If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Matching-adjusted comparisons demonstrate better clinical outcomes with SC peginterferon beta-1a every two weeks than with SC interferon beta-1a three times per week
SC peginterferon beta-1a and SC IFN beta-1a were compared.
•
Annualized relapse rate was lower for SC peginterferon beta-1a.
•
24-week CDW was significantly lower for SC peginterferon beta-1a.
•
Clinical-NEDA was significantly higher for SC peginterferon beta-1a.
Abstract
Background
Subcutaneous (SC) peginterferon beta-1a and SC interferon beta-1a (IFN beta-1a) have demonstrated efficacy in treating relapsing-remitting multiple sclerosis (RRMS) but have never been compared in direct head-to-head clinical trials, the gold-standard comparison. A well-balanced matching-adjusted comparison of weighted individual patient data on SC peginterferon beta-1a, and aggregate data from published phase 3 clinical trials of SC IFN beta-1a, was conducted to provide additional information on the comparative efficacy of these two agents.
Methods
Individual patient data from a study of SC peginterferon beta-1a 125 mcg every two weeks (ADVANCE) and pooled summary data from four published studies of SC IFN beta-1a 44 mcg three times per week (OPERA I and II, CARE-MS I and II) with similar populations were utilized. A comparison was conducted by weighting individual peginterferon beta-1a–treated patients, using estimated propensity of enrolling in SC IFN beta-1a treatment to match multiple key aggregate baseline characteristics of SC IFN beta-1a–treated patients. After matching, weighted annualized relapse rate (ARR), 24-week confirmed disability worsening (CDW), and clinical no evidence of disease activity (clinical-NEDA) were calculated and compared for peginterferon beta-1a and SC IFN beta-1a.
Results
After matching, baseline characteristics were well balanced across treatment groups. At 2 years, ARR after matching was 0.256 for patients receiving peginterferon beta-1a (effective n = 376) and 0.335 for those receiving SC IFN beta-1a (n = 1218) (P = 0.0901). The percentage of patients who were relapse free over 2 years was significantly higher with peginterferon beta-1a than with SC IFN beta-1a (75.1% vs. 57.4% [after matching], P < 0.0001). The peginterferon beta-1a treatment group had a significantly lower proportion of patients with 24-week CDW compared with SC IFN beta-1a (after matching 6.5% vs. 13.2%; P = 0.0007). Clinical-NEDA occurred in a significantly higher proportion of patients treated with SC peginterferon beta-1a versus SC IFN beta-1a (74.1% vs. 48.1%; P < 0.0001).
Conclusions
This matching-adjusted comparison using data from four phase 3 trials with SC IFN beta-1a formulations demonstrated that patients with RRMS treated with SC peginterferon beta-1a 125 mcg every two weeks achieved better clinical outcomes than patients who received SC IFN beta-1a 44 mcg three times per week.
The management of RRMS involves the use of disease-modifying therapies (DMTs), such as IFN beta-based therapies that reduce disease activity and disability by modulating the host immune response (
Therapeutics and technology assessment subcommittee of the American academy of neurology and the MS council for clinical practice guidelines. Disease modifying therapies in multiple sclerosis: report of the therapeutics and technology assessment Subcommittee of the American Academy of Neurology and the MS Council for clinical practice Guidelines.
European Medicines Agency, 2005. Guideline on clinical investigation of medicinal products for the treatment of multiple sclerosis. London, 15 September.
). SC IFN beta-1a is a well-established RRMS treatment that has demonstrated efficacy in terms of reducing relapse frequency and disability worsening (
Therapeutics and technology assessment subcommittee of the American academy of neurology and the MS council for clinical practice guidelines. Disease modifying therapies in multiple sclerosis: report of the therapeutics and technology assessment Subcommittee of the American Academy of Neurology and the MS Council for clinical practice Guidelines.
