Sage Journals: Discover world-class research

Abstract

Background:

A proportion of patients with psoriatic arthritis (PsA) have their disease onset after the age of 60 (Late-onset PsA-LoPsA).

Objectives:

To examine whether this subgroup exhibits distinct characteristics, disease outcomes, and treatment patterns.

Design:

Cross-sectional, observational study.

Methods:

Data were collected cross-sectionally from 805 PsA patients in a real-life, multicenter, nationwide study. Among them, we identified 130 patients (16%) with LoPsA (disease onset ⩾60 years). Univariable analyses were conducted to compare the characteristics of LoPsA with the rest of the cohort and multivariable logistic regression was performed to identify factors independently associated with LoPsA.

Results:

After controlling for potential confounders, LoPsA patients were less likely to report a family history of psoriasis (OR: 0.18, 95% CI: 0.05–0.64). At diagnosis, the LoPsA group displayed more than 2-fold higher probability of presenting with polyarthritis (OR: 2.13, 95% CI: 1.11–4.09) while during their disease course, they had a 50% lower probability of developing enthesitis (OR: 0.50, 95% CI: 0.26–0.98). Patients with LoPsA exhibited a higher rate of comorbidities such as diabetes mellitus (OR: 3.17, 95% CI: 1.56–6.43) and hypertension (OR: 4.03, 95% CI: 2.02–8.04), and were more likely to develop major adverse cardiovascular events (MACEs), (OR: 4.83, 95% CI: 1.56–14.92). There were no differences in treatment patterns, disease outcomes, and damage indices between the two patient groups.

Conclusion:

Patients with LoPsA were more likely to present with polyarthritis, had lower odds of developing enthesitis, and were more frequently observed to have comorbidities and MACEs. Overall, their treatment patterns and disease outcomes did not differ from patients with early onset PsA.

Plain language summary

Psoriatic arthritis starting after the age of 60. Data from a nation-wide study

In some patients with PsA, disease starts after the age of 60 (Late-onset PsA-LoPsA). In this study, we examined whether this subgroup exhibits distinct characteristics, disease outcomes and treatment patterns. Data were collected from 805 PsA patients in a real-life, multicenter, nationwide study. Among them, we identified 130 patients (16%) with LoPsA. Comparing this subset with the rest of the cohort, and after adjusting for potential confounders, people living with LoPsA were less likely to report a family history of psoriasis (adOR 0.18; 95% CI: 0.05–0.64). Also,they displayed more than 2-fold higher probability of presenting with polyarthritis (adOR 2.13; 95% CI: 1.11–4.09) while during their disease course, they had a 50% lower probability of developing enthesitis (adOR 0.50; 95% CI: 0.26–0.98). Individuals with LoPsA had a higher prevalence of comorbidities such as diabetes mellitus (adOR 3.17; 95% CI: 1.56–6.43) and hypertension (adOR 4.03; 95% CI: 2.02–8.04), and a higher risk for developing major adverse cardiovascular events (MACEs, adOR 4.83; 95% CI: 1.56–14.92). There were no differences in treatment patterns, disease outcomes and other indices between the two groups. In conclusion, people with LoPsA, presented more commonly with polyarthritis, developed less often enthesitis, had more comorbidities and MACEs but overall, their treatment patterns and disease outcomes did not differ from patients with younger-onset PsA.

Keywords

late-onset psoriatic arthritis psoriatic arthritis

Introduction

Psoriatic arthritis (PsA) is a chronic inflammatory disease classified within the group of spondylarthritis.¹ Its prevalence is estimated at 0.1%–1.0% in the general population and about 20%–30% among patients with psoriasis (PsO), yet these rates vary across different geographic regions.² Males and females are equally affected at any age, although the disease onset is usually between 30 and 60 years old.³ PsA is a heterogeneous condition characterized by musculoskeletal symptoms, including peripheral arthritis, enthesitis, dactylitis, axial disease, along with skin involvement. Extra-musculoskeletal manifestations, such as uveitis and inflammatory bowel disease, may also occur.⁴ In addition, PsA can be associated with comorbidities, such as cardiometabolic disease and mental health disorders.^5–7

The prevalence of PsA rises steadily until the age of 60, after which it declines.⁸ Previous studies have reported that individuals with disease onset after the age of 60 experience more aggressive disease, with a higher number of active joints, higher inflammation markers at baseline, and worse outcomes after 2 years, compared to those with disease onset at a younger age.⁹ In this context, the concept of late-onset PsA (LoPsA) was introduced to further investigate the characteristics of this specific population. Although a clear consensus is lacking,¹⁰ adopting the age of 60 years old as the cut-off for defining LoPsA is the most common approach followed by most investigators in existing literature.^10–12

There are a few studies comparing early- and late-onset PsA, highlighting differences in clinical presentation, treatment approaches¹² and underlying pathogenetic mechanisms.⁹ LoPsA has been associated with higher body mass index (BMI),¹³ more prominent enthesitis and peripheral arthritis, higher inflammatory markers, and increased comorbidity burden, whereas other associations, such as with smoking or gender, remain inconclusive. Regarding treatment, several studies indicate that LoPsA patients receive biologics less frequently, often due to physician hesitancy,¹⁴ despite evidence showing similar efficacy and safety for some of them in elderly patients.¹²

The increasing prevalence of LoPsA¹⁵ combined with the complexity of managing elderly patients due to the challenges of polypharmacy and multiple comorbidities,¹⁶ underscores the need for further research. In this contemporary, real-world, nationwide study, we aimed to investigate the characteristics of patients with LoPsA regarding clinical presentation, comorbidity profile, long-term outcomes, and treatment response.

