Abstract
Background:
In resource-constrained settings where biologic agents are not widely available, there are limited therapeutic options for patients with rheumatoid arthritis (RA) refractory to other synthetic disease modifying antirheumatic (DMARD) therapies. The aim of this study is to evaluate the effectiveness and safety of leflunomide (LEF) with methotrexate (MTX) in refractory RA.
Methods:
A retrospective record review of adult RA patients treated with LEF/MTX. Demographic details, adverse reactions, and the 3-variable 28 joint disease activity score (DAS28-3) were recorded at initiation of LEF/MTX therapy, and after 4 and 12 months of treatment.
Results:
Of 194 patients, most were middle-aged seropositive Black African females, with established disease [mean (standard deviation, SD) disease duration 9.4 (8.2) years] and time on previous DMARDs of 7.0 (5.5) years. Before adding LEF, the mean (SD) dose of MTX was 21.7 (3.5) mg/week, and 87.6% of patients used low dose oral corticosteroids.
A good or moderate EULAR response was achieved by 44% and 42% of patients, and the retention rate was 71%. Major infections were seen in 6 patients: comprising 2 deaths, 3 cases of leucopaenia and septicaemia and 1 case of tuberculosis. Hepatotoxicity (
Conclusions:
LEF/MTX was effective in the majority of patients in this cohort of mainly Black African women who failed other combination synthetic DMARDs, particularly in those with moderate disease activity at the time of addition of LEF. Infections and hypertension were important complications. In a setting where biologic DMARDs are not readily accessible, the combination of LEF/MTX is a cost-effective approach.
Introduction
Adequate control of disease activity in rheumatoid arthritis (RA) is achieved in approximately one-third of patients with synthetic disease-modifying antirheumatic drugs (sDMARDs) including methotrexate (MTX), chloroquine (CQ) and sulfasalazine (SSZ) prescribed either as a monotherapy or combination therapy [Khanna et al. 2007; Hodkinson et al. 2012]. In resource-constrained settings, access to biologic drugs is limited, presenting a major challenge in managing refractory RA.
One approach in patients who have failed triple combination therapy (MTX, CQ and SSZ) is to combine MTX and leflunomide (LEF). These two sDMARDs have different modes of action: MTX inhibits 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR) and purine synthesis, and LEF inhibits pyrimidine synthesis [Kremer, 1999]. LEF was initially registered as a monotherapy for the treatment of active RA and several clinical trials and observational studies have demonstrated that it has equivalent efficacy and tolerability to other sDMARDs, resulting in control of disease activity, improvement in functional disability and retarding radiographic progression [Li et al. 2004; Osiri et al. 2003]. However, there are concerns and documented cases of hepatotoxicity, bone marrow suppression, pneumonitis and an increased risk of infection with combination therapy [Toth and Bernd, 2014; Jenks et al. 2007; Curtis et al. 2010; Rutanen et al. 2014].
The European League Against Rheumatism (EULAR) recommendations suggest that patients with poor prognostic factors (including seropositivity, high disease activity and early radiographic damage) failing to respond to one sDMARD should be switched to a biologic DMARD (bDMARD) rather than combination sDMARDs [Smolen et al. 2010]. The rationale is that such patients often do not respond sufficiently well to combination sDMARD therapy without addition of biological agents. However, in resource-constrained settings, bDMARDs are not widely available because of their cost, and because of the risk of infections, in particular, tuberculosis (TB). In South Africa, biologics are recommended for patients who have an inadequate response to least three sDMARDs over 6 months. Vaccination is recommended for all patients initiating DMARD therapy [Hodkinson et al. 2013]. At present, however, state hospital patients have very limited access to biologic therapy, and vaccines are not consistently available.
We undertook a retrospective record review to evaluate the clinical response, retention rates and toxicity of LEF co-prescribed with MTX (LEF/MTX) in RA patients who had either failed or are intolerant to triple therapy.
