| Summary: | Psoriatic arthritis (PsA) is a chronic inflammatory disease, integrated in the spondylarthritis (SpA) nosologic entity, that exhibits marked phenotypic heterogeneity and complex underlying inflammatory pathways. Tumor necrosis factor (TNF) is the most central inflammatory cytokine implicated in PsA articular and extra-articular manifestations, and TNF inhibitors (TNFi) the most successful therapeutic intervention breakthrough in the last decades for the treatment of this debilitating disease. Despite the remarkable benefits of TNFi for the management of PsA patients, optimal longterm control of disease activity is difficult to achieve at the population level. Furthermore, due to PsA heterogenous phenotypes, some manifestations have been scarcely studied, leaving uncovered evidence gaps for dactylitis, enthesitis and axial PsA symptoms. In this dissertation, we sought to provide new perspectives on TNFi for the treatment algorithm of PsA. Our specific aims were: 1) to describe TNFi positioning in PsA treatment, through a literature review and national rheumatologists consensus; 2) to identify predictors of TNFi’ effectiveness (persistence and response) in PsA patients, using realworld data from the Rheumatic Diseases Portuguese Register (Reuma.pt); 3) to provide high level evidence on TNFi’ efficacy for the treatment of PsA dactylitis (and enthesitis) through an investigator-initiated multicentric randomized placebo-controlled trial; 4) to analyze the effect of TNFi on bone remodeling at the synovio-entheseal complexes using transmembrane TNF transgenic mice (TgA86). Encompassed by the first aim of this doctoral thesis, we performed a comprehensive literature review that included TNFi evidence of efficacy for PsA musculoskeletal manifestations: peripheral arthritis, axial disease, enthesitis and dactylitis; that was integrated in the 2015 update of the recommendations for the use of biologics diseasemodifying antirheumatic drugs (bDMARDs) in the treatment of PsA patients, endorsed by the Portuguese Society of Rheumatology. This review, followed by the assessment of consensus within Portuguese rheumatologists for the established recommendations, reinforced the need to identify clinical predictors of effectiveness that could help clinicians to best select candidates for TNFi in the context of recently approved bDMARDs (non-TNFi) for PsA, aiming at optimal disease control. Furthermore, the lack of consensus on dactylitis and enthesitis treatment indications and the absence of evidence of TNFi efficacy in these manifestations, assessed as primary endpoints in randomized controlled clinical trials, was identified as one of the most relevant knowledge gaps in PsA medical field. The Exchange PsA study addressed the second aim of this thesis and showed that PsA patients registered at Reuma.pt, starting a first TNFi between 2001 and 2017, experienced reduced effectiveness (drug survival and response) when switched within TNFi therapeutic class, for both peripheral and axial phenotypes. We also identified the female gender as a common unfavorable predictor of persistence and response to a first TNFi in this population. We further explored possible interactions and acknowledged that pain, but not disease activity at baseline, contributed to modulate the effect of gender on drug retention of the first-line TNFi, nevertheless gender remaining an independent predictor of discontinuation. The results from our work suggest that the poorer responses to a first-TNFi are not exclusively dependent on patient-reported outcomes. In fact, significantly higher baseline objective measures of peripheral disease activity were observed in females, and these responses were not modified by obesity or disease phenotype, supporting gender-dependent patterns of response. Obesity, a commonly associated PsA comorbidity, was further identified as a predictor of worse response to TNFi, suggesting that concomitant weight reduction plans for obese patients can be determinant to potentiate TNFi effectiveness. The results from the Exchange PsA study created awareness for the early identification of female PsA patients who suffer from delayed therapeutic (TNFi) initiation and are at risk of poor treatment survival and response to TNFi. In line with this, patients’ education combined with nutrition clinics access should be broadly implemented in daily practice and weight reduction programs included in PsA recommendations. The GO-DACT trial was developed under the scope of the third aim of the thesis and assessed dactylitis for the first time as the primary endpoint of a double-blind, randomized, placebo-controlled trial. The results from this multicentre investigator-initiated trial provided first time evidence that the combination of a TNFi (golimumab) and methotrexate (MTX) is superior to MTX monotherapy in improving PsA dactylitis, assessed both clinically and by high-resolution magnetic resonance imaging. Considering the high heterogeneity of outcome measures used to report dactylitis activity in PsA trials, the results from GO-DACT showed that the application of the innovative Dactylitis Severity Score (DSS) and Leeds Dactylitis Index (LDI) response indices (DSS20, 50, 70 and LDI20, 50,70), developed in the framework of this trial, consistently allowed discrimination between treatment arms. The good performance of these newly developed response indices supports their potential usefulness for future PsA trials and for standardization of PsA dactylitis outcomes. In light of the high disease burden of PsA dactylitis, including the risk of structural damage, the lower chance of achieving minimal disease activity and the early benefits from TNFi intervention, it seems reasonable to argue that PsA patients with active dactylitis could benefit from first line TNFi in combination with MTX. The fourth objective of this dissertation was centered on the effect of TNFi at the synovioentheseal complexes, in particular those related to local dichotomic bone remodeling disturbances observed in PsA pathogenesis (erosions and new bone formation). It required the identification of a mouse model developing SpA-like manifestations and, specifically, axial new bone formation. This was possible through a literature review on mouse models mimicking SpA characteristics, which lead to the selection of transmembrane TNF transgenic mice (TgA86) as the best option to address this research question. In an in vivo experimental setting, we were able to demonstrate that tmTNF-dependent pathways induce a SpA-like phenotype with peripheral and axial inflammation and deformity, and axial new bone formation in mice. Furthermore, we showed that the inhibition of tmTNF signaling, using a TNFi like mouse equivalent antibody, reduced not only inflammation but also new chondrogenesis/bone formation in the axial spine of tmTNF transgenic mice (TgA86), and this effect was more notorious in an early therapeutic intervention. Despite the limitations in showing a clear effect of TNFi in hampering syndesmophytes progression in human PsA/SpA, blocking new chondrogenesis/new bone formation in mice seems an attainable objective, especially in early disease interventions. With this doctoral thesis, we provide new evidence for the use of TNFi in clinical practice, which is likely to have implications for future guidelines and recommendations in the treatment of PsA patients.
|
|---|