Methods

We analyzed the baseline data from an ongoing, multicenter, nationwide study of patients with PsA in Greece, under the oversight of the Greek Rheumatologic Society.¹⁷

Patients fulfilling the CASPAR (ClASsification criteria for Psoriatic ARthritis) criteria followed in the participating clinics and practices from January 1, 2022, to December 31, 2022 (time of assessment) were enrolled. Individuals with unknown age at disease onset, missing data, or loss to follow-up were excluded from the analysis. The study size was determined by the total number of patients fulfilling the inclusion criteria who were enrolled across the participating centers in the predefined period. The patient cohort has been described in more detail elsewhere.¹⁷

The patients were divided into two subgroups according to the age of PsA onset. LoPsA was defined as a disease onset after the age of 60. Comparisons were made between LoPsA patients (disease onset ⩾60 years) and the rest of the cohort (disease onset <60 years). The following variables were recorded: (a) Demographic data including gender, age at the time of assessment, obesity (defined as BMI >30 kg/m²), current or past smoking, family history of PsO or PsA, disease duration (the time between diagnosis and the last follow-up); (b) Clinical features at the time of diagnosis and at any time during the disease course, including type of peripheral arthritis (monoarthritis, oligoarthritis or polyarthritis), axial disease (sacroiliac joint and/or spinal pain with confirmatory X-ray or magnetic resonance imaging (MRI) findings—the definition of axial disease is further explained elsewhere¹⁸), enthesitis (as assessed by Leeds enthesitis index) and dactylitis (as assessed during clinical examination by rheumatologists, based on the number of swollen digits and without the use of validated scoring system), skin psoriasis, nail involvement, uveitis (confirmed by ophthalmologist), inflammatory bowel disease (confirmed by a gastroenterologist), family history of psoriasis (PsO) or PsA; (c) Laboratory findings: HLA-B27 (due to missing data (available data for 352/805 patients) HLA-B27 analyses were not included in the results), C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR) levels at the time of diagnosis and at the time of assessment. (d) Comorbidities: diabetes mellitus (defined as fasting blood sugar >126 mg/dL and/or prescription of anti-diabetic medication), coronary vascular disease (angina, myocardial infarction, stent placement), dyslipidemia (defined as total cholesterol > 200 mg/dL and/or triglyceride > 150 mg/dL and/or prescription of lipid-lowering therapy), hypertension (defined as systolic blood pressure > 140 mmHg and/or diastolic blood pressure > 90 mmHg in two measurements and/or prescription of antihypertensive medication), coronary vascular disease (angina, myocardial infarction, revascularization procedure), depression (confirmed by the use of antidepressants prescribed by psychiatrist) and hyperuricemia (defined as serum uric acid >6.8 mg/dL or use of uric-acid lowering therapy). Comorbidities were considered positive if they had been diagnosed at or before the time of assessment. (e) Long-term outcomes: history of hospitalization and serious infections (defined as requiring hospitalization and/or iv antibiotics) over the last year from the time of assessment, arthroplasty, major adverse cardiovascular events (MACE: myocardial infarction, angina, and/or stroke), presence of erosions and new bone formations in hands X-rays, presence of syndesmophytes in spine X-rays (based on clinical assessment, without a standardized structural damage scoring system), as well as the classification of difficult-to-manage (D2M), refractory PsA (cross-sectionally assessed as per EULAR Points to Consider and Consensus Definitions for Difficult-to-Manage and Treatment-Refractory Psoriatic Arthritis)¹⁹ and PsA with minimal disease activity (MDA).²⁰ (f) Treatment history: use of glucocorticoids, nonsteroidal anti-inflammatory drugs (NSAIDs), conventional synthetic disease-modifying antirheumatic drugs (csDMARDs), biologic (b) and targeted synthetic (ts) DMARDs since the PsA diagnosis. Clinical, laboratory, and imaging data were collected following the same procedures and criteria for all patients regardless of their disease onset so that comparability of assessment methods between the two groups is ensured.

For statistical analysis, categorical features were compared with two-sided chi-square, and numerical data were compared with an unpaired t-test for parametric or a Mann-Whitney test for non-parametric variables. The Kolmogorov-Smirnov test was used to assess normal distribution. Statistical significance was set at a p-value less than 0.05. Given the exploratory nature of this study, no formal correction for multiple testing was applied. Multivariable logistic regression analysis was performed, in which all variables showing significant association with LoPsA status (p-value <0.05) in the univariable analyses were included as independent variables, and LoPsA as the dependent variable. For all independent variables, analyses were performed using only available data, and no imputation methods were employed. Multicollinearity diagnostics were conducted using linear regression (Tolerance and Variance Inflation Factor (VIF)), with a VIF <2 defined as the threshold for acceptable collinearity. The adequacy and predictive performance of the multivariable logistic regression were assessed with the Hosmer–Lemeshow goodness-of-fit test and the calculation of the AUC value (area under the curve) of the ROC (Receiver operating characteristic) curve, respectively. For the former test, a non-significant result (p-value >0.05) indicated a satisfactory agreement between observed and predicted outcomes, whereas, for the latter, an AUC value approaching 1 was interpreted as excellent discrimination, with the acceptable cut-off being set at 0.7. Data analysis was performed using GraphPad Prism 5.00 (GraphPad Software, Inc., Boston, MA, USA) and SPSS 24.0 (SPSS Inc., Chicago, IL, USA). Institutional Review Board approval was provided by the Joint Rheumatology Program (“Laiko Hospital” No.: 780/2021) and by the local institutional boards of participating centers. Informed consent was obtained from all patients. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement (Supplemental Material).²¹

Results

Patients’ demographics

Out of 805 patients with PsA enrolled in this study, 130 patients (16%) had LoPsA (Table 1). Patients with LoPsA were less commonly females compared to patients with younger disease onset (45.4% vs 56.4%, p = 0.021), were older (71.1 vs 53.1 years, p = 0.0001), and had less often a family history of psoriasis (23% vs 33.5%, p = 0.028). Demographic characteristics are depicted in Table 1.

