Delay to institution of therapy and induction of remission using single-drug or combination–disease-modifying antirheumatic drug therapy in early rheumatoid arthritis

Abstract

Objective

To study the impacts of 1) the delay from the onset of symptoms to the institution of disease-modifying antirheumatic drug (DMARD) therapy, 2) two treatment strategies (treatment with a combination of DMARDs or with a single drug), and 3) the presence of HLA–DRB1 alleles (shared epitope) on the prediction of disease remission after 2 years in patients with early rheumatoid arthritis (RA).

Methods

In the FIN-RACo (FINnish Rheumatoid Arthritis Combination therapy) trial, 195 patients with recent-onset RA (median duration 6 months) were randomly assigned to receive either 1) a combination of DMARDs (sulfasalazine, methotrexate, hydroxychloroquine, and prednisolone) or 2) a single DMARD with or without prednisolone. The presence of a shared epitope was tested for in 165 of the 178 patients completing the study. The additional variables of age, sex, presence of rheumatoid factor, number of fulfilled American College of Rheumatology criteria for the classification of RA, and length of delay from onset of symptoms to institution of therapy were entered into a logistic regression model to determine the significant predictors for remission at 2 years.

Results

The delay to therapy (cut point of 4 months) was the only significant predictor for remission in patients treated using the single-DMARD strategy, while no variable was a significant predictor for remission in those treated using the combination-DMARD strategy. The frequency of achieving remission in the combination-DMARD group after 2 years was similar in patients with short (0–4 months) and long (>4 months) delay periods (11 of 26 patients and 22 of 53 patients, respectively [∼42% in each group]), while the corresponding frequencies in the single-DMARD group were 8 of 23 patients (35%) and 7 of 63 patients (11%) (P = 0.021). The presence of a shared epitope was not related to the induction of remission.

Conclusion

The delay of a few months from the onset of symptoms to institution of therapy decreases the ability of the traditional single-drug strategy to induce remission in early RA.

Spontaneous remissions are rare in established rheumatoid arthritis (RA) and, in most cases, the disease shows a resistant course with progressive joint damage and increasing disability (1). Accumulating evidence indicates that the course of RA is determined early and that the optimal management of RA requires a swift diagnosis and immediate aggressive treatment (2). Randomized clinical trials have shown that conventional single-drug therapy with disease-modifying antirheumatic drugs (DMARDs) and glucocorticoids can alter the clinical course of RA (3, 4). Early, rather than late, institution of therapy with DMARDs has also been suggested to be more effective in prevention of joint damage and in induction of clinical remission (5, 6). Further, the induction of remission, the ultimate target of RA therapy, may be facilitated by the use of more aggressive treatment policies in the early phases of disease (e.g., by simultaneous therapy with multiple DMARDs) (7). However, it is of utmost interest whether the delay from the onset of symptoms to the institution of DMARD therapy can affect the prognosis of RA. Further, it has also been suggested that the presence of HLA–DR4 or the “shared epitope” may impact negatively on outcome in RA patients (8).

In the FIN-RACo (FINnish Rheumatoid Arthritis Combination therapy) trial, therapy using the combination strategy with sulfasalazine (1–2 gm/day), methotrexate (7.5–15 mg/week), hydroxychloroquine (0.3 gm/day), and prednisolone (5–10 mg/day) induced more remissions than did therapy using the single-drug strategy with or without prednisolone (37% versus 18% at 2 years; P = 0.03) in DMARD-naive patients with early RA (median duration 6 months) (9). In the present study, we analyzed the impacts of the delay from the onset of symptoms to the institution of therapy with DMARDs and the presence of HLA–DRB1 alleles on the prediction of disease remission in the patients enrolled in the FIN-RACo trial.

PATIENTS AND METHODS

Selection of patients and study design.

From April 1993 to May 1995, 199 DMARD-naive patients with recent-onset RA (duration <2 years) were recruited for a multicenter, parallel-group, randomized study comparing the efficacy and tolerability of combination-DMARD therapy (simultaneous sulfasalazine, methotrexate, hydroxychloroquine, and prednisolone) with those of single-DMARD therapy (initially, sulfasalazine 1,000 mg twice daily with or without low-dose prednisolone). Combination therapy was started with sulfasalazine 500 mg twice daily, methotrexate 7.5 mg/week, hydroxychloroquine 300 mg/day, and prednisolone 5 mg/day, but the protocol allowed flexible subsequent dose adjustments to mimic clinical practice. Oral prednisolone was prescribed for 63 patients in the single-DMARD therapy group (according to the clinicians' decisions). The details of the study have been described previously (9).

