IQWIG DECISION (translated version)
Dossier evaluation A14-14
Version 1.0
Dimethyl fumarate - benefit assessment pursuant to § 35a SGB V
07/30/2014
Institute for Quality and Efficiency in Health Care (IQWiG)
- 3 -
2 benefit assessment
2.1 Summary of the benefit assessment
Background
The G-BA, IQWiG has in accordance with the benefit assessment of the active substance dimethyl fumarate
§ 35a SGB V commissioned. The evaluation was made on the basis of a dossier of
Marketing Authorisation Holder (pU). The dossier was IQWiG on 30/04/2014
transmitted.
Question
The aim of this report is to assess the added value of
Dimethyl fumarate (DMF) at in comparison to the appropriate comparator therapy
adult patients with relapsing-remitting multiple sclerosis (relapsing
remitting multiple sclerosis [RRMS]).
The G-BA has for this application, the appropriate comparator therapy as follows
set: Beta-interferon (1a or 1b) or glatiramer acetate (GA).
The pU selects interferon beta-1a (IFN β-1a) under the conditions specified by the G-BA alternatives
from, but his choice is limited to IFN β-1a, 44 mcg subcutaneously (sc) (Rebif), one of the
Preparations containing this substance. Are as defined by the G-BA on drug level
However, all IFN β-1a preparations to be considered independently of the application form,
ie in addition also another product with this ingredient - IFN β-1a,
30 micrograms intramuscularly (im) (Avonex). The study pool of pU for direct comparison
is not affected by this procedure (no direct comparative studies
available). However, as a result of the limitation of the comparative treatment of pU sets a
content of incomplete indirect comparison for comparison of DMF compared to the
appropriate comparator therapy (IFN β-1a) ago.
The present benefit assessment was compared with the appropriate comparator therapy
IFN β-1a performed.
The evaluation was made with respect to patient-relevant endpoints.
Results
Direct Comparison
There are no direct comparative studies of DMF compared to the appropriate
Comparative therapy IFN β-1a ago.
Page 2
Dossier evaluation A14-14
Version 1.0
Dimethyl fumarate - benefit assessment pursuant to § 35a SGB V
07/30/2014
Institute for Quality and Efficiency in Health Care (IQWiG)
- 4 -
Indirect comparison
The pU sets over a network meta-analysis for indirect comparison of DMF
IFN β-1a, 44 mcg sc (Rebif) in Module 4 of the dossier before. For the pU researched a
Network of DMF, IFN β-1a (sc and im), IFN β-1b, GA and placebo. In the network
Meta-analysis, a total of 14 studies were included. This study pool contained
Treatment arms with DMF, IFN β-1a, IFN β-1b, GA and placebo. The various preparations-
rate with the active IFN β-1a (sc [Rebif] and [Avonex]) and its possible
Doses (44 mcg or 22 mcg sc [Rebif]) were considered separately in the network.
The indirect comparison is presented for statements about the additional benefit of DMF over
IFN β-1a for the following reasons unsuitable:
§ The indirect comparison is incomplete in content.
The statistical model used § network meta-analysis is not suitable.
§ The three basic assumptions of network meta-analysis - Similarity,
Homogeneity and consistency - have not been reviewed by the pU adequately. Beyond
is the similarity of the included studies doubtful.
Indirect comparison of content incomplete
Although the network comparing DMF compared to the appropriate comparative
therapy principle in its entirety enables (IFN β-1a in all application
form), presents the pU in Module 4 of the dossier only results for comparison
of DMF to IFN β-1a, 44 mcg sc (Rebif) and thus forms the expedient
Comparative therapy (IFN β-1a) from only partially. The presented indirect comparison is therefore
content incomplete.
Network meta-analyzes are based on an inappropriate statistical model
The network meta-analyzes were mixed on the basis of generalized linear
Models (generalized linear mixed models [GLMMs]) performed. It was in the
GLMMs, the treatment effect presented as a fixed effect and the study effect as
random effect models. The modeling of the study as a random effect can lead to a
lead cross-level bias (also called ecological bias) and to an underestimation
Standard errors lead, so much so that they are smaller than in a meta-
analytical model with only fixed effects. The presented network meta-
Analysis thus not based on an adequate statistical model.
Unsuitable review of similarity, uniformity and consistency
The three basic assumptions of network meta-analysis - similarity, homogeneity
and consistency - have not been reviewed by the pU adequately.
To check the similarity assumption of pU compares the methodology of the studies and
the characteristics of the patient populations of the included studies qualitatively
Page 3
Dossier evaluation A14-14
Version 1.0
Dimethyl fumarate - benefit assessment pursuant to § 35a SGB V
07/30/2014
Institute for Quality and Efficiency in Health Care (IQWiG)
- 5 -
each other. From this consideration of pU infers that the indirect comparison in the
included studies essentially comparable study populations (as well as a
have comparable methodology). This assessment is not followed. For example, speaks
large span of the proportions of patients with at least a boost in the
Placebo arms of the studies (39% to 84%) against a similarity of
Studies. In addition, on the basis of patient characteristics, pretreatment,
Disease severity or duration and previous relapse activity not of a sufficient
Similarity of the included in the network meta-analysis study populations
be assumed. The similarity assumption is therefore contrary to the assessment of pU
upset.
Homogeneity was the pU using correlation and regression analyzes to
Identify potential effect modifiers tested. This approach is inappropriate. In
Furthermore, the verification of the consistency assumption was not appropriate because there is no
Were mentioned criteria for the violation of the assumptions and consistency checking
was only for 2 selected comparisons and thus performed incomplete.
Executive Summary
There are no adequate data to assess the added value of DMF compared to the
appropriate comparator therapy before, neither for direct nor an indirect
Comparison.
Likelihood and extent of added benefit, patient groups with
therapeutically bedeutsamem additional benefits
Based on the results shown, the magnitude and probability of be
Additional benefits of the active ingredient Dimethyl fumarate compared to the appropriate
Comparative therapy valued as follows:
Table 2: dimethyl fumarate - extent and probability of the value added
Field of application
Expedient
Comparative therapy
a
Extent and
Probability of
Added value
Adult patients with
relapsing-remitting
Multiple sclerosis
Interferon beta (1a or 1b) or
Glatiramer acetate
Additional benefits not covered
a: In each case, the set of G-BA appropriate comparator therapy. In the cases where
pU due to the determination of the appropriate comparator therapy by the G-BA of several
Alternatives, a comparison therapy can select the appropriate selection of pU is marked in bold. In the
the present case, the pU the appropriate comparator therapy to beta interferon 1a 44 mcg sc (Rebif)
restricted. This restriction is not followed.
G-BA: Federal Joint Committee; pU: pharmaceutical company; sc: subcutaneous
The procedure for deriving an overall statement for additional benefit is a proposal
IQWiG dar. About the added value decided by the G-BA.