An early start of a disease-modifying therapy and achieving early remission are two of the basic principles of the therapy concept for RA (4). Early treatment significantly improves the long-term outcome and is therefore an important principle of the EULAR recommendations (4). The treat-to-target (T2T) strategy has become almost universally established in order to make therapy more efficient (25). A clear therapeutic goal, usually remission, is pursued. If remission or at least low disease activity cannot be archived, the patient should be escalated according to EULAR guidelines at an early stage regarding the T2T concept (25). The introduction of biologic (b) and recently targeted synthetic (ts) DMARDs has opened up new therapeutic possibilities beyond conventional synthetic (cs) DMARD therapy, in particular for patients who have not responded to the conventional drugs or have responded only insufficiently (26).
In this double-blinded, placebo-controlled, randomized clinical trial, the effect of Adalimumab in combination with MTX vs. MTX monotherapy on joint cartilage integrity of MCP2 and MCP3 in patients with early RA was examined with compositional MRI of the cartilage by using dGEMRIC (27). To substantiate our results, further parameters describing disease activity were assessed, including cartilage thickness, the RAMRIS, specific bio- respectively serum cartilage markers and clinical and functional outcome. To the best of our knowledge this is the first randomized trial evaluating the effect on cartilage of Adalimumab therapy plus MTX versus MTX monotherapy using dGEMRIC and high-field MRI.
Our most important finding was that Adalimumab plus MTX therapy leads to a significant increase of dGEMRIC values over time. In contrast to the Adalimumab group, there was no significant increase of dGEMRIC values in the placebo group, which indicates that controlling inflammation with Adalimumab leads to an improvement of cartilage quality. These results are in line with Beals et al., who documented reduced cartilage damage under bDMARD therapy using Infliximab (28). But the results of this study thus point to an even clearer positive effect and indicate that even regeneration of cartilage composition seems to be possible under Adalimumab treatment. The slightly different results between our study and the results of Beals et al. could occur due to the different bDMARDs used or to the different cartilage measurement MRI techniques (DCE-MRI vs. dGEMRIC). Moreover, it has to be mentioned that the study populations differ since the CAR-ERA trial aimed at very early, therapy naïve RA patients. However, both studies support the notion of a positive effect of TNF-alpha inhibitors on cartilage integrity in RA. The results of this study are in contrast to the only further study that investigated the effect of bDMARDs on cartilage using dGEMRIC. Tiderius et al. investigated knee joints in 7 chronic RA patients and found cartilage deterioration after 22 weeks (29). A possible explanation is that patients with early – not established - RA respond better to bDMARDs than patients at a chronic stage of the disease, which in turn supports the current treatment principle of early and T2T therapy. Another potential explanation is a varying response in different joints. This theory can be supported by findings that point out a difference of cartilage regeneration between knee and ankle joints. Kuettner and Cole showed that in response to damage, knee chondrocytes synthesize proteoglycans at a lower rate than in ankle cartilage chondrocytes, which suggests a lower capacity for repair and regeneration (30). As expected, the results of Kuettner and Cole indicated an increased degree of cartilage surface disorder with increasing body weight. In our cohort, however, there were no significant differences with respect to body weight.
Recently, the suitability of MRI for therapy monitoring was critically questioned based on the control parameter bone edema. A superiority could not be demonstrated for the MRI-guided treat-to-target group compared to the conventionally guided group (31). In this study, the focus was explicitly placed on the compositional cartilage quality measurable with dGEMRIC, since compositional cartilage imaging is expected to be able to detect cartilage damage at an early stage (14), which in turn may be a sign of progressive RA (12). Therefore, we consider MRI to be a valuable diagnostic tool. However, cartilage changes may be a more sensitive control parameter than bone marrow edema.
This study had several strengths. It is a double-blinded, placebo-controlled, randomized clinical trial and the first published report of compositional MRI’s ability to point out cartilage regeneration in RA. This is significant since the exclusion of the assessment of cartilage in previous clinical trials was an obstacle to the acceptance of MRI as a substitute for radiography (32). However, the use of MRI is preferable as previous studies have shown that cartilage loss is at least as relevant to long-term outcome as bone erosion (32). One of the further strengths of this study at once is the use of the dGEMRIC for cartilage visualization. Until recently, the MRI technique dGEMRIC was considered the reference standard to evaluate extracellular matrix components of hyaline cartilage (17). For dGEMRIC the application of gadolinium-based contrast agent is obligatory. Due to new indications of gadolinium ablation in the brain and a subsequent recommendation of the European Medicines Agency (EMA) to discontinue the use of linear contrast agents based on gadolinium, subject safety was prioritized, and the study was stopped prematurely (24). A limitation therefore arises due to the discontinuation of the study in the unequal group size and in the relatively small number of control subjects. Despite a reduced number of subjects, the results, which were collected until the end of the study, are highly valuable.
Further limitations of this study include manually selected ROIs for dGEMRIC measurement in the MCP finger joints instead of automatically algorithm-based ROIs. Due to the small size of the MCP joint and the resulting low contrast, automated cartilage recognition is not yet reliable enough to be used in this study. Manual evaluation by an experienced radiologist was therefore preferred, despite the possible human interference factor. For future studies, the increased automated evaluation of clinical image data should be pursued.