Multiple Sclerosis Resource Centre

Welcome to the Multiple Sclerosis Resource Centre. This website is intended for international healthcare professionals with an interest in Multiple Sclerosis. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Predictors of long-term interferon discontinuation in newly diagnosed relapsing multiple sclerosis

Multiple Sclerosis and Related Disorders, November 2016, Pages 90 - 96



Interferon-β has long-term safety and efficacy profiles for Relapsing Remitting Multiple Sclerosis (RRMS). However, the increasing number of available treatments requires to improve patient profiling and to perform individualized clinical decisions. Therefore, the present study investigated predictors of Interferon-β discontinuation.


The present retrospective observational cohort study included 499 newly diagnosed, drug naïve RRMS subjects receiving Interferon-β as first disease modifying treatment (DMT), during a 7.9±3.8 year period, up to treatment discontinuation. Possible markers of interest were recorded at the time of diagnosis (age, gender, disease duration, baseline EDSS) or during follow-up as variables of disease evolution (relapse occurrence, annualized relapse rate -ARR-, 1-point EDSS progression, reaching of EDSS 4.0) or of treatment (high-dose Interferon-β1a, low-dose Interferon-β1a, or Interferon-β1b).


217 patients (43.5%) discontinued the treatment during the follow-up period, with an incidence of 5% person-years (95%CI=4.6–5.9%). A multivariate Cox regression model showed an increased rate of Interferon-β discontinuation for female gender (p=0.019; HR=1.428), higher baseline EDSS (p=0.026; HR=1.346), relapse occurrence (p=0.009; HR=1.618), higher ARR (p<0.001; HR=5.269), and Interferon-β1b treatment (p=0.019; HR=1.506); and a reduced rate for occurrence of EDSS progression (p<0.001; HR=0.299).


Most of the factors associated with Interferon-β discontinuation are not modifiable, and are part of demographic features (i.e. gender), or of disease characteristics (i.e. disability at diagnosis), but should be taken into account when prescribing the first DMT for MS. Noteworthy, the use of Interferon-β1b is associated with 50% increased risk of discontinuation, compared with high-dose Interferon-β1a, highlighting the importance of drug formulations in treatment persistence.


  • Predictors of persistence on DMT can help physicians in individualizing the treatment for MS.
  • An incidence of 5% person-years has been found for Interferon-β discontinuation.
  • Female gender, higher baseline EDSS and relapse occurrence were associated with an increased rate of discontinuing Interferon-β1a/b.
  • EDSS progression was associated with a reduced rate of discontinuing Interferon-β1a/b.
  • The risk of discontinuation was 50% higher for Interferon-β1b, and 30% higher for low-dose Interferon-β1a, compared to high-dose Interferon-β1a.

Keywords: Multiple sclerosis, Interferon, Persistence, Discontinuation, Longitudinal.

1. Introduction

Clinical trials and real life studies have been showing that Interferon-β1a and Interferon-β1b provide multiple sclerosis (MS) patients with remarkable benefits on long-term disability outcomes (Trojano et al, 2009, Bates, 2011, and Plosker, 2011), and have acceptable safety profiles (Reder et al, 2010, Sørensen, 2011, Saida et al, 2016, and Signori et al, 2016). Accordingly, neurologists are usually confident in prescribing Interferon-β as a first-line treatment for the vast majority of patients who develop Relapsing-Remitting (RR) MS (Sørensen, 2011). Nevertheless, Interferon-β possibly might also reduce the risk of conversion to secondary progressive MS (SPMS) (Trojano et al., 2007).

However, most patients may still present signs of disease activity, sometimes requiring more aggressive and expensive therapeutic strategies (Goodin, 2008, Sørensen, 2011, and Moccia et al, 2016b). Furthermore, considering the need for long term use of disease modifying treatments (DMTs), tolerability and adverse events can represent a significant concern, with substantial effects on treatment adherence (Goodin et al, 2012, Reder et al, 2014, Lanzillo et al, 2015, and Moccia et al, 2015). In addition, the development of new therapies for MS, with different benefits and risks, requires an accurate evaluation of factors possibly predicting optimal long-term response, in order to use the most appropriate and individualized treatment (Sørensen, 2011, Sormani and De Stefano, 2013, and Hartung et al, 2015). Indeed, if the presence of a therapeutic window of opportunity is assumed, the identification of early predictors (i.e. demographic features or clinical characteristics) of poor response to Interferon-β is expected to facilitate treatment decisions, possibly also reducing future disability accrual (Bermel et al, 2010, Du Pasquier et al, 2014, Hartung et al, 2015, and Ziemssen et al, 2016).

Therefore, the possibility to predict Interferon-β discontinuation remains an unmet need in MS clinical care and research, and, so far, has been investigated only in two long-term studies (Evans et al, 2012 and Zhornitsky et al, 2015). However, these investigations recruited heterogeneous MS populations with different courses of the disease, with the use of variable diagnostic criteria, and occasionally dating from periods without many therapeutical options. Nevertheless, only a limited number of demographic features (age, gender) and disease characteristics (disease duration, EDSS at first DMT) have been evaluated (Evans et al, 2012 and Zhornitsky et al, 2015), whereas nowadays different markers of long-term disease evolution and treatment response have been identified (Sormani and De Stefano, 2013). In view of this, the present retrospective study has been conducted in a longitudinal cohort of newly diagnosed, drug naïve RRMS patients with an average follow-up of 8 years, and aims at evaluating predictors of Interferon-β discontinuation that have been determined at the time of the diagnosis and during the course of MS.

