Arthritis afflicts over 40 million Americans, approximately 21 million of whom have clinical signs and symptoms of osteo-arthritis (OA) and 2.1 million have rheumatoid arthritis (RA). Seven million people with arthritis have activity limitations; and for more than three million patients, symptoms are so severe as to interfere with ordinary daily functions. Collectively, Americans with arthritis total 1.5 billion days of restricted activity, 315 million visits to physicians, and eight million hospital admissions
annually. Both prevalence and activity limitation are increasing and expected to increase significantly by 2020.
The therapeutic goal in arthritis is to reduce pain and inflammation in order to restore activity and minimize hospitalization. A succession of new agents has replaced acetylsalicylic acid as first-line treatment, with the pharmaceutical industry attempting to improve efficacy and toxicity profiles. Not until the first blockading agent specific to cyclooxygenase-2 receptors (celecoxib, Celebrex®) gained FDA approval in 1993 were these twin goals achieved. With the subsequent release of rofecoxib (Vioxx®), physicians now a have choice of twoCOX-2-specific inhibitors.
This program was sponsored by the Center for Health Care Education, Inc., and supported by an unrestricted educationalgrant from Pharmacia.
The Role of the Family Physician in Treating Arthritis
A study published in 1995 concluded that the annual direct cost of musculoskeletal conditions in the United States was $72.3 billion in 1992 dollars and that an additional $77.1 billion was lost in wages, equating the total cost to 2.5 percent of the gross national product. (Yelin. Arthritis Rheum. 1995;38:1351.) Only one percent of direct cost was for the purchase of drugs, with most of the expense consumed by physicians and other providers, and by hospitals and long-term care facilities.
Age and gender are important risk factors for OA. Robert E. Rakel, MD (Baylor College of Medicine) reported that between 80 and 90 percent of individuals show some X-ray evidence after age 65 and that more than 80 percent have symptoms of OA after age 75. Women are affected approximately twice as often as men. Patients with OA typically present with morning pain usually lasting not more than 30 minutes, and with stiffness and limitation of motion. Common clinical signs include crepitus, bony hypertrophy or tenderness, limitation of range of motion, malalignment, and altered gait. The peak period of discomfort is likely to be after prolonged use. OA has no systemic signs or symptoms.
RA is a chronic, progressive, systemic inflammatory disease that has a peak incidence between ages 20 and 50 years. Women are affected two to three times as often as men. Prevalence is high among low-income populations and among individuals who have less education. Articular signs and symptoms include polyarticular involvement that is often symmetrical, joint swelling and tenderness, limitation of motion, malalignment of joints, and long-lasting pain, often at rest. Common extra-articular signs and symptoms are rheumatoid nodules, vaculitis, carditis, pericarditis, pulmonary fibrosis, and ocular disease. At the systemic level, fever, involuntary weight loss, fatigue, anemia, and whole-body stiffness are common. Peak discomfort is likely to follow prolonged inactivity.
The two diseases differ in anatomical pattern. RA is much more likely to involve feet. OA is somewhat more common in knees. Both involve the cervical spine and shoulders, but OA is more likely to involve the lumbar spine and hips. Hand involvement in RA is predominately in the proximal inter-phalangeal and metacarpal joints, whereas OA is more likely to involve proximal and distal inter-phalangeal joints, but not metacarpal joints.
An Overview of COX-2 Science and Technology: Pharmacology and Efficacy
Traditional non-steroidal anti-inflammatory drugs (NSAIDs) have been linked to as many as 105,000 hospitalizations and 16,500 deaths per year from gastrointestinal bleeding and from a variety of hepatic, renal, and other toxicities. Between 10 and 30 percent of individuals are intolerant to NSAIDs. Fifteen percent of users have abnormal liver function tests, and ulcerations are seen endoscopically in 10 to 30 percent after two to four weeks of use. "The half-life of an NSAID correlates better with its toxicity profile dose than with its efficacy, dose schedule, or COX-2 inhibition," according to John A. Magaldi, MD of Torrington, CT.
Unlike opioid agents, which relieve nociceptive pain centrally through the spinal thalamic tracts, NSAIDs operate predominately at the periphery. Activation of COX-2 receptors by pain stimuli up-regulates the receptors, producing inflammation and pain. NSAIDs work through the breakdown or arachidonic acid and cyclooxygenase to produce prostaglandins that have homeostatic effects on platelets, GI mucosa, kidneys, and so forth that are mediated by COX-1 receptors. Because traditional NSAIDs have affinity for both receptor sub-types, they serve as analgesics by blockade of pain mediation at COX-2 receptors but offset prostaglandin-mediated homeostasis by also blockading COX-1 receptors. The newer COX-2-specific inhibitors spare the COX-1 receptors, reducing inflammation and pain while preserving the constitutive and homeo- static functions. This is achieved by designing a molecule that exploits subtle structural differences in the receptor sub-types. Figure 1 compares celecoxib with ibuprofen and naproxen with respect to receptor affinity.
