Psoriatic disease is a chronic inflammatory condition affecting the skin, the entheses, and the joints (1). The disease has a robust genetic basis linked to HLA class I alleles and the interleukin-23 (IL-23) receptor and is associated with the development of a specific immune pathology characterized by innate immune cell and T cell activation (1). Clinical studies have shown that, in addition to tumor necrosis factor, IL-17 and IL-23 are pivotally involved in the pathogenesis of psoriatic disease (2, 3). Experimental (4, 5) as well as clinical studies (6, 7) suggest that environmental factors such as mechanical load essentially contribute to the pathology of psoriatic disease (“mechanoinflammation”). Hence, psoriatic skin disease usually develops at mechanically exposed areas of the skin, and psoriatic plaque formation based on physical irritation of the skin (Koebner phenomenon) has been described (8). Similarly, entheseal sites, which typically become inflamed in psoriatic disease, are highly mechanically exposed tissues that are essential for the transduction of physical forces (9-12).
It is well-known that clinical manifestations of psoriatic skin disease usually precede joint disease. While this concept appears correct when considering clinically visible disease, it also has some shortcomings, since psoriatic skin is much easier to detect than joint disease. Given these considerations, skin and joint disease may indeed arise from the same underlying process (e.g., mechanoinflammation) at the same time, with patients developing similar inflammatory lesions in the skin and entheses, while the speed of detection of these lesions is different and allows skin changes to be detected earlier. Independent from this, stable monomorphic psoriatic disease manifestations exist, in which skin disease indeed remains the only manifestation of the disease (most frequent), or isolated joint, or even entheseal, disease can occur. Conceptually, however, skin and entheseal manifestations of psoriatic disease may arise simultaneously, at least in a subset of patients, but may be recognized at different times.
In support of this concept, several studies have suggested that subclinical inflammatory lesions can be detected in the joints and the entheses of a subset of psoriasis patients without signs of clinical arthritis (13-16). Notably, discrete structural entheseal lesions, sometimes also referred to as “deep Koebner” phenomenon, are a key finding in the joints of patients with psoriasis, suggesting that similar mechanoinflammation-induced tissue responses can be found in the skin and the joints in psoriasis patients (17). Structural entheseal lesions emerge from resident tissue responses, are similar to psoriatic plaques at entheseal sites, and are characterized by periosteal proliferation and new bone formation (18).
While structural entheseal lesions in patients with psoriasis may represent the first sign of musculoskeletal involvement of psoriatic disease, it is unclear whether patients with such lesions face a higher risk of developing psoriatic arthritis (PsA). To test whether structural entheseal lesions are an indicator of the later development of PsA, we assessed a cohort of patients with psoriasis for the presence or absence of structural entheseal lesions and other signs of bone changes in the hand joints and prospectively followed up this cohort for later development of PsA. We hypothesized that the risk of developing PsA is higher in patients who exhibit structural entheseal lesions, indicating that the clinical manifestation of psoriatic disease is already determined early on in the disease process, i.e., before the development of clinical joint disease.
DISCUSSIONOur data show that the presence of structural entheseal lesions in patients with psoriasis represents a robust and independent marker for the later development of PsA. These findings support the concept of mechanoinflammation in the development of PsA, with the initial development of entheseal lesions in the disease process (5). Structural entheseal lesions are highly typical of psoriatic joint disease and part of the CASPAR criteria for the classification of PsA (23). They are anatomically different from bony spurs observed in osteoarthritis (26) and progressively increase during the course of PsA (34). Even more importantly, such lesions can already be observed in a subset of psoriasis patients without PsA, suggesting that the initial insult to joints in patients with psoriatic disease may be enthesitis associated with localized tissue responses (17). As such, these lesions appear to represent a very early feature of joint involvement in psoriatic disease and are thus found to be associated with the later development of signs and symptoms of PsA.
The observation of localized reduction in bone mass at the entheseal region of the joints in patients with psoriasis progressing to PsA is another notable finding of this study. This link between BMD and progression to PsA is specific to the entheseal region and is not apparent in the intraarticular region of the joint. After multiple imputation, the association of decreased cortical vBMD with progression to PsA remained a robust finding. Furthermore, the concomitant presence of structural entheseal lesions and low cortical vBMD substantially increased the risk of developing PsA, suggesting a state of imminent PsA. These results provide evidence of early localized cortical bone changes in psoriasis patients who progress to PsA and translate preclinical observations showing that inflammatory cytokines involved in psoriatic disease induce a negative net balance of bone homeostasis (35, 36).