2. Methods

2.1. Study design

The present retrospective observational cohort study evaluated markers associated with the risk of discontinuing Interferon-β treatment among newly diagnosed, drug naïve MS subjects.

In compliance with current Italian applicable laws and regulations, considering that all clinical assessments were part of clinical practice in a university setting and that the retrospective analysis included anonymized data, specific ethics approval was not required. All subjects signed the general informed consent form, authorizing the use of personal data for research purposes. The study was performed in accordance with good clinical practice and Declaration of Helsinki.

2.2. Patients

Inclusion criteria were: 1) new diagnosis of RRMS from January 2001 to January 2010, at the MS Clinical Care and Research Centre of the Federico II University Hospital (Naples, Italy), according to McDonald 2001 criteria (McDonald et al, 2001 and Lublin et al, 2014); 2) prescription of Interferon-β1a (high-dose or low-dose) or Interferon-β1b as first DMT for MS; 3) presence of at least 5 year clinical follow-up or discontinuation of the first DMT before that period; 4) age at baseline > 18.

Exclusion criteria were: 1) progressive forms of MS at the time of enrolment (Lublin et al., 2014); 2) pregnancy during the study period. Indeed, pregnancy does not only cause Interferon-β discontinuation, but can also result in a more benign clinical course, possibly affecting study results (Jokubaitis et al., 2016).

Due to the retrospective nature of the study, all MS subjects underwent DMTs, which were possibly discontinued during the study period, in accordance with the European Medicines Agency indications for clinical practice (EMA, 2015), and, so, as a consequence not only of suboptimal treatment effect but also of adverse events or low tolerability. Follow-up visits were scheduled at three-month intervals, or on the occasion of relapse.

2.3. Discontinuation of Interferon-β1a/b

Discontinuation of Interferon-β1a/b treatment was defined as a >90-day interruption in therapy, a switch to another first- or second-line therapy (i.e. Glatiramer acetate, Natalizumab, or Fingolimod), or complete discontinuation (i.e. no further record of medication initiation). The cohort was categorized as discontinuing or not the first DMT. To be more precise, switching from one Interferon-β1a dose to another was considered as discontinuation (i.e. from low-dose once weekly regimen, to high-dose three times a week). Switching between different branded versions of Interferon-β1b was not considered as discontinuation. Switching from Interferon-β1a to β1b formulations was considered as discontinuation (Evans et al, 2012 and Zhornitsky et al, 2015).

Within 1 month from the diagnosis, patients were suggested to start a DMT by the physician, and received their first drug supply with instructions for the administration from a trained nurse, according to clinical practice (baseline visit). The time occurring between DMT start and discontinuation was calculated (time to discontinuation). The discontinuation date was the last day that a patient received a dose, as reported in clinical records (Evans et al, 2012 and Zhornitsky et al, 2015).

2.4. Markers of interest for discontinuation of Interferon-β1a/b

Possible predictors of Interferon-β1a/b discontinuation were either recorded at the time of the diagnosis (early markers), or during follow-up visits (long-term markers).

In particular, at the time of the diagnosis, demographic characteristics (age, gender), and disease duration were collected. The Kurtzke’s Expanded Disability Status Scale (EDSS) was recorded to estimate the MS-related disability at diagnosis (baseline) (Kurtzke, 1983).

In addition, patients were classified into three groups according to the first DMT received (high-dose Interferon-β1a, low-dose Interferon-β1a, or Interferon-β1b).

Finally, MS subjects were evaluated for 4 different long-term markers of MS evolution, recorded in the most recent visit or, conversely, up to Interferon-β1a/b discontinuation:

  • Occurrence of clinical relapse: number of relapses occurring during the study period was recorded, and annualized relapse rate (ARR) was subsequently calculated; patients were further categorized as being relapse free or not during the treatment period; relapsing patients presented a range of motor/sensory symptoms and met commonly used standards for relapse as determined by clinical neurologists (McDonald et al., 2001);
  • 1-point EDSS progression: the cohort was subsequently categorized as experiencing or not EDSS progression;
  • Reaching of EDSS 4.0: the cohort was subsequently categorized according to the reaching of EDSS 4.0 (EDSS≥4.0) or not (EDSS<4.0);
  • Transition from RR to SP course: MS was retrospectively considered SP when a progressive accumulation of disability occurred after an initial relapsing course, and was associated with a worsening of the same functional system, independently from relapse activity (Lublin et al., 2014); the cohort was subsequently categorized in converting to SP (RR→SP) or not (RR→RR).

Long term disability markers (EDSS progression, reaching of EDSS 4.0 and converting to SP), were chosen because clinically important and unlikely to remit once sustained (Runmarker and Andersen, 1993, Sormani and De Stefano, 2013, and Moccia et al, 2016a). For their evaluation, the observation period was extended to 12 months in order not to misestimate disability accrual (Kalincik et al., 2015).

2.5. Statistical analyses

Means and proportions of demographics and clinical features were calculated for MS patients discontinuing or not Interferon-β1a/b treatment, and compared with t-test, χ2 test, or Fisher's exact test, as appropriate. The median time to discontinuation was estimated by using Kaplan-Meier survival analyses.

