[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.161.241.199. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Download PDF
Figure 1.
Calculation of the percentage of change from the baseline curve for blood lymphocyte counts during the dosing interval (EAUC). For each patient, the percentage of change from baseline counts at each time point was calculated from the first observation after beginning treatment to 2 weeks after the last injection in each course. This change was then plotted against the actual sampling time, and the points were joined to form a polygon. The area between the resulting polygon and the line of no change (0% line) was calculated using the trapezoidal rule. The EAUC therefore represents the cumulative attrition of these T-cell counts over time (A). For a patient whose count was greater than at baseline, the area above the line of no change was not included in the calculation. The second sample after baseline provided a count above baseline. The area above the line of no change would not be included, but the 2 triangles (shaded) would be counted in the calculation of the EAUC. The intersections with the line of no change were calculated by means of interpolation (B).

Calculation of the percentage of change from the baseline curve for blood lymphocyte counts during the dosing interval (EAUC). For each patient, the percentage of change from baseline counts at each time point was calculated from the first observation after beginning treatment to 2 weeks after the last injection in each course. This change was then plotted against the actual sampling time, and the points were joined to form a polygon. The area between the resulting polygon and the line of no change (0% line) was calculated using the trapezoidal rule. The EAUC therefore represents the cumulative attrition of these T-cell counts over time (A). For a patient whose count was greater than at baseline, the area above the line of no change was not included in the calculation. The second sample after baseline provided a count above baseline. The area above the line of no change would not be included, but the 2 triangles (shaded) would be counted in the calculation of the EAUC. The intersections with the line of no change were calculated by means of interpolation (B).

Figure 2.
Levels of CD4+ memory T cells were reduced by alefacept (A), whereas CD4+ naive T cells were spared (B). The effect of alefacept was consistent when administered for 1 or 2 treatment courses. Patients are included who underwent a baseline assessment and at least 1 postbaseline assessment in both courses. Similar results were seen for CD8+ memory and naive T cells (data not shown). Cohorts are described in the "Study Design" subsection of the "Methods" section and the "Baseline Characteristics" subsection of the "Results" section.

Levels of CD4+ memory T cells were reduced by alefacept (A), whereas CD4+ naive T cells were spared (B). The effect of alefacept was consistent when administered for 1 or 2 treatment courses. Patients are included who underwent a baseline assessment and at least 1 postbaseline assessment in both courses. Similar results were seen for CD8+ memory and naive T cells (data not shown). Cohorts are described in the "Study Design" subsection of the "Methods" section and the "Baseline Characteristics" subsection of the "Results" section.

Figure 3.
Alefacept-induced reductions in CD4+ memory T-cell levels, as measured by the area under the percentage of change from the baseline curve for blood lymphocyte counts during the dosing interval (EAUC), were related to clinical response rates in courses 1 (A) and 2 (B) of the study. In general, the greater the magnitude of reduction in memory T-cell counts, the greater the clinical response. Asterisk indicates that, in course 1, among alefacept-treated patients (cohorts 1 and 2 combined), there was a significant trend for increasing response rates (≥75% reduction in Psoriasis Area Severity Index [PASI]) with increasing CD4+CD45RO+ EAUC quartile (Cochran-Armitage trend test, P<.001). Cohorts are described in the "Study Design" subsection of the "Methods" section. PGA indicates Physician Global Assessment; Q, quartile of EAUC (Q1, 0-2209, cumulative reduction in T cells [CRT]; Q2, 2235-3394 CRT; Q3, 3404-4712 CRT; and Q4, 4723-7660 CRT). Similar results were seen when CD8+ memory T cells were analyzed (data not shown).

Alefacept-induced reductions in CD4+ memory T-cell levels, as measured by the area under the percentage of change from the baseline curve for blood lymphocyte counts during the dosing interval (EAUC), were related to clinical response rates in courses 1 (A) and 2 (B) of the study. In general, the greater the magnitude of reduction in memory T-cell counts, the greater the clinical response. Asterisk indicates that, in course 1, among alefacept-treated patients (cohorts 1 and 2 combined), there was a significant trend for increasing response rates (≥75% reduction in Psoriasis Area Severity Index [PASI]) with increasing CD4+CD45RO+ EAUC quartile (Cochran-Armitage trend test, P<.001). Cohorts are described in the "Study Design" subsection of the "Methods" section. PGA indicates Physician Global Assessment; Q, quartile of EAUC (Q1, 0-2209, cumulative reduction in T cells [CRT]; Q2, 2235-3394 CRT; Q3, 3404-4712 CRT; and Q4, 4723-7660 CRT). Similar results were seen when CD8+ memory T cells were analyzed (data not shown).