Table 1.

Demographics of patients included in the study.

Demographics	Whole cohort (n = 805)	Age at disease onset <60 (n = 675)	Age at disease onset ⩾60 (n = 130)	p-Value
Female gender, n (%)	440 (54.7)	381 (56.4)	59 (45.4)	0.021
Age (years), mean ± SD	56 ± 12.2	53.1 ± 10.9	71.1 ± 6.4	0.0001
Smoking,^a n (%)	414 (54.4) n = 761	347 (54.3) n = 639	67 (54.9) n = 122	0.921
Disease duration (months), mean ± SD	113 ± 95.6	121.7 ± 68.7	67.6 ± 59.7	0.0001
Obesity,^b n (%)	281 (38.7) n = 723	238 (38.9) n = 612	43 (38.7) n = 111	1.000
BMI (kg/m²), mean ± SD	29.3 ± 6 n = 723	29.2 ± 6.1 n = 612	29.5 ± 5.5 n = 111	0.290
Family history of PsO, n (%)	232 (31.9) n = 728	206 (33.5) n = 615	26 (23) n = 113	0.028
Family history of PsA, n (%)	54 (7.4) n = 731	48 (7.8) n = 615	6 (5.3) n = 113	0.437

Statistically significant values are denoted with bold fonts.

Current or past.

BMI >30 kg/m².

BMI, body mass index; N, number; PsA, psoriatic arthritis; PsO, psoriasis; SD, standard deviation.

Clinical characteristics, comorbidities, disease damage and treatment patterns

At diagnosis, patients with LoPsA presented more frequently with polyarthritis (66.3% vs 49.6%, p = 0.003), higher mean ESR (34.1 ± 23.4 vs 28.2 ± 20.9, p = 0.020) while during their disease course, they developed less frequently enthesitis (15.4% vs 31.7%, p = 0.0002) and dactylitis (9% vs 17.2%, p = 0.022, Table 2).

Table 2.

Patients’ clinical characteristics at diagnosis and during the disease course.

Clinical characteristics	Age at disease onset <60 (n = 675)	Age at disease onset ⩾60 (n = 130)	p-Value
At diagnosis
Peripheral arthritis, n (%)	528 (78.2)	102 (78.5)	1.000
Monoarthritis, n (%)	31 (6.2) n = 498	2 (2.1) n = 95	0.142
Oligoarthritis, n (%)	196 (39.4) n = 498	29 (30.5) n = 95	0.108
Polyarthritis, n (%)	247 (49.6) n = 498	63 (66.3) n = 95	0.003
Axial disease, n (%)	115 (17)	17 (13)	0.301
Enthesitis, n (%)	191 (29.3) n = 652	30 (24.4) n = 123	0.327
Dactylitis, n (%)	140 (21.3) n = 656	24 (19.7) n = 122	0.719
Nail involvement, n (%)	217 (33.5) n = 647	35 (28.7) n = 122	0.344
Uveitis, n (%)	10 (1.6) n = 636	1 (0.8) n = 121	1.000
Ulcerative colitis, n (%)	3 (0.5) n = 643	0 (0) n = 121	1.000
Crohn disease, n (%)	2 (0.3) n = 634	2 (1.7) n = 119	0.123
CRP, mean ± SD	2.3 ± 2.3 n = 446	2.0 ± 2.1 n = 94	0.149
ESR, mean ± SD	28.2 ± 20.9 n = 478	34.1 ± 23.4 n = 94	0.020
During the disease course
Peripheral arthritis, n (%)	270 (40)	52 (40)	1.000
Axial disease, n (%)	95 (14.1)	20 (15.4)	0.682
Enthesitis, n (%)	207 (31.7) n = 652	19 (15.4) n = 123	0.0002
Dactylitis, n (%)	113 (17.2) n = 656	11 (9) n = 122	0.022
Nail involvement, n (%)	151 (23.3) n = 647	20 (16.4) n = 122	0.097
Uveitis, n (%)	11 (1.7) n = 636	3 (2.5) n = 121	0.478
Ulcerative colitis, n (%)	5 (0.8) n = 643	1 (0.8) n = 121	1.000
Crohn disease, n (%)	9 (1.4) n = 634	1 (0.8) n = 119	1.000

Statistically significant values are denoted with bold fonts.

CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; SD, standard deviation.

Regarding comorbidities, diabetes, dyslipidemia, hypertension, and hyperuricemia were significantly more common in patients with LoPsA (34.4% vs 13.8%, p < 0.0001; 65.5% vs 38.3%, p < 0.0001; 73.8% vs 42.2%, p < 0.0001; 24.2% vs 14.3%, p = 0.009, respectively, Table 3).

Table 3.

Comorbidities, long-term outcomes, and treatment patterns.

