Effect of low dose methotrexate as an add-on therapy in patients with palindromic rheumatism unresponsive to hydroxychloroquine: An observational study

Abstract

Objective

Palindromic rheumatism (PR) is characterized by repetitive attacks of arthritis and/or periarthritis. There is inadequate data regarding the role of low dose methotrexate in patients with PR. The aim of this study was to determine the efficacy of low dose methotrexate add-on therapy in patients with PR with inadequate response to hydroxychloroquine (HCQ) monotherapy.

Methods

In this observational study, between August 2016 and July 2019, 15 patients with seropositive PR with inadequate response to HCQ (≥ 2 attacks per month even after 6 months of therapy) were included. All patients were treated with oral low dose methotrexate in addition to HCQ. The responses to therapy were monitored 3 and 6 months after adding methotrexate.

Results

In our study, of the 15 patients with PR, two-third were female, with a mean age of 45.73±9.18 years. All were seropositive (Rheumatoid factor positive in eight patients; anti-cyclic citrullinated peptide [CCP] positive in 11 patients; both Rheumatoid factor and anti-CCP positive in four patients). The average number of attacks at baseline was 5.6±2.8, and the median duration of attacks was 2 years. All patients had significant response to add-on methotrexate therapy at follow-up after 3 and 6 months; none developed Rheumatoid arthritis (RA) after 2 years of follow-up.

Conclusion

Low dose methotrexate is an effective add-on therapy in patients with PR having inadequate response to HCQ.

Introduction

Palindromic rheumatism (PR) is characterized by repeated painful attacks of arthritis often located in and around the joints, which are typically severe and unpredictable. Initially described by Hench and Rosenberg (1) in the year 1944, PR typically resolves with no apparent residual deformities. Most patients with seropositive PR may eventually progress to Rheumatoid arthritis (RA) if untreated (2).

The treatment of PR includes chloroquine and hydroxychloroquine (HCQ) with a good response in reduction in frequency and duration of attacks (3). There are reports of delay of progression to RA in the presence of chloroquine (4). There are reports of use of sulfasalazine, D-penicillamine, and gold salts for PR in the past (5–7).

There is one described case of a patient with PR with hypertrophic osteoarthropathy who responded to methotrexate. Literature regarding the use of leflunomide and other combination Disease Modifying Anti-Rheumatic Drugs (DMARDs) are sparse (8). There is one retrospective study regarding the usage of rituximab in PR in refractory cases (9).

There are no published guidelines regarding the treatment and disease monitoring for PR in literature. This study was undertaken to observe response to methotrexate, an add-on therapy, of patients with seropositive PR who had inadequate response to HCQ.

Methods

This prospective observational study was conducted in the Department of Rheumatology, Pradyumna Bal Memorial Hospital, Kalinga Institute of Medical Sciences, KIIT University, Bhubaneswar, India. All patients fulfilled diagnostic criteria set by Hannonen et al. (10). After informed consent, 15 patients with seropositive PR (either Rheumatoid factor positive and/or anti-cyclic citrullinated peptide [CCP] positive), between August 2016 and July 2019, unresponsive to treatment with HCQ (HCQS; Ipca Laboratories Ltd, Mumbai, India) for at least 6 months after therapy were included in the study. Active disease was defined as two or more attacks of arthritis per month, requiring intermittent non-steroidal anti-inflammatory drugs and/or low dose prednisolone. For all the patients with active PR on HCQ, after baseline work up, methotrexate (Folitrax; Ipca Laboratories Ltd, Mumbai, India) was started as an addon therapy in the dose ranging from 7.5 mg to 25 mg per week as per tolerability. The number and duration of painful attacks were recorded after 3 and 6 months of starting methotrexate. If there were one or no attacks, we defined the response as good response to methotrexate. The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were monitored at baseline, at 3 months, and at 6 months. The study was approved by Kalinga Institute of Medical Sciences, Bhubaneswar, India and was performed in accordance with the Declaration of Helsinki (Approval Date: April 6, 2016; Approval Number: KIMS/KIIT/IEC/35/2016).