Figure 4.
Mean percentage of change from baseline in Psoriasis Area Severity Index (PASI) and CD4+ T-cell count in course 1 for alefacept-treated patients (cohorts 1 and 2 combined). Patients are included who underwent assessments of PASI and CD4+ T-cell counts at baseline and at least 1 postbaseline visit in course 1 (n = 366). The decline in CD4+ T-cell count preceded the improvement in PASI. The peak effect of alefacept on CD4+ T cells occurred during the treatment period, whereas the peak effect on PASI occurred well after treatment cessation. Similar results were seen when CD8+ T cells were analyzed (data not shown).

Mean percentage of change from baseline in Psoriasis Area Severity Index (PASI) and CD4+ T-cell count in course 1 for alefacept-treated patients (cohorts 1 and 2 combined). Patients are included who underwent assessments of PASI and CD4+ T-cell counts at baseline and at least 1 postbaseline visit in course 1 (n = 366). The decline in CD4+ T-cell count preceded the improvement in PASI. The peak effect of alefacept on CD4+ T cells occurred during the treatment period, whereas the peak effect on PASI occurred well after treatment cessation. Similar results were seen when CD8+ T cells were analyzed (data not shown).

Figure 5.
The mechanism of action of alefacept causes a remission that appears to be related to its effects on CD4+ and CD8+ memory T cells. This figure shows the duration of at least a 50% reduction in Psoriasis Area Severity Index (PASI) by EAUC quartile of CD4+ (A) and CD8+ (B) memory T cells in cohort 2 patients who achieved at least a 75% reduction from baseline PASI at any time during treatment or follow-up in course 1. In this subset of 52 patients, those who experienced the greatest reductions in CD4+ and CD8+ memory T-cell counts (ie, Q4) seemed to have a longer median duration of response to alefacept relative to patients who experienced less of a reduction in these memory T-cell subset counts. EAUC indicates the area under the percentage of change from the baseline curve for blood lymphocyte counts during the dosing interval; Q, quartile of EAUC (for CD4+ memory T cells, Q1, ≤2979, cumulative reduction in T cells [CRT]; Q2, 2980-4561 CRT; Q3, 4562-5669 CRT; and Q4, >5669 CRT; for CD8+ memory T cells, Q1, ≤3840 CRT; Q2, 3841-5621 CRT; Q3, 5622-6680 CRT; and Q4, >6680 CRT).

The mechanism of action of alefacept causes a remission that appears to be related to its effects on CD4+ and CD8+ memory T cells. This figure shows the duration of at least a 50% reduction in Psoriasis Area Severity Index (PASI) by EAUC quartile of CD4+ (A) and CD8+ (B) memory T cells in cohort 2 patients who achieved at least a 75% reduction from baseline PASI at any time during treatment or follow-up in course 1. In this subset of 52 patients, those who experienced the greatest reductions in CD4+ and CD8+ memory T-cell counts (ie, Q4) seemed to have a longer median duration of response to alefacept relative to patients who experienced less of a reduction in these memory T-cell subset counts. EAUC indicates the area under the percentage of change from the baseline curve for blood lymphocyte counts during the dosing interval; Q, quartile of EAUC (for CD4+ memory T cells, Q1, ≤2979, cumulative reduction in T cells [CRT]; Q2, 2980-4561 CRT; Q3, 4562-5669 CRT; and Q4, >5669 CRT; for CD8+ memory T cells, Q1, ≤3840 CRT; Q2, 3841-5621 CRT; Q3, 5622-6680 CRT; and Q4, >6680 CRT).