). Use of SC IFN beta-1a requires injections several times per week, which has been linked to poor treatment compliance leading to reduced treatment effects (
). SC peginterferon beta-1a has also shown efficacy in treating RRMS, with safety and tolerability in line with the IFN class and a reduced injection frequency (once every two weeks) compared with SC IFN beta-1a (
To date, there have been no head-to-head trials comparing the efficacy and tolerability of SC peginterferon beta-1a with SC IFN beta-1a in patients with RRMS. While head-to-head randomized clinical trials remain the gold standard for comparative effectiveness research, indirect methods for comparing treatment effects can provide useful evidence (
Comparative effectiveness without head-to-head trials: a method for matching-adjusted indirect comparisons applied to psoriasis treatment with adalimumab or etanercept.
). Indirect comparison approaches (Fig. 1) use an evidence base of clinical trials that individually do not compare all treatment options, but can estimate the comparative effects of multiple treatment options (
Indirect comparisons are statistical methods used to synthesize publicly available evidence and compare therapies using results from published randomized controlled trials. These are typically utilized in the absence of direct head-to-head evidence to generate data for comparative effectiveness and cost-effectiveness modelling. The most widely utilized methods for indirect comparison use aggregate data, such as from published clinical trials, and use common comparator arms to adjust for any differences between trials. This approach, first introduced by Bucher in 1997, allows data to be incorporated from multiple trials per treatment (
). An extension of the Bucher method to simultaneously compare multiple treatments linked through a network of indirect comparisons is known as a mixed treatment comparison (or network meta-analysis;
Comparative effectiveness without head-to-head trials: a method for matching-adjusted indirect comparisons applied to psoriasis treatment with adalimumab or etanercept.
A network meta-analysis of efficacy and evaluation of safety of subcutaneous pegylated interferon beta-1a versus other injectable therapies for the treatment of relapsing-remitting multiple sclerosis.
). Some limitations of using aggregate data can be mitigated by using individual patient data from clinical trials for some treatments and aggregate data for the remaining treatments—the matching-adjusted comparison method or re-weight matching-adjusted method (
Comparative effectiveness without head-to-head trials: a method for matching-adjusted indirect comparisons applied to psoriasis treatment with adalimumab or etanercept.
). This is the method we have used in the current analysis. If individual patient data are available for all clinical trials of the various treatments being compared, potential bias due to cross-trial differences can be addressed using propensity scoring; however, individual patient data are seldom available for all clinical trials (
We describe in this analysis how the matching-adjusted comparison approach was used to determine the efficacy at 2 years of SC peginterferon beta-1a versus SC IFN beta-1a in patients with RRMS.
2. Materials and methods
Individual patient data from a phase 3 study of SC peginterferon beta-1a 125 mcg every two weeks (ADVANCE) were compared with pooled summary data from four published studies of SC IFN beta-1a 44 mcg three times per week (CARE-MS I and II, OPERA I and II) with similar patient populations (
CARE-MS I investigators. Alemtuzumab versus interferon beta 1a as first-line treatment for patients with relapsing-remitting multiple sclerosis: a randomised controlled phase 3 trial.
CARE-MS II investigators. Alemtuzumab for patients with relapsing multiple sclerosis after disease-modifying therapy: a randomised controlled phase 3 trial.
). Two other clinical trials with SC IFN beta-1a, PRISMS and TENERE, were excluded from the analyses. While PRISMS is the pivotal phase 3 study for SC IFN beta-1a, the study is outdated, enrolled patients had more severe baseline disease characteristics, and the medical practice (e.g. diagnostic criteria, standard care, patient life style, etc.) is different from more recent studies (
PRISMS (Prevention of Relapses and Disability by Interferon beta-1a Subcutaneously in Multiple Sclerosis) Study Group, 1998. Randomised double-blind placebo-controlled study of interferon beta-1a in relapsing/remitting multiple sclerosis. Lancet;352(9139):1498–1504.
A matching-adjusted comparison was conducted by weighting individual peginterferon beta-1a–treated patients, using estimated propensity of enrolling in SC IFN beta-1a treatment to match multiple key aggregate baseline characteristics of SC IFN beta-1a–treated patients (
Comparative effectiveness without head-to-head trials: a method for matching-adjusted indirect comparisons applied to psoriasis treatment with adalimumab or etanercept.