Celecoxib reaches peak plasma levels in three hours or less, achieves steady-state concentration within five days, and can be administered for OA and RA without regard to timing of meals. Celecoxib causes a 17 percent increase in lithium levels, and serum levels may be doubled by fluconazole. The FDA's caution against using celecoxib along with sulfonamide agents is based not on known adverse reactions, but on the fact that the reaction has not been tested. It is likely that the combination will prove safe at least in non-arylamine sulfonamide-allergic patients.
Rofecoxib can also be administered without regard to meal timing. Its longer half-life is not a significant feature because half-life has minimal effect on the pharmacodynamics of NSAIDs. When taken at a dose of 25 mg concurrently with ACE inhibitors, rofecoxib causes a mean arterial pressure increase of 3 mmHg. Rifampin decreases rofecoxib serum concentrations by approximately half. A dose of 75 mg of rofecoxib taken for 10 days may increase methotrexate concentration by as much as 23 percent, so patients should be monitored for methotrexate levels when the drugs are used concurrently.
COX-2-specific inhibitors have been submitted to several efficacy trials. In a 24-week comparison trial of celecoxib 200 mg twice daily and diclofenac SR 75 mg twice daily by an RA population, patient assessment of the two agents was virtually identical, as was the mean reduction in number of tender joints. Similar results emerged from a 6-week OA knee trial using celecoxib 100 mg twice daily or diclofenac 50 mg twice daily. In a dose-comparison trial in which OA patients reported responses to knee pain after six weeks of treatment with celecoxib 100mg, celecoxib 200 mg, or placebo, both doses were significantly more efficacious than placebo but not significantly different from one another. From the standpoints of cost and convenience, however, it appears that 200 mg daily is preferable to 100 mg twice daily.
A 52-week Phase III trial compared OA patient response to rofecoxib 12.5 mg, rofecoxib 25 mg, and diclofenac 150 mg based on improved pain scores when walking on a flat surface. No significant differences were recorded from the three regimens, as shown in Figure 2.
The first head-to-head comparison of celecoxib and rofecoxib was a 6-week trial of pain relief from OA knee flare using celecoxib 200 mg daily, rofecoxib 25 mg daily, or placebo. Mean pain levels decreased significantly with both trial drugs compared with placebo, but there was no significant efficacy difference between the two agents at three weeks or six weeks. (Data on file, Searle Corpor-ation, a division of Pharmacia.)
This and other studies have also produced tolerability data, particularly with respect to upper GI distress. Doses of celecoxib ranging from 50 mg twice daily to 400 mg twice daily induced significantly fewer incidents of distress (abdominal pain, nausea, dyspepsia) than naproxen 500 mg twice daily in a 12-week trial. A 6-month OA trial comparing three doses of rofecoxib with ibuprofen 2400 mg and diclofenac 150 mg revealed a dose-related incidence of distress (acid reflux, dyspepsia, epigastric discomfort, heartburn, nausea, vomiting) from rofecoxib that was significantly lower than from ibuprofen and comparable to diclofenac at low and intermediate doses. In a 6-week, placebo-controlled OA knee trial, the incidence of upper GI events from celecoxib 200 mg daily was comparable to placebo, but significantly more events occurred with rofecoxib 25 mg daily.
Dr. Magaldi concluded that celecoxib 200 mg daily, in either single or divided doses, is effective for treating the pain and inflammation associated with OA and RA, that rofecoxib 12.5 mg per day or 25 mg per day is effective in OA, and that celecoxib 200 mg and rofecoxib 25 mg daily have comparable efficacy in OA. Celecoxib has similar upper GI tolerability to placebo and is better tolerated than both naproxen and rofecoxib. He also emphasized the safety of COX-2-specific inhibitors in patients with gastroesophageal reflux disease (GERD) and irritable bowel syndrome. Because they do not affect platelet aggregation as do the traditional NSAIDs, future trials will probably demonstrate the safety of COX-2-specific inhibitors in patients on anticoagulant therapy. They may also prove to be both safe and efficacious in patients with asthma
Arthritis and Beyond: Long-Term GI Safety and Tolerability of COX-2-Specific Inhibitors
As reported by Jay L. Goldstein, MD (University of Illinois at Chicago), 2 to 4 percent of individuals who take traditional NSAIDs continually develop symptomatic ulcers or ulcer complications (perforation, obstruction, and bleeding episodes) yearly. It has been estimated that up to 81 percent of ulcer complications are asymptomatic up to the moment of the complication, contributing dramatically to the morbidity and mortality associated with NSAID use. The NSAID-associated incidence of ulcer complications in patients with RA or OA are not different. Importantly, the risk of ulceration is constant over time: It is no greater on the last day of the year than it was on the first. The major risk factors for predicting ulcer complications in NSAID users are age over 65 years, history of GI bleeding, prior upper GI-related hospitaliza-tion, extent of arthritis-related disability, NSAID dose, concurrent use of NSAIDs and prednisone, and cardiovascular disease.
Interestingly, although certain patients are at higher risk for NSAID-associated complications, the greatest number of ulcer complications occurs among patients of average risk because they are more numerous than high-risk patients. Consequently, targeting only high-risk patients for prophylaxis may not be the best approach for large populations.