Methods
Case records of all RA patients attending Chris Hani Baragwanath Academic Hospital in Soweto, South Africa aged 18 years or older who fulfilled the 1987 American College of Rheumatology (ACR) criteria for RA [Arnett et al. 1988] were reviewed. This hospital is a state sector academic centre servicing the indigent citizens of Soweto, South Africa. Patients who had either failed or were intolerant to triple therapy and subsequently treated with LEF/MTX for ⩾4 months between 2007 and 2012 were included in the study. The study was approved by the medical ethics committee of the University of the Witwatersrand.
Data extracted from case records included demographic details, symptom duration, presence of rheumatoid factor (RF), subcutaneous nodules and radiographic erosions (hands or feet) at diagnosis of RA, details of previous and current DMARD use, and disease activity at the baseline point (LEF/MTX-B) of initiating combination LEF/MTX, at 4 months (LEF/MTX-4), 12 months LEF/MTX-12) and last available review (LEF/MTX-L), and included the 28-joint tender joint count (TJC), swollen joint count (SJC), C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR). Because patient and physician global assessment scores were not available in all cases, the DAS28-3 (CRP) was used as the disease activity metric [Madsen, 2011]. The DAS28-3 scores were categorized into high disease activity (HDA), moderate disease activity (MDA), low disease activity (LDA) or remission, using cutoffs of 2.4, 3.6 and 5.5 [Aletaha et al. 2005]. Additionally, the EULAR response at 12 months was calculated [van Gestel et al. 1996]. Safety features were recorded at regular, frequent intervals (initially monthly, then every 4 months), including patient-expressed side effects, new-onset hypertension and laboratory investigations, comprising a full blood count, serum alanine (AST) and aspartate aminotransferase (AST) levels.
The Student’s
Results
Of more than 4000 RA patient case records reviewed, 194 met the inclusion criteria. The patients were mostly middle-aged seropositive Black African females:180 (92.8%) females, 192 (99.0%) Black Africans, mean (SD) age 40.1 (12.2) years, and 92.3% RF positive (Table 1). The mean (SD) time between symptom onset and RA diagnosis when DMARDs were initiated was 2.3 (2.0) years. In keeping with established RA, 67 (34.5%) patients had erosions on plain radiographs (hands or feet) and 39 (20.1%) patients had subcutaneous nodules at diagnosis. The vast majority (94%) had one or more poor prognostic factor as defined by EULAR.The combination of LEF/MTX therapy was initiated a mean (SD) 7.0 (5.5) years after the diagnosis of RA. Prior to this, all patients had been on MTX, and most were also treated with CQ and SSZ (91.2% and 84.0% respectively), either as a monotherapy or in combination with MTX. All patients were HIV negative at the start of MTX/LEF therapy, and were tested annually. No patients received pneumococcal or influenza vaccines.
Baseline characteristics and details of therapy of 194 patients at start of leflunomide/methotrexate combination therapy.
CQ, chloroquine; DMARD, disease modifying antirheumatic drugs; LEF/MTX, combination leflunomide with methotrexate; MTX, methotrexate; SD, standard deviation; SSZ, sulfasalazine.
At LEF/MTX-B, patients had active disease: the mean (SD) DAS28-3 was 4.9 (0.9), and the majority of patients were in MDA (75.8%) or HDA (23.2%) categories. Patients were prescribed LEF 20 mg daily, and no loading dose was given. At LEF/MTX-B, the dose of MTX was 10–15 mg weekly, and then escalated as necessary: at LEF/MTX-L the mean (SD) dose of MTX was 21.7 (3.5) mg/week. The majority (87.6%) of patients were also on low dose (⩽7.5 mg/day) oral corticosteroids and all received folate supplementation.
At LEF/MTX-L 137 (70.6%) were still on LEF/MTX therapy and the mean (SD) duration on this regime was 30.7 (19.2) months. LEF was discontinued in 39 (20.1%) patients, two patients (1.0%) died whilst on LEF therapy, and 16 (8.3%) patients were lost to follow up.
Response to LEF
There was a significant decline in disease activity over the first 4 months of LEF/MTX therapy (DAS28-3 of 4.9 at LEF/MTX-B
Changes in disease activity of patients on leflunomide/methotrexate combination therapy.
for changes from baseline (time of addition of LEF) to 12 months.
