The safety and efficacy of a JAK inhibitor in patients with active rheumatoid arthritis: Results of a double-blind, placebo-controlled phase IIa trial of three dosage levels of CP-690,550 versus placebo

Authors

  • Joel M. Kremer,

    1. Albany Medical College, Albany, New York
    Search for more papers by this author
    • Dr. Kremer has received consulting fees from Pfizer (more than $10,000) and from Bristol-Myers Squibb, Centocor, Roche, and UCB (less than $10,000 each), and he served as a paid consultant to Pfizer in connection with this study.

  • Bradley J. Bloom,

    1. Pfizer, New London, Connecticut
    Current affiliation:
    1. Novartis Pharmaceuticals, Florham Park, New Jersey
    Search for more papers by this author
    • Drs. Bloom and Gruben own stock or stock options in Pfizer.

  • Ferdinand C. Breedveld,

    1. Leiden University Medical Centre, Leiden, The Netherlands
    Search for more papers by this author
    • Dr. Breedveld has received consulting fees, speaking fees, and/or honoraria from Pfizer, Wyeth, Centocor, Abbott, and Roche (less than $10,000 each) and served as a paid consultant to Pfizer in connection with this study.

  • John H. Coombs,

    1. Pfizer, New London, Connecticut
    Current affiliation:
    1. Novartis Pharmaceuticals, Florham Park, New Jersey
    Search for more papers by this author
  • Mark P. Fletcher,

    1. Pfizer, Ann Arbor, Michigan
    Search for more papers by this author
    • Dr. Fletcher owns stock or stock options in Pfizer and holds patents related to the experimental therapy described in this report.

  • David Gruben,

    1. Pfizer, New London, Connecticut
    Search for more papers by this author
    • Drs. Bloom and Gruben own stock or stock options in Pfizer.

  • Sriram Krishnaswami,

    1. Pfizer, New London, Connecticut
    Search for more papers by this author
  • Rubén Burgos-Vargas,

    1. Hospital General de Mexico, Mexico City, Mexico
    Search for more papers by this author
  • Bethanie Wilkinson,

    1. Pfizer, New London, Connecticut
    Search for more papers by this author
  • Cristiano A. F. Zerbini,

    1. Hospital Heliopolis, Sao Paulo, Brazil
    Search for more papers by this author
    • Dr. Zerbini has received speaking fees from Bristol-Myers Squibb, Sanofi-Aventis, Roche, and Lilly (less than $10,000 each).

  • Samuel H. Zwillich

    Corresponding author
    1. Pfizer, New London, Connecticut
    • Pfizer Global Research and Development, 50 Pequot Avenue, New London, CT 06320
    Search for more papers by this author
    • Dr. Zwillich owns stock or stock options in Pfizer and is named as an inventor on unpublished patent applications assigned to Pfizer Products Inc. relating to the subject matter of this study.

Errata

This article is corrected by:

  1. Errata: The safety and efficacy of a JAK inhibitor in patients with active rheumatoid arthritis: Results of a double-blind, placebo-controlled phase IIa trial of three dosage levels of CP-690,550 versus placebo Volume 64, Issue 5, 1487, Article first published online: 26 April 2012

  • ClinicalTrials.gov identifier: NCT00147498.

  • Presented in part at the 70th Annual Scientific Meeting of the American College of Rheumatology, Washington, DC, November 2006.

Abstract

Objective

To determine the efficacy, safety, and tolerability of 3 different dosages of CP-690,550, a potent, orally active JAK inhibitor, in patients with active rheumatoid arthritis (RA) in whom methotrexate, etanercept, infliximab, or adalimumab caused an inadequate or toxic response.

Methods

Patients (n = 264) were randomized equally to receive placebo, 5 mg of CP-690,550, 15 mg of CP-690,550, or 30 mg of CP-690,550 twice daily for 6 weeks, and were followed up for an additional 6 weeks after treatment. The primary efficacy end point was the American College of Rheumatology 20% improvement criteria (ACR20) response rate at 6 weeks.

Results

By week 6, the ACR20 response rates were 70.5%, 81.2%, and 76.8% in the 5 mg, 15 mg, and 30 mg twice daily groups, respectively, compared with 29.2% in the placebo group (P < 0.001). Improvements in disease activity in CP-690,550–treated patients compared with placebo were seen in all treatment groups as early as week 1. ACR50 and ACR70 response rates significantly improved in all treatment groups by week 4. The most common adverse events reported were headache and nausea. The infection rate in both the 15 mg twice daily group and the 30 mg twice daily group was 30.4% (versus 26.2% in the placebo group). No opportunistic infections or deaths occurred. Increases in mean low-density lipoprotein cholesterol and high-density lipoprotein cholesterol levels, and increases in mean serum creatinine level (0.04–0.06 mg/dl) were seen in all CP-690,550 treatment arms.

Conclusion

Our findings indicate that CP-690,550 is efficacious in the treatment of RA, resulting in rapid, statistically significant, and clinically meaningful reductions in the signs and symptoms of RA. Further studies of CP-690,550 in RA are warranted.

Rheumatoid arthritis (RA) presents a significant health and socioeconomic burden (1). Disease-modifying antirheumatic drugs (DMARDs) are the cornerstone of RA treatment. However, not all available DMARDs prevent progressive joint damage, and they do not always significantly improve quality of life.