Cox regression models were utilized to estimate the hazard ratios (HR) of Interferon-β1a/b discontinuation in relation to different markers of interest, recorded either at the time of the diagnosis (age, gender, disease duration, baseline EDSS), or during the follow-up period as variables of disease evolution (relapse occurrence, ARR, EDSS progression, reaching of EDSS 4.0, SP conversion) or of treatment (high-dose Interferon-β1a, low-dose Interferon-β1a, or Interferon-β1b). First of all, a univariate model was fit to assess associations between each marker and the selected outcome. Subsequently, each model was adjusted for different covariates (age, gender, disease duration, baseline EDSS, ARR and DMT), and, finally, a fully adjusted multivariate logistic model included all the possible markers, as well as the covariates. HR and 95% confidence interval (95%CI) were calculated.

Stata 12.0 has been used for data processing and analysis. Normal distribution of the variables or residuals was tested by both statistical and graphical approaches, when appropriate. Variables were tested for multicollinearity and all the variance inflation factors (VIF) were lower than 2.5 indicating a reasonable assumption of independence among predictor variables. Results have been considered statistically significant if p<0.05.

3. Results

685 subjects received a new diagnosis of MS at the MS Clinical Care and Research Centre from January 2001 to January 2010. Among them, 499 RRMS patients were included in the statistical analyses (full details are reported in Fig. 1).

Fig. 1

Fig. 1

Patient disposition flow diagram.

Patient disposition flow diagram showing patients included, excluded or lost-to-follow-up within the study cohort.


During a mean follow-up time of 7.9±3.8 years, 217 patients (43.5%) discontinued the treatment, with an incidence of 5% person-years (95%CI=4.6–5.9%). The median time to discontinuation of Interferon-β1a/b prescribed as first DMT, was 11.98 years (95%CI=10.78–13.30). After treatment discontinuation, 196 patients switched to a different DMT (90.3%), whereas for 21 patients this represented the overall discontinuation of DMTs (9.7%).

Demographics and clinical features of the MS cohort are reported in Table 1, with particular regard to patients discontinuing or not Interferon-β1a/b.

Table 1

Demographics and clinical features of MS patients discontinuing or nor Interferon-β1a/b treatment.


Interferon-β1a/b Persistence(n=282) Interferon-β1a/b Discontinuation(n=217) P-values
Age (at baseline), years 33.5±8.5 30.8±7.5 <0.001*
Gender, female 164 (58.1%) 151 (69.5%) 0.009*
Disease duration (at baseline), years 2.8±2.8 2.5±2.6 0.176
Baseline EDSS 2.0±0.5 2.1±0.6 0.190
Patients with relapse occurrence 173 (61.1%) 163 (75.1%) 0.001*
ARR 0.19±0.28 0.55±0.68 <0.001*
1-point EDSS progression 211 (74.5%) 132 (60.8%) 0.001*
Reaching of EDSS 4.0 92 (32.5%) 65 (29.9%) 0.542
SP conversion 23 (8.1%) 13 (5.9%) 0.360
DMT 0.946
High-dose Interferon β1a 114 (57.2%) 85 (42.8%)
Low-dose Interferon β1a 88 (56.0%) 69 (44.0%)
Interferon β1b 80 (55.9%) 63 (44.1%)

*p<0.050. Results are presented as number (%) or as mean±standard deviation.

MS: Multiple Sclerosis; ARR: Annualized Relapse Rate; EDSS: Expanded Disability Status Scale; SP: Secondary Progressive; DMT: Disease Modifying Treatment.

Demographics and clinical features of MS patients discontinuing or nor Interferon β-1a/b treatment are reported. P-values are shown from t-test, χ2 test, Fisher's exact test, as appropriate.

The fully adjusted Cox regression model showed an increased rate of discontinuing Interferon-β1a/b for female gender (p=0.019; Adj. HR=1.428), higher baseline EDSS (p=0.026; Adj. HR=1.346) (Fig. 2), relapse occurrence (p=0.009; Adj. HR=1.618), higher ARR (p<0.001; Adj. HR=5.269) (Fig. 3), and Interferon-β1b treatment (p=0.019; Adj. HR=1.506) (Fig. 4); and a reduced rate for occurrence of EDSS progression (p<0.001; Adj. HR=0.299) (Fig. 3). Full results from unadjusted, partially adjusted and fully adjusted models are reported in Table 2.

Fig. 2

Fig. 2

Kaplan-Meier curves for Interferon-β1a/b discontinuation in relation to possible markers of interests recorded at the time of MS diagnosis. Kaplan-Meier plots estimating the probability of discontinuing Interferon-β1a/b in relation to age (A), gender (B), disease duration (C), and baseline EDSS (D) (for graphical purposes, continuous variables are represented as a solid line if higher than the median value, and as a dashed line if lower than the median value). P-values are shown from the multivariate Cox regression model.


Fig. 3

Fig. 3

Kaplan-Meier curves for Interferon-β1a/b discontinuation in relation to possible markers of interests for MS evolution. Kaplan-Meier plots estimating the probability of discontinuing Interferon-β1a/b in relation to relapse occurrence (A), ARR (B) (for graphical purposes, ARR is represented as a solid line if higher than the median value, and as a dashed line if lower than the median value), EDSS progression (C), reaching of EDSS 4.0 (D), and conversion to SP (E). P-values are shown from the multivariate Cox regression model.