Table 1. 
Patient Demographics and Baseline Characteristics*
Patient Demographics and Baseline Characteristics*
Table 2. 
Mean Maximum Percentage Reductions During the Dosing Period*
Mean Maximum Percentage Reductions During the Dosing Period*
1.
Wrone-Smith  TNickoloff  BJ Dermal injection of immunocytes induces psoriasis. J Clin Invest. 1996;981878- 1887
PubMedArticle
2.
Gottlieb  SLGilleaudeau  PJohnson  R  et al.  Response of psoriasis to a lymphocyte-selective toxin (DAB389IL-2) suggests a primary immune, but not keratinocyte, pathogenic basis. Nat Med. 1995;1442- 447
PubMedArticle
3.
Bos  JDHagenaars  CDas  PKKrieg  SRVoorn  WJKapsenberg  ML Predominance of "memory" T cells (CD4+, CDw29+) over "naïve" T cells (CD4+, CD45R+) in both normal and diseased human skin. Arch Dermatol Res. 1989;28124- 30
PubMedArticle
4.
Szabo  SKHammerberg  CYoshida  YBata-Csorgo  ZCooper  KD Identification and quantitation of interferon-γ producing T cells in psoriatic lesions: localization to both CD4+ and CD8+ subsets. J Invest Dermatol. 1998;1111072- 1078
PubMedArticle
5.
Austin  LMOzawa  MKikui  TWalters  IBKrueger  JG The majority of epidermal T cells in psoriasis vulgaris lesions can produce type I cytokines, interferon-γ, interleukin-2, and tumor necrosis factor-α, defining TC1 (cytotoxic T lymphocyte) and TH1 effector populations: a type 1 differentiation bias is also measured in circulating blood T cells in psoriatic patients. J Invest Dermatol. 1999;113752- 759
PubMedArticle
6.
Friedri  MKrammig  SHenze  MDöcke  WDSterry  WAsadullah  K Flow cytometric characterization of lesional T cells in psoriasis: intracellular cytokine and surface antigen expression indicates an activated, memory/effector type 1 immunophenotype. Arch Dermatol Res. 2000;292519- 521
PubMedArticle
7.
Jeffes 3rd  EWLee  GCSaid  S  et al.  Elevated numbers of proliferating mononuclear cells in the peripheral blood of psoriatic patients correlate with disease severity. J Invest Dermatol. 1995;105733- 738
PubMedArticle
8.
Sigmundsdóttir  HGudjónsson  JEJónsdóttir  ILúdvíksson  BRValdimarsson  H The frequency of CLA+ CD8+ T cells in the blood of psoriasis patients correlates closely with the severity of their disease. Clin Exp Immunol. 2001;126365- 369
PubMedArticle
9.
Singri  PWest  DPGordon  KB Biologic therapy for psoriasis: the new therapeutic frontier. Arch Dermatol. 2002;138657- 663
PubMedArticle
10.
Lebwohl  MAli  S Treatment of psoriasis, I: topical therapy and phototherapy. J Am Acad Dermatol. 2001;45487- 498
PubMedArticle
11.
Lebwohl  MAli  S Treatment of psoriasis, I: systemic therapies. J Am Acad Dermatol. 2001;45649- 661
PubMedArticle
12.
Wingren  AGParra  EVarga  M  et al.  T cell activation pathways: B7, LFA-3, and ICAM-1 shape unique T cell profiles. Crit Rev Immunol. 1995;15235- 253
PubMedArticle
13.
Miller  GTHochman  PSMeier  W  et al.  Specific interaction of lymphocyte function-associated antigen 3 with CD2 can inhibit T cell responses. J Exp Med. 1993;178211- 222
PubMedArticle
14.
Majeau  GRMeier  WJimmo  BKioussis  DHochman  PS Mechanism of lymphocyte function-associated molecule 3-Ig fusion proteins inhibition of T cell responses: structure/function analysis in vitro and in human CD2 transgenic mice. J Immunol. 1994;1522753- 2767
PubMed
15.
Sanders  MEMakgoba  MWSharrow  SO  et al.  Human memory T lymphocytes express increased levels of three cell adhesion molecules (LFA-3, CD2, and LFA-1) and three other molecules (UCHL1, CDw29, and Pgp-1) and have enhanced IFN-γ production. J Immunol. 1988;1401401- 1407
PubMed
16.
Majeau  GRWhitty  AYim  KMeier  WHochman  PS Low affinity binding of an LFA-3/IgG1 fusion protein to CD2+ T cells is independent of cell activation. Cell Adhes Commun. 1999;7267- 279
PubMedArticle
17.
Ellis  CNKrueger  GGfor the Alefacept Clinical Study Group, Treatment of chronic plaque psoriasis by selective targeting of memory effector T lymphocytes. N Engl J Med. 2001;345248- 255
PubMedArticle
18.
Fredriksson  TPettersson  U Severe psoriasis—oral therapy with a new retinoid.  Dermatologica. 1978;157238- 244
PubMedArticle
19.
Ellis  CKrueger  GShrager  Dfor the Alefacept Clinical Study Group, Repeated courses of alefacept therapy in chronic plaque psoriasis provide consistent efficacy and safety [abstract]. J Eur Acad Dermatol Venereol. 2001;15 (suppl 2) 246Abstract P24-38
20.
Morganroth  GSChan  LSWeinstein  GDVoorhees  JJCooper  KD Proliferating cells in psoriatic dermis are comprised primarily of T cells, endothelial cells, and factor XIIIa+ perivascular dendritic cells. J Invest Dermatol. 1991;96333- 340
PubMedArticle
21.
Krueger  GGPapp  KAStough  DB  et al.  A randomized, double-blind, placebo-controlled phase III study evaluating efficacy and tolerability of 2 courses of alefacept in patients with chronic plaque psoriasis. J Am Acad Dermatol. 2002;47821- 833
PubMedArticle
22.
Gyulai  RSugiyama  HKobayashi  S  et al.  Alefacept treatment for psoriasis reduces the number of activated T cells in lesional skin and correlates with reduced PASI scores.  Poster presented at: International Psoriasis Symposium, European Congress on Psoriasis June 22, 2001 San Francisco, Calif
23.
Kobayashi  SSugiyama  HGyulai  R  et al.  Alefacept treatment for psoriasis reduces the number of infiltrating IFNγ+-producing T cells in lesional skin [abstract]. J Invest Dermatol. 2001;117546Abstract 938
24.
Sugiyama  HKobayashi  SGyulai  R  et al.  Alefacept inhibits IFNγ production by normal and psoriatic peripheral blood T cells and enhances the action of UVB [abstract]. J Invest Dermatol. 2001;117465Abstract 455
25.
Gottlieb  ABCasale  TBFrankel  E  et al.  CD4+ T-cell–directed antibody responses are maintained in patients with psoriasis receiving alefacept: results of a randomized study.  J Am Acad Dermatol. 2003;49816- 825
PubMedArticle
Study
December 2003