H2-receptor blockers such as ranitidine and cimetidine may prevent ulcer symptoms, but they do not prevent gastric ulcers or ulcer complications. Proton pump inhibitors reduce the incidence of both NSAID-associated gastric and duodenal ulcers, and although no study has yet proved that they prevent complications, clinical experience suggests that they are efficacious.
In a trial in which 918 RA patients were demonstrated by endoscopy to be free of ulcers and then randomized to naproxen 500 mg twice daily, placebo, or one of three doses of celecoxib (100 mg to 400 mg twice daily), the incidence of gastroduodenal ulcer from all doses of celecoxib was comparable to placebo when endoscopy was repeated 12 weeks later. The incidence among those treated with naproxen was significantly higher.
Similarly, trials comparing rofecoxib (12.5 mg and 25 mg per day) with ibuprofen (800 mg three times daily) found the cumulative incidences of gastric and duodenal ulcers to be comparable to placebo when treated with rofecoxib, but significantly higher in the ibuprofen trial arm at 12 and 24 weeks, respectively.
A pooled analysis of data from 14 blinded and controlled trials involving 11,008 OA and RA patients compared ulcer complication rates among patients taking celecoxib, traditional NSAIDs, and placebo. The celecoxib group had a statistically significant seven-fold lower incidence of complications (0.2 percent vs 1.68 percent) than did the NSAIDs group. An analysis of data from a single long-term open-label trial involving 5,002 patient years of celecoxib use revealed a 0.18 percent complication rate in celecoxib-treated patients. The 6-month Celecoxib Long-Term Arthritis Safety Study (CLASS) was a prospective trial that enrolled approximately 8,000 patients, with equal numbers taking celecoxib at supra-therapeutic doses and traditional NSAIDs at normal doses. The results of this large scale, blinded, prospective outcomes trial showed that the ulcer complication rate associated with celecoxib treatment was significantly (two-fold) lower than in patients taking diclofenac or ibuprofen. Similar findings of an approximately two-fold reduction in ulcer complications were seen in VIGOR, another large-scale outcomes trial with rofecoxib.
In a recent report of 12-month data from a pooled analysis of blinded and randomized studies involving 4,921 patients who took either rofecoxib or traditional NSAIDs at therapeutic doses, the reported rate of symptomatic ulcers and ulcer complications in the rofecoxib trial arm was 1.5 percent at 12 months. The rate in the traditional NSAIDs arm was significantly higher (3.5 percent), with the incidence increasing steadily from day one through 12 months. These analyses suggest that long-term use of rofecoxib or celecoxib significantly reduces risk of ulcer complications relative to non-COX-1-sparing NSAIDs.
Cardiovascular and Renal Effects of COX-2-Specific Inhibitors
One of the unresolved issues regarding COX-2-specific inhibitors is the effect of COX-1-sparing agents on cardiovascular safety compared to traditional NSAIDs. This was addressed in a review of data from 51 studies enrolling 9,463 patients taking celecoxib, 2,443 of them for a year of more, for a total of 1,020 patient years in controlled trials and 5,002 patient years in open-label extension trials. The rate of adverse cardiovascular events including myocardial infarction was comparable to that of patients taking traditional NSAIDs.
In CLASS, a prospective outcomes trial, incidences of myocardial infarction, cardiovascular death, and all-cause death among celecoxib patients were not different than among patients taking diclofenac or ibuprofen. While the incidence of cerebrovascular events was significantly lower in the celecoxib trial arm, the clinical significance of this finding is unclear. There were no differences in rates of myocardial infarction or cerebral vascular accidents for either OA or RA patients by treatment with celecoxib as compared to NSAIDs. The results suggest that celecoxib is not associated with any change in incidence of cardiovascular or cerebrovascular events compared with traditional NSAIDs. In the VIGOR trial, a statistically significant increase in myocardial infarction was seen in a rofecoxib study arm (50 mg four times daily) compared with naproxen (500 mg twice daily). While many opinions have been offered to explain the differences between the two COX-2-specific inhibitors with respect to their comparison NSAIDs, none have been tested in clinical trials.
Analysis of data from the North American Arthritis Trial showed that celecoxib is not associated with a dose-dependent increase in peripheral edema. In a 6-week head-to-head randomized, double-blind comparison of celecoxib (200 mg daily) and rofecoxib (25 mg daily) in a population of OA patients with stable, treated hypertension, rofecoxib patients had a two-fold increase in the rate of peripheral edema at four and six weeks. The rofecoxib study arm had a mean increase of 2 to 3 mmHg in systolic blood pressure, a magnitude of change potentially associated with morbid cardiovascular events in this high-risk population.
Dr. Goldstein concluded that although celecoxib and rofecoxib are comparable with respect to efficacy and gastrointestinal toxicity, they potentially differ considerably in their side-effects profiles, particularly with regard to blood pressure and peripheral edema.
All contents Copyright © 1999 - 2001 Medical Association Communications