Comorbidities, long-term outcomes, treatment patterns	Age at disease onset <60 (n = 675)	Age at disease onset ⩾60 (n = 130)	p-Value
Comorbidities
Diabetes, n (%)	88 (13.8) n = 637	42 (34.4) n = 122	<0.0001
Dyslipidemia, n (%)	238 (38.3) n = 622	78 (65.5) n = 119	<0.0001
Hypertension, n (%)	271 (42.2) n = 642	93 (73.8) n = 126	<0.0001
Depression, n (%)	138 (21.8) n = 633	35 (29.4) n = 119	0.075
Hyperuricemia, n (%)	90 (14.3) n = 629	29 (24.2) n = 120	0.009
Disease damage patterns
Bone erosions (hand X-rays), n (%)	145 (32) n = 453	29 (33.3) n = 87	0.803
New bone formation (hand X-rays), n (%)	235 (52) n = 452	45 (51.7) n = 87	1.000
Syndesmophytes (spine X-rays)	75 (26.5) n = 283	16 (29.1) n = 55	0.740
History of arthroplasty, n (%)	26 (3.9)	3 (2.3) n = 128	0.606
Serious events
Hospitalizations,^a n (%)	28 (4.2) n = 669	6 (4.7) n = 128	0.810
MACEs, n (%)	27 (4.3) n = 630	19 (15.8) n = 120	<0.0001
Disease outcomes
MDA, n (%)	273 (29.9) n = 578	47 (43.1) n = 109	0.464
D2M,^b n (%)	127 (19.6) n = 647	29 (24.4) n = 119	0.264
Refractory,^b n (%)	59 (9.2) n = 639	15 (12.7) n = 118	0.264
Treatment patterns
csDMARDs, n (%)	578 (85.6)	114 (87.7)	0.584
Number of b-ts-DMARDs, median (IQR)	1 (0–2)	1 (0–2)	0.002
b-ts-DMARDs, n (%)	484 (71.7)	76 (58.5)	0.004
Glucocorticoids, n (%)	366 (55) n = 666	71 (55) n = 129	1.000
NSAIDs, n (%)	370 (56.1) n = 659	49 (38.3) n = 128	0.0002

Statistically significant values are denoted with bold fonts.

Hospitalization occurred over the last year from the time of assessment.

Cross-sectionally assessed as per EULAR points to consider and consensus definitions for difficult-to-manage and treatment-refractory psoriatic arthritis.

b-ts-DMARDs, biologic-targeted synthetic-disease-modifying antirheumatic drugs; csDMARDs, conventional synthetic disease-modifying antirheumatic drugs; D2M, difficult-to-manage; IQR, interquartile range; MACE, major adverse cardiovascular events; MDA, minimal disease activity; NSAIDs, nonsteroidal anti-inflammatory drugs; SD, standard deviation.

Patients with LoPsA received NSAIDs less frequently (38.3% vs 56.1%, p = 0.0002), and a lower number of b- and ts-DMARDs (1 [0–2] vs 1 [0–2], p = 0.002). The use of glucocorticoids and csDMARDs was similar between the two groups (Table 3).

Regarding long-term outcomes, the radiographic and disease damage (bone erosions, new bone formation, syndesmophytes, and arthroplasties) as well as the disease outcomes (achievement of MDA, refractory or difficult-to-manage disease) did not differ between the two groups. Concerning serious events, MACEs developed more frequently in patients with LoPsA (15.8% vs 4.3%, p < 0.0001) (Table 3).

Multicollinearity diagnostics

Multicollinearity diagnostics indicated acceptable levels, with all VIF values below 2, confirming the absence of multicollinearity among the variables.

Multivariable logistic regression

The multivariable logistic regression analysis identified factors independently associated with LoPsA (Table 4). LoPsA was independently associated with a shorter disease duration (odds ratio (OR): 0.987, confidence interval (CI): 0.981–0.992) and a higher likelihood of presenting with polyarthritis at diagnosis (OR: 2.13, 95% CI: 1.11–4.09). Patients with LoPsA had lower odds of having a family history of PsO (OR: 0.18, 95% CI: 0.05–0.64), lower odds of developing enthesitis (OR: 0.50, 95% CI: 0.26–0.97), and were less likely to use NSAIDs (OR: 0.483, 95% CI: 0.25–0.93). Furthermore, patients with LoPsA were more frequently observed to have comorbidities such as diabetes mellitus type 2 (OR: 3.16, 95% CI: 1.59–6.29), hypertension (OR: 4.03, 95% CI: 2.03–7.97), and MACEs (OR: 4.82, 95% CI: 1.46–15.91). Other variables, including gender, dactylitis history, dyslipidemia, hyperuricemia, and biologic treatment, were not significantly associated with the age of PsA onset after adjustment.

Table 4.

Multivariable logistic regression analysis for factors associated with late-onset psoriatic arthritis.

Multivariable logistic regression analysis	Adjusted OR (95% CI)	p-Value
Female gender	0.76 (0.39–1.48)	0.416
Family history of PsO	0.18 (0.05–0.64)	0.008
Family history of PsA, SpA, or IBD	1.65 (0.54–5.04)	0.378
Disease duration (months)	0.99 (0.98–0.99)	<0.001
Polyarthritis at diagnosis	2.13 (1.11–4.10)	0.023
Enthesitis (ever)	0.50 (0.26–0.98)	0.042
Dactylitis (ever)	1.05 (0.53–2.09)	0.902
Diabetes	3.17 (1.53–6.56)	0.001
Dyslipidemia	1.73 (0.89–3.38)	0.105
Hypertension	4.03 (1.90–8.55)	<0.001
Hyperuricemia	0.69 (0.26–1.84)	0.463
MACE	4.83 (1.56–14.91)	0.006
bDMARDs (total number)	0.84 (0.63–1.13)	0.285
On TNF inhibitor therapy	1.41 (0.66–2.98)	0.386
On IL-17/23 inhibitor therapy	1.18 (0.39–3.60)	0.765
NSAIDs (ever)	0.48 (0.25–0.92)	0.027

Statistically significant values are denoted with bold fonts.

bDMARDs, biological disease-modifying antirheumatic drugs; CI, confidence interval; IBD, inflammatory bowel disease; IL-17/23, Interleukin-17 and Interleukin-23; MACE, major adverse cardiovascular events; NSAIDs, nonsteroidal anti-inflammatory drugs; OR, odds ratio; PsA, psoriatic arthritis; PsO, psoriasis; SpA, spondyloarthritis; TNF, tumor necrosis factor.