The patient selection criteria were as follows: 1) fulfillment of the American College of Rheumatology (ACR; formerly, the American Rheumatism Association) 1987 revised criteria for RA (10), 2) age 18–65 years, 3) duration of symptoms <2 years, and 4) active disease with ≥3 swollen joints and at least 3 of the following: (a) erythrocyte sedimentation rate (ESR) of ≥28 mm/hour or C-reactive protein (CRP) level of >19 mg/liter, (b) morning stiffness of ≥29 minutes, (c) >5 swollen joints, or (d) >10 tender joints. Disease duration was defined as the time from the onset of symptoms and was identical to the delay to the institution of first DMARD therapy in this study.

One hundred ninety-five patients started therapy; 97 received combination DMARD therapy and 98 received single-drug therapy. Ten patients in the combination group and 7 patients in the single-DMARD group withdrew from the study. The HLA–DRB1*04 and *01 alleles (shared epitope) were determined by a sequence-specific polymerase chain reaction amplification for 165 of the 178 patients completing the FIN-RACo trial (79 patients in the combination-treatment arm and 86 patients in the single-treatment arm). The data reported here are derived from the completers. The patients were assessed clinically at the beginning of the study and after 1, 3, 4, 5, 6, 9, 12, 18, and 24 months. Swollen and tender joint counts, as well as several other variables, including ESR and CRP level, were used to assess disease activity. Rheumatoid factor (RF) was measured at the beginning of the study. Radiographs of the hands and feet were obtained at baseline and at 24 months (available for 78 combination-treated and 81 single drug–treated patients), and erosions were scored by the method of Larsen et al (11).

The main end point was the induction of remission. The ACR preliminary criteria for remission (12) were applied. While the fatigue and duration criteria were excluded, all other criteria had to be fulfilled, including 1) morning stiffness of ≤15 minutes, 2) no joint pain (by history), 3) no joint tenderness or pain on motion, 4) no soft-tissue swelling in joint or tendon sheaths, and 5) ESR of <30 mm/hour in women and <20 mm/hour in men. The extent of radiographic joint damage was our secondary end point.

The study was performed according to the principles of the Declaration of Helsinki. The protocol was approved by the national health authorities and ethics committees in all 18 participating hospitals. All patients gave written informed consent.

Statistical analysis.

The descriptive values of variables were expressed as the mean ± SD and as the median and interquartile range (IQR). Variables with normal distribution were analyzed by Student's t-test. Variables with non-normal or ordinal distribution were analyzed with the Mann-Whitney test. Categorical data were analyzed by chi-square test or Fisher's exact test. For predicting remission in each treatment group at 2 years, a logistic regression model was used. No adjustments were made for multiple testing.

RESULTS

The median delay to the institution of DMARD therapy was 6 months (IQR 4, 9) for the patients in the combination-treatment arm and 7 months (IQR 4, 11) for the patients in the single-treatment arm. Patients with short delay (<4 months) included 26 of 79 (33%) in the combination-treatment arm and 23 of 86 (27%) in the single-treatment arm.

The baseline demographic, clinical, and laboratory characteristics of the patients with short (0–4 months) and long (>4 months) delays to each treatment strategy were comparable, as shown in Table 1. Further, the median Larsen score was significantly higher in the long-delay group than in the short-delay group in each treatment arm (P < 0.01), while the median ESR was significantly higher in the short-delay group only within the combination-treatment arm (42 mm/hour versus 26 mm/hour; P = 0.031) (Table 1).