Statistical analysis

All the continuous parameters were presented with mean and standard deviation as well as median and interquartile range (25^th percentile (p25) and 75^th percentile (p75)). Categorical parameters were presented as frequency (n) and percentage (%). To assess the treatment effect over time, we used three different population average models with exchangeable correlation structure, namely generalized Poisson regression model for count outcome (number of attacks), generalized Gamma regression model with log link function for gamma distributed outcome (duration of attack), and generalized Gaussian linear model with identity link function for normally distributed outcomes, namely ESR and CRP. We have reported unadjusted and adjusted effect for incidence rate ratios for count outcome (number of attacks) and beta coefficients for rest of the outcomes.

Main Points.

• Some patients of Palindromic Rheumatism (PR) can be refractory to Hydroxychloroquine (HCQ) monotherapy.

• Low dose methotrexate is an effective add-on therapy in PR patients with inadequate response to HCQ.

• Low dose methotrexate can prevent development of rheumatoid arthritis (RA) in PR patients.

A p value of <0.05 is considered as statistically significant. All the statistical analysis was done using the standard statistical software Stata 15.1. (Stata Corp College Station; Texas, USA).

Results

In our study, of the 15 patients with PR, two-third were female, with a mean age of 45.73±9.18 years. The average number of attacks at baseline was 5.6±2.8, and the median duration of attacks was 2 years. Four patients had a family history of RA; two of them were male patients who were smokers. The summary of baseline characteristics of the patients is depicted in Table 1. Table 2 shows the changes in clinical parameters (number and duration of attacks) and laboratory parameters (ESR and CRP) at baseline, at 3 months, and at 6 months. Table 3 shows the effect of methotrexate on clinical and laboratory parameters of the patients with PR at 3 and 6 months.

Table 1.

Summary of the background characteristics, symptoms, and treatment.

Characteristics	Description
Age in years, mean±SD	45.73±9.18
Female, n (%)	10 (66.7)
Disease duration (in months), median (p25, p75)	24 (12, 96)
HCQ dose (in mg/day), mean±SD	253.3 (83.38)
Duration of attack (in hours), median (p25, p75)	12 (6, 12)
Methotrexate dose (in mg/week), mean±SD	14.5±4.65
Family history of RA, n (%)	4 (26.67)
Smoker, n (%)	2 (13.33)
RF positivity, n (%)	8 (53.33)
Anti-CCP antibodies positivity, n (%)	11 (73.33)

SD: standard deviation; p25: 25^th percentile; p75: 75^th percentile; HCQ: hydroxychloroquine; RA: rheumatoid arthritis; RF:rheumatoid factor; anti-CCP: anti-cyclic citrullinated peptide.

Table 2.

Changes in the clinical and laboratory parameters over the time period.

Clinical parameters	Baseline	At 3 months	At 6 months
Number of attacks, N (%)
0	0 (0)	0 (0)	11 (73.33)
1–2	1 (6.67)	8 (53.33)	4 (26.66)
≥3	14 (93.33)	7 (46.67)	0 (0.00)
Duration of attack (in hours), mean±SD	34.27±21.49	5.73±3.83	0.2±0.37
Duration of attack (in hours), median (p25–p75)	36 (12–48)	6 (2–8)	0 (0–0.5)
ESR (mm in the first hour), mean±SD	42.72±17.23	36.33±10.82	28.67±8.94
ESR (mm in the first hour), median (p25–p75)	43 (34–52)	38 (30–42)	28 (20–36)
CRP (mg/L), mean±SD	4.12±1.68	3.59±1.31	2.61±0.94
CRP (mg/L), median (p25–p75)	4.5 (2.6–4.8)	3.3 (2.2–4.2)	2.5 (2–3.5)

N: number of patients; SD: standard deviation; p25: 25^th percentile; p75: 75^th percentile; ESR: erythrocyte sedimentation rate; CRP: C-reactive protein.