Entheses insert into the cortical bone, which is strongly vascularized (37) with a multitude of bone channels (“transcortical vessels”) shuttling immune cells in and out of the bone marrow and harboring bone-resorbing osteoclasts (38, 39). Inflammation along these transcortical vessels may activate osteoclasts and thereby trigger bone loss. This concept is supported by 1) the observation of subclinical inflammatory changes in the joints of patients with psoriasis (15), 2) IL-17 dependency of such preclinical inflammatory changes predominantly affecting entheseal segments (40), and 3) the presence of localized periarticular bone loss already in psoriasis, while systemic bone loss is confined to patients with PsA (41).
By observing structural changes in the joints of psoriasis patients that are associated with later development of PsA, this study highlights the early phase of musculoskeletal involvement in a subset of patients with psoriatic disease. These data support the importance of the concept of early disease interception (42) to prevent damage, disability, and poor outcomes in patients with PsA (43, 44). Since structural joint changes occur very early in a subgroup of psoriasis patients, the benefit of systemic treatment may be based on the prevention of progression to more severe and clinically overt joint disease. It is particularly important to consider that the progression to PsA in an unselected psoriasis population is rather limited, and hence the identification of an at-risk population is of importance (45).
One intriguing finding of our study was the lack of a positive association between PsA and nail involvement. Nail dystrophy is among the CASPAR criteria (18) for the classification of PsA and was associated with CASPAR-defined PsA in previous prospective studies (45, 46). This association, however, is necessitated by design, such that when any 2 individuals with a given clinical finding suggesting PsA are compared, one that has nail dystrophy is more likely to be classified as having PsA by the CASPAR criteria, compared to one that does not. Interestingly, in one of those observational studies, the nail dystrophy association did not persist when psoriasis severity was included in a multivariable model (46), suggesting that nail dystrophy is likely a correlate of psoriasis severity, which could be considered an intermediate along the causal pathway from psoriasis to PsA. Including psoriasis severity as such an intermediate in our causal models did not necessitate any further adjustment to our estimates. We consider the lack of this association, which we had also previously observed, to be a result of sampling variation (15).
A limitation of the study is the rather small number of patients who did develop PsA during the observation period relative to the number of adjustment variables. This might impact the precision of our effect estimates. Another limitation is the potential overestimation of the risk of PsA explained by exposure. It is therefore necessary that our observations are confirmed by others using similar methods. In addition, there was some data loss due to the meticulous quality standards required to reliably measure volumetric bone densities using high-resolution peripheral QCT. We tried to address this issue using multiple imputation to test the robustness of our findings to such missing data and thereby verified the association between PsA risk and cortical vBMD at entheseal segments. Finally, our study sample consisted of patients conditionally selected among a psoriasis population attending a tertiary dermatology center, and as can be inferred from the high incidence of PsA, does not necessarily reflect an overall psoriasis population. Since our analyses are causal inference oriented, the generalizability and predictive potential of our results for daily practice should be considered limited. Our findings, however, can be considered among potential predictors in a PsA risk prediction exercise.
In summary, our findings show that specific structural features of the joints in psoriasis patients, namely, structural entheseal lesions as well as cortical bone loss at entheseal segments, are associated with the later development of PsA. These findings not only substantiate the concept of mechanoinflammation in the pathogenesis of psoriatic disease but also show that the presence of entheseal lesions predicts the onset of clinical joint disease. Based on these data, interventions with high efficacy in controlling entheseal inflammation appear to be a particularly valuable strategy for interfering with the onset of PsA in patients with psoriatic disease.
ACKNOWLEDGMENTWe thank the staff of Siemens Healthineers for creating images with Cinematic Rendering (Figure 1 and Supplementary Video 1).
AUTHOR CONTRIBUTIONSAll authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Schett had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and designSimon, Tascilar, Kleyer, Rech, Sticherling, Schett.
Acquisition of dataSimon, Kleyer, Bayat, Kampylafka, Sokolova, Zekovic, Hueber, Rech, Schuster, Sticherling.
Analysis and interpretation of dataSimon, Tascilar, Kleyer, Bayat, Hueber, Rech, Engel, Sticherling, Schett.
ADDITIONAL DISCLOSURESAuthor Engel is an employee of Siemens Healthineers.
Comments (0)