Background: Psoriatic arthritis (PsA) is a clinically and biologically heterogeneous inflammatory disease, characterized by marked variability in disease course, clinical phenotype, and treatment response. Despite the availability of multiple therapeutic options targeting distinct inflammatory pathways, a substantial proportion of patients fail to achieve early remission or sustained low disease activity, even when treated promptly after diagnosis. The lack of validated biomarkers capable of predicting early treatment response in treatment-naïve patients represents a major unmet need, limiting the implementation of personalized treatment strategies in PsA [1]. B cells are recognized as active contributors to PsA pathogenesis, beyond their traditional role in autoantibody production. Through antigen presentation, cytokine secretion and interactions with T cells and stromal components, B cells may shape both peripheral immune responses and synovial inflammation. However, the prognostic relevance of baseline peripheral B-cell subsets, particularly in treatment-naïve PsA, and their relationship with synovial histopathology and extra-articular disease manifestations remain poorly defined.

Objectives: To evaluate whether baseline peripheral B-cell subsets predict early clinical response at 6-month follow-up in treatment-naïve PsA patients, and to explore their association with extra-articular manifestations and synovial histology.

Methods: Treatment-naïve PsA patients presenting with oligoarticular or polyarticular disease were included. Peripheral blood B-cell subsets were assessed at baseline by multiparametric flow cytometry and classified according to IgD and CD27 expression. (2) Subsets were expressed as percentages of total circulating B cells. Clinical assessment was performed at baseline (T0) and at 6-month follow-up (T6). Clinical response at T6 was defined as achievement of remission or low disease activity (LDA) according to the Disease Activity index for Psoriatic Arthritis (DAPSA). (3) Baseline extra-articular manifestations, including dactylitis, enthesitis and skin involvement, were systematically recorded as part of routine clinical evaluation. Ultrasound-guided synovial biopsies were obtained at baseline and evaluated using the Krenn synovitis score (KSS). (4) Associations between immunological, clinical and histological variables were explored using non-parametric statistics. Multivariable logistic regression models were constructed to identify independent predictors of response at T6, adjusting for age, sex, baseline DAPSA and pain score. Missing data were handled as missing without imputation.

Results: Fifty-five treatment-naïve PsA patients with oligoarticular or polyarticular involvement were included; 24 (43.6%) achieved remission or LDA at 6-month follow-up. Baseline DAPSA did not differ between responders and non-responders, whereas baseline pain scores were significantly higher in non-responders (p=0.004), suggesting that pain-related mechanisms rather than baseline inflammatory burden contributed to early outcome discrimination. Responders displayed a distinct baseline peripheral B-cell profile, characterized by higher proportions of naïve CD27 ⁻ IgD ⁺ B cells (median 68.9% vs 53.2%, p<0.001) and lower proportions of memory CD27 ⁺ IgD ⁺ B cells (13.4% vs 20.0%, p<0.001), CD27 ⁻ IgD ⁻ double-negative B cells (3.18% vs 4.15%, p=0.018) and plasma cells (0.21% vs 0.52%, p=0.005). In multivariable logistic regression models, baseline naïve B-cell frequency remained independently associated with response at T6 (OR per 10% increase 1.96, p=0.006), while higher memory, double-negative and plasma cell proportions were independently associated with a reduced probability of response, even after adjustment for baseline disease activity and pain. Among extra-articular manifestations, baseline dactylitis identified a subgroup of patients with a more differentiated B-cell profile, enriched in double-negative B cells and plasma cells, and was associated with a lower likelihood of achieving remission or LDA at T6 (p=0.037). Regarding synovial histology, lining layer hyperplasia at baseline was significantly higher in non-responders (p=0.016) and independently associated with lack of response. In addition, total KSS showed significant correlations with systemic inflammatory markers, particularly C-reactive protein levels, supporting a link between tissue-level inflammation and systemic inflammatory burden.

Conclusions: In treatment-naïve PsA, a baseline peripheral B-cell immunophenotype skewed toward naïve CD27 ⁻ IgD ⁺ B cells strongly predicts clinical remission or low disease activity at 6 months, independently of baseline disease activity and pain. Conversely, enrichment of antigen-experienced and plasma cell–lineage B-cell subsets identify patients with a reduced likelihood of early response. Extra-articular disease features, particularly dactylitis, further define biologically distinct subgroups characterized by a more differentiated B-cell compartment and less favourable outcomes. Synovial histology provides complementary prognostic information, linking tissue inflammation to systemic inflammatory burden. Altogether, these findings support baseline B-cell profiling as a promising prognostic tool in early PsA and highlight the biological heterogeneity underlying treatment response, with potential implications for precision medicine approaches.

REFERENCES: [1] Zabotti A, et al.Delineating inflammatory from non-inflammatory mechanisms for therapy optimization in psoriatic arthritis. Nat Rev Rheumatol. 2025 Apr;21(4):237-248. 2-Sanz I, et al.Phenotypic and functional heterogeneity of human memory B cells. Semin Immunol. 2008 Feb;20(1):67-82. 3-Schoels M, et al.Application of the DAREA/DAPSA score for assessment of disease activity in psoriatic arthritis. Ann Rheum Dis. 2010 Aug;69(8):1441-7. 4-Krenn V, et al.Synovitis score: discrimination between chronic low-grade and high-grade synovitis. Histopathology. 2006 Oct;49(4):358-64.

Acknowledgments: NIL.

Disclosure of Interests: None declared.