Treatment of Psoriasis With AlefaceptCorrelation of Clinical Improvement With Reductions of Memory T-Cell Counts

Author Affiliations

From the Department of Medicine, Division of Dermatology, Loyola University Medical Center, Maywood, Ill (Dr Gordon); Biogen, Inc, Cambridge, Mass (Drs Vaishnaw and O'Gorman); Coley Pharmaceutical Group, Inc, Wellesley, Mass (Mr Haney); and Department of Dermatology, Baylor University Medical Center, Dallas, Tex (Dr Menter). Members of the Alefacept Clinical Study Group are listed on page 1569. Drs Gordon and Menter have received research support from Biogen, Inc.

Arch Dermatol. 2003;139(12):1563-1570. doi:10.1001/archderm.139.12.1563
Abstract

Objective  To examine the relationship between the pharmacodynamic and antipsoriatic effects of alefacept, a biologic agent that targets CD4+ and CD8+ memory T cells.

Design  Randomized, double-blind, placebo-controlled study of 3 parallel groups.

Setting  Fifty-one study centers.

Patients  Five hundred fifty-three patients with chronic plaque psoriasis.

Interventions  Patients were randomized (1:1:1) to 1 of the following 3 cohorts: alefacept, 7.5 mg, in both courses (cohort 1); alefacept, 7.5 mg, in the first course and placebo in the second course (cohort 2); or placebo in the first course and alefacept, 7.5 mg, in the second course (cohort 3). In each course, alefacept or placebo was administered by intravenous bolus once weekly for 12 weeks, followed by 12 weeks of observation.

Main Outcome Measures  Circulating lymphocyte levels and the Psoriasis Area Severity Index.