Fig. 4

Fig. 4

Kaplan-Meier curves for Interferon-β1a/b discontinuation in relation to disease modifying treatments. Kaplan-Meier plots estimating the probability of discontinuing current disease modifying treatment in relation to the use of high dosage Interferon-β1a (green), of low-dose Interferon-β1a (red), or of Interferon-β1b (blue). P-values are shown from the multivariate Cox regression model, using high dosage Interferon-β1a as reference. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


Table 2

Factors associated with the risk of Interferon-β1a/b discontinuation.


Unadjusted model Partially adjusted model Fully adjusted model
HR 95% CI P-value HR 95% CI P-value HR 95% CI P-value
Lower Upper Lower Upper Lower Upper
Age 0.974 0.957 0.991 0.004* 0.982 0.963 1.001 0.070 0.986 0.967 1.005 0.151
Gender 1.505 1.126 2.011 0.006* 1.507 1.122 2.023 0.006* 1.428 1.060 1.923 0.019*
Disease duration 0.948 0.900 0.999 0.047* 0.966 0.915 1.020 0.218 0.952 0.901 1.006 0.084
Baseline EDSS 1.255 1.012 1.557 0.039 1.243 0.981 1.576 0.071 1.346 1.036 1.750 0.026*
Relapse occurrence 1.278 0.937 1.742 0.120 1.571 1.398 8.280 0.002* 1.618 1.432 8.860 0.009*
ARR 3.724 2.990 4.636 <0.001* 3.576 2.841 4.501 <0.001* 5.269 4.079 6.806 <0.001*
EDSS progression 0.330 0.247 0.440 <0.001* 0.253 0.187 0.342 <0.001* 0.299 0.213 0.419 <0.001*
Reaching of EDSS 4.0 0.666 0.497 0.893 0.007* 0.431 0.313 0.596 <0.001* 0.720 0.495 1.047 0.086
SP conversion 0.611 0.348 1.072 0.086 0.494 0.278 0.879 0.017* 0.842 0.455 1.556 0.584
Low-dose Interferon-β1a 1.289 0.932 1.783 0.124 1.328 0.957 1.844 0.089 1.325 0.952 1.844 0.095
Interferon-β1b 1.502 1.077 2.095 0.016* 1.651 1.176 2.319 0.004* 1.506 1.068 2.124 0.019*


MS: Multiple Sclerosis; ARR: Annualized Relapse Rate; EDSS: Expanded Disability Status Scale; SP: Secondary Progressive; DMT: Disease modifying treatment.

Different possible markers of interest are associated with the risk of Interferon β1a/b discontinuation during follow-up period. Results are shown from Cox regression analyses; each model was adjusted for different covariates (age, gender, disease duration, baseline EDSS, ARR and DMT -high-dose Interferon-β-1a as reference-), and, then, a fully adjusted multivariate Cox regression model included all the possible markers, as well as the covariates.

4. Discussion

Long-term DMTs are the main reason for a considerable improvement in clinical outcomes among MS patients and, thus, evaluating predictors of treatment persistence or, conversely, discontinuation is of primary importance to help physicians in prescribing the most appropriate and individualized medication (Bates, 2011 and Sørensen, 2011). Real-life studies are warranted to evaluate this issue, since clinical trials include highly motivated patients and cannot account for the long-term management in the everyday clinical setting. Hence, this 8-year monocentric cohort study showed a 43.5% discontinuation rate for the use of Interferon-β as the initially prescribed injectable DMT in adults with RRMS, with an incidence of 5% person-years and a median time to discontinuation of 11.98 years. Present findings are in line with the study of Zhornisky and colleagues, who reported a median time to discontinuation of 11.1 years during a 6-year follow-up period on a larger sample (Zhornitsky et al., 2015). On the other hand, Hansen and colleagues found an extremely high discontinuation rate after 2 years (67.7%), possibly because their definition of discontinuation was derived from the medication possession rate rather than on clinical evaluation (Hansen et al., 2015). Similarly, Evans and colleagues reported a shorter median time to discontinuation, compared with the present study (6.3 vs 11.99 years), but they included DMTs other than Interferon-β (Evans et al., 2012). Noteworthy, adherence rates might change in relation to geographical and cultural factors which cannot be excluded when comparing these results (Moccia et al., 2015). Overall, DMT discontinuation frequently occurs during the course of MS, possibly as a consequence of different factors, such as suboptimal treatment effect, adverse events, and low tolerability. Therefore, the present analysis accounted for all possible causes by evaluating the overall discontinuation rates for Interferon-β treatment.

Different factors recorded at diagnosis have been associated with the risk of discontinuing Interferon-β during the following 8 years, and, in particular, females presented 50% higher risk of treatment discontinuation compared to males. Although discordant results have been reported on gender effect in treatment persistence (Zhornitsky et al, 2015 and Bove et al, 2016), females might undergo DMT escalation more frequently, since they have dominant inflammatory aspects, which are the specific target of currently available treatments, whereas males seem to be more at risk of a progressive course (Runmarker and Andersen, 1993, Ribbons et al, 2015, and Bove et al, 2016). Conceivably, age might account for a similar explanation, with younger patients being more likely to discontinue Interferon-β treatment because of dominant inflammatory components (Trojano et al, 2009, Sormani and De Stefano, 2013, and Hartung et al, 2015).