CRP, C-reactive protein; DAS28-3 disease activity score for 28 joints (3 variable version); ESR, erythrocyte sedimentation rate; HDA, high disease activity defined as DAS28-3> 5.5; LDA, low disease activity defined as DAS28-3 ⩽ 3.6; LEF, leflunomide; MDA, moderate disease activity defined as DAS28-3 ⩽ 5.5; Remission, defined as DAS28-3 ⩽ 2.4; SJC, swollen joint count; TJC, tender joint count.
Predictors of good response to leflunomide/methotrexate combination at 12 months. Remission/LDA was defined as a DAS28-3 score ⩽ 3.6, and MDA/HDA was defined as a DAS28-3 score > 3.6.
CRP, C-Reactive Protein; DAS28-3, disease activity score for 28 joints (3 variable version); DMARD, disease-modifying antirheumatic drug; ESR, erythrocyte sedimentation rate; HDA, high disease activity; LDA, low disease activity; LEF, leflunomide; MDA, moderate disease activity; MTX, methotrexate; ns, not significant; RF, rheumatoid factor; SJC, swollen joint count, TJC, tender joint count.
Predictors of response to LEF
The only significant predictor of response to LEF/MTX therapy was baseline disease activity (Table 3). Patients who achieved remission or LDA at LEF/MTX-12 had lower baseline DAS28-3 compared with those in MDA or HDA (4.8 (0.9)
Adverse events and discontinuation
A total of two patients died while using LEF/MTX of lower respiratory tract infection, after 24 and 28 months of therapy. Neither patient was noted to have leucopaenia nor abnormal liver function tests at the visits prior to these events.
LEF treatment was permanently discontinued in 39 (20.1%) patients. Reasons for discontinuation (Table 4) included perceived drug toxicity in 17 patients, inadequate response to therapy, nonadherence to treatment and patients wanting to become pregnant.
Reasons for discontinuation of leflunomide/methotrexate combination therapy.
Hepatotoxicity defined as AST or ALT >3 times upper limit of normal.
Severe vomiting and diarrhea.
Other side effects included: headache, hot flushes, paresthesia, and sleep disorders.
Amongst the 17 patients who discontinued LEF due to toxicity, leucopaenia, hepatotoxicity, and intolerable gastrointestinal symptoms were the most common problems. Overall, three patients (1.5%) developed hepatotoxicity (defined as persistent elevated AST/ALT >3 the upper limit of normal) and required discontinuation of LEF. There were three patients (1.5%) that developed leucopaenia with sepsis (all after completing 12 months of LEF/MTX), but all recovered after LEF was stopped. One patient was diagnosed with pulmonary TB. There was one patient that presented with acute onset dyspnoea, with chest X-ray infiltrates in keeping with pneumonitis after 14 months of LEF/MTX therapy, and this resolved with antibiotic therapy and discontinuation of LEF/MTX. There were no significant predictors of toxicity. The use of corticosteroids did not have an impact on infection, hypertension, nor any other adverse event.
In the remaining patients LEF/MTX was generally well tolerated. Mild transient gastrointestinal complaints were experienced in the initial weeks of LEF/MTX therapy by 13 (6.7%) of patients. Transient mild leucopaenia was seen in two patients, and transient mild transaminitis was seen in one patient. New-onset hypertension was noted in 7 patients after 4 months and 10 patients after 12 months. No patients had accelerated hypertension, and all responded to standard antihypertensive therapy.
A total of 16 patients were lost to follow up. Of these, five were ‘early defaulters’ (within 12 months of starting LEF therapy), and they showed a poor response to LEF/MTX at their last visit (four had HDA, one MDA). The 11 late defaulters (mean (SD) 24.4 (14.5) months of LEF/MTX were doing better: 5 were in remission or LDA, and 3 had MDA. None had documented complaints in their files.
