In a first of its kind study addressing the unmet need for biomarkers in the field, the study aimed to investigate protease-mediated markers of inflammation and tissue remodeling as biomarkers of flares in PsA patients. These biomarkers were measured in serum and SF from PsA patients experiencing flares defined as an acutely swollen and tender joint requiring intra-articular injection with corticosteroids. The serum levels were compared to controls and PsA patients without flares (who might have active disease, but were not experiencing flares that is any change in disease activity), and SF levels were compared to OA, a commonly used less inflammatory arthritic control.

In serum, levels of PRO-C3 and C3M, which are related to formation and degradation of the interstitial matrix, were found to be elevated in PsA with flares compared to controls and PsA without flare. The remodeling marker of the basement membrane, PRO-C4, was elevated in PsA-flare compared to PsA without flare. The inflammation and immune cell activity related markers, CRPM, VICM, and CPa9-HNE were highly elevated in PsA-flare patients compared to both controls and PsA without flare patients. Furthermore, PRO-C3 and CPa9-HNE showed best discriminatory performance for separating PsA patients with flares from PsA without flare patients. In the multivariable analysis, a combination of the biomarkers displayed superior value as a serum biomarker panel to distinguish between PsA-flare from PsA without flares, thus as a tool for monitoring flares and by extension, elevated disease activity in PsA patients.

A likely explanation for CPa9-HNE being one of the best markers to discriminate elevated disease activity in PsA patients may be that neutrophils are playing an important role in psoriatic diseases. Neutrophils can release a program of the neutrophil extracellular trap formation (NETosis) which may take part in the pathogenesis of psoriasis and PsA [10,11,12]. Recently, a NETosis marker has been investigated as a potential disease activity biomarker in PsA, the myeloperoxidase–DNA (MPO–DNA) complex, one of the neutrophil extracellular trap components. In this study, the level of MPO–DNA complex was markedly elevated in serum samples from patients with PsA and psoriasis and further shown to positively correlate with psoriatic disease burden [12]. Another research group investigated immature neutrophils as disease activity biomarker in PsA. It was shown that the immature neutrophil score was elevated in PsA patients compared to controls and significantly correlated with disease activity based on clinical parameters (ASDAS, SPARCC, PASI, ESR, CRP, enthesitis, dactylitis) [13]. Lastly, serum calprotectin has been explored as biomarkers of disease activity in PsA patients and found to be significantly elevated in patients with psoriasis and PsA compared with controls, and further positively correlated with psoriasis area and severity PASI, DAPSA, TJC, and SJC [14]. Overall, evidence from previous findings indicates that neutrophils play a key role in psoriatic disease, thus this may explain why CPa9-HNE levels are particularly elevated in PsA patients with flares and may be a useful biomarker to monitor elevated disease activity. Additionally, PRO-C3 also showed to be one of the best markers to discriminate elevated disease activity in PsA patients. PRO-C3 measures type III collagen formation and has been reported to be a marker of fibrogenesis [15]. In previous studies, PRO-C3 showed to be connected with the activation of dermal and lung fibroblasts and upregulated in patients with systemic sclerosis (SSc), associated with progressive disease [16, 17]. Hence, it can be questioned whether PRO-C3 is upregulated during flares in this study as a result of progressive fibrosis or because of an increased repair mechanism that is activated to counteract the upregulated joint tissue degradation mirrored by the elevated collagen degradation, in particular the type III degradation (C3M).

Besides assessing the elevated disease activity systemically in the serum from PsA patients, it was further investigated whether the elevated disease activity could be observed in the local joint microenvironment using SF. Two panels of biomarkers were chosen in the SF samples, where one panel consisted of markers reflecting inflammation and the other reflecting cartilage remodeling. It was hypothesized that the inflammatory biomarkers would be elevated in PsA known to be an inflammatory disease, while the cartilage degradation markers would be elevated in OA, known to be a degenerative disease, where cartilage degradation is a hallmark. Interestingly, one of the inflammation markers reflecting macrophage activity, VICM, showed to be elevated in PsA-flare, whereas the marker of neutrophil activity, CPa9-HNE, was equally expressed in PsA-flare compared to OA patients. Neutrophils are known to play an important role in the PsA pathology at the site of local inflammation where enhanced IL-23 production and Th17 activation trigger neutrophil infiltration and increased concentrations of calprotectin [18]. Therefore, it is notable that in this study the neutrophil activity levels are not significantly elevated in the PsA-flare patients compared to OA. On the other hand, the macrophage activity was highly elevated in PsA-flare patients compared to OA. These elevated levels may reflect the synovial tissue macrophages, which are recognized as pro-inflammatory cells previously shown to have a key role in rheumatoid arthritis by producing TNF, in turn driving chronic pathology [19]. SF macrophages have also been recently shown to play key roles in joint inflammation [20]. Altogether, these results may suggest distinct functions of neutrophils and macrophages during PsA-flare, where macrophages may be driving the disease activity both locally and systemically.