Model validation indicated an overall good fit and satisfactory discriminative ability, as shown by the Hosmer–Lemeshow goodness-of-fit test (χ² = 7.57, p = 0.48) and the AUC value of the ROC curve (AUC = 0.881).

Sensitivity analysis excluding disease duration

A sensitivity analysis excluding disease duration from the independent values was also performed. In this analysis, all associations previously reported (namely in the model including disease duration) remained consistent in terms of significance, except for the ever presence of enthesitis, which was no longer statistically significant (OR: 0.60, 95% CI: 0.32–1.12), although it continued to show a negative trend. Additionally, a lower total number of bDMARDs was independently associated with LoPsA (OR: 0.72, 95% CI: 0.54–0.97).

Discussion

In this large, real-life, multicenter, cross-sectional study, we found that approximately one out of six patients with PsA had their disease onset after the age of 60. Compared to patients with earlier disease onset (<60 years), these patients were less likely to have a family history of PsO, more likely to present with polyarthritis, less likely to develop enthesitis, and exhibited higher rates of comorbidities and MACEs.

Patients with LoPsA experienced enthesitis less frequently during their disease course. This observation aligns with previous studies,^10,14,22 including one utilizing data from the RESPONDIA and REGISPONSER registries.¹¹ Thus, it could be suggested that a lower enthesitis rate may be a distinguishing feature of LoPsA, potentially indicating a distinct clinical phenotype within the PsA spectrum. However, it is important to note that the association with enthesitis may be confounded by shorter disease duration. To address this, we conducted a sensitivity analysis excluding disease duration as a variable. In this analysis, the presence of enthesitis remained less likely in individuals with LoPsA, although the association did not reach statistical significance.

Regarding other disease manifestations, dactylitis appeared less likely in the LoPsA group compared to the rest of the cohort, aligning with previous studies such as that by Kobak et al.²³ However, after adjusting for confounding factors, this association was no longer statistically significant. At diagnosis, patients with LoPsA were more likely to present with polyarthritis, compared to patients with earlier disease onset. This finding is consistent with the studies by Queiro et al. and López-Montilla et al.,^22,24 who also identified polyarthritis as the predominant form of onset among PsA patients with later disease onset, using age cut-offs of 40 and 60 years, respectively.

Additionally, we found that individuals with LoPsA were more likely to have a shorter disease duration compared to individuals with earlier disease onset. This association does not imply a predictive relationship, as this difference is expected given that disease duration follows disease onset and patients with late-onset disease are, by definition, older at diagnosis. However, disease duration was included in the multivariable logistic regression model to control for its potential confounding effect, since it can alter the manifestation of other clinical characteristics throughout the disease course, such as the ever presence of enthesitis or dactylitis, as well as the development of various comorbidities. Interestingly, when a sensitivity analysis excluding the disease duration was conducted, no differences were observed between this model and the one including the variable, except for its effect on ever presence of enthesitis and the total number of bDMARDs used in individuals with LoPsA.

Regarding demographic features, data have been conflicting so far, suggesting that there is no strong association between LoPsA and characteristics such as gender and BMI. Polachek et al., in a study enrolling 566 patients, have linked female gender and elevated BMI to LoPsA.²⁵ On the other hand, studies by Puche-Larrubia et al. and López-Montilla et al. identified a male predominance among patients with LoPsA,^11,24 while other studies found no significant differences.^14,22 In our cohort, although the percentage of female patients in the LoPsA group was lower (45% vs 56%), this did not reach statistical significance. Of note, our data revealed a significantly higher frequency of a history of PsO among PsA patients with earlier disease onset. In this context, family history and genetic predisposition should be taken into consideration when interpreting differences between LoPsA and earlier-onset patients, with previous studies associating family history of PsA or PsO,²⁶ and HLA-B27 positivity²⁷ with later disease onset. This association further supports the notion that LoPsA may represent a distinct disease phenotype within the spectrum of PsA, potentially driven by specific genetic factors.

Comorbidities, particularly those of a cardiometabolic nature, are at the center of care for patients with PsA and may be intrinsically linked to the disease’s underlying pathogenic mechanisms.⁵ Individuals with LoPsA demonstrated significantly higher odds of having comorbidities, including diabetes mellitus, hypertension, dyslipidemia, hyperuricemia, and MACEs. These findings are consistent with our previous bicentric study involving 281 patients, in which LoPsA was associated with a significantly increased frequency of dyslipidemia and MACE,¹⁰ as well as other studies,^11,23,28 Specifically, in our analysis, LoPsA patients were three times more likely to have type 2 diabetes mellitus, four times more likely to have hypertension, and nearly five times more likely to have MACEs. The interpretation of the elevated cardiovascular risk observed in the LoPsA group, as compared to the rest of the cohort, presents a complex challenge. It cannot be excluded that this could be mainly due to the more advanced age of patients in this group. Adjustments, however, for disease duration and other parameters that were different in the univariable analyses were made. Given the inflammatory nature of PsA at disease onset, it may act as a catalyst in initiating or intensifying preexisting cardiovascular risk factors associated with aging.²⁹ In any case, early cardiovascular risk assessment at the time of diagnosis, close monitoring, and vigilant management are of utmost significance in individuals with LoPsA, whereas in the general context of the comorbidities, treatment approaches should be tailored and personalized, considering age-related factors.