Results  One or 2 courses of alefacept reduced CD4+ and CD8+ memory T-cell counts, while sparing the naive population. At 12 weeks after the last dose of alefacept in courses 1 and 2, 88% and 83% of patients, respectively, had CD4+ cell counts greater than the lower limit of normal. In course 1, alefacept-treated patients with the largest decreases in memory T-cell counts experienced the greatest reductions in disease activity (P<.001). The duration of clinical benefit seemed to be longer among patients who had the greatest reduction in CD4+ and CD8+ memory T-cell counts.

Conclusions  One or 2 courses of intravenous alefacept reduced circulating memory T-cell counts while sparing the naïve T-cell population. The reductions in memory T-cell counts were related to all measures of disease activity evaluated and the duration of response to therapy, suggesting that prolonged remissions of psoriasis can be attained with reduction of the pathogenic T-cell count.

SCIENTIFIC AND clinical evidence now points to the pivotal role of the T cell in the pathogenesis of psoriasis. Experiments using the severe combined immunodeficiency disorder (SCID) mouse model strongly suggested a central role for T cells.1 Elimination of T cells using the selective lymphocyte toxin, DAB389IL-2, improved disease severity and reversed the epidermal alterations that define this chronic disease.2 The lesional skin of patients with psoriasis contains an abundance of CD4+ and CD8+ T cells that belong to the memory subset (CD4+CD45RO+ and CD8+CD45RO+) and express an activated phenotype (CD25+ and CD69+),36 therefore rendering these cells as attractive targets for therapeutic intervention. In the blood, a correlation has been found between the numbers of circulating CD4+ and CD8+ activated T cells and the severity of psoriasis.7,8 The CD4+ and CD8+ T cells are capable of producing type I inflammatory cytokines (eg, interferon-γ, interleukin 2) that are purported to be important for keratinocyte hyperproliferation and disease.4 Recent clinical research in psoriasis has therefore focused on biologic therapies that interfere with 1 or more of the key steps necessary for perpetuating T-cell immune responses.9 More selective targeting of the immune system may improve tolerability compared with currently available treatment options.10,11

Alefacept (Amevive; Biogen Inc, Cambridge, Mass) is a fully human fusion protein consisting of the first extracellular domain of leukocyte function–associated antigen 3 fused to the hinge, CH2, and CH3 domains of IgG1. The interaction between leukocyte function–associated antigen 3 on antigen-presenting cells and CD2 on T cells represents a key costimulatory pathway for T-cell activation.12 The leukocyte function–associated antigen 3 domain of alefacept blocks this interaction by binding CD2 on T cells,13,14 whereas engagement of FcγRIII receptors on accessory cells by the IgG1 domain of alefacept induces T-cell apoptosis in vitro.14 Furthermore, it has been shown that T cells belonging to the memory subset, which express the highest levels of CD2 cells on their surface, are preferentially targeted by alefacept.15,16 In a phase 2 study, treatment of patients with psoriasis with a once-weekly regimen of intravenous alefacept for 12 weeks had no significant effect on naïve (CD45RA+) T cells, but significantly decreased the level of circulating memory T cells.17 Patients with clear or almost clear results after treatment cessation did not require retreatment with antipsoriatic therapy for a median of 10 months.17 It has been theorized that reductions in levels of circulating memory T cells would correlate with clinical efficacy and the prolonged responses seen with alefacept.9 The present study examined the relationship between the pharmacodynamic and antipsoriatic effects of intravenous alefacept in a large phase 3 trial involving patients with chronic plaque psoriasis.

METHODS

Institutional review boards approved the study protocol. Standard written consent was obtained from all patients, and the study was conducted in accordance with the ethics principles outlined in the Declaration of Helsinki.

PATIENTS

Patients eligible for enrollment were 16 years or older with moderate to severe stable plaque psoriasis for at least 12 months that involved at least 10% of body surface area and normal CD4+ lymphocyte counts. Patients with erythrodermic, guttate, or generalized pustular forms of psoriasis, serious local or systemic infection within the previous 3 months, or a history of malignancy other than basal cell carcinomas or fewer than 3 squamous cell carcinomas were excluded. The following treatments were prohibited within 4 weeks before study drug administration and throughout the study: phototherapy, systemic retinoids, systemic steroids, systemic fumarates, methotrexate, cyclosporine, azathioprine, thioguanine, or high-potency topical corticosteroids. Use of moderate-potency topical corticosteroids, vitamin D analogues, keratolytics, and coal tar was prohibited within 2 weeks before study drug administration and throughout the study, except on the scalp, palms, groin, anal fold, and soles. Sparing use of low-potency topical corticosteroids and emollients was permitted, except within 12 hours of efficacy assessments.