) that may have an impact to response variables (e.g. ARR, CDW, and NEDA). These baseline characteristics included age, gender, time since MS symptom onset, EDSS score prior to treatment initiation, and the number of relapses in the year prior to the study. After matching, weighted ARR and weighted 24-week CDW for peginterferon beta-1a were compared with average outcomes from the four pooled SC IFN beta-1a studies.
Protocol-defined objective relapses were used for peginterferon beta-1a. ARR for peginterferon beta-1a was analyzed with a negative binomial regression model adjusted for baseline EDSS score (< 4 vs. ≥ 4), baseline ARR (the number of relapses in the 3 years prior to study entry divided by 3), and age (< 40 vs. ≥ 40). The proportion of patients with 24-week CDW at 2 years was compared using the Rao-Scott chi-square test. In addition, clinical-NEDA, defined as no relapses and no onset of 24-week CDW, for patients receiving peginterferon beta-1a, was compared with pooled data from the studies in which it was available, and analyzed using the Rao-Scott chi-square test (
CARE-MS I investigators. Alemtuzumab versus interferon beta 1a as first-line treatment for patients with relapsing-remitting multiple sclerosis: a randomised controlled phase 3 trial.
), allowing a subanalysis examining outcomes with SC peginterferon beta-1a and SC IFN beta-1a in treatment-naïve patients using individual patient data from ADVANCE and aggregate data from CARE-MS I. A matching-adjusted comparison was conducted, with the intent-to-treat population from ADVANCE who met the inclusion-exclusion criteria in CARE-MS I weighted such that their average baseline characteristics matched those for patients in the CARE-MS I trial. After matching for baseline characteristics, efficacy outcomes for patients treated with peginterferon beta-1a were weighted based on the model that was used for SC IFN beta-1a in the CARE-MS I trial. Results were compared with summary efficacy outcomes for SC IFN beta-1a.
3. Results
3.1 Study patient characteristics
Baseline characteristics before matching were relatively similar across the four studies of SC IFN beta-1a (
CARE-MS I investigators. Alemtuzumab versus interferon beta 1a as first-line treatment for patients with relapsing-remitting multiple sclerosis: a randomised controlled phase 3 trial.
CARE-MS II investigators. Alemtuzumab for patients with relapsing multiple sclerosis after disease-modifying therapy: a randomised controlled phase 3 trial.
). After matching, key baseline characteristic values for the peginterferon beta-1a group, including age, gender, years since MS symptom onset, EDSS score, and number of relapses in the previous 12 months, were identical to the IFN beta-1a group (Table 1).
Table 1Baseline characteristics of SC peginterferon beta-1a and SC IFN beta-1a before and after matching.
Peginterferon beta-1a
IFN beta-1a from pooled studies (n = 1218)
Before matching (n = 512)
After matching (n = 376a)
Age, years
36.9 (9.8)
36.3 (9.1)
36.3 (9.1)
Female, %
71
66
66
Time from onset of MS symptoms, years
6.9 (6.6)
5.5 (5.4)
5.5 (5.4)
EDSS score
2.47 (1.25)
2.66 (1.29)
2.66 (1.29)
Number of relapses in previous 12 months
1.6 (0.7)
1.4 (0.7)
1.4 (0.7)
Values Female, % are mean (SD).
EDSS = Expanded Disability Status Scale; IFN = interferon; SC = subcutaneous; SD = standard deviation; aEffective n.
Over 2 years of treatment, ARR was significantly lower in patients receiving peginterferon beta-1a compared with patients who received SC IFN beta-1a before matching (0.24 vs. 0.34; P = 0.0317) but only numerically lower after matching (0.26 vs. 0.34; P = 0.0901; Fig. 2). However, the percentage of patients who were relapse free over 2 years was significantly higher among those treated with peginterferon beta-1a than among those treated with SC IFN beta-1a both before matching (74.4% vs. 57.5%, P < 0.0001) and after matching (75.1% vs. 57.5%, P < 0.0001; Fig. 3).
Fig. 2Comparison of ARR over 2 years with SC peginterferon beta-1a and SC IFN beta-1a before and after matching baseline characteristics. ARR = annualized relapse rate; CI = confidence interval; IFN = interferon; RR = rate ratio; SC = subcutaneous. * Effective n.
Fig. 3Comparison of the percentages of patients who were relapse free over 2 years with SC peginterferon beta-1a and SC IFN beta-1a before and after matching baseline characteristics. IFN = interferon; SC = subcutaneous. * Effective n.
At 2 years, the proportion of patients with 24-week CDW was significantly lower in patients receiving peginterferon beta-1a compared with those receiving SC IFN beta-1a both before (6.6% vs. 13.2%; P < 0.0001) and after (6.5% vs. 13.2%; P = 0.0007) matching (Fig. 4).
Fig. 4Comparison of 24-week CDW over 2 years with SC peginterferon beta-1a and SC IFN beta 1a before and after matching baseline characteristics. CDW = confirmed disability worsening; IFN = interferon; SC = subcutaneous. * Effective n.
For the assessment of clinical-NEDA, only data from CARE-MS I and II could be used because OPERA I and II did not report clinical-NEDA separately from overall NEDA (which included MRI assessment). Therefore, for the analysis of clinical-NEDA, peginterferon beta-1a–treated patients were weighted to match the aggregate baseline characteristics of SC IFN beta-1a–treated patients in the CARE-MS I and II studies. After matching, the key baseline characteristic values for the peginterferon beta-1a group (age, gender, years from MS symptom onset, EDSS score, and number of relapses in previous 12 months) were identical to the IFN beta-1a group (data not shown). A significantly higher proportion of peginterferon beta-1a–treated patients than SC IFN beta-1a–treated patients achieved clinical-NEDA both before (72.5% vs. 48.1%; P < 0.0001) and after (74.1% vs. 48.1%; P < 0.0001) matching (Fig. 5).
Fig. 5Summary of clinical-NEDA analysis over 2 years’ treatment with SC peginterferon beta-1a or SC IFN beta-1a. IFN = interferon; NEDA = no evidence of disease activity; SC = subcutaneous. * Effective n.
In the subanalysis evaluating only treatment-naive patients, some imbalance in baseline characteristics between patients randomized to peginterferon beta-1a in ADVANCE and those randomized to SC IFN beta-1a in CARE-MS I was observed before matching. After matching, all baseline characteristics were balanced across the SC peginterferon beta-1a (effective sample size, n = 193) and SC IFN beta-1a (n = 187) treatment groups.
In treatment-naive patients, ARR at 2 years was significantly lower before matching for patients treated with SC peginterferon beta-1a than for those treated with SC IFN beta-1a (0.21 vs. 0.39; P = 0.0013). ARR remained significantly lower after matching with SC peginterferon beta-1a than with SC IFN beta-1a (0.14 vs. 0.39; P < 0.0001; Fig. 6). Similarly, at 2 years there were significant differences in the proportion of patients with 24-week CDW (4.9% vs. 10.7%; P = 0.030) and the proportion of patients with clinical-NEDA (75.1% vs. 55.6%; P = 0.0001), both favoring SC peginterferon beta-1a (Fig. 6).
Fig. 6Summary of clinical outcomes in treatment-naïve patients after 2 years treatment with SC peginterferon beta-1a or SC IFN beta-1a. CDW = confirmed disability worsening; CI = confidence interval; IFN = interferon; NEDA = no evidence of disease activity; RR = rate ratio; SC = subcutaneous. * Effective n.
This matching-adjusted comparison demonstrated that patients with RRMS treated with SC peginterferon beta-1a 125 mcg every two weeks achieved better clinical outcomes than patients who received SC IFN beta-1a 44 mcg administered three times per week. This is in agreement with data from a network meta-analysis that showed a similar or numerically improved efficacy profile for SC peginterferon beta-1a when compared with other first-line DMTs in patients with RRMS (
A network meta-analysis of efficacy and evaluation of safety of subcutaneous pegylated interferon beta-1a versus other injectable therapies for the treatment of relapsing-remitting multiple sclerosis.
