Results of the IMAGINE-RA study
show no added benefit of using
magnetic resonance imaging as
part of a treat-to-target strategy for
At 2 years, similar percentages
of patients achieved the coprimary
endpoints of Disease Activity Score
in 28 joints using C-reactive protein
(DAS28-CRP) remission or no radiographic progression regardless of
whether MRI was used. Indeed, 85%
versus 88% (P = .958) of patients
achieved a DAS28-CRP of less than
2. 6, and 66% and 62% exhibited
no radiographic changes (P = .922)
with the MRI-guided or conventional treat-to-target strategies.
“Despite patients achieving a
target of clinical remission, we still
see erosive progression in about
20%-30%,” study investigator Dr.
Signe Møller-Bisgaard said at the
Congress. That’s regardless of the
definition of remission that you
use, she added.
Dr. Møller-Bisgaard, a resident
in rheumatology and postdoctoral
researcher who works at Rigshos-
pitalet and Frederiksberg Hospital
in Copenhagen, observed that both
synovial inflammation and bone
marrow oedema seen on MRI had
been shown to predict progression
in patients with rheumatoid arthritis.
What was not known, however,
was whether there was any value
in specifically targeting MRI remission in patients who had already
achieved clinical remission. This
is what the IMAGINE-RA study set
out to address. It was a 2-year trial
of 200 patients with rheumatoid
arthritis in clinical remission who
were recruited and randomised
to either an MRI or conventional
treat-to-target strategy. The study
involved nine rheumatology and
eight radiological departments, Dr.
The protocol for the study (
Trials. 2015;16:178) defined clinical
remission as a DAS28-CRP of 3. 2
or lower and no swollen joints.
Patients had to have erosions on
x-ray, be anti–cyclic citrullinated
peptide positive, and be treated
only with conventional synthetic
drugs (csDMARDs) at the time of
During the study, patients were
assessed every 4 months via the
DAS28 or DAS28 plus MRI of the
dominant hand and wrist, with radiographs of the hands and feet performed annually in both groups and
MRI also performed yearly in the
conventional treat-to-target group.
“Treatment was intensified in
both arms if the DAS28-CRP was
above 3. 2 and there was at least
one clinical swollen joint,” Dr.
Møller-Bisgaard explained. Treatment was also intensified in the
MRI group if bone marrow oedema
was observed. Treatment intensification involved maximal doses of
csDMARDs alone or in combinations and then addition of biologic
treatments, such as a tumour necrosis factor inhibitor.
“Targeting absence of MRI
bone marrow oedema in addition
to a conventional treat-to-target
strategy in RA patients in clinical
remission had no effect on the
probability of achieving DAS28-
CRP remission or halting radio-
graphic progression,” she said.
However, there were some pos-
itive effects on several predefined
secondary endpoints. For instance,
more patients in the MRI group
than in the conventional treat-to-
target group achieved American
College of Rheumatology/EULAR
remission (49% vs. 32%; P = .017).
There was a significant improve-
ment in the number of swollen
joints and a patient and physician
global assessment. “There was
also more improvement in HAQ
[Health Assessment Question-
naire], with a difference between
the groups of . 14 [P less than
.001],” Dr. Møller-Bisgaard reported.
The IMAGINE-RA study is fund-
ed by grants from the Danish
Rheumatism Association and the
Research Fund of Region Zealand.
Funding is also provided by AbbVie
via a nonrestricted grant, and adal-
imumab is provided free of charge.
Dr. Møller-Bisgaard and her coau-
thors had no personal conflicts of
interest to declare.
Dr. Signe Møller-Bisgaard
IMAGINE-RA: No need for MRI with treat-to-
BY SARA FREEMAN
“Targeting absence of MRI
bone marrow oedema in
addition to a conventional
in RA patients in clinical
remission had no effect
on the probability of