With regard to disease severity at diagnosis, subjects with higher baseline EDSS presented a 30% increased risk of discontinuing Interferon-β treatment, also after adjusting for disease duration, supporting the current practice of using more incisive treatments in more aggressive disease courses from the onset. In fact, EDSS at diagnosis has been significantly related to the development of both physical disability and cognitive decline (Coppola et al, 2006, Freedman, 2011, Goodin et al, 2012, and Kappos et al, 2015). Nevertheless, the present study longitudinally recorded possible markers of Interferon-β persistence associated with inflammatory evolution, and found that the occurrence of a single relapse was associated with 60% increased risk of discontinuing Interferon-β, and that the occurrence of an additional relapse within the same year determined a five-fold risk of discontinuing Interferon-β. Therefore, acute demyelinating events are apparently an important element for therapeutical decisions in the clinical practice (Fox et al, 2013 and Hartung et al, 2015), and, in fact, a single relapse has been shown to predict future relapses and, generally, the inflammatory evolution of the disease (Sormani and De Stefano, 2013 and Hartung et al, 2015). In line with this, it is worth discussing that switching from Interferon-β to more active treatments might have had a positive impact in the present population. Indeed, patients with higher Interferon-β persistence presented a higher rate of disability accrual at follow-up evaluations. However, we did not include data after Interferon-β discontinuation and, so, further analyses are warranted.

After 15 years from the diagnosis, half of the subjects with RRMS are expected to convert to SPMS, characterized by a steadily increasing neurologic disability and a decline in inflammatory activity (Moccia et al., 2016a). From a clinical perspective, it is still debated whether patients who have been treated during the relapsing phase, should discontinue their DMTs when converting to a progressive form. Intriguingly, there is recent evidence suggesting an efficacy of DMTs in reducing the long-term disability accrual, and highlighting the importance of continuous treatment also in patients with reduced inflammatory component (Trojano et al, 2007, Tobin and Weinshenker, 2015, Tsivgoulis et al, 2015, and Kister et al, 2016). In view of this, considering that patients with a single point increase in the EDSS presented 35% reduced risk of Interferon-β discontinuation, it is possible that Interferon-β was utilized as a reliable and safe treatment in long-term real-life experience for MS forms characterized by predominantly progressive alterations or, at least, by reduced inflammatory aspects. However, other markers of disease progression (i.e. reaching of EDSS 4.0 or converting to SP) were not statistically significant and, so, future studies with longer follow-up periods are warranted to evaluate these long-term outcomes.

The current research also evaluated the importance of different Interferon-β formulations in predicting treatment discontinuation, and found high-dose Interferon-β1a to be particularly suitable in order to achieve longer persistence on treatment. In particular, the risk of discontinuation was 50% higher for Interferon-β1b, and 30% higher for low-dose Interferon-β1a, as compared with high-dose Interferon-β1a. This result is in line with the findings of Reynolds and colleagues (Reynolds et al., 2010), and is particularly relevant since Interferon-β formulations are the only modifiable factor possibly affecting long-term persistence. As a possible explanation, instead of a difference in DMT efficacy which has never been demonstrated so far, a difference in treatment satisfaction and adherence might be hypothesized because of the DMT by itself or by its injection system (Hupperts et al, 2015 and Moccia et al, 2015). Intriguingly, this result was particularly evident in longitudinal analyses, suggesting the need for long-term follow-up studies in order to assess discontinuation rates, possibly missed by previous studies with shorter observation periods (Halpern et al, 2011b and Hansen et al, 2015).

Finally, the present study had to deal with different limitations. Above all, due to its retrospective design, present analysis did not account for the specific reasons for treatment discontinuation. Notwithstanding, causes for poor persistence in the present population might be similar to those previously reported, and, so, might be different for each treatment (Evans et al, 2012, Fox et al, 2013, Kalincik et al, 2013, and Tobin and Weinshenker, 2015). In fact, there are many justifications for discontinuing and/or changing a treatment, which are related not only to disease activity, but also to patient's preference, tolerance and adherence, to the presence of neutralizing antibodies, to the injection system or to the nursing support provided (Malucchi et al, 2008, Depont et al, 2015, and Lanzillo et al, 2015). For instance, we cannot exclude that neutralizing antibodies are at least in part responsible for non-response to Interferon-β treatment (Malucchi et al, 2008, Magyari et al, 2013, and Paolicelli et al, 2013). In addition, the development of side effects from Interferon-β administration is another well-known reason for treatment discontinuation (Depont et al, 2015 and Lanzillo et al, 2015). Nevertheless, subjects treated with Glatiramer Acetate have not been included in the present study, since this injectable first-line treatment possibly presents different persistence rates as compared with Interferon-β (Zhornitsky et al., 2015). However, current results might provide an overall scenario considering all possible determinants of treatment discontinuation, since persistence accounts for all these reasons. Moreover, it has to be acknowledged that the current research included traditional variables (i.e. disability progression), whereas newer endpoints (i.e. cognitive dysfunction or quality of life) should have been evaluated as well (Bandari et al, 2013, Foley et al, 2013, and Moccia et al, 2016a). Furthermore, the sample size should have been larger in order to support present findings, but it might be considered appropriate if compared with similar previous studies (Reder et al., 2010). In addition, a longer follow-up would have certainly been useful and, in particular, would have allowed the study of persistence on newer treatments prescribed after first DMT discontinuation (Jokubaitis et al., 2013), or on newer formulations of previously available DMTs (Halpern et al., 2011a).