). Furthermore, an additional network meta-analysis found greater improvements in ARR and disability worsening at 6 and 9 months with SC peginterferon beta-1a 125 mcg compared with SC IFN beta-1a 44 mcg (
Comparative efficacy of disease-modifying therapies for patients with relapsing remitting multiple sclerosis: systematic review and network meta-analysis.
Data from the COMPARE study, which investigated the pharmacokinetic and tolerability differences between SC peginterferon beta-1a once every two weeks and SC IFN beta-1a three times per week, demonstrated that drug exposure over two weeks was 60% higher with peginterferon beta-1a (
). An exposure-response model using data from ADVANCE found that the superior efficacy of peginterferon beta-1a every two weeks compared with every four weeks, as measured by ARR, was a result of greater exposure in the every-two-weeks group (Hu et al., 2017). These data support the possibility that the improvement in clinical outcome observed in the current analysis is related to the increased exposure seen with SC peginterferon beta-1a compared with IFN beta-1a.
The current study used a matching-adjusted comparison to compare the efficacy of SC peginterferon beta-1a and SC IFN beta-1a. The method balances observed baseline characteristics, is robust to model misspecification, and has been used to compare drugs in previous studies (
Comparative efficacy of nilotinib and dasatinib in newly diagnosed chronic myeloid leukemia: a matching-adjusted indirect comparison of randomized trials.
Comparative efficacy of vildagliptin and sitagliptin in Japanese patients with type 2 diabetes mellitus: a matching-adjusted indirect comparison of randomized trials.
Comparative effectiveness research using matching-adjusted indirect comparison: an application to treatment with guanfacine extended release or atomoxetine in children with attention-deficit/hyperactivity disorder and comorbid oppositional defiant disorder.
Fox R.J., Cutter G., Chan A., Xiao J., Okwuokenye M., Levison D., Lewin J.B., Edwards M.R., Marantz J.L., 2015. Comparative effectiveness using a matching-adjusted indirect comparison between delayed-release dimethyl fumarate and fingolimod for the treatment of relapsing-remitting multiple sclerosis. Poster presentation at the 31st Congress of the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS).
). In the absence of head-to-head randomized trials, matching-adjusted comparisons can provide timely and reliable comparative evidence even when very few trials are available for the treatments of interest. However, matching-adjusted comparisons are generally less powerful due to the use of aggregated data for the comparator (SC IFN beta-1a in this study) and due to the smaller sample sizes after matching. The results should be interpreted with caution, as there is still the potential for some residual bias despite the design of the analyses to minimize bias, mainly due to assumptions that the patient populations were sufficiently similar for pooling as aggregate data. Thus, matching-adjusted comparisons should be considered as observational studies. Another limitation of this method is inherent biases from included patient populations. For example, the requirement in some studies that patients be nonresponding would bias the results towards patients less responsive to given therapies. Because patients are weighted based on select baseline characteristics, the interpretation of the results after matching should be limited to patients whose baseline profile is similar to that of the analyzed patients. In addition, the exclusion of historic studies, while reducing heterogeneity in the populations, may be a source of selection bias. Finally, due to variability in adverse event reporting across trials, the safety of SC peginterferon beta-1a and SC IFN beta-1a was not compared in this study.
5. Conclusions
This matching-adjusted comparison using individual patient data from a phase 3 study of SC peginterferon beta-1a 125 mcg every two weeks and pooled data from four phase 3 trials of SC IFN beta-1a 44 mcg three times per week indicates better clinical outcomes are achieved with SC peginterferon beta-1a in patients with RRMS. This is in agreement with previous evidence indicating SC peginterferon beta-1a is associated with improved clinical, pharmacokinetic, and tolerability outcomes compared with SC IFN beta-1a. Furthermore, the reduced dosing schedule with SC peginterferon beta-1a makes it a convenient alternative to more frequently injected DMTs.