Treatment strategies were generally comparable between the two groups, except for the lower likelihood of NSAID use in the LoPsA group. Patients with LoPsA were about half as likely to use NSAIDs compared to the rest of the cohort. This reduced use may reflect clinicians’ hesitancy to prescribe NSAIDs to older patients, due to the increased risk of adverse effects associated with advanced age, as well as interactions with other medications, given that older patients are often subject to polypharmacy.³⁰ Another possible explanation could be the decreased patient tolerability associated with advanced age. The use of b-ts-DMARDs did not differ between the two groups. These findings are in alignment with our own previous study,¹⁰ and similar other studies.^11,23,25 Two studies from Queiro et al.²⁸ and Yamamoto et al.¹⁴ found that older patients with PsA received fewer biological therapies. However, both studies employed an age cut-off of 65 years, higher than the 60-year threshold used in our analysis. Furthermore, Queiro et al. included a much smaller sample of 227 patients.

Another important aspect that should be taken into consideration when interpreting the above-mentioned findings is immunosenescence, as it can affect both disease phenotype and management. As the immune system of elderly individuals weakens, they are more susceptible to infection, cancer, and inflammatory diseases.^31,32 This is due to a decline in adaptive immune function, reﬂected by increased blood concentrations of several inﬂammatory biomarkers, including CRP, IL-18, TNF-α, and IL-6.³³ This state of chronic inflammation, called inflammaging, may contribute to different disease expression and comorbidity burden in LoPsA patients,^32,34 whereas the polypharmacy associated with the advanced age also drives the different facets of the disease.

Interestingly, the proportion of patients with D2M or refractory disease was similar between the LoPsA group and the rest of the cohort. This observation suggests that the age at disease onset may not significantly impact therapeutic outcomes. In other inflammatory diseases, such as rheumatoid arthritis (RA), researchers have previously examined the predictive factors associated with D2T disease. The ANSWER cohort study identified younger age at disease onset, high rheumatoid factor (RF) titers, and the presence of comorbidities as significant predictors for the development of D2T RA.³⁵ This suggests that elderly-onset rheumatoid arthritis (EORA) could act as a protective factor against the development of D2T RA. Similar findings have been reported by Roodenrijs et al., who linked younger age at disease onset to D2T RA,³⁶ and by Bécède et al., who found an association between younger disease onset and refractory RA.³⁷ In contrast, the Kurama cohort study found no significant association between D2T RA and age at disease onset.³⁸ Regarding PsA, on the other hand, no associations had previously been established between these outcomes (such as D2T or refractory disease) and age at disease onset or other predictive variables. To our best knowledge, this study is the first to utilize the newly updated definitions of D2T and refractory disease by the European Alliance of Associations for Rheumatology in the context of PsA.³⁹

Late-onset presentation has been investigated in other immune-mediated inflammatory diseases, including axial spondyloarthritis (AxSpA) and RA. In a multicenter nationwide study conducted in Portugal,⁴⁰ late-onset AxSpA (defined as onset after age 45) accounted for approximately 13% of cases, a proportion comparable to the 16% observed in our cohort of LoPsA patients. Similar to LoPsA, individuals with late-onset AxSpA more frequently present with peripheral arthritis and demonstrate a lower prevalence of family history of spondyloarthritis—paralleling the reduced familial psoriasis history seen in LoPsA. Seeing that the opposite flip of the coin, one could hypothesize that genetic factors might contribute to the earlier disease onset. Interestingly, despite differences in clinical features, treatment patterns among late-onset AxSpA patients remain largely consistent with those in younger patients, which is similar to LoPsA patients, apart from the use of NSAIDs. Furthermore, as reported by Victor D. Acuña-Rocha et al.,⁴¹ EORA patients exhibit lower rates of RF and anti-citrullinated protein antibody positivity compared to their younger-onset counterparts. Notably, both LoPsA and EORA patients share a higher burden of comorbidities and elevated inflammatory markers, including ESR and CRP, suggesting overlapping immunoinflammatory and systemic profiles in late-onset disease presentations.

Limitations

We acknowledge that our study has certain limitations. First, the use of 60 years old as a cut-off to define LoPsA is arbitrary. While there is no universally accepted threshold, we opted for this cut-off to maintain consistency with our previous bicentric study¹⁰ and to enhance comparability with existing studies, which commonly utilize this age as a threshold.^9,12,24 This choice is further supported by epidemiological data indicating a decline in PsA incidence after the age of 60,⁸ as well as by the United Nations’ classification of individuals aged 60 and above as older persons.⁴² A second limitation of our study is the number of missing data. However, these are typically within the reasonable limit of 10% and are not expected to have substantially affected the outcomes of the analysis. Third, the assessment of clinical features during the disease course might have been influenced by the shorter disease duration characterizing the LoPsA group. To address this, we adjusted for disease duration in our multivariable analyses. Fourth, dactylitis was assessed clinically without the use of a validated scoring system, whereas for imaging evaluation, no central reading and no standardized structural damage scoring systems were followed. Another limitation is the absence of JAKi in the treatment regimens, as they were not widely available in Greece when the patients of the cohort were assessed. Furthermore, an inherent limitation of the study is the cross-sectional design, which allows to make associations but cannot confidently infer causality, whereas methodologically, the absence of power analysis for sample size calculation is an additional limitation.