Furthermore, among the markers of cartilage remodeling, PRO-C2 reflecting type II collagen formation, was the only marker being elevated in OA patients compared to PsA-flare. The elevated levels of type II collagen formation may reflect an imbalance between the formation and degradation processes of type II collagen. While no biomarkers were available for the assessment of type II collagen degradation in SF, the elevated type II collagen levels may be a result of an imbalance in the tissue homeostasis, where elevated degenerative processes compensate with enhanced formation of type II collagen to maintain the tissue homeostasis. Collectively, these findings show that there are overlaps in the pathology of OA and PsA patients, where some OA patients present with inflammatory symptoms, whereas PsA patients additionally have increased cartilage degradation. Hence, these results may highlight the difficulty in telling the two types of arthritis apart in the clinic.

The matched serum and synovial fluid samples from the PsA-flare patients were analyzed to investigate potential marker differences in the circulation versus the local joint microenvironment. The biomarkers CPa9-HNE and VICM were analyzed since these biomarker assays are technically validated for measurement in both serum and SF matrix. Results from the analysis showed that the mean levels of CPa9-HNE and VICM were significantly elevated in serum compared to synovial fluid. The higher levels observed in the serum compared to SF may indicate higher systemic macrophage and neutrophil activity that is not confined locally to the knee joints. Notably, no correlation was observed between the systemic and local levels of the PsA-flare patients, which may indicate the heterogeneity characterizing the PsA disease.

Interestingly, in this study no significant difference was observed in biomarkers levels between controls and PsA without flare patients. One possible explanation for this may be that the controls are self-reported healthy donors with limited clinical data available; thus, some donors may have underlying conditions affecting systemic marker levels. In one study, where CPa9-HNE was assessed in serum from healthy donors, the median value of CPa9-HNE was 30 ng/mL where patients with ulcerative colitis or Crohn’s disease had levels all above 100 ng/mL [21]. In this work, the controls presented a median value of 125 ng/mL, which is a level more similar to the patients presenting ulcerative colitis and Crohn’s disease. Another explanation for the similar biomarker levels observed between controls and PsA without flare may be that these biomarkers are end products of elevated protease activity as a result of extensive tissue inflammation, which may be significantly lower in the PsA patients without flare. The PsA without flare patients have only been diagnosed with PsA disease within 1.9 years in average compared to PsA-flare patients who have had the disease for 11.5 years in average. Interestingly, when only observing the clinical parameters of the PsA without flare patients and PsA-flare patients, the hsCRP levels were similar between the two groups, although the PsA without flare patients have significantly higher PASI and DAPSA scores. The tender joint count and active joint criteria were significantly higher in PsA without flare compared to PsA-flare. PsA-flares were defined by the presence of an acutely swollen and tender knee joint whereas PsA without flare patients did not have such acute changes in disease activity. None of the PsA without flare patients were receiving t-DMARD, while 40% of PsA-flare patients were treated with t-DMARD. The wide range in joint counts seen for the PsA patients are likely due to the effect of treatment and the criteria for study entry. However, highly elevated levels of the inflammation markers including the fibrotic marker were found in serum from the PsA-flare patients when compared to PsA without flare and controls. Based on these data, the ability of hsCRP as a good marker to monitor disease activity may be questioned. Furthermore, in the PsA without flare patients, marginally significant correlation was observed between hsCRP and CPa9-HNE.

Limitations of this study include the cross-sectional design and the fact that the number of serum samples from PsA-flare group (n = 30) was three times lesser than the controls (n = 99) and PsA without flare group (n = 98). Given the unmet need for disease activity and flare biomarkers in the management of PsA, there is an immediate need to conduct longitudinal internal and external validation studies with an expanded cohort of patients to confirm the findings of this study.