) with interest. reasoning for this position below.
The evidence for commencing IFNβ or glatiramer
Potential Failure of Trial Blinding
acetate in patients presenting with CIS is based on data
The BENEFIT trial acknowledges that blinding was
from the ETOMS, CHAMPS, and BENEFIT studies,
unsuccessful, with 67% of patients correctly guessing
as well as the recently completed but as yet unpub-
that they were receiving the active drug, versus 51% in
lished PRECISE study.1-3 Each of these trials random-
the placebo cohort.3 Unfortunately, the other CIS trial
ized patients with “positive” baseline brain magnetic
reports do not provide this essential information on the
resonance imaging (MRI) studies to either placebo or
potential for failure of blinding. Both CHAMPS and
IFNβ (at a range of dosages) or glatiramer acetate
ETOMS, however, do report a significantly higher inci-
(PRECISE). All four treatment cohorts show similar
dence of well-recognized and highly publicized side
rates of relapse reduction in the region of approximate-
effects (as also seen in the BENEFIT trial) in the IFNβ
ly 15% over a 2-year period but no significant alter-
group compared with the placebo group, including
ation in disability as compared with the placebo
injection-site reactions (60% vs 12% in ETOMS), flu-
cohorts. Indeed, the longest extension trial presented to like symptoms (54% vs 26% in CHAMPS), and
date, the CHAMPIONS 10-year data (American Acad- depression (20% vs 13% in CHAMPS).1,2
emy of Neurology, 2009) continues to show no signifi- The side-effect profiles of IFNβ and glatiramer
cant difference in either the Expanded Disability Status acetate are readily available to all with Internet access.
Scale (EDSS) or Multiple Sclerosis Functional Com- Therefore, loss of trial blinding is perhaps unavoidable
posite (MSFC) between the immediate and delayed given the frequently unpleasant side effects that often
treatment cohorts. These trial data also demonstrate a accompany these agents.
slightly worrying trend, with 14% of the immediate It is certainly widely recognized that relapse rates are
not immune to placebo effects. Indeed, significant
From the Department of Neurology, Frenchay Hospital, Bristol, Unit- apparent positive effects on relapse rates occur regularly
ed Kingdom. Correspondence: David A. Cottrell, MB ChB, MRCP,
BSc, PhD, Department of Neurology, Frenchay Hospital, Bristol, in the placebo arms of recent RRMS clinical trials. If
BS16 1LE, United Kingdom; e-mail: David.Cottrell@nbt.nhs.uk. blinding is not successfully achieved, it does call into
International Journal of MS Care
question the validity of such trials, particularly those not due to pharmacologic intervention, as only 4 of the
with small apparent treatment effects. Could this lack 107 patients who were followed up (including patients
of blinding explain the remarkably consistent apparent with normal as well as abnormal baseline MRI studies)
clinical treatment effect seen despite the wide range of were treated with disease-modifying therapies at some
IFNβ dosages used in the CIS trials (22 µg once weekly point over the 20 years of follow-up. This natural histo-
to 250 µg on alternate days), not to mention the simi- ry of CIS in patients with abnormal baseline MRI stud-
lar response rate seen with the unrelated glatiramer ies is in striking contrast to the median EDSS score of 8
acetate? found in the entire London, Ontario, cohort at 20.6
If not, these trials suggest that there is no dose years from the time of first clinical onset, or the median
response to IFNβ in the reduction of relapse rates in EDSS score of 7 found at 20 years from the first event
CIS. Pharmacologically, it is quite unusual to see no in those patients who presented with RRMS.7
apparent dose response to an active treatment. Although In addition, ETOMS reported a surprisingly low
I accept that this is unlikely to be the full explanation annual relapse rate (ARR): 0.43/year in the placebo
for the trial results and certainly could not explain the arm despite enrolling a highly selected patient group
MRI findings, I believe that all such trials in cohorts of with MRIs laden with demyelination (median T2 load,
MS patients should be assessed for blinding to aid clini- 24).1 A review of the data provided in the CHAMPS
cians in the interpretation of such important data. and BENEFIT trials suggests that the relapse rates in
the placebo arms of these CIS trials do not differ sub-
Natural History of CIS stantially from that in ETOMS. These ARRs are con-
The principal aim of starting a treatment early in siderably lower than those in the original RRMS piv-