Table 1. Baseline demographic, clinical, and laboratory characteristics of 165 patients with early RA classified according to short (0–4 months) and long (>4 months) delays to therapy with combination-DMARD and single-DMARD regimens*

Variable	Delay to therapy with DMARDs
	Combination treatment		Single treatment
	Short (n = 26)	Long (n = 53)	Short (n = 23)	Long (n = 63)
* RA = rheumatoid arthritis; DMARD = disease-modifying antirheumatic drug; IQR = interquartile range; HAQ = Health Assessment Questionnaire; ESR = erythrocyte sedimentation rate; RF = rheumatoid factor; ACR = American College of Rheumatology.
Women, no. (%)	14 (54)	32 (60)	15 (65)	42 (67)
Age in years, mean ± SD	48 ± 10	46 ± 10	50 ± 9	47 ± 11
Swollen joint count, median (IQR)	14 (8, 16)	13 (9, 16)	14 (10, 17)	13 (10, 15)
Tender joint count, median (IQR)	18 (12, 25)	15 (12, 19)	17 (13, 24)	17 (13, 25)
HAQ score, median (IQR)	1.00 (0.50, 1.40)	0.75 (0.38, 1.00)	0.62 (0.38, 1.25)	0.88 (0.38, 1.25)
ESR in mm/hour, median (IQR)	42 (20, 63)	26 (16, 46)	38 (27, 70)	35 (21, 51)
RF positive, no. (%)	20 (77)	38 (72)	16 (70)	44 (70)
HLA–DRB1*04 or *01 present, no. (%)	19 (73)	43 (81)	15 (65)	50 (79)
ACR RA criteria fulfilled, median (IQR)	5 (5, 6)	5 (5, 6)	5 (4, 5)	5 (4, 6)
Larsen score, median (IQR)	0 (0, 0)	2 (0, 6)	0 (0, 1)	2 (0, 10)

At the end of the study, 33 of 79 patients (42%) and 15 of 86 patients (17%) originally assigned to the combination- and single-treatment arms, respectively, were in remission (P = 0.001). The logistic regression analysis was used to predict remission in each treatment regimen. The independent variables of age, sex, RF positivity, number of other ACR RA criteria fulfilled, presence of shared epitope, and delay to therapy with DMARDs (0–4 months versus >4 months) were entered into the model.

The delay to therapy was the only variable that was highly significant in predicting remission at the 2-year visit in the single drug–treated patients, while no variable was a significant predictor for remission in the combination-treated patients. The frequencies of patients with remission of their disease were similar in the long- and short-delay groups treated with the combination regimen (22 of 53 and 11 of 26, respectively; ∼42% in each group), while the respective frequencies were 11% (7 of 63) and 35% (8 of 23) (P = 0.021) in the single-treatment arm (Figure 1). The difference remained statistically significant after adjustment for genetic (shared epitope) and baseline (RF positivity, age, sex, number of other ACR RA criteria fulfilled) variables, as shown in Figure 1. This finding did not disappear when the cut point of delay was set at any of the time points from 3 months to 6 months (data not shown). Inclusion of data from the Health Assessment Questionnaire (HAQ) (13) in the logistic regression analysis did not change the main result. Further, the HAQ scores at baseline were not significant predictors for remission either in the single- or in the combination-treatment arm (data not shown).

Figure 1. Crude and adjusted frequency of achieved remission in the short (0–4 months)– and long (>4 months)–delay groups of patients with early rheumatoid arthritis (RA) treated with combination or single disease-modifying antirheumatic drugs. Data are adjusted for age at baseline, sex, presence of shared epitope, rheumatoid factor positivity, and number of other American College of Rheumatology RA criteria fulfilled.

The median Larsen score increase in the single drug–treated patients (7.0; IQR 1.0, 16.0) was more pronounced than that in the combination-treated patients (1.5; IQR 0, 10.5) (P < 0.001). The median Larsen score increased during the study in the combination-treatment group by 2 (IQR 0, 11) and 2 (IQR 0, 9) in patients with short and long delays, respectively, to DMARD treatment (P = 0.66). The corresponding median Larsen score increases in the single-treatment group were 3 (IQR 0, 18) and 10 (IQR 2, 16) (P = 0.41). At the end of the study, the increase in Larsen score was less in the combination-treated patients whose disease was in remission (0; IQR 0, 2) than in the other combination-treated patients (2; IQR 0, 12.5) (P = 0.005) (Table 2).