Acknowledgments
The authors wish to thank the patients who volunteered for this study and the many site staff members who helped to conduct the study. The authors were assisted in the preparation of the manuscript by Jenna Steere and Nicholas White of CircleScience (New York, NY, USA). Writing and editorial support were funded by the study sponsor, Biogen.
All authors were involved in reviewing the manuscript critically for important intellectual content and provided final approval of the submitted version.
Disclosures and conflicts of interest
PKC receives consulting fees from Accordant, Acorda, Bayer, Biogen, Celgene, Genentech/Roche, Genzyme/Sanofi, Novartis, Serono, Teva; and research support from Actelion, Alkermes, Genentech/Roche, MedDay, NINDS, and Novartis. QD and SS were employees of Biogen at the time of the study and are Biogen stockholders. XZ and CC-V are employees and stockholders of Biogen.
Funding
This study was funded by Biogen (Cambridge, MA, USA).
Submission declaration
We confirm that the work described in this manuscript is consistent with the journal's position on ethical publication. The work has not been published previously (except in the form of an abstract), is not under consideration for publication elsewhere, is approved for publication by all authors and tacitly or explicitly by the responsible authorities where the work was carried out and, if accepted, will not be published elsewhere in the same form, in English or in any other language, including electronically, without the written consent of Multiple Sclerosis and Related Disorders.
References
Bucher H.C.
Guyatt G.H.
Griffith L.E.
Walter S.D.
The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials.
CARE-MS I investigators. Alemtuzumab versus interferon beta 1a as first-line treatment for patients with relapsing-remitting multiple sclerosis: a randomised controlled phase 3 trial.
CARE-MS II investigators. Alemtuzumab for patients with relapsing multiple sclerosis after disease-modifying therapy: a randomised controlled phase 3 trial.
European Medicines Agency, 2005. Guideline on clinical investigation of medicinal products for the treatment of multiple sclerosis. London, 15 September.
Comparative efficacy of disease-modifying therapies for patients with relapsing remitting multiple sclerosis: systematic review and network meta-analysis.
Fox R.J., Cutter G., Chan A., Xiao J., Okwuokenye M., Levison D., Lewin J.B., Edwards M.R., Marantz J.L., 2015. Comparative effectiveness using a matching-adjusted indirect comparison between delayed-release dimethyl fumarate and fingolimod for the treatment of relapsing-remitting multiple sclerosis. Poster presentation at the 31st Congress of the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS).
Therapeutics and technology assessment subcommittee of the American academy of neurology and the MS council for clinical practice guidelines. Disease modifying therapies in multiple sclerosis: report of the therapeutics and technology assessment Subcommittee of the American Academy of Neurology and the MS Council for clinical practice Guidelines.
PRISMS (Prevention of Relapses and Disability by Interferon beta-1a Subcutaneously in Multiple Sclerosis) Study Group, 1998. Randomised double-blind placebo-controlled study of interferon beta-1a in relapsing/remitting multiple sclerosis. Lancet;352(9139):1498–1504.
Comparative effectiveness research using matching-adjusted indirect comparison: an application to treatment with guanfacine extended release or atomoxetine in children with attention-deficit/hyperactivity disorder and comorbid oppositional defiant disorder.
Comparative effectiveness without head-to-head trials: a method for matching-adjusted indirect comparisons applied to psoriasis treatment with adalimumab or etanercept.
Comparative efficacy of nilotinib and dasatinib in newly diagnosed chronic myeloid leukemia: a matching-adjusted indirect comparison of randomized trials.
Comparative efficacy of vildagliptin and sitagliptin in Japanese patients with type 2 diabetes mellitus: a matching-adjusted indirect comparison of randomized trials.
A network meta-analysis of efficacy and evaluation of safety of subcutaneous pegylated interferon beta-1a versus other injectable therapies for the treatment of relapsing-remitting multiple sclerosis.
30074-9&id=gr1.jpg){kind=link}
30074-9&id=gr2.jpg){kind=link}
30074-9&id=gr3.jpg){kind=link}
30074-9&id=gr4.jpg){kind=link}
30074-9&id=gr5.jpg){kind=link}
30074-9&id=gr6.jpg){kind=link}