the course of MS must be to avoid long-term disabili- otal studies, which ranged from 0.9 to 1.28. Given that
ty. We are all well versed in the natural history of these patients presumably have already had numerous
CDMS, but the natural history of CIS, with or with- episodes of clinically silent demyelination, as evidenced
out MRI evidence of disseminated demyelination, can- by their MRI brain scans, could these studies have
not be assumed to be the same. inadvertently selected a more benign clinical pheno-
The London, Ontario, and Lyon long-term prospec- type? We have all seen highly abnormal MRI scans sug-
tive natural history cohort studies both enrolled patients gestive of demyelination in patients with either no
diagnosed using Poser’s criteria. These cohorts therefore symptoms suggestive of MS or surprisingly little in the
represent the natural history of patients who go on to way of disability. Certainly despite their significant
have a second clinical episode of disability or progres- median T2 lesion loads these trial cohorts show surpris-
sive disease. The natural history of CIS has not been ingly little in the way of disability as measured by the
studied to the same extent. Two smaller long-term CIS EDSS either at the beginning or end of the CIS trials,
follow-up trials, however, suggest that as many as 20% in addition to their low clinical relapse rates.
of patients with CIS will remain free of any further Many would argue that we need not wait for the
new neurologic symptoms suggestive of a relapse or second clinical episode since the baseline MRI is a reli-
progression for 20 to 25 years.5,6 able indicator of developing CDMS and long-term dis-
The recently published data from the more compa- ability. The Optic Neuritis Treatment Trial (ONTT),8
rable long-term study of CIS patients by Fisniku et al.6 national hospital study,6 and Barcelona study9 all pro-
recruited for a baseline MRI scan between 1984 and vide evidence that having a positive baseline MRI scan
1987 show a much more favorable prognosis than significantly increases the risk of future conversion to
demonstrated in the London, Ontario, and Lyon stud- CDMS compared with having a normal baseline MRI
ies. Twenty years after the diagnosis of CIS, of the scan. However, even having a baseline MRI scan with a
patients with abnormal baseline MRI studies, 18% had large number of T2 lesions does not nearly guarantee
not developed CDMS, and an additional 39% had conversion to CDMS even after many years of follow-
EDSS scores of 3 or less. Therefore, at least 57% (23% up in any of these cohorts.
of the cohort were lost to long-term follow-up) of the In addition, and perhaps more importantly, neither
CIS patients with abnormal baseline MRI studies had a the 15-year data from the ONTT8 nor the recently
good or an excellent outcome over 20 years. This was reported 20-year data from the national hospital study6
International Journal of MS Care
provide evidence to support the belief that baseline mented that even CIS patients with a positive base-
MRI lesion load translates into that all-important reli- line MRI scan have a reasonably good chance, about
able predictor of long-term disability. The ONTT 20%, of not developing further clinical symptoms of
found no correlation between the number of T2 lesions MS over a 20-year period,6 and pre-MRI natural his-
at baseline and disability at 15 years. Fisniku and col- tory suggests that 1 in 5 will have no further symp-
leagues6 found that 18% of patients with 10 or more toms after 25 years.5 Therefore, if one started IFNβ
T2 lesions at baseline had not developed CDMS at 20 or glatiramer acetate in a patient with CIS, when
years and that another 35% had minimal disability would one consider stopping treatment? Would the
(EDSS ≤3). Therefore, more than half of the patients CIS patient continue on the drug for 20 years if he or
in their cohort with highly active baseline scans had a she remained relapse-free because of the perceived ben-
reassuringly favorable outcome even at 20 years after efit to the patient or the physician? At a drug cost
their first relapse. alone (based on UK prices) of approximately £120,000
Therefore, patients with CIS appear to have both a to £180,000 (approximately $192,000 to $288,000)
low ARR and a significantly better long-term prognosis for 20 years of administration, any error in this deci-
than those with clinically definite RRMS even when sion would be a costly one, not to mention the 1040 to
their baseline scans are highly abnormal. It also seems 7300 injections the patient would have to endure, with
clear from the natural history evidence that those diag- the accompanying “medicalization.”