In the strengths of our study, one could note that this is the largest study in the field, focused specifically on the LoPsA population, conducted within a well-established, multicenter, nationwide cohort. This enhances the validity and generalizability of our findings and lays the foundation for future research in this understudied area.

Conclusion

In conclusion, in the largest study in the literature so far, we showed that patients with LoPsA were more likely to present with polyarthritis and exhibited higher rates of cardiovascular comorbidities and MACEs. Despite receiving similar DMARD therapies with earlier-onset PsA, they were less likely to develop enthesitis, while their disease damage and outcomes did not differ from the younger group. These findings emphasize the need for increased vigilance and better cardiovascular protection in this patient population.

Supplemental Material

sj-docx-1-tab-10.1177_1759720X251405104 – Supplemental material for Late-onset psoriatic arthritis: data from a nationwide cross-sectional study

Supplemental material, sj-docx-1-tab-10.1177_1759720X251405104 for Late-onset psoriatic arthritis: data from a nationwide cross-sectional study by Sofia Karagianni, Anastasia Eleftheraki, Charalampos Papagoras, Sousana Gazi, Evangelia Mole, Michael Krikelis, Paraskevi V. Voulgari, Evripidis Kaltsonoudis, Nikolaos Koletsos, Dimitrios Boumpas, Pelagia Katsimpri, Dimitrios Katsifis-Nezis, Theodoros Dimitroulas, Nikolaos Kougkas, Petros P. Sfikakis, Maria G. Tektonidou, Konstantinos D. Vassilakis, Dimitrios Bogdanos, Theodora Simopoulou, Christos Koutsianas, Eugenia Mavrea, Gkikas Katsifis, Konstantinos Kottas, Maria Konsta, Matthoula Tziafalia, Evangelia Kataxaki, Eleni Kalavri, Kalliopi Klavdianou, Anastasios Karamanakos, Ioannis Xynogalas, Dimitrios Daoussis, George Iliopoulos, Ilias Bournazos, Constantinos Georganas, Dimos Patrikos, Dimitrios Vassilopoulos and George E. Fragoulis in Therapeutic Advances in Musculoskeletal Disease

Footnotes

Acknowledgements

We would like to thank late Dimitrios Karokis for his support and contribution.

Declarations

ORCID iDs

Maria G. Tektonidou

Christos Koutsianas

Dimitrios Daoussis

Dimitrios Vassilopoulos

George E. Fragoulis

Supplemental material

Supplemental Material for this article is available online.

References

López-Medina

Molto

Sieper

, et al. Prevalence and distribution of peripheral musculoskeletal manifestations in spondyloarthritis including psoriatic arthritis: results of the worldwide, cross-sectional ASAS-PerSpA study. RMD Open 2021; 7(1): e001450.

Scotti

Franchi

Marchesoni

, et al. Prevalence and incidence of psoriatic arthritis: a systematic review and meta-analysis. Semin Arthritis Rheum 2018; 48(1): 28–34.

Ogdie

Langan

Love

, et al. Prevalence and treatment patterns of psoriatic arthritis in the UK. Rheumatology 2013; 52(3): 568–575.

Gladman

Antoni

Mease

, et al. Psoriatic arthritis: epidemiology, clinical features, course, and outcome. Ann Rheum Dis 2005; 64: ii14–ii17.

Panagiotopoulos

Fragoulis

. Comorbidities in psoriatic arthritis: a narrative review. Clin Ther 2023; 45(2): 177–189.

Karmacharya

Ogdie

Eder

. Psoriatic arthritis and the association with cardiometabolic disease: a narrative review. Ther Adv Musculoskelet Dis 2021; 13: 1759720X21998279.

Gialouri

Evangelatos

Zhao

, et al. Depression and anxiety in a real-world psoriatic arthritis longitudinal study: should we focus more on patients’ perception? Clin Exp Rheumatol 2023; 41: 159–165.

Deike

Brinks

Meller

, et al. Risk of psoriatic arthritis depending on age: analysis of data from 65 million people on statutory insurance in Germany. RMD Open 2021; 7(3): e001975.

Punzi

Pianon

Rossini

, et al. Clinical and laboratory manifestations of elderly onset psoriatic arthritis: a comparison with younger onset disease. Ann Rheum Dis 1999; 58(4): 226–229.

10.

Gialouri

Evangelatos

Iliopoulos

, et al. Late-onset psoriatic arthritis: are there any distinct characteristics? A retrospective cohort data analysis. Life 2023; 13(3): 792.

11.

Puche-Larrubia

MÁ

Ladehesa-Pineda

López-Montilla

, et al. Differences between early vs. late-onset of psoriatic arthritis: data from the RESPONDIA and REGISPONSER registries. Joint Bone Spine 2023; 90(4): 105563.

12.

Caso

Tasso

Chimenti

, et al. Late-onset and elderly psoriatic arthritis: clinical aspects and management. Drugs Aging 2019; 36(10): 909–925.

13.

Herédi

Csordás

Clemens

, et al. The prevalence of obesity is increased in patients with late compared with early onset psoriasis. Ann Epidemiol 2013; 23(11): 688–692.

14.