STUDY DESIGN

This was a multicenter (51 sites in the United States and Canada), double-blind, parallel-group study consisting of 2 treatment courses. Patients were randomized (1:1:1) to 1 of the following 3 cohorts: alefacept, 7.5 mg, in both courses (cohort 1), alefacept, 7.5 mg, in the first course and placebo in the second course (cohort 2), or placebo in the first course and alefacept, 7.5 mg, in the second course (cohort 3). Patients were randomized to treatment based on initial Psoriasis Area Severity Index (PASI) (≤20 or >20) and previous treatment history (received or did not receive systemic therapy/phototherapy). The study drug inventory and patient randomization and tracking was the responsibility of ICTI, Inc, Lambertville, NJ. In each course, treatment was administered by means of 30-second intravenous bolus once weekly for 12 weeks, followed by 12 weeks of observation. Patients were eligible for course 2 if the Physician Global Assessment (PGA) was worse than clear (see the following subsection, "Evaluations"), the CD4+ lymphocyte count was at or above the lower limit of normal, and prohibited medications (see the "Patients" subsection in the "Methods" section) had not been received. In each course, dosing was withheld for 2 weeks for fever or evidence of clinically significant infection. The scheduled alefacept dosage was substituted with placebo if the CD4+ lymphocyte count from the previous week was below 250 cells/µL. A CD4+ lymphocyte count below 250 cells/µL for 4 consecutive visits resulted in permanent placebo substitution.

Alefacept and placebo were identical in appearance. An unmasked pharmacist prepared, coded, and dispensed the treatments and maintained drug accountability at each study site. Also at each site, a physician who had no contact with the patients in this study reviewed all laboratory results and other safety data as needed and interacted with the pharmacist to modify the treatment schedule, if necessary. The examining physician or study coordinator, both of whom were masked to cohort assignment, administered the treatments. The examining physician performed all clinical evaluations.

EVALUATIONS

Serial blood samples were collected at screening, weekly during treatment, and every 2 to 4 weeks during follow-up for determination of levels of circulating total lymphocytes and the following lymphocyte subsets by means of flow cytometry: CD4+ and CD8+ memory and naïve T cells, B cells (CD19+), and natural killer cells (CD16+/CD56+). Baseline values for all lymphocyte counts were measured at the same time, just before the first dose of alefacept or placebo in course 1. Analysis was performed by Covance Central Laboratories, Indianapolis, Ind.

To relate the cumulative effect of alefacept on memory T cells with antipsoriatic efficacy, the area under the percentage of change from the baseline curve for blood lymphocyte counts during the dosing interval (EAUC) was calculated for each patient, as described in Figure 1. In each course, each cohort of patients was divided into quartiles based on the EAUC (Q1 indicates the lowest quartile; Q4, the highest), and the percentages of patients achieving at least a 75% reduction from baseline PASI or a PGA of clear or almost clear at any time during the treatment and follow-up periods, without the use of phototherapy or systemic therapy, were expressed graphically. The PASI, which ranges from 0 (no psoriasis) to 72 (most severe disease possible), combined scores for erythema, induration, desquamation, and body surface area affected.18 The PGA used a 7-point scale ranging from clear (no signs of psoriasis) to severe (very marked plaque elevation, scaling, and/or erythema). In each course, PASI and PGA evaluations were performed before treatment, every other week during treatment, and every 2 to 4 weeks during follow-up.

SAMPLE SIZE DETERMINATION

The primary goal of the study was to demonstrate the efficacy of alefacept relative to placebo in reducing the severity of chronic plaque psoriasis. The sample size estimate was calculated using data from a previous study approximating the 75% reduction from baseline in PASI response rate at 1 week after dosing for the alefacept (0.075 mg/kg) group (25%) and the placebo group (10%).17 Using a 2-sided χ2 test with 95% power and a type I error of 5%, 370 patients (2 of 3 receiving active drug and 1 of 3 receiving placebo) were needed to demonstrate the efficacy of alefacept in course 1. Assuming that 50% of patients would have a PGA of clear, withdraw from treatment, or fail to achieve CD4+ T-cell counts above 250 cells/µL between the 2 dosing periods, an additional 185 patients were required to determine the efficacy of alefacept in course 2, for a total accrual of 555 patients.