nosed with CDMS as defined by Poser’s criteria have a The BENEFIT study states that the “patient num-
significantly worse prognosis in terms of both relapse ber needed to be treated in order to prevent one case of
rates and long-term disability. CDMS (ie, one relapse) within the study period of 2
years is estimated to be 5.9.”3 This equates to approxi-
Costs for Patients and Taxpayers mately 12 years of treatment (624 to 4380 injections) at
Another issue related to treatment timing is the cost a drug cost alone of approximately £72,000 to £90,000
involved for both patients and taxpayers. Those of us ($115,000 to $144,000) to stop one relapse in 12
who work within the UK National Health Service are patients over a 2-year period. If one assumes that the
made acutely aware of the cost/benefit ratio of poten- average length of disability caused by a relapse is 2
tial treatments. The Association of British Neurologists months, this equates to a drug cost of between £432,000
released revised guidelines in 2007 regarding the sug- and £576,000 (approximately $691,000 to $921,000)
gested use of disease-modifying treatments for MS cli- per year of patient disability benefit. Given that many of
nicians in the UK. These included approval for treating these relapses will be mild and often do not affect activi-
patients with CIS who met stipulated MRI criteria. ties of daily living and there are additional costs in terms
However, perhaps following controversy, these guide- of physician consultations, blood tests, and so on, the
lines were subsequently withdrawn and are awaiting real cost will be substantially higher still. In addition to
further revision.10 Importantly, the decision whether to this financial cost, and perhaps more importantly, the
treat CIS patients has yet to be addressed by our Nation- other 11 patients will experience no benefit, but they
al Institute for Clinical Excellence (NICE). Perhaps as will often experience unpleasant side effects and the
a result, beginning IFNβ or glatiramer acetate treat- accompanying “medicalization” that starting treatment
ment in CIS patients has not been widely accepted in with IFNβ or glatiramer acetate entails.
the UK. Additionally, and extremely importantly, there is the
Each of the CIS treatment trials reported that over a possibility of misdiagnosis. In my experience, the diag-
2- to 3-year period on treatment approximately 15% of nosis of MS tends to become easier and more secure
patients avoided a relapse, as compared with placebo. with increasing duration of the illness. The differential
The ETOMS, CHAMPS, BENEFIT, and PRECISE diagnosis of CIS is wide, and no matter how experi-
data therefore suggest that approximately 85% of enced or talented the physician, it is unavoidable for
patients starting IFNβ or glatiramer acetate will experi- some patients to be misdiagnosed and thus potentially
ence no benefit. In addition, 50% to 55% of these mistreated. If one includes the CIS patients who will
patients would not have had a relapse whether treated not convert to CDMS, the potential for misdiagnosis
or not. Indeed, as already highlighted, it is well docu- and mistreatment is very large—approximately 50% at
International Journal of MS Care
2 years and 20% at 20 years. These figures are substan- mencing these agents in CIS patients we may inadver-
tially higher than the 15% of patients whose second tently do more harm than good. Hence, until we have
relapse is delayed. a reliable prognostic indicator and a treatment with
proven long-term efficacy, is it not best to wait until at
Many patients might argue, rightly, that the finan- least that second clinical relapse to avoid potentially
cial cost is immaterial if treatment will improve their unnecessary, costly, and harmful (in terms of side effects
quality of life and reduce their disability over the 40 to and medicalization) treatment, since this is clearly a bet-
50 years of their illness. In addition, many MS special- ter predictor than a positive MRI study, or any other
ists have suggested that treating patients early may test, of possible future deterioration?
modify their disease in the long term. Yet one could
Financial Disclosures: Dr. Cottrell has received unrestricted
argue that the beta-interferons and glatiramer acetate research grants from Merck Serono, Teva Neuroscience, Biogen Idec,
even in the responding CIS patients do not prevent the and Bayer-Schering (in total, less than £250,000, or approximately
process of MS, except simply by means of “definitions” $400,000), as well as direct honoraria for lectures and participation
over a short period of time, which may not be a clini- in advisory board meetings from Bayer-Schering, Biogen Idec,
cally meaningful way to think of therapy for a chronic, Novartis, and Merck Serono (totaling less than £10,000, or approxi-
Indeed, each of the CIS treatment trials includes References
patients with highly active baseline MRI scans (median 1. Comi G, Filippi M, Barkhof F, et al. Effect of early interferon treat-
T2 lesion load, between 5 and 24). This suggests that ment on conversion to definite multiple sclerosis: a randomised
study. Lancet. 2001;357:1576–1582.
the process of central nervous system demyelination 2. Jacobs LD, Beck RW, Simon JH, et al. Intramuscular interferon beta-
has been present for some considerable time prior to 1a therapy initiated during a first demyelinating event in multiple
their clinical presentation. Therefore, the argument sclerosis. N Engl J Med. 2000;343:898–904.
3. Kappos L, Polman CH, Freedman MS, et al. Treatment with interfer-
that these trials represent “early treatment” appears on beta-1b delays conversion to clinically definite and McDonald
flawed, and so to propose that we must avoid a relative- MS in patients with clinically isolated syndromes. Neurology.
ly arbitrary delay in one relapse during a 40- to 50-year 2006;67:1242–1249.
4. Ebers GC, Heigenhauser L, Daumer M, Lederer C, Noseworthy JH.
illness by introducing a treatment that appears to have Disability as an outcome in MS clinical trials. Neurology. 2008;
no impact on disability is perhaps unsustainable. 71(9):624–631.
Intuitively, I agree that there is an attractive hypo- 5. Eriksson M, Andersen O, Runmarker B. Long-term follow up of
patients with clinically isolated syndromes, relapsing-remitting and
thetical link between early treatment and delay in long- secondary progressive multiple sclerosis. Mult Scler. 2003;9:
term disability. However, in MS this presupposes that 260–274.
the treatments in question are effective in preventing 6. Fisniku LK, Brex PA, Altmann DR, et al. Disability and T2 MRI
lesions: a 20-year follow-up of patients with relapse onset of multiple
disability. Sadly, there is no convincing clinical or trial sclerosis. Brain. 2008;131;808–817.
evidence yet to support early treatment on grounds of 7. Kremenchutzky M, Rice GP, Baskerville J, Wingerchuk DM, Ebers
preventing long-term disability or, crucially, delaying GC. The natural history of multiple sclerosis: a geographically
based study 9: observations on the progressive phase of the dis-
the onset of progressive disease. ETOMS reported inter- ease. Brain. 2006;129:584–594.
feron to have no significant effect on the accumulation 8. The Optic Neuritis Study Group. Multiple sclerosis risk after optic
of disability; likewise, the 10-year open-label extension neuritis: final optic neuritis treatment trial follow-up. Arch Neurol.
study of the CHAMPS cohort showed a lack of effect 9. Tintoré M, Rovira A, Río J, et al. Baseline MRI predicts future attacks
on disability as measured by the EDSS or MSFC. and disability in clinically isolated syndromes. Neurology. 2006;
As an MS physician, I would like nothing better for 67:968–972.
10. Scolding N, Wilkins A, Cottrell D. New guidelines for MS treat-
my patients than an early, safe treatment modality with ment—no cause for celebration. Adv Clin Neurosci Rehabil. 2007;
long-term efficacy. My fear, however, is that by com- 7:17–18.
International Journal of MS Care