Yamamoto

Ohtsuki

Sano

, et al. Late-onset psoriatic arthritis in Japanese patients. J Dermatol 2019; 46(2): 169–170.

15.

Bournia

Fragoulis

Mitrou

, et al. Increased prevalence of inflammatory arthritis, systemic lupus erythematosus and systemic sclerosis, during 2020–2023 versus 2016–2019 in a Nation-Wide Cohort Study. Rheumatol Int 2024; 44(12): 2837–2846.

16.

Dovjak

. Polypharmacy in elderly people. Wiener Medizinische Wochenschrift 2022; 172(5–6): 109–113.

17.

Fragoulis

Papagoras

Gazi

, et al. Disease profile and achievement of therapeutic goals in a modern, nationwide cohort of 923 patients with psoriatic arthritis. Mediterr J Rheumatol 2023; 34(4): 418–426.

18.

Vassilakis

Papagoras

Gazi

, et al. Identification and characteristics of patients with axial psoriatic arthritis: clinical, phenotypic and imaging associations. Rheumatology 2025; 64(8): 4539–4545.

19.

Marzo-Ortega

Harrison

Fragoulis

, et al. POS1269 Eular points to consider for the definitions of difficult-to-manage and treatment-refractory psoriatic arthritis. Ann Rheum Dis 2025; 84: 1319–1320.

20.

Gossec

McGonagle

Korotaeva

, et al. Minimal disease activity as a treatment target in psoriatic arthritis: a review of the literature. J Rheumatol 2018; 45(1): 6–13.

21.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med 2007; 147(8): 573–577.

22.

Queiro

Alperi

Lopez

, et al. Clinical expression, but not disease outcome, may vary according to age at disease onset in psoriatic spondylitis. Joint Bone Spine 2008; 75(5): 544–547.

23.

Kobak

Yildiz

Karaarslan

, et al. Characteristics of Turkish patients with elderly onset psoriatic arthritis. Medicine 2017; 96(33): e7833.

24.

López-Montilla

. Clinical features of late onset psoriatic arthritis. Exp Gerontol 2002; 37(2–3): 441–443.

25.

Polachek

Al-Johani

, et al. Late onset psoriatic arthritis in a longitudinal cohort: Disease presentation, activity over time and prognosis. Semin Arthritis Rheum 2019; 48(5): 834–839.

26.

Queiro

Torre

González

, et al. HLA antigens may influence the age of onset of psoriasis and psoriatic arthritis. J Rheumatol 2003; 30(3): 505–507.

27.

Alonso

Tejón

Sarasqueta

, et al. Age at disease onset may help to further characterize the disease phenotype in psoriatic arthritis. Joint Bone Spine 2016; 83(5): 533–537.

28.

Queiro

Pardo

Charca

, et al. Analysis by age group of disease outcomes in patients with psoriatic arthritis: a cross-sectional multicentre study. Drugs Aging 2020; 37(2): 99–104.

29.

Puspitasari

Ministrini

Schwarz

, et al. Modern concepts in cardiovascular disease: inflamm-aging. Front Cell Dev Biol 2022; 10: 882211.

30.

Coskun Benlidayi

Gokce Kutsal

. Antirheumatic drugs in older adults and polypharmacy issues. Z Gerontol Geriatr 2022; 55(6): 507–512.

31.

Fülöp

Larbi

Hirokawa

, et al. Immunosupportive therapies in aging. Clin Interv Aging 2007; 2(1): 33–54.

32.

Ventura

Casciaro

Gangemi

, et al. Immunosenescence in aging: between immune cells depletion and cytokines up-regulation. Clin Mol Allergy 2017; 15(1): 21.

33.

Bektas

Schurman

Sen

, et al. Human T cell immunosenescence and inflammation in aging. J Leukoc Biol 2017; 102(4): 977–988.

34.

Michaud

Balardy

Moulis

, et al. Proinflammatory cytokines, aging, and age-related diseases. J Am Med Dir Assoc 2013; 14(12): 877–882.

35.

Watanabe

Ebina

Gon

, et al. Predictive factors and treatment outcomes associated with difficult-to-treat rheumatoid arthritis conditions: the ANSWER cohort study. Rheumatology 2024; 63(9): 2418–2426.

36.

Roodenrijs

NMT

van der Goes

Welsing

PMJ

, et al. Difficult-to-treat rheumatoid arthritis: contributing factors and burden of disease. Rheumatology 2021; 60(8): 3778–3788.

37.

Bécède

Alasti

Gessl

, et al. Risk profiling for a refractory course of rheumatoid arthritis. Semin Arthritis Rheum 2019; 49(2): 211–217.

38.

Watanabe

Hashimoto

Murata

, et al. Prevalence and predictive factors of difficult-to-treat rheumatoid arthritis: the KURAMA cohort. Immunol Med 2022; 45(1): 35–44.

39.

Marzo-Ortega

Harrison

Fragoulis

, et al. EULAR points to consider for the definitions of difficult-to-manage and treatment-refractory psoriatic arthritis. Ann Rheum Dis. 2025; 84: 1319–1320.

40.

Lucas Rocha

Ramiro

Pinheiro Torres

, et al. Late-onset axial spondyloarthritis: a multicentre nationwide study. Joint Bone Spine 2025; 92(1): 105804.

41.

Acuña-Rocha

Regalado-Ceballos

Salcedo-Soto

, et al. Clinical characteristics of young-onset versus elderly-onset rheumatoid arthritis: a systematic review and meta-analysis. Cureus 2024; 16: e74148.

42.

The UN Refugee Agency. https://emergency.unhcr.org/entry/43935/older-persons

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB