Management of Established Rheumatoid Arthritis

CASE 1:

A 27-year-old school teacher with a 6-year history of rheumatoid arthritis (RA) is seen for routine follow-up. She was recently married, and her disease had previously been well controlled with 20 mg of weekly oral methotrexate and folic acid 1 mg daily. She currently reports 90 minutes of morning stiffness that is new, and she is having difficulty writing on the chalkboard at school due to worsening pain and stiffness in her hands and wrists. On examination, she has marked synovitis involving the wrists and small joints in both hands.

Suboptimal Response to Methotrexate

CONSIDERATIONS WITH A SUBOPTIMAL RESPONSE TO METHOTREXATE

Methotrexate has revitalized the treatment of RA over the past 20 years, and is widely considered to be the cornerstone of active therapy for the disease. It is frequently the first disease-modifying antirheumatic drug (DMARD) prescribed following RA diagnosis. Although an effective treatment for many, a substantial number of patients (20%–53%) do not adequately respond to methotrexate monotherapy. A number of factors may be implicated in the etiology of an incomplete methotrexate response, including poor gastrointestinal (GI) absorption and reduced bioavailability, possible antagonism by concomitantly administered folate or caffeine intake, and poor patient adherence.

Although a direct relationship exists between methotrexate dose and clinical response, the bioavailability of oral methotrexate has been thought to be a limiting factor in its efficacy. There is significant intra- and interpatient variability in the absorption of orally administered methotrexate. Although much of the pharmacokinetic data specific to methotrexate relate to its use in the treatment of solid tumors, it has been suggested that at a dose of 20 to 25 mg per week or greater (typical dosing in RA) oral administration results in limited bioavailability. With higher doses of methotrexate, the intestinal folate transport system is saturated and a parenteral route (typically subcutaneous) may be required to achieve higher circulating drug levels. The problem of decreased absorption with high-dose oral methotrexate can also potentially be overcome through the use of a split oral dose regimen (e.g., half of prescribed dose in morning and remaining half on evening of same day), a dosing strategy that has been demonstrated to be beneficial in the treatment of solid tumors. Hoekstra et al. demonstrated in 10 RA patients treated with high doses of oral methotrexate (median weekly dose of 30 mg) that improvements in bioavailability could be achieved with a split-dose regimen, recognizing that data supporting the clinical efficacy of this approach are limited.

Alternatively, increased bioavailability of methotrexate can be achieved by switching from an oral route to subcutaneous or intramuscular administration. A small improvement in disease control was achieved by switching to intramuscular administration in patients with active disease despite 15 to 20 mg of oral weekly methotrexate. Using parenteral methotrexate in patients with suboptimal response to maximum oral dosages, may obviate the need for biologics or other forms of step-up therapy. In a retrospective study of 61 patients with juvenile inflammatory arthritis who failed oral methotrexate, a majority (76%) demonstrated significant clinical improvement after being switched to subcutaneous methotrexate. In patients with a suboptimal response to weekly methotrexate at dosages of 20 mg or more, consideration should be given to switching to a split-dose regimen or switching to a parenteral route of administration.

Rheumatologists often prescribe folate supplementation concomitant with methotrexate therapy to minimize or prevent dose-related toxicity. Although the mechanism of action for methotrexate in RA is incompletely understood, some of its adverse effects (e.g., stomatitis, alopecia, and so on) are believed to be related to its effect on folate antagonism. Although folate administration has been associated with a lower incidence of methotrexate-related liver transaminase elevations, and GI and mucosal side effects, it is unknown whether folate supplementation reduces the effectiveness of methotrexate in the treatment of RA. Folate supplementation (1 mg folic acid or 2.5 mg folinic acid) was not associated with significant differences in treatment response with weekly methotrexate in the trial by Van Ede and colleagues, but this study (with 434 patients divided among the three treatment arms) may have been underpowered to detect small, but potentially relevant differences in disease activity. In this study, the group treated with folate supplementation was characterized by a statistically significant higher dose of methotrexate achieved at the end of the study, suggesting that with folate supplementation, higher dosages of methotrexate may be required to achieve a clinical response similar to that without folate supplementation. A separate post hoc analysis of two randomized control trials demonstrated that 9% to 21% fewer patients receiving methotrexate in conjunction with folic acid, achieved an American College of Rheumatology -20, -50 or -70 response when compared with patients who did not receive folic acid. This analysis was limited by the fact that neither of the two studies included in the analysis were performed with the primary goal of determining the effect of folic acid supplementation on disease activity. A randomized, double-blind, placebo-controlled study examining the effect of two different dosages of supplemental folic acid versus placebo on disease activity and toxicity found that the folic acid groups had similar response rates to those receiving placebo. Although there is no consensus on the effect of folate supplementation on the efficacy of methotrexate, there is evidence that appropriate supplementation minimizes the risk of select toxicities. If there is a detrimental impact of folate on the efficacy of methotrexate in RA, it appears to be small and this risk is likely outweighed by the protection conferred against side effects.

Methylxanthines, such as caffeine, have been hypothesized to attenuate the anti-inflammatory effects of methotrexate due to their inhibitory effect on extracelluar adenosine. Although the mechanism of action of methotrexate is not completely understood, it is thought to work in part by increasing the extracellular expression of adenosine, which has potent anti-inflammatory effects. Data on caffeine consumption in a series of 39 newly diagnosed RA patients receiving initial weekly methotrexate suggested that caffeine in excess of 180 mg/day (equivalent to approximately one to two cups of coffee or four servings of caffeinated soda) was associated with decreased response relative to methotrexate-treated patients with a daily caffeine intake of less than 120 mg. In contrast, caffeine consumption in a larger cohort of 264 RA patients was not shown to impact methotrexate response, although there was a trend toward higher disease activity scores in the moderate and high caffeine consumption groups. Although the evidence has not consistently demonstrated a relationship between caffeine consumption and the efficacy of methotrexate, it may be reasonable to counsel patients on the need for moderation in terms of caffeine intake (especially in those patients consuming in excess of 180 mg of caffeine per day).

With suboptimal response to methotrexate (and other DMARDs), issues of patient adherence must be taken into consideration. In a recent study of the Tennessee Medicaid database, Grijalva and colleagues examined the long-term use of several DMARD/DMARD combinations in the treatment of RA, examining patient adherence using a medication possession ratio (MPR; total days supply of medication/total time of observation). Although adherence with methotrexate compared favorably with other DMARDs and DMARD combinations, approximately half of all patients treated with methotrexate monotherapy were nonadherent (corresponding to an MPR ≤ 0.8). These data underscore the importance of patient education and optimal patient-provider communication to successfully address and overcome the potential barriers of treatment adherence.

When faced with a patient with RA who has not achieved the desired clinical response to methotrexate monotherapy, it is important to take into consideration the aforementioned factors that might deleteriously affect the effectiveness of this treatment. Even after careful consideration of these factors, many patients will still not achieve an optimal treatment response to methotrexate and additional DMARD/DMARD combinations will be required to achieve improved disease control.

STRATEGIES FOR PHARMACOLOGIC MANAGEMENT OF PATIENTS WITH RHEUMATOID ARTHRITIS WITH A SUBOPTIMAL RESPONSE TO METHOTREXATE

Although the goal of treatment for patients with established RA was once merely improvement, treatment expectations have expanded with the availability of new and highly effective therapies. Increasingly, the treatment goal in established RA is best characterized by clinical remission. Recognizing that a significant proportion of patients treated with initial DMARD monotherapy will not achieve remission, it is important that RA patients undergo frequent assessments of disease activity in order to best optimize treatment. Several therapeutic strategies, including escalation to triple therapy (methotrexate, sulfasalazine, and hydroxychloroquine) and the addition of biologic agents have been associated with improved clinical responses in patients with a suboptimal response to methotrexate ( Fig. 3-1 ). The availability of agents targeting tumor necrosis factor-α (TNF-α) (etanercept, infliximab, adalimumab) has revolutionized the treatment approach for such patients, because these agents appear to be particularly potent in protecting patients from radiographic disease progression and its associated disability. The treatment armamentarium in established disease has expanded even further in recent years with the availability of agents targeting both B-cell (rituximab) and T-cell costimulation (abatacept)—the first agents shown to be effective in anti-TNF failures.

Although there are many effective treatment strategies from which to choose, recent studies have emphasized that the goal of therapy is as important or more important than the specific therapies used. The goal of therapy should be remission—a disease state characterized by the elimination of joint pain and swelling, the maintenance of employability, a normalization of disability and mortality risk, and the slowing or even reversal of joint damage. The Tight Control in the Treatment of RA Study demonstrated that an intense management strategy with predefined goals and thresholds for escalation of therapy, resulted in substantial improvement in disease activity. Patients exposed to this intensive regimen were approximately 10 times more likely (odds ratio = 9.7; 95% confidence interval [CI] 3.9 to 23.9) than RA patients randomized to routine care to achieve clinical remission. The BeSt (Dutch acronym for Behandel-Strategieen, “treatment strategies”) study expanded on this approach, demonstrating that with implementation of an intensive strategy of monitoring disease activity and rapid escalation to combination DMARD/biologic therapies, 80% of patients achieved a target Disease Activity Score (DAS) of less than 2.4, and 42% achieved clinical remission (DAS < 1.6). Once disease activity was controlled, most patients in this study were able to tolerate a de-escalation of therapy with either elimination of prednisolone or infliximab.

Despite the recent advances in the treatment of established RA, several questions remain. What is the optimal initial strategy (e.g., triple therapy versus the addition of biologic versus other drugs) for patients with established RA and a suboptimal response to methotrexate or other DMARD monotherapy? What is the best strategy for patients with suboptimal treatment responses to second-line agents and various DMARD/biologic combinations? What is the role for newly available and evolving therapies in RA for patients with established disease? Is there an accurate and reproducible way to tailor therapy for individual patients? Can we predict efficacy and toxicity of available treatments? The answers to these and other questions are urgently needed, underscoring the need for additional investigations in this patient population.

CASE 2:

A 61-year-old man with long-standing seropositive RA is seen for follow-up examination. He is on subcutaneous weekly methotrexate (25 mg) in addition to daily sulfasalazine and hydroxychloroquine. He was recently hospitalized with newly diagnosed congestive heart failure (CHF: New York Heart Association Class II) with impaired left ventricular function (ejection fraction of 35%). He was successfully treated for a nosocomial pneumonia during his hospitalization. His renal function has also declined (Cr. 1.8 mg/dL, estimated GFR 42 mL/min). His RA is active with six tender and four swollen joints, an elevated erythrocyte sedimentation rate (54 mm/h), and several hours of morning stiffness.

Addressing Comorbid Illness

RHEUMATOID ARTHRITIS–ASSOCIATED COMORBIDITIES AND COMPLICATIONS

Several RA-associated comorbidities have been identified including cardiovascular disease (coronary artery disease and CHF), infection, osteoporosis leading to fracture, and lymphoproliferative malignancies. RA-related comorbidities and preventive strategies to address them are summarized in Table 3-1 (although not all will be directly discussed here). Management of RA patients is further complicated by treatment-related toxicities including glucocorticoid-induced hyperglycemia, peptic ulcer disease related to nonsteroidal anti-inflammatory agents (NSAIDs), infections related to immunosuppression, and glucocorticoid-induced osteoporosis (GIOP). Additionally, comorbid illnesses not directly related to RA, such as renal or hepatic dysfunction, can substantially affect the available therapeutic options.

Table 3-1

Rheumatoid Arthritis (RA)–Related Comorbidities and Preventive Strategies Used to Minimize Their Impact

	RA Specific Factors	Preventive Measures
Cardiovascular disease	Severe disease with elevated inflammatory markers	Lifestyle modification (diet, exercise, smoking cessation) Prophylactic low-dose ASA Aggressive anti-inflammatory treatments Folate supplementation for methotrexate users Treatment of diabetes, dyslipidemia, hypertension Minimize glucocorticoid and NSAID use
Infection	Severe long-standing disease, immunosuppressive therapy	Minimize immunosuppressive exposure Routine influenza and pneumococcal vaccines TB screening before TNF inhibition Routine surveillance Patient education
Osteoporosis	Glucocorticoid use, severe disease, and immobility	Lifestyle modification (diet, exercise, smoking cessation) Minimize glucocorticoid exposure Calcium/Vitamin D supplementation BMD measurement Bisphosphonate and human recombinant parathyroid hormone (teriparatide) as appropriate
Peptic ulcer disease	NSAID use with or without glucocorticoid use	Minimize NSAID exposure Proton pump inhibitors High-dose H2 receptor antagonists Misoprostol COX-2 selective NSAIDs
Lymphoproliferative disease	Severe long-standing disease	Aggressive anti-inflammatory treatments, disease control (?)

ASA, aspirin; BMD, bone mineral density; COX, cyclooxygenase; H2, histamine type 2; NSAID, nonsteroidal anti-inflammatory drug; RA, rheumatoid arthritis; TB, tuberculosis; TNF, tumor necrosis factor.

Cardiovascular disease (CVD) is the leading contributor to excess mortality in RA, accounting for one third to one half of all RA-related deaths. The risk for incident myocardial infarction and CHF are significantly increased in patients with RA (40%, 60% respectively). The increased morbidity and mortality of CVD in the setting of RA cannot be accounted for by an increase in the prevalence of traditional cardiac risk factors including dyslipidemia, hypertension, diabetes mellitus, family history, and cigarette smoking. Atherosclerosis (and resulting CVD) has been shown to be an inflammatory disease, or at least to have a core inflammatory component in its complex pathogenesis. Proinflammatory cytokines, elevated serum levels of acute phase reactants, neoangiogenesis, T-cell activation, expression of leukocyte adhesion molecules and endothelin, and collagen degradation with localized metalloproteinase expression via activated monocytes are all common pathogenic features to both RA and atherosclerosis. The clinical associations of these two diseases, coupled with their histopathologic similarities, have led some authors to suggest that CVD should be considered an extra-articular manifestation of RA.

In addition to the increased CVD risk inherent to RA, treatments used in managing RA have also been speculated to play a role in the development of CVD. For instance, methotrexate use leads to increased levels of circulating homocysteine, an independent risk factor for accelerated atherosclerosis and thrombotic events. Both NSAID and glucocorticoid use may result in hypertension or worsen pre-existing hypertension, the former being associated with the risk of incident thromboembolic event and heart failure. TNF inhibitors have been associated with increased rates of mortality and need for hospitalization when given to patients with moderate to severe heart failure. In addition to the effects on blood pressure, glucocorticoids may also promote hyperglycemia, increased adiposity, and hyperlipidemia-all independent risk factors for CVD. Recognizing the seminal role of inflammation in CVD, the benefit of select disease-modifying therapies may far outweigh any risk relevant to long-term cardiovascular outcomes. For instance, methotrexate use has been associated with a 70% reduction in cardiovascular mortality (HR = 0.3; 95% CI 0.2 to 0.7) after accounting for differences in RA disease severity, emphasizing the potential importance of disease control in RA patients as a means of reducing CVD burden. Whether other DMARDs and biologic agents share the protective properties of methotrexate remains to be seen.

The increased risk for CVD in patients with RA should prompt aggressive primary and secondary preventive measures including smoking cessation, aggressive control of hyperlipidemia, the appropriate management of hypertension and co-morbid diabetes, and the routine use of prophylactic low-dose aspirin. There is evidence that 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) have anti-inflammatory properties in addition to their lipid-lowering effect, underscoring that these treatments are important adjunctive therapies in RA. Attempts should be made to achieve maximal control of the patient’s RA, while minimizing the use of NSAIDs (particularly those that are highly selective inhibitors of cyclooxygenase-2) and glucocorticoids. Additionally, patients should be frequently screened for symptoms of CVD, helping to identify and intervene early in the course of this comorbid condition.

A significant increase in infection-related morbidity and mortality has been described in RA with an increased disease-related incidence in pulmonary infections, septic arthritis, cellulitis and soft tissue infections, osteomyelitis, and systemic sepsis. The increased infection risk observed in established RA does not appear to be completely attributable to known risk factors including leukopenia, diabetes mellitus, smoking, lung disease, and glucocorticoid use. Alterations in cellular immunity, including decreased numbers of T-suppressor and natural killer cells, which may predispose to infection, are characteristic of RA. Infection risk in RA has been associated with increasing age, extra-articular disease, leukopenia, and select comorbidity (alcoholism, diabetes mellitus, organic brain disease, chronic lung disease).

Similar to CVD, distinguishing between the effects of RA disease severity and its treatments in RA-related infection risk is difficult. Conflicting data have emerged regarding the increased risk of infection with TNF inhibition. A meta-analysis of randomized clinical trials demonstrated an increased risk for serious infections in RA patients treated with monoclonal anti-TNF antibodies, adalimumab or infliximab (relative risk [RR] 2.0; 1.3–3.1) versus placebo. In contrast, large observational registries have not demonstrated an overall increased risk of serious infection related to TNF inhibition, although suggesting that there may be a period of heightened infection risk during the first three to six months of anti-TNF therapy ( Fig. 3-2 ). Although the effects of anti-TNF therapy on risk of infection are not currently fully understood, it is prudent to counsel patients on the risk of infection while receiving any immunomodulators. TNF-α plays a critical role in the development and maintenance of granulomas that are central in host defense against disseminated fungal and mycobacterial infections. All patients should be screened for latent tuberculosis prior to initiation of anti-TNF therapies, and appropriate surveillance should be undertaken during therapy. With first signs of infection, TNF inhibitors and other potent immunosuppressants should be discontinued until the infection has been satisfactorily treated.