Table 3.

Effect of methotrexate on clinical and laboratory parameters of the patients with palindromic rheumatism.

Clinical parameters	Effect size (95% CI)
	Unadjusted	p	Adjusted	p
Number of attacks, IRR
At 3 months	0.50 (0.40–0.62)	<0.001	0.50 (0.44–0.57)	<0.001
At 6 months	0.07 (0.04–0.14)	<0.001	0.07 (0.04–0.13)	<0.001
Duration of attack, β
At 3 months	−1.79 (−2.59 to −0.98)	<0.001	−2.00 (−3.00 to −1.00)	<0.001
At 6 months	−5.14 (−5.95 to −4.34)	<0.001	−7.60 (−8.60 to −6.61)	<0.001
ESR (mm in the first hour), β
At 3 months	−6.40 (−12.71 to −0.09)	0.047	−6.40 (−12.71 to −0.09)	0.047
At 6 months	−14.07 (−20.38 to −7.75)	<0.001	−14.07 (−20.38 to −7.75)	<0.001
CRP (mg/L), β
At 3 months	−0.53 (−0.94 to −0.11)	0.013	−0.53 (−0.94 to −0.11)	0.013
At 6 months	−1.51 (−1.93 to −1.10)	<0.001	−1.51 (−1.93 to −1.10)	<0.001

IRR: incidence rate ratio; β: regression coefficient; CI: confidence interval: ESR: erythrocyte sedimentation rate; CRP: C-reactive protein.

Discussion

In this study, we found good therapeutic response to add-on methotrexate in patients with PR unresponsive to HCQ. There was a significant reduction in the number and duration of PR attacks at follow-up at 3 and 6 months. A significant reduction in inflammatory markers (ESR and CRP) at 6 months was also observed. At 2-years short-term follow-up, none developed RA. This study highlights the role of methotrexate as a disease modifying agent in PR similar to RA. A retrospective study by Khabbazi et al. (11) has shown that tight control of PR unresponsive to HCQ with combination DMARDs using either methotrexate, sulfasalazine, leflunomide, or azathioprine provides good symptom control and can prevent progression to RA. To date, there are no consensus guidelines for the treatment of PR owing to paucity of literature and guidelines for the randomized controlled trials owing to its rarity, as opined by other authors (12). Once considered benign, this disease can significantly affect the quality of life owing to the sudden onset of unpredictable painful episodes associated with this condition. A role of B-cell depletion therapy with rituximab has also been found to be an effective therapy for refractory PR by Raghavan et al. (9). However, such a therapy may not be feasible in many developing nations owing to the high cost of therapy as well as high risk of infections associated with rituximab. We suggest that parenteral methotrexate could be an option in patients with PR having inadequate response to oral methotrexate, similar to RA (13).

The risk of evolution of PR to RA is high in patients who are seropositive, especially for those who are anti-CCP positive at onset (14). Treatment with methotrexate can prevent or delay the onset of RA in these patients as we found in our study. Similarly, the use of combination DMARDs or rituximab could be an option as suggested in other studies (9, 11). Hence, we recommend a routine use of methotrexate in patients with PR who are seropositive, especially for those who are anti-CPP positive, to prevent the progression to RA. Recent review by Mankia and Emery (15) suggest that a subgroup of PR may have an autoinflammatory etiopathogenesis along with autoimmunity with options of colchicine and interleukin-1 blockade as therapies to prevent progression to RA. Despite encouraging findings, our study has limitations such as lack of a control group, small sample size, and limited follow-up data. A multicentric randomized controlled trial in a larger cohort is warranted to validate our findings.

In patients with PR with inadequate response to HCQ, an add-on therapy with low dose methotrexate is effective. Other combination DMARDs may be considered in cases of refractory PR.