STATISTICAL METHODS

A Cochran-Armitage trend test evaluated the relationship between the magnitude of reduction in memory T-cell levels and the prospectively defined end point of the percentage of patients who achieved at least a 75% reduction in PASI at any time during treatment or follow-up. This analysis was performed for alefacept-treated patients in course 1 (cohorts 1 and 2 combined). A log-rank test was used to compare Q4 with Q1 through Q3 (pooled) with respect to the prospectively defined end point of median duration of at least a 50% reduction in PASI among cohort 2 patients who achieved at least a 75% reduction in PASI at any time during treatment or follow-up in course 1. This analysis was performed for cohort 2 because these patients were followed for up to 36 months after the last alefacept dose (ie, 12 months of follow-up, 12 months of placebo, and another 12 months of follow-up).

RESULTS
BASELINE CHARACTERISTICS

Study enrollment began on November 21, 1999; the last date of follow-up was March 22, 2001. A total of 553 patients were randomized and received at least 1 injection (cohort 1, n = 183; cohort 2, n = 184; cohort 3, n = 186). In course 1, 490 patients (89%) completed the 12-week treatment period, and 482 (87%) completed the 12-week follow-up period. A total of 449 patients (cohort 1, n = 154; cohort 2, n = 142; cohort 3, n = 153) received treatment in course 2. Of these, 417 (93%) completed the 12-week treatment period and 401 (89%) completed the 12-week follow-up period. Demographic and baseline characteristics were well balanced across the 3 cohorts in course 1 (Table 1).

Of the 553 patients in course 1, 551 underwent 1 or more assessments of levels of total lymphocytes or lymphocyte subsets after dosing and were included in the analysis of pharmacodynamic effects in course 1. All of the 449 patients who received treatment in both courses underwent 1 or more assessments of levels of total lymphocytes or lymphocyte subsets after dosing and were included in the analysis of pharmacodynamic effects in both courses.

PHARMACODYNAMICS
Course 1

Consistent with its mechanism of action, alefacept reduced circulating total lymphocyte counts and CD4+ and CD8+ lymphocyte subset counts (Table 2). Reductions were selective for memory T cells with relative sparing of naïve T cells (Figure 2 and Table 2). The median time of most marked reduction of CD4+CD45RO+ and CD8+CD45RO+ T-cell levels during the dosing period was 71 days for cohorts 1 and 2 (alefacept), and 36 and 43 days, respectively, for cohort 3 (placebo). Thus, reductions in lymphocyte counts were evident in both active treatment groups, as expected based on the mechanism of action of alefacept. In the placebo group, 6% and 1% of patients had at least 1 CD4+ T-cell count below 400 and 300 cells/µL, respectively, demonstrating the normal variation in T-cell counts in patients with psoriasis.

In contrast to memory T-cell subsets, changes in naïve T-cell counts were negligible. The mean percentage of changes in CD4+CD45RA+ T-cell counts from baseline to 12 weeks after the last dose in course 1 for cohorts 1 and 2 were 3% and 0%, respectively. Similar results were observed for CD8+CD45RA+ T-cell counts. Alefacept had no notable effect on CD19+ B cells or CD16+/CD56+ natural killer cells.

In cohort 1, 98% of patients had total lymphocyte counts, 87% had CD4+ T-cell counts, and 70% had CD8+ T-cell counts above the lower limit of normal at 12 weeks after the last alefacept dose in course 1. Corresponding results for cohort 2 patients were similar: 98% for total lymphocyte counts, 89% for CD4+ T-cell counts, and 64% for CD8+ T-cell counts. At the end of the 12-week follow-up interval of course 1, mean total lymphocyte counts relative to baseline were −12% for cohort 1, −17% for cohort 2, and 5% for cohort 3. Corresponding percentages were −23%, −29%, and 3%, respectively, for CD4+ T-cell counts and −24%, −35%, and 2%, respectively, for CD8+ T-cell counts.

During course 1, placebo was substituted for alefacept treatment at least once in 10% of patients in cohort 1 and 11% of patients in cohort 2. There were 7 (1.9%) permanent placebo substitutions in course 1