In conclusion, different predictors of long-term persistence on Interferon-β have been investigated and, more precisely, males with older age, lower disability scores and reduced inflammatory aspects are less at risk of discontinuing Interferon-β treatment, in particular if receiving high dose Interferon-β1a from the time of the diagnosis. Present finding might be helpful in the clinical practice when prescribing the first DMT for MS, since the evaluation of simple factors (i.e. age, gender, Interferon-β formulation) might allow a personalized treatment with reduced risk of long-term discontinuation. As a future perspective, the identification of patterns of Interferon-β persistence in MS patients might be helpful not only for developing support strategies targeted at patients at risk for poor persistence, but also for predicting long term impact of different treatments for MS on social and medical costs.


The present study received no specific funding.

Conflict of interest statement

Authors declare no conflict of interest in relation to the current work.


  • Bandari et al., 2013 D. Bandari, D. Wynn, T. Miller, B. Singer, S. Wray, R. Bennett, et al. Rebif (R) Quality of Life (RebiQoL): a randomized, multicenter, Phase IIIb study evaluating quality-of-life measures in patients receiving the serum-free formulation of subcutaneous interferon beta-1a for the treatment of relapsing forms of multiple scler. Mult. Scler. Relat. Disord.. 2013;2(1):45-56 Crossref
  • Bates, 2011 D. Bates. Treatment effects of immunomodulatory therapies at different stages of multiple sclerosis in short-term trials. Neurology. 2011;76(Suppl 1):S14-S25 Crossref
  • Bermel et al., 2010 R. Bermel, B. Weinstock-Guttman, D. Bourdette, P. Foulds, X. You, R. Rudick. Intramuscular interferon beta-1a therapy in patients with relapsing-remitting multiple sclerosis: a 15-year follow-up study. Mult. Scler.. 2010;16(5):588-596 Crossref
  • Bove et al., 2016 R. Bove, A. McHenry, K. Hellwig, M. Houtchens, N. Razaz, P. Smyth, et al. Multiple sclerosis in men: management considerations. J. Neurol.. 2016;263(7):1263-1273
  • Coppola et al., 2006 G. Coppola, R. Lanzillo, C. Florio, G. Orefice, P. Vivo, S. Ascione, et al. Long-term clinical experience with weekly interferon beta-1a in relapsing multiple sclerosis. Eur. J. Neurol.. 2006;13(9):1014-1021 Crossref
  • Depont et al., 2015 F. Depont, F. Berenbaum, J. Filippi, M. Le Maitre, H. Nataf, C. Paul, et al. Interventions to improve adherence in patients with Immune-Mediated Inflammatory Disorders: a systematic review. PLoS One. 2015;10(12):e0145076
  • EMA, 2015 EMA, 2015. European public assessment reports. Available from: 〈〉 (accessed 20.08.15).
  • Evans et al., 2012 C. Evans, J. Tam, E. Kingwell, J. Oger, H. Tremlett. Long-term persistence with the immunomodulatory drugs for multiple sclerosis: a retrospective database study. Clin. Ther.. 2012;34(2):341-350 Crossref
  • Foley et al., 2013 J.F. Foley, K. Zerkowski, K. Nair V. Evidence for long-term use of intramuscular interferon beta-1a: an overview of relapse, disability, and MRI data from selected clinical trials. J. Manag. Care Pharm.. 2013;19(1 Suppl A) (S4–15)
  • Fox et al., 2013 R.J. Fox, A.R. Salter, T. Tyry, J. Sun, X. You, G. Laforet, et al. Treatment discontinuation and disease progression with injectable disease-modifying therapies. Int. J. MS Care. 2013;15(4):194-201 Crossref
  • Freedman, 2011 M.S. Freedman. Long-term follow-up of clinical trials of multiple sclerosis therapies. Neurology. 2011;76(Suppl 1):S26-S34 Crossref
  • Goodin, 2008 D.S. Goodin. Disease-modifying therapy in multiple sclerosis update and clinical implications. Neurology. 2008;71(Suppl 3):S8-13
  • Goodin et al., 2012 D.S. Goodin, A. Traboulsee, V. Knappertz, A.T. Reder, D. Li, D. Langdon, et al. Relationship between early clinical characteristics and long term disability outcomes: 16 year cohort study (follow-up) of the pivotal interferon-1b trial in multiple sclerosis. J. Neurol. Neurosurg. Psychiatry. 2012;83(3):282-287 Crossref
  • Halpern et al., 2011a R. Halpern, S. Agarwal, L. Borton, K. Oneacre, M.V. Lopez-Bresnahan. Adherence and persistence among multiple sclerosis patients after one immunomodulatory therapy failure: retrospective claims analysis. Adv. Ther.. 2011;28:761-775 Crossref
  • Halpern et al., 2011b R. Halpern, S. Agarwal, C. Dembek, L. Borton, M. Lopez-Bresnahan. Comparison of adherence and persistence among multiple sclerosis patients treated with disease-modifying therapies: a retrospective administrative claims analysis. Patient Preference Adherence. 2011;5:73-84 Crossref