Discussion
In this retrospective study of predominantly Black South African patients, we found the LEF/MTX combination therapy to be effective, with minimal toxicity. Almost two-thirds of patients achieved remission or LDA, and all disease activity parameters and the overall DAS28-3 improved significantly. These results are notable, particularly given that the majority of patients (94%) had one or more poor prognostic factor as defined by EULAR [Smolen et al. 2010], and all had established disease (>2 years symptom duration).
Disease activity improved rapidly on the LEF/MTX combination within 4 months and was sustained at 12 months and at the last visit. This rapid response to LEF/MTX has been shown elsewhere [Kellner et al. 2010]. The retention rate on LEF/MTX therapy was good, with more than two thirds (70.6%) of patients still on the combination therapy at the end of the observation period, which is somewhat better than reported previously elsewhere [Chopra et al. 2008; Aletaha et al. 2003].
In previous studies in other populations, the main concern with combination LEF/MTX therapy has been potentially life-threatening side effects, in particular hepatotoxicity and leucopaenia (Table 5) [Gupta et al. 2011; Antony et al. 2006; Curtis et al. 2010; Kremer et al. 2002; Weinblatt et al. 1999; Bird et al. 2013]. Mild elevation of liver enzymes has been reported in up to 45% of patients, with marked elevation necessitating discontinuation of therapy in <10% and some cases of death due to liver failure [Weinblatt et al. 2000]. In the present study, we observed very little hepatotoxicity. Similarly, no hepatotoxicity was encountered in a recent prospective study of predominantly Black South African RA patients where 14% of patients were on LEF/MTX [Hodkinson et al. 2015]. One likely explanation is that liver toxicity from MTX is seen less in Black Africans, possibly related to the rarity of MTHFR gene polymorphism that has been shown to be associated with MTX in Whites [Tikly et al. 2003].
Severe side effects of combination LEF/MTX therapy.
ILD, interstitial lung disease; LEF, leflunomide; LFT, liver function tests; MTX, methotrexate; PsA, psoriatic arthritis; RA, rheumatoid arthritis; Rct, randomised controlled trial; sDMARD, synthetic disease-modifying antirheumatic drug; ULN, upper limit of normal.
Infection caused death in two patients, discontinuation in three patients, and one patient was diagnosed with pulmonary TB. In keeping with this, infection-related deaths occurred in two patients in each of two previous studies at our centre, and LEF-associated TB has been reported elsewhere [Hodkinson et al. 2012, 2015; Verma et al. 2013]. The majority of our patients were prescribed low-dose oral corticosteroids, and this may have contributed to the high infection rate, although corticosteroids as a risk for infection did not reach statistical significance. Of note, MTX/biologic combination therapy is also associated with an increased a risk of serious bacterial infection. Thus, physicians treating severe RA in the developing world setting need to screen and remain vigilant for infections. Adherence to immunization guidelines, in particular pneumococcal and influenza vaccines, is likely reduce the incidence of infection.
Hypertension was newly diagnosed in 8.8% of patients and is a well-documented side-effect of LEF [Rozman et al. 2002], highlighting the importance of blood pressure monitoring.
As with most retrospective case record reviews, our study has limitations. Patient and physician global assessments were not recorded at all visits; hence the DAS28-3 was used to assess disease activity. In addition, we did not crosscheck extracted data. A high number of patients (8.2%) patients were lost to follow up. The early defaulters were responding poorly to LEF, and may have sought treatment elsewhere. A further limitation is that we did not evaluate X-rays before and after LEF/MTX therapy, and thus have no data on prevention of radiographic damage in these patients.
In summary, LEF in combination with MTX is effective for RA patients who do not respond to MTX or other sDMARDs, particularly for patients with moderate disease activity at the time of addition of LEF. The LEF/MTX combination was relatively well-tolerated with a good retention rate, although infections were the most significant problem, and two deaths and three cases of sepsis were observed. We suggest LEF/MTX is a useful strategy in a resource-constrained setting where biologic drugs are not widely available. Our findings also highlight the need for biologic therapy for a sizeable proportion of patients with established RA that is refractory to all sDMARDs.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement
The authors declare that there is no conflict of interest.