Table 2. Median increase of Larsen score during 2 years in patients with early RA achieving or not achieving remission with combination-DMARD or single-DMARD strategy at 2-year visit*

Treatment strategy (n)	Increase in Larsen score, median (IQR)		Difference (95% CI)†	P
	Patients achieving remission	Patients not achieving remission
* 95% CI = 95% confidence interval (see Table 1 for other definitions). † Hodges-Lehman estimation of shift in location between two groups.
Combination DMARD (78)	0 (0, 2)	2 (0, 12.5)	−2 (−10, 0)	0.005
Single DMARD (81)	2 (0, 10)	10 (2, 16.5)	−4 (−10, 0)	0.057

DISCUSSION

Treatment of RA should aim at early clinical remission. There is a lack of data from randomized trials regarding the impact of the treatment strategies on induction of clinical remissions in early RA. In addition to different treatment strategies, the use of various single DMARDs may impact outcome in patients with early RA (14). Recently, the FIN-RACo trial and the COBRA (Combinatietherapie Bÿ Reumatoide Artritis) study have shown that therapy with a combination of several DMARDs and prednisolone results in better outcomes in patients with early RA than does treatment with a single DMARD (9, 15). However, there exists no common consensus as to whether a combination therapy with several DMARDs should be instituted as the first therapy in all patients newly diagnosed as having RA. In contrast, it has been recommended that a combination therapy with DMARDs should be considered only in patients with residual inflammation after maximum doses of single agents (4). On the one hand, our present data indicate that if the delay to the institution of therapy is >3–6 months from the onset of symptoms, the single-drug strategy is less potent for inducing remissions than in patients with shorter periods of symptoms. On the other hand, the ability of the combination regimen to induce remission was not related to the delay from the initial symptoms to the start of therapy.

The delay to therapy was reasonably short (median 6 months) in our patients. The results were most evident when the cut point of the delay to therapy was set at 4 months. Unfortunately, our own experience and other reported data indicate that a patient with recent-onset arthritis can seldom be admitted to specialist care before 3–6 months have elapsed from the start of symptoms (16). Thus, our data suggest that the combination of DMARDs as an initial therapy in early RA should be considered more frequently than it is at present.

The primary end point of the present study was clinical remission at the 2-year visit. To confirm that the disease course was more favorable in the patients with remission at 2 years than in the others, the extent of radiographic joint damage was assessed. The joint damage score in the patients with remission at 2 years was significantly lower than in those not reaching remission. The difference was especially prominent for patients in the combination-treatment group. Thus, the impact of the choice of the initial treatment strategy on the disease course may be sustained, although in some patients the clinical remission will be lost in the future. Nevertheless, the median Larsen score increases within either the single-treatment group or the combination-treatment group did not differ with regard to the delay of DMARD therapy.

In our study, the single-drug therapy was started with sulfasalazine, but the protocol required that it had to be replaced with methotrexate (7.5–15 mg/week) either if an adverse event occurred or if the clinical response was <25% at 6 months. Thus, 51 of the 98 original patients in the single-treatment arm were also treated with methotrexate during the study (9). Two previous randomized, double-blind, prospective studies did not show any significant difference in efficacy among methotrexate, sulfasalazine, or the combination of both of these drugs in early RA (see ref. 7). Thus, we doubt whether the efficacy of our single-treatment regimen would have been better if methotrexate had been instituted instead of sulfasalazine as the first DMARD.

There are conflicting data regarding the role of HLA–DRB1 alleles in disease course and outcome in early RA. In this study, the prediction of remission on an individual basis at baseline was not possible. Only the patients allocated to the combination-treatment arm and those with a short delay to therapy (<6 months) in the single-treatment arm proved to be inclined to achieve clinical remission more frequently than the others. In accordance with results reported previously, the presence of a shared epitope had no impact on the disease outcome in early RA (17).

The prerequisite for early initiation of therapy is the rapid admission of patients to specialist care. Thus, postgraduate education of general practitioners concerning the diagnosis, treatment, and outcome options of RA should be promoted actively. We conclude that patients with active RA and a period of symptoms exceeding 3–6 months are candidates for treatment with a combination of DMARDs, including methotrexate, sulfasalazine, hydroxychloroquine, and prednisolone.