4. Discussion
We performed a systematic review of all of the randomized clinical
trials evaluating use of VDZ as treatment for IBD which mentioned the
development of joint-related complaints. Our review encompassed large
trials in which close to 3,000 patients with both UC and CD with
evidence of active disease were assessed. While previous reports have
suggested that VDZ may induce new or worsening arthralgias [7-22],
our analysis did not identify concrete evidence of an association
between use of VDZ in IBD and the development of arthralgias.
Three of the trials included indeed showed an increased concentration of
arthralgias in patients with CD receiving VDZ rather than placebo as
part of the induction protocol [32,33] and while receiving a
maintenance dose [4]. Similarly, the other two studies showed that
there were increased arthralgias in patients with UC treated with VDZ
compared with placebo during the induction and maintenance phases
[34,3]. The OR’s comparing these two groups ranged from just over 1
to more than 10, in the case of a UC cohort receiving VDZ maintenance
treatment, highly suggestive of a trend toward a risk of arthralgias in
those treated with VDZ. However, because none of our analyses reached
statistical significance, these studies do not prove that there is a
causal link between use of VDZ and the development of joint pains.
Post-hoc analyses of the GEMINI trials offer mixed conclusions. On the
one hand, they found that when the participants of GEMINI 2 were
stratified into 3 groups, dividing participants into those who received
VDZ alone, placebo alone, and VDZ during induction followed by placebo
maintenance, they did indeed find that there was an increased incidence
in the development of arthritis/arthralgias in the 2 groups who received
VDZ during the trial compared with those received placebo throughout
[36]. On the other hand, they found that the composite endpoint of
both new and worsening arthritis/arthralgias was less likely to be met
in those who received VDZ and VDZ followed by placebo compared with
participants who received placebo alone.
Prior papers have suggested that the tapering of anti-TNF therapy in
conjunction with initiation of VDZ may largely explain the joint
complaints that develop in those receiving VDZ [37]. The post-hoc
analysis stratifying the patients with CD included in the GEMINI 2 trial
into those who had previously received anti-TNF treatment and those who
were naïve found that during the maintenance phase alone, there was
increased risk of developing arthralgias in those receiving VDZ compared
with those given placebo among those who had previously been exposed to
anti-TNF treatment [35]. These findings however, lacked statistical
significance and therefore only hint at the presence of a link. The
presence of many confounders with regard to disease activity, duration
of disease, and use of corticosteroids in those who had previously been
treated with anti-TNF agents, may explain the absence of a clearer
correlation.
Previous research has demonstrated that use of VDZ alters the
trafficking of gut-homing T lymphocytes, leaving them to continue
circulating in the periphery [38]. While signaling alterations
impact both effector and regulatory T-cell subsets, Th1 effector cells
are most specifically targeted by VDZ, preventing their entrance into
the GIT while the permitting the entry to other lymphocytic populations,
ultimately leading to greater immunoregulation within the
gastrointestinal mucosa [39]. At the same time, the resulting
imbalance within the systemic circulation does strongly suggest an
immunologic explanation for the subsets of patients who developed
arthralgias with VDZ treatment as seen in the trials reviewed in our
paper. Simultaneously, the development of arthralgias in smaller numbers
in those who were treated with placebo may atleast in part be related to
a parallel process taking place along the interwoven pathways of the
psycho-neuro-endocrine-immune axes [40].
Our study has several limitations. Firstly, our search revealed a small
number of RCTs which was heterogeneous in the included study
populations, dosing regimens, and both time-and data-points analyzed.
Another limitation is the lack of a statistical procedure for combining
numerical data (such as a meta-analysis) from the different studies due
to the limited number of studies and the differences in study
characteristics, particularly in sample size, dose regimen, and
co-medications. In addition, some of the studies also failed to
distinguish the incidence of joint complaints induced by VDZ usage
during induction and during maintenance and also did not reveal at which
time point new-onset arthralgias first manifested. In addition, two of
the studies, when enumerating the AEs, grouped together participants who
received placebo during the maintenance phase regardless of what they
had received for induction. Moreover, data was limited in terms of the
characteristics of those who developed arthralgias including baseline
disease, EIMs especially presence of an associated spondyloarthropathy,
other immunosuppressive treatments, and clinical response which may have
helped to stratify those at risk for developing arthralgias. Finally, in
these papers and in the supplementary appendices provided, no
information was provided regarding whether arthralgias led to drug
discontinuation and whether resolution was achieved following VDZ
cessation.
Despite these limitations, we were able to analyze 5 moderate to high
quality randomized controlled trials involving a large number of
patients with both CD and UC. We did not detect any statistically
significantly increased risk of arthralgias in those treated with VDZ
compared with placebo, indicating that there is no clear link. At the
same time, our analysis yielded a trend toward increased arthralgias in
those who received VDZ. These findings suggest that more studies are
needed which evaluate the development of arthralgias in VDZ-treated IBD
patients who are stratified according to baseline disease features and
response to VDZ. Moreover, additional studies measuring the responses of
different T-cell subsets together with evaluation of clinical and
histological parameters of the gut and the assessment of EIM, will
provide a clearer understanding of the relationship between the
immunologic shifts induced by VDZ and the possible generation of
arthralgias.