  • Hansen et al., 2015 K. Hansen, K. Schüssel, M. Kieble, J. Werning, M. Schulz, R. Friis, et al. Adherence to disease modifying drugs among patients with multiple sclerosis in Germany: a retrospective cohort study. PLoS One. 2015;10(7):e0133279 Crossref
  • Hartung et al., 2015 H.-P. Hartung, L. Kappos, D.S. Goodin, P. O’Connor, M. Filippi, B. Arnason, et al. Predictors of disease activity in 857 patients with MS treated with interferon beta-1b. J. Neurol.. 2015;262(11):2466-2471
  • Hupperts et al., 2015 R. Hupperts, V. Becker, J. Friedrich, C. Gobbi, A.V. Salgado, B. Sperling, et al. Multiple sclerosis patients treated with intramuscular IFN-β−1a autoinjector in a real-world setting: prospective evaluation of treatment persistence, adherence, quality of life and satisfaction. Expert Opin. Drug Deliv.. 2015;12(1):15-25 Crossref
  • Jokubaitis et al., 2016 V.G. Jokubaitis, T. Spelman, T. Kalincik, J. Lorscheider, E. Havrdova, D. Horakova, et al. Predictors of long-term disability accrual in relapse-onset multiple sclerosis. Ann. Neurol.. 2016;80(1):89-100
  • Jokubaitis et al., 2013 V.G. Jokubaitis, T. Spelman, J. Lechner-Scott, M. Barnett, C. Shaw, S. Vucic, et al. The Australian Multiple Sclerosis (MS) immunotherapy study: a prospective, multicentre study of drug utilisation using the MSBase platform. PLoS One. 2013;8(3):e59694
  • Kalincik et al., 2013 T. Kalincik, T. Spelman, M. Trojano, P. Duquette, G. Izquierdo, P. Grammond, et al. Persistence on therapy and propensity matched outcome comparison of two subcutaneous interferon beta 1a dosages for multiple sclerosis. PLoS One. 2013;8(5):e63480 Crossref
  • Kalincik et al., 2015 T. Kalincik, G. Cutter, T. Spelman, V. Jokubaitis, E. Havrdova, D. Horakova, et al. Defining reliable disability outcomes in multiple sclerosis. Brain. 2015;138(Pt 11):3287-3298
  • Kappos et al., 2015 L. Kappos, J. Kuhle, J. Multanen, M. Kremenchutzky, E. Verdun di Cantogno, P. Cornelisse, et al. Factors influencing long-term outcomes in relapsing–remitting multiple sclerosis: PRISMS-15. J. Neurol. Neurosurg. Psychiatry. 2015;86(11):1202-1207
  • Kister et al., 2016 I. Kister, T. Spelman, R. Alroughani, J. Lechner-Scott, P. Duquette, F. Grand’Maison, et al. Discontinuing disease-modifying therapy in MS after a prolonged relapse-free period: a propensity score-matched study. J. Neurol. Neurosurg. Psychiatry. 2016;
  • Kurtzke, 1983 J.F. Kurtzke. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology. 1983;33:1444-1452
  • Lanzillo et al., 2015 R. Lanzillo, M. Moccia, A. Carotenuto, V. Vacchiano, B. Satelliti, V. Panetta, et al. Vitamin K cream reduces reactions at the injection site in patients with relapsing-remitting multiple sclerosis treated with subcutaneous interferon beta - VIKING study. Mult. Scler.. 2015;21(9):1215-1216 Crossref
  • Lublin et al., 2014 F.D. Lublin, S.C. Reingold, J.A. Cohen, G.R. Cutter, A.J. Thompson, J.S. Wolinsky, et al. Defining the clinical course of multiple sclerosis: the 2013 revisions. Neurology. 2014;83(3):278-286 Crossref
  • Magyari et al., 2013 M. Magyari, H. Bach Søndergaard, F. Sellebjerg, P. Soelberg Sørensen. Preserved in vivo response to interferon-alpha in multiple sclerosis patients with neutralising antibodies against interferon-beta (REPAIR study). Mult. Scler. Relat. Disord.. 2013;2(2):141-146 Crossref
  • Malucchi et al., 2008 S. Malucchi, F. Gilli, M. Caldano, F. Marnetto, P. Valentino, L. Granieri, et al. Predictive markers for response to interferon therapy in patients with multiple sclerosis. Neurology. 2008;70:1119-1127 Crossref
  • McDonald et al., 2001 W.I. McDonald, A. Compston, G. Edan, D. Goodkin, H.P. Hartung, F.D. Lublin, et al. Recommended diagnostic criteria for multiple sclerosis. Ann. Neurol.. 2001;50:121-127 Crossref
  • Moccia et al., 2015 M. Moccia, R. Palladino, C. Russo, M. Massarelli, A. Nardone, M. Triassi, et al. How many injections did you miss last month? A simple question to predict interferon β-1a adherence in multiple sclerosis. Expert Opin. Drug Deliv.. 2015;12(12):1829-1835
  • Moccia et al., 2016a M. Moccia, R. Lanzillo, R. Palladino, K. Chang, T. Costabile, C. Russo, et al. Cognitive impairment at diagnosis predicts 10-year multiple sclerosis progression. Mult. Scler.. 2016;22(5):659
  • Moccia et al., 2016b M. Moccia, R. Palladino, R. Lanzillo, M. Triassi, V. Brescia Morra. Predictors of the 10-year direct costs for treating multiple sclerosis. Acta Neurol. Scand.. 2016;
  • Paolicelli et al., 2013 D. Paolicelli, M. D’Onghia, F. Pellegrini, V. Direnzo, P. Iaffaldano, V. Lavolpe, et al. The impact of neutralizing antibodies on the risk of disease worsening in interferon B-treated relapsing multiple sclerosis: a 5 year post-marketing study. J. Neurol.. 2013;260(6):1562-1568 Crossref
  • Du Pasquier et al., 2014 R.A. Du Pasquier, D.D. Pinschewer, D. Merkler. Immunological mechanism of action and clinical profile of disease-modifying treatments in multiple sclerosis. CNS Drugs. 2014;28(6):535-558 Crossref
  • Plosker, 2011 G.L. Plosker. Interferon-β-1b: a review of its use in multiple sclerosis. CNS Drugs. 2011;25(1):67-88 Crossref
  • Reder et al., 2010 A. Reder, G. Ebers, A. Traboulsee, D. Li, D. Langdon, D. Goodin, et al. Cross-sectional study assessing long-term safety of interferon-beta-1b for relapsing-remitting MS. Neurology. 2010;74(23):1877-1885 Crossref
  • Reder et al., 2014 A.T. Reder, J.F. Oger, L. Kappos, P. O’Connor, M. Rametta. Short-term and long-term safety and tolerability of interferon β-1b in multiple sclerosis. Mult. Scler. Relat. Disord.. 2014;3(3):294-302 Crossref
  • Reynolds et al., 2010 M.W. Reynolds, R. Stephen, C. Seaman, K. Rajagopalan. Persistence and adherence to disease modifying drugs among patients with multiple sclerosis. Curr. Med. Res. Opin.. 2010;26(3):663-674 Crossref
  • Ribbons et al., 2015 K.A. Ribbons, P. McElduff, C. Boz, M. Trojano, G. Izquierdo, P. Duquette, et al. Male sex is independently associated with faster disability accumulation in Relapse-onset MS but not in primary progressive MS. PLoS One. 2015;10(6):e0122686 Crossref
  • Runmarker and Andersen, 1993 B. Runmarker, O. Andersen. Prognostic factors in a multiple sclerosis incidence cohort with twenty-five years of follow-up. Brain. 1993;116(Pt 1):117-134
  • Saida et al., 2016 T. Saida, J. Kira, Y. Ueno, N. Harada, T. Hirakata. Long-term efficacy and safety of intramuscular interferon beta-1a: randomized postmarketing trial of two dosing regimens in Japanese patients with relapsing-remitting multiple sclerosis. Mult. Scler. Relat. Disord.. 2016;7:102-108
  • Signori et al., 2016 A. Signori, F. Gallo, F. Bovis, N. Di Tullio, I. Maietta, M.P. Sormani. Long-term impact of interferon or Glatiramer acetate in multiple sclerosis: a systematic review and meta-analysis. Mult. Scler. Relat. Disord.. 2016;6:57-63
  • Sørensen, 2011 P.S. Sørensen. Balancing the benefits and risks of disease-modifying therapy in patients with multiple sclerosis. J. Neurol. Sci.. 2011;311(Suppl 1):S29-S34
  • Sormani and De Stefano, 2013 M.P. Sormani, N. De Stefano. Defining and scoring response to IFN-β in multiple sclerosis. Nat. Rev. Neurol.. 2013;9(9):504-512 Crossref
  • Tobin and Weinshenker, 2015 W.O. Tobin, B.G. Weinshenker. Stopping immunomodulatory medications in MS: frequency, reasons and consequences. Mult. Scler. Relat. Disord.. 2015;4(5):437-443 Crossref
  • Trojano et al., 2007 M. Trojano, F. Pellegrini, A. Fuiani, D. Paolicelli, V. Zipoli, G.B. Zimatore, et al. New natural history of interferon-beta-treated relapsing multiple sclerosis. Ann. Neurol.. 2007;61(4):300-306 Crossref
  • Trojano et al., 2009 M. Trojano, F. Pellegrini, D. Paolicelli, A. Fuiani, G.B. Zimatore, C. Tortorella, et al. Real-life impact of early interferon beta therapy in relapsing multiple sclerosis. Ann. Neurol.. 2009;66(4):513-520 Crossref
  • Tsivgoulis et al., 2015 G. Tsivgoulis, A.H. Katsanos, N. Grigoriadis, G.M. Hadjigeorgiou, I. Heliopoulos, C. Kilidireas, et al. The effect of disease modifying therapies on brain atrophy in patients with relapsing-remitting multiple sclerosis: a systematic review and meta-analysis. PLoS One. 2015;10(3):e0116511 Crossref
  • Zhornitsky et al., 2015 S. Zhornitsky, J. Greenfield, M.W. Koch, S.B. Patten, C. Harris, W. Wall, et al. Long-term persistence with injectable therapy in relapsing-remitting multiple sclerosis: an 18-Year observational cohort study. PLoS One. 2015;10(4):e0123824 Crossref
  • Ziemssen et al., 2016 T. Ziemssen, T. Derfuss, N. de Stefano, G. Giovannoni, F. Palavra, D. Tomic, et al. Optimizing treatment success in multiple sclerosis. J. Neurol.. 2016;263(6):1052-1065


a Multiple Sclerosis Clinical Care and Research Center, Department of Neuroscience, Reproductive Science and Odontostomatology, Federico II University, Naples, Italy

b Department of Primary Care and Public Health, Imperial College, London, UK

c Department of Public Health, Federico II University, Naples, Italy

Corresponding author.