Biologic agents, such as tumor necrosis factor α antagonists, are generally more efficacious than traditional DMARDs in terms of controlling the progression of radiologically assessed joint damage, and, when combined with methotrexate, are superior to methotrexate alone in clinical outcomes (2–4). The combination of DMARDs and biologic response modifiers can increase response rate, but even with these treatments, no response at all is achieved in ∼30% of patients, and some patients stop treatment due to adverse events (AEs) or loss of efficacy after an initial response (2–5).

JAK-3 is critical for signal transduction from the common γ-chain of the receptors for interleukin-2 (IL-2), IL-4, IL-7, IL-9, IL-15, and IL-21 on the plasma membrane to the nuclei of immune cells. These interleukins are integral to lymphocyte activation, function, and proliferation. JAK-3 is predominantly expressed in cells of the immune system (6). Expression has been reported in other cell types on rare occasions (7, 8); however, the functional significance of this is not fully understood at present. JAK-3–knockout mice have defects in T lymphocytes, B lymphocytes, and natural killer (NK) cells, with no other defects reported (9–11), and mutations in JAK-3 have been identified as a cause of autosomal-recessive severe combined immunodeficiency disorder (12, 13). Therefore, agents that selectively inhibit JAK-3 have the potential to mediate potent immune modulation, affecting T lymphocytes, B lymphocytes, macrophages, and NK cells, without significantly affecting other organ systems.

CP-690,550 is an orally available JAK antagonist in development for the treatment of RA and other autoimmune conditions, as well as for the prevention of renal allograft rejection. CP-690,550 is a potent, selective inhibitor of the JAK family of kinases with an ∼1,000-fold selectivity over 82 other kinases tested in a selectivity panel compared with the potency of JAK-3 (1 nM). In cell-based assays, CP-690,550 potently inhibited JAK-1/3– and JAK-1–dependent STAT activities with 50% inhibition concentration (IC50) values in the 26–63 nM range, whereas IC50 values for JAK-2–mediated pathways ranged from 129 to 501 nM (14).

The pharmacokinetic profile of CP-690,550 in RA patients is linear, and is characterized by rapid absorption and rapid elimination with a half-life of ∼3 hours (data on file, Pfizer). CP-690,550 produced dose-dependent decreases in clinical scores in collagen-induced arthritis and adjuvant-induced arthritis in mice and rats, with >90% disease reduction observed with the 15 mg/kg/day dosage. The 24-hour area under the curve of this dosage in rodents (1,680–6,216 ng/hour/ml) was within 2-fold of that observed with dosages of 15 mg and 30 mg twice daily in humans. Histologic evidence of inflammation was reduced in both models, with histologic parameters equivalent to those in naive mice at the 15 mg/kg/day dosage in the collagen-induced arthritis model (15).

Herein we report the results of the first randomized, double-blind, placebo-controlled, phase IIa trial of 3 dosage levels of the orally administered JAK inhibitor CP-690,550 versus placebo for the treatment of active RA.

PATIENTS AND METHODS

Patients.

This study was conducted in 60 centers in 11 countries (the US, Canada, Brazil, Mexico, Austria, Spain, Germany, Italy, Belgium, Norway, and Slovakia), and was conducted in accordance with the Declaration of Helsinki and in compliance with all International Conference on Harmonisation Guidelines for Good Clinical Practice. The study was approved by the international review boards or the ethics committees for each participating center in each country.

The primary objective of this phase IIa, randomized, double-blind, placebo-controlled, parallel-group study was to compare the efficacy of CP-690,550 (in dosages of 5 mg, 15 mg, and 30 mg twice daily) with placebo for the treatment of active RA over 6 weeks, plus an additional 6 weeks of followup. Secondary objectives included assessing safety and tolerability, and evaluating the pharmacokinetics of CP-690,550.

Key inclusion criteria were a diagnosis of RA based on the American College of Rheumatology (ACR) 1987 revised criteria (16) and active disease at both the screening and baseline visits. Active RA was defined as ≥9 tender or painful joints on motion, based on a 68-joint count; ≥6 swollen joints, based on a 66-joint count; and 2 of the following 3 criteria: at least 45 minutes' duration of morning stiffness; an erythrocyte sedimentation rate of ≥28 mm/hour; or C-reactive protein (CRP) level ≥10 mg/liter (upper limit of normal 4.0 mg/liter). All patients met the ACR 1991 revised criteria for global functional status in RA of class I, II, or III (17). In addition, patients had an inadequate response to, or discontinued therapy due to unacceptable toxicity from, methotrexate, etanercept, infliximab, or adalimumab. Patients also discontinued all DMARD and immunosuppressive/immunomodulatory therapy for at least 4 weeks prior to the first dose of the study drug (8 weeks for adalimumab and infliximab). Patients who had previously taken other biologics were not excluded, provided they had undergone suitable washout.

Key exclusion criteria included the following: evidence of hematopoietic disorders (hemoglobin <9.0 mg/dl, hematocrit <32%, white blood cell count <3.0 × 109/liter, or platelet count <100 × 109/liter); total bilirubin, aspartate aminotransferase, or alanine aminotransferase >1.2 times the upper limit of normal; an estimated glomerular filtration rate of 60 ml/minute or less; a history of clinically significant infection within the past 6 months; infection with human immunodeficiency virus, hepatitis B, or hepatitis C; and a history of malignancy, other than adequately treated squamous cell or basal cell skin cancer or cervical carcinoma in situ.

Study treatment.

Eligible patients were randomized equally to receive 5 mg of CP-690,550, 15 mg of CP-690,550, or 30 mg of CP-690,550 twice daily or placebo for 6 weeks, with a 6-week followup period. Dosage adjustment was not allowed, and patients in whom treatment was interrupted for >3 consecutive days, or a total of 5 days, were withdrawn from the study. Patients continued to receive stable background therapy for arthritis (nonsteroidal antiinflammatory drugs, selective cyclooxygenase 2 inhibitors, opioids, acetaminophen, and/or oral corticosteroids [≤10 mg of prednisone or equivalent daily]).

Study assessments.

Efficacy assessments were carried out at baseline and at weeks 1, 2, 4, 6 (end of treatment), and 8. The arthritis assessments were the components of the ACR response criteria, including a 68-joint tender/painful joint count, 66-joint swollen joint count, CRP level, and the disability index (DI) of the Health Assessment Questionnaire (HAQ) (18). Patient's assessment of arthritis pain, patient's assessment of disease activity, and physician's assessment of disease activity were measured using a visual analog scale.

Safety evaluations.

All AEs were recorded and graded, in terms of relationship to study treatment and severity, as mild (does not interfere with patient's usual function), moderate (interferes to some extent with patient's usual function), or severe (interferes significantly with patient's usual function). Laboratory safety evaluations, including assessment of blood Epstein-Barr virus (EBV) DNA levels (average value from 2 separate assays expressed as copies/500 mg of DNA) by quantitative polymerase chain reaction of total DNA, as previously described (19), and vital-sign assessments were carried out at screening (14–28 days prior to baseline), baseline, weeks 1, 2, 4, and 6, and at the 2 followup visits (weeks 8 and 12). Electrocardiograms were obtained at screening and at weeks 1, 2, 4, and 6. Complete physical examinations were carried out at screening and at weeks 6 and 12.

Statistical analysis.

We used the normal approximation to the binomial distribution for the difference in response rates (with no pooled standard deviation and no continuity correction), with pairwise comparisons with placebo for each of the ACR criteria (ACR20, ACR50, and ACR70). Patients who withdrew from the study were counted as nonresponders. The study was designed to maintain an overall Type I error rate of 10% in 1-sided testing, with 80% power for the primary end point of ACR20 response at week 6 for the 2 comparisons with placebo, involving 30 mg twice daily and 15 mg twice daily. A modified step-down procedure that allowed for declaring the comparison with 15 mg twice daily to be significant, even if the comparison with 30 mg twice daily was not, was used. P values less than or equal to 0.05 were considered significant for the 5 mg twice daily dosage. All other analyses involving different times or ACR responses were secondary analyses, and P values less than or equal to 0.05 were considered significant. There were no adjustments for multiple end points or multiple time points.

RESULTS

Patient disposition and study treatment.

A total of 264 patients with active RA were randomized to receive study treatment. All 264 patients received study medication and were evaluated for safety and efficacy measures. In the CP-690,550 5 mg, 15 mg, and 30 mg twice daily treatment groups, 95%, 87%, and 75% of patients completed the study, respectively, compared with 74% in the placebo group (Figure 1).

Figure 1.

Disposition of the patients. Laboratory data were obtained in 63 of the 65 patients in the placebo group (1 patient discontinued due to an adverse event, and 1 patient no longer wanted to participate in the study). BID = twice daily.

Table 1 summarizes the demographic distribution of the study population. Treatment groups were comparable with regard to demographic and baseline characteristics. At baseline, the mean duration of RA was similar between the groups, ranging from 8.7 to 10.2 years. The mean number of DMARDs (including methotrexate) that had previously led to an inadequate response ranged from 2.4 to 2.8, and ∼60% of the subjects were taking glucocorticoids.

Table 1. Baseline characteristics of the study population
 CP-690,550 5 mg bid (n = 61)CP-690,550 15 mg bid (n = 69)CP-690,550 30 mg bid (n = 69)Placebo (n = 65)
  • *

    One patient in the 5 mg twice daily (bid) treatment group, 4 patients in the 15 mg twice daily treatment group, and 2 patients in the placebo group had neither methotrexate (MTX) nor tumor necrosis factor (TNF) inhibitors recorded as failed medications. DMARDs = disease-modifying antirheumatic drugs

Sex, no. (%) female53 (86.9)58 (84.1)60 (87.0)55 (84.6)
Age, mean ± SD years47.9 ± 10.851.8 ± 13.051.1 ± 10.651.3 ± 12.1
Race, no. (%)    
 White42 (68.9)47 (68.1)44 (63.8)47 (72.3)
 Black3 (4.9)4 (5.8)5 (7.2)2 (3.1)
 Asian1 (1.6)2 (2.9)2 (2.9)0 (0)
 Hispanic15 (24.6)14 (20.3)18 (26.1)15 (23.1)
 Other0 (0)2 (2.9)0 (0)1 (1.5)
Disease duration, mean (range) years10.2 (1.2–35.2)9.6 (0.6–48.2)9.9 (0.8–40.0)8.7 (0.7–26.8)
Prior failed DMARDs, no. (%)*    
 Failed MTX and TNF inhibitors10 (16.4)15 (21.7)14 (20.3)11 (16.9)
 Failed MTX but not TNF inhibitors47 (77.0)48 (69.6)54 (78.3)52 (80.0)
 Failed TNF inhibitors but not MTX3 (4.9)2 (2.9)1 (1.4)0 (0)
Rheumatoid factor positive, no. (%)56 (91.8)65 (94.2)60 (87.0)57 (87.7)
Glucocorticoid use at baseline, no. (%)39 (63.9)43 (62.3)44 (63.8)40 (61.5)

Efficacy assessments.

All patients were included in the primary, intent-to-treat analysis. Treatment with CP-690,550 resulted in significant, dose-dependent improvements in ACR20 response rates. At week 6, the primary end point, the ACR20 response rate, was 70.5%, 81.2%, and 76.8% in the 5 mg, 15 mg, and 30 mg twice daily groups, respectively, compared with 29.2% in the placebo group (P < 0.001 for all comparisons at week 6) (Figure 2a). The percentage-point differences from placebo were 41%, 52%, and 48% in the CP-690,550 5 mg, 15 mg, and 30 mg twice daily groups, respectively, and were highly significant for all CP-690,550 treatment groups compared with placebo (P < 0.0001).

Figure 2.

Mean ± SEM response rate in patients with active rheumatoid arthritis who were treated with 5 mg of CP-690,550, 15 mg of CP-690,550, or 30 mg of CP-690,550 twice daily (BID) or placebo. a, American College of Rheumatology 20% improvement criteria (ACR20) response rate. b, ACR50 response rate. c, ACR70 response rate. ∗ = P < 0.05; ∗∗ = P < 0.01; ∗∗∗ = P < 0.001, versus placebo.

Improvements in disease activity (as measured by ACR20 response) in CP-690,550–treated patients compared with placebo were seen across all treatment groups as early as week 1. All CP-690,550 treatment groups also exhibited increased ACR50 and ACR70 response rates relative to placebo at all time points (Figures 2b and c). These differences were statistically significant versus placebo at week 2 for the CP-690,550 30 mg twice daily treatment group and in all CP-690,550 groups by week 4. A breakdown of ACR20 response by geographic region is shown in Figure 3; response rates within each subgroup were consistent with the overall results.

Figure 3.

Mean and SEM American College of Rheumatology 20% improvement criteria (ACR20) response rate in patients with rheumatoid arthritis who were treated with 5 mg of CP-690,550, 15 mg of CP-690,550, or 30 mg of CP-690,550 twice daily (BID) or placebo, by geographic distribution. ∗ = P < 0.05; ∗∗ = P < 0.01; ∗∗∗ = P < 0.001, versus placebo.

Table 2 shows the values at baseline and week 6, and the difference between these values, for the 7 components of the ACR response criteria for each treatment group. Statistically significant improvements were observed in all components by week 6 in each CP-690,550 treatment group as compared with placebo. While a clear difference in response was seen between the 5 mg twice daily dosage and the 2 higher dosages, there was not a clear separation of response between the 15 mg and 30 mg twice daily dosages in any of the individual ACR component scores.

Table 2. ACR20 component scores at baseline and week 6*
 Tender/painful joint countSwollen joint countPatient's assessment of pain, 0–100-mm VASPatient's global assessment of disease activity, 0–100-mm VASPhysician's global assessment of disease activity, 0–100-mm VASHAQ DICRP, mg/liter
 No. of patientsMean ± SDNo. of patientsMean ± SDNo. of patientsMean ± SDNo. of patientsMean ± SDNo. of patientsMean ± SDNo. of patientsMean ± SDNo. of patientsMean ± SD
  • *

    To calculate change from baseline, the value at baseline was subtracted from the value at week 6. The descriptive statistics were then calculated. ACR20 = American College of Rheumatology 20% improvement criteria; VAS = visual analog scale; HAQ = Health Assessment Questionnaire; DI = disability index; CRP = C-reactive protein; bid = twice daily.

CP-690,550 5 mg bid              
 Baseline6132.3 ± 15.86121.1 ± 11.75968.3 ± 24.15769.4 ± 23.66068.7 ± 16.5601.7 ± 0.65031.3 ± 35.0
 Week 65612.0 ± 10.8567.5 ± 7.25534.3 ± 27.15535.9 ± 26.85530.6 ± 20.4541.1 ± 0.84512.6 ± 21.0
 Change −20.0 ± 14.6 −13.4 ± 11.2 −34.0 ± 29.2 −34.9 ± 28.9 −37.9 ± 22.6 −0.6 ± 0.7 −17.8 ± 35.9
CP-690,550 15 mg bid 
 Baseline6926.7 ± 12.76916.2 ± 7.66867.2 ± 22.96769.3 ± 21.16963.9 ± 18.0691.6 ± 0.75725.5 ± 31.1
 Week 6598.5 ± 8.5594.5 ± 5.55928.2 ± 23.25928.6 ± 23.15821.9 ± 19.2590.9 ± 0.6496.6 ± 16.4
 Change −18.4 ± 11.8 −11.8 ± 8.0 −39.9 ± 30.1 −40.8 ± 29.0 −43.1 ± 24.4 −0.7 ± 0.6 −18.1 ± 26.0
CP-690,550 30 mg bid 
 Baseline6829.3 ± 14.16819.5 ± 10.26667.3 ± 21.86563.8 ± 24.06666.4 ± 16.8671.6 ± 0.65828.1 ± 28.4
 Week 6536.1 ± 7.6535.2 ± 6.35320.8 ± 21.55222.5 ± 22.05317.3 ± 15.2530.8 ± 0.6416.1 ± 11.8
 Change −22.3 ± 13.9 −15.0 ± 10.6 −44.2 ± 27.1 −39.2 ± 28.8 −47.5 ± 23.1 −0.7 ± 0.5 −17.6 ± 24.8
Placebo 
 Baseline6530.3 ± 14.36520.1 ± 10.46463.9 ± 24.86365.3 ± 24.06467.6 ± 16.2621.7 ± 0.65322.3 ± 18.5
 Week 65018.0 ± 12.7509.5 ± 9.15049.1 ± 24.14946.4 ± 24.05044.1 ± 21.7501.3 ± 0.64122.9 ± 20.8
 Change −11.9 ± 15.9 −9.2 ± 12 −9.9 ± 29.1 −15.1 ± 29.3 −22.4 ± 27.2 −0.3 ± 0.5 0.7 ± 16.8

In this study, the Disease Activity Score in 28 joints (DAS28) (20) was included as an exploratory end point and was based on measurements of CRP levels and the 28-count subsets of tender/painful joints and swollen joints (DAS28-3 [CRP]). At baseline the mean DAS28-3 (CRP) scores for the placebo and CP-690,550 5 mg, 15 mg, and 30 mg twice daily treatment groups were 6.0, 6.2, 5.7, and 5.9, respectively, indicating high disease activity. At week 6, DAS28-3 (CRP) scores for the placebo and CP-690,550 5 mg, 15 mg, and 30 mg twice daily treatment groups were 4.8, 4.2, 3.4, and 3.1, respectively. Decreases in DAS28-3 (CRP) were dose-dependent, indicating that treatment with CP-690,550 was associated with an improvement in disease activity.

Using the European League Against Rheumatism criteria (21), the percentage of patients in whom a moderate response was achieved in the CP-690,550 treatment arms was numerically superior to placebo at all time points (56%, 80%, 86%, and 92% for placebo and CP-690,550 5 mg, 15 mg, and 30 mg twice daily, respectively, at week 6). Similarly, the percentage of patients in whom a good response was achieved in the CP-690,550 treatment arms was numerically superior to placebo at all time points (3%, 24%, 43%, and 46% for placebo and CP-690,550 5 mg, 15 mg, and 30 mg twice daily, respectively, at week 6). The percentages of patients in whom a moderate or good response was achieved in the CP-690,550 15 mg and 30 mg twice daily arms were also numerically superior to the percentages in the CP-690,550 5 mg twice daily group at every time point.

Safety.

The incidences of AEs in the CP-690,550 5 mg twice daily and placebo groups were similar, with 59% and 58.5% of patients experiencing ≥1 AE, respectively. There were dose-related increases in the number of patients reporting treatment-emergent AEs in the CP-690,550 15 mg and 30 mg twice daily groups, in which 75.4% and 76.8% of patients, respectively, experienced ≥1 AE. Six and 9 patients withdrew from the study due to treatment-emergent AEs in the CP-690,550 15 mg and 30 mg twice daily treatment groups, respectively. Of these, 3 and 6 AEs, respectively, were thought by the investigator to be related to the study drug. Only 1 patient in the CP-690,550 5 mg twice daily treatment group withdrew from the study due to an AE that was also considered to be treatment related. In comparison, 3 patients in the placebo group withdrew from the study due to AEs, 1 of which was thought to be related to treatment (Figure 1). There were no deaths during the study.

Table 3 summarizes AEs that occurred at a frequency of ≥5% in any one treatment group during the study. Headache was the most frequently reported AE among all subjects. Patients receiving 30 mg of CP-690,550 twice daily had higher incidences of anemia, leukopenia, neutropenia, lymphopenia, and thrombocytopenia than did other treatment groups. These events were generally rated mild to moderate in severity.

Table 3. Treatment-emergent adverse events occurring at an incidence of ≥5%*
System organ class/ preferred term (MedDRA)CP-690,550 5 mg bid (n = 61)CP-690,550 15 mg bid (n = 69)CP-690,550 30 mg bid (n = 69)Placebo (n = 65)
All causesTreatment- relatedAll causesTreatment- relatedAll causesTreatment- relatedAll causesTreatment- related
  • *

    Incidence of ≥5% based on all causes. MedDRA = Medical Dictionary for Regulatory Activities; bid = twice daily.

  • Other infections that occurred in 2 or more patients across all treatment groups were bronchitis (1 patient), gastroenteritis (2 patients), herpes zoster (1 patient), nasopharyngitis (3 patients), and tinea versicolor (2 patients) in the 5 mg group; bronchitis (2 patients), conjunctivitis (1 patient), cystitis (1 patient), gastroenteritis (4 patients), herpes simplex (2 patients), nasopharyngitis (1 patient), respiratory tract infection (1 patient), and tinea pedis (3 patients) in the 15 mg group; bronchitis (1 patient), candidiasis (1 patient), conjunctivitis (1 patient), cystitis (1 patient), gastroenteritis (1 patient), herpes simplex (2 patients), nasopharyngitis (1 patient), and tinea pedis (1 patient) in the 30 mg group; and candidiasis (1 patient), cystitis (1 patient), gastroenteritis (1 patient), herpes zoster (1 patient), nasopharyngitis (1 patient), respiratory tract infection (1 patient), and sinusitis (4 patients) in the placebo group.

Any adverse event36 (59)23 (38)52 (75)29 (42)53 (77)33 (48)38 (59)21 (32)
Blood and lymphatic system disorders6 (9.8)4 (6.6)6 (8.7)3 (4.3)12 (17.4)10 (14.5)3 (4.6)1 (1.5)
 Anemia1 (1.6)0 (0)2 (2.9)1 (1.4)4 (5.8)3 (4.3)3 (4.6)1 (1.5)
 Leukopenia2 (3.3)1 (1.6)2 (2.9)2 (2.9)7 (10.1)6 (8.7)1 (1.5)1 (1.5)
Gastrointestinal disorders13 (21.3)12 (19.7)15 (21.7)11 (15.9)15 (21.7)9 (13.0)15 (23.1)5 (7.7)
 Abdominal pain, upper4 (6.6)4 (6.6)2 (2.9)2 (2.9)2 (2.9)1 (1.4)0 (0)0 (0)
 Nausea4 (6.6)4 (6.6)3 (4.3)3 (4.3)6 (8.7)3 (4.3)3 (4.6)2 (3.1)
Infections and infestations15 (24.6)6 (9.8)21 (30.4)13 (18.8)21 (30.4)10 (14.5)17 (26.2)9 (13.8)
 Influenza3 (4.9)1 (1.6)2 (2.9)0 (0)4 (5.8)2 (2.9)2 (3.1)0 (0)
 Upper respiratory tract infection0 (0)0 (0)3 (4.3)1 (1.4)1 (1.4)0 (0)4 (6.2)3 (4.6)
 Urinary tract infection1 (1.6)1 (1.6)0 (0)0 (0)5 (7.2)4 (5.8)2 (3.1)1 (1.5)
Metabolism and nutritional disorders2 (3.3)2 (3.3)5 (7.2)3 (4.3)4 (5.8)3 (4.3)1 (1.5)0 (0)
 Hypercholesterolemia2 (3.3)2 (3.3)4 (5.8)3 (4.3)0 (0)0 (0)0 (0)0 (0)
Musculoskeletal and connective tissue disorders7 (11.5)2 (3.3)17 (24.6)0 (0)7 (10.1)2 (2.9)13 (20.0)2 (3.1)
 Arthralgia3 (4.9)0 (0)0 (0)0 (0)3 (4.3)2 (2.9)6 (9.2)1 (1.5)
 Rheumatoid arthritis1 (1.6)1 (1.6)4 (5.8)0 (0)2 (2.9)0 (0)4 (6.2)0 (0)
Nervous system disorders11 (18)8 (13.1)13 (18.8)11 (15.9)21 (30.4)11 (15.9)6 (9.2)5 (7.7)
 Dizziness0 (0)0 (0)4 (5.8)3 (4.3)3 (4.3)3 (4.3)1 (1.5)1 (1.5)
 Headache10 (16.4)8 (13.1)8 (11.6)7 (10.1)17 (24.6)17 (24.6)6 (9.2)5 (7.7)

In the CP-690,550 15 mg twice daily group, 1 patient withdrew from the study due to severe leukopenia and 1 patient withdrew from the study due to moderate leukopenia and moderate neutropenia. In the CP-690,550 30 mg twice daily treatment group, 1 patient withdrew from the study due to 2 occurrences of moderate neutropenia; 1 patient withdrew due to leukopenia (moderate), neutropenia, lymphopenia, and thrombocytopenia; and 1 patient withdrew due to moderate leukopenia. The incidence of these events was generally similar between the CP-690,550 5 mg twice daily treatment group and the placebo group.

Nonhematologic AEs leading to discontinuation in the CP-690,550 15 mg treatment group were lymphadenopathy, bronchitis (moderate), pregnancy, and myocardial ischemia. In the CP-690,550 30 mg twice daily treatment group patients also withdrew due to myocardial ischemia (moderate); blurred vision and balance disorder (moderate); abdominal pain, nausea, malaise, and headache (severe); Whipple's disease (moderate); liver disorder and pyelonephritis; and cystitis. All AEs, except the myocardial ischemia in the CP-690,550 15 mg twice daily treatment group, resolved after cessation of study treatment. The pregnancy in the CP-690,550 15 mg treatment group ended in spontaneous abortion.

Infections occurred in 26%, 25%, 30%, and 30% of patients in the placebo and CP-690,550 5 mg, 15 mg, and 30 mg twice daily treatment groups, respectively. Most infections occurred no more than once in any treatment group, responded promptly to standard therapy and, with the exception of 1 patient in the CP-690,550 15 mg group who had a severe episode of gastroenteritis, were mild to moderate in severity.

Three patients experienced serious AEs that were considered possibly or probably related to blinded study treatment: infectious gastroenteritis in 1 patient receiving 15 mg of CP-690,550 twice daily; severe leukopenia in 1 patient receiving 30 mg of CP-690,550 twice daily; and Staphylococcus aureus pneumonia in 1 patient receiving placebo.

Abnormal laboratory findings.

Mean absolute neutrophil counts decreased in a dose-dependent manner during 6 weeks of active treatment, with maximum mean decreases of 940, 1,620, and 2,190 neutrophils/mm3 in the CP-690,550 5 mg, 15 mg, and 30 mg twice daily treatment groups, respectively. The incidence of neutropenia at week 6, as defined by the Outcome Measures in Rheumatology Proposed Rheumatology Common Toxicity Criteria (22), was 3.6% (2 patients) in the 5 mg twice daily group (both mild [1.5–1.9 × 103 neutrophils/mm3]), 3.4% (2 patients) in the 15 mg twice daily group (both mild), and 7.8% (4 patients) in the 30 mg twice daily group (2 moderate [1.0–1.4 × 103 neutrophils/mm3] and 2 severe [0.5–0.9 × 103 neutrophils/mm3]). Mean absolute neutrophil count recovered to baseline levels by week 12. In a total of 3 patients (1 each in the placebo, 5 mg twice daily, and 30 mg twice daily groups), absolute neutrophil counts did not recover by week 12.

At baseline the mean hemoglobin level in the subjects ranged from 12.7 to 13.1 gm/dl and decreased in patients assigned to the placebo group, with a mean change from baseline of −0.28 gm/dl by week 6. Hemoglobin was stable in the CP-690,550 5 mg twice daily treatment group (mean change from baseline 0.03 gm/dl by week 6) and decreased in a dose-dependent manner in the 15 mg and 30 mg twice daily treatment groups (mean change from baseline −0.19 gm/dl and −0.40 gm/dl at week 6, respectively).

Although the largest increases in blood EBV DNA levels were observed in the CP-690,550 30 mg twice daily group, median increases of ≤7 copies per 500 mg of DNA were seen in all groups, including placebo. Mean EBV levels (copies per 500 mg of DNA) at baseline, week 6, and week 12 were 2.57, 3.74, and 1.66, respectively, for CP-690,550 5 mg twice daily; 2.09, 5.11, and 2.60, respectively, for CP-690,550 15 mg twice daily; 2.05, 12.23, and 2.49, respectively, for CP-690,550 30 mg twice daily; and 1.77, 8.19, and 1.66, respectively, for placebo.

Mean serum total cholesterol, high-density lipoprotein (HDL), and low-density lipoprotein (LDL) levels all increased in a dose-dependent manner between baseline and week 6. Serum lipid levels were not assessed beyond week 6. The ratios of total cholesterol to HDL cholesterol, which is a measure of atherogenic risk (23), were unchanged (Table 4).

Table 4. Mean levels of total cholesterol, HDL, ratio of total cholesterol to HDL, and LDL at baseline and week 6*
Cholesterol parameterBaselineWeek 6Percentage change from baseline, least squares mean ± SD
No. of patientsMean ± SD mg/dlNo. of patientsMean ± SD mg/dl
  • *

    HDL = high-density lipoprotein; LDL = low-density lipoprotein; bid = twice daily.

  • Calculated from a linear model that included treatment, visit, center, and the baseline cholesterol parameter value as fixed effects, and subjects as a random effect, with the cholesterol parameter expressed as a percentage change from baseline as the dependent variable.

CP-690,550 5 mg bid     
 Total cholesterol61193 ± 4356220 ± 5315 ± 2
 HDL6157 ± 175664 ± 1613 ± 3
 Ratio of total cholesterol to HDL613.5 ± 0.1563.6 ± 0.116 ± 3
 LDL61108 ± 3656122 ± 4316 ± 4
CP-690,550 15 mg bid     
 Total cholesterol69202 ± 4055249 ± 4623 ± 2
 HDL6963 ± 195576 ± 2126 ± 4
 Ratio of total cholesterol to HDL693.5 ± 0.2553.5 ± 0.221 ± 3
 LDL69111 ± 3555144 ± 4328 ± 4
CP-690,550 30 mg bid     
 Total cholesterol69187 ± 3952238 ± 6326 ± 2
 HDL6958 ± 155276 ± 2029 ± 4
 Ratio of total cholesterol to HDL693.4 ± 0.1523.3 ± 0.126 ± 3
 LDL69106 ± 3252138 ± 5429 ± 4
Placebo     
 Total cholesterol65195 ± 3745196 ± 413 ± 3
 HDL6559 ± 134561 ± 134 ± 4
 Ratio of total cholesterol to HDL653.4 ± 0.1453.3 ± 0.13 ± 3
 LDL65109 ± 3145108 ± 364 ± 5

Mean serum creatinine levels increased (0.04–0.06 mg/dl over baseline) in all CP-690,550 treatment groups, with these changes observed from week 1. The mean percent change from baseline in serum creatinine at week 12 was 0.73, 3.19, 5.47, and 2.09 for CP-690,550 5 mg, 15 mg, and 30 mg twice daily and placebo, respectively. Of the 51 patients with increases in serum creatinine values, values in 31 had returned to within 10% of their baseline value at weeks 8 and 12. However, increased serum creatinine level did not fully resolve in all patients by the end of the study (2 patients in the 5 mg twice daily group, 7 patients in the 15 mg twice daily group, 8 patients in the 30 mg twice daily group, and 1 patient in the placebo group had increased serum creatinine level), with a maximum increase of 0.25 mg/dl at week 12. No followup data were available for these patients after week 12. The 2 remaining patients were not available for followup.

DISCUSSION

The results of this phase IIa trial showed that CP-690,550, administered at dosages of 5 mg, 15 mg, and 30 mg twice daily, was efficacious for all primary and secondary end point measures in the treatment of active RA during this 6-week study. Statistically significant and clinically meaningful improvements compared with placebo were seen across all doses in the primary efficacy measure of ACR20 response rate, and in the secondary measures of ACR50 and ACR70 response rates. When individual components of the ACR response criteria were considered, both tender joint counts and swollen joint counts showed dosage-related improvements over time; improvements in both physician- and patient-reported outcomes supported the improvement seen in disease activity; and the DI of the HAQ indicated that patients taking CP-690,550 experienced clinically significant improvements in functional and health status. In many patients, these improvements were apparent at the first study assessment at the end of week 1.

While all 3 doses of CP-690,550 were associated with AEs, many AEs were treatable or reversible with drug discontinuation. Some of these AEs (i.e., infections) could be anticipated from the known pharmacology of JAK inhibition; others, such as increased serum creatinine and serum lipid concentrations, are currently less well understood. The incidence of blood and lymphatic system events was low and generally similar between the CP-690,550 5 mg and 15 mg groups and the placebo group. Consistent, though clinically small, increases were observed in serum creatinine levels in all CP-690,550 groups. Increases in serum creatinine levels were not observed in phase I studies in healthy volunteers, and the mechanism underlying these changes is unknown (24).

Laboratory evaluations revealed dose-related increases in total cholesterol, HDL cholesterol, and LDL cholesterol. The ratios of total cholesterol to HDL cholesterol, a measure of atherogenic risk (23), remained consistent from baseline. The mechanism of the increase in lipid levels associated with CP-690,550 is presently unclear; however, increases in blood lipid levels have been reported with the use of several other biologic agents with disease-modifying activity in RA (25–27). Increases in serum lipid levels after treatment with biologic agents are thought to be related to reversal of the inhibitory effects of inflammation on hepatic cholesterol production, but the precise mechanisms of lipid changes with CP-690,550 have not yet been elucidated. EBV DNA titers were assessed as an exploratory marker of immunosuppression, based on the practice in the CP-690,550 transplantation clinical trials program, and because lymphoproliferative disorders and lymphomas are theoretical risks associated with the clinical use of any potent immunosuppressive drug. Most blood EBV DNA levels remained below 10 copies per 500 ng of DNA, a level of potential clinical concern.

CP-690,550 is ∼20-fold more selective for JAK-3 than for JAK-2 (28). JAK-2 is essential for signaling by many hematopoietic growth factors, with mutations in JAK-2 causing embryonic death in mice due to the absence of erythropoiesis. Significant inhibition of JAK-2 could, therefore, cause anemia, leukopenia, and thrombocytopenia. The observed hematologic AEs, particularly the complex changes in hemoglobin level, may be due to a combination of competitive factors, including inhibition of disease activity as well as some measure of JAK-2 inhibition. However, despite the potentially additive risks of infectious AEs from neutropenia and immunosuppression, the incidence of infections typically associated with neutropenia was not increased relative to placebo, nor was the incidence of infection increased in those patients who experienced neutropenia during this 6-week study. Studies of longer duration will be necessary to adequately assess the risk of neutropenia and infection.

We have demonstrated that at dosages of 5 mg, 15 mg, and 30 mg twice daily for 6 weeks, CP-690,550 is efficacious in the treatment of RA, resulting in rapid and clinically meaningful reductions in the signs and symptoms of RA. Longer-duration, dose-ranging studies of CP-690,550 in the treatment of RA are warranted.

AUTHOR CONTRIBUTIONS

Dr. Zwillich had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Kremer, Bloom, Fletcher, Gruben, Wilkinson, Zwillich.

Acquisition of data. Kremer, Bloom, Coombs, Burgos-Vargas, Wilkinson, Zerbini.

Analysis and interpretation of data. Kremer, Bloom, Breedveld, Coombs, Fletcher, Gruben, Krishnaswami, Burgos-Vargas, Wilkinson, Zwillich.

Manuscript preparation. Kremer, Bloom, Breedveld, Coombs, Fletcher, Gruben, Burgos-Vargas, Wilkinson, Zerbini, Zwillich, Sarah Feaver (nonauthor; Complete Medical Communications [funded by Pfizer]).

Statistical analysis. Gruben, Wilkinson.

Acknowledgements

We gratefully acknowledge the contributions of the following individuals to this study: Dr. Ethan Weiner, Dr. Stanley B. Cohen, Dr. Edward Keystone, Dr. Michael E. Weinblatt, Dr. Eugene Boling, Dr. Stephen Bookbinder, Dr. Jane Box, Dr. Ronald Collins, Dr. Andres Quiceno, Dr. Robert Ettlinger, Dr. Alan Kivitz, Dr. Reynold Karr, Dr. Clarence Legerton, Dr. Jeffrey Poiley, Dr. Jaime Pachon, Dr. Joel Silverfield, Dr. S. Tanner, Dr. James Trice, Dr. Sanford Wolfe, Dr. Farrukh Zaidi, Dr. Mark Niemer, Dr. Dayton Payne, Dr. Ignacio Garcia-De-La Torre, Dr. Virginia Pascual, Dr. Carlos Abud-Mendoza, Dr. Guillermo Huerta-Yanez, Sebastiao Radominski, Dr. Prem Chatpar, Dr. Flora Marcolino, Prof. Dr. Hubert Nuesslein, Dr. Majed Khraishi, Milton Baker, Dr. H. Jones, Dr. Arthur Bookman, Dr. Juan Gomez Reino, Dr. Jesus Tornero Molina, Dr. Jordi Carbonell, Dr. Juan Sanchez Burson, Dr. Emilio Martin Mola, Dr. Antonio Ximenes, Dr. Wolfgang Bolten, Prof. Dr. Gerd-Ruediger Burmester, Prof. Dr. Herbert Kellner, Dr. Juergen Wollenhaupt, Dr. med. Thilo Klopsch, Dr. Ulrich von Hinueber, Prof. Dr. Holm Haentzschel, Dr. Filip Van den Bosch, Dr. Carlo Maurizio Montecucco, Prof. Maurizio Cutolo, Prof. Marco Matucci Cerinic, Dr. Jozef Lukac, Dr. Anna Sabova, Pavol Polak, Zelmira Macejova, and Dr. Omid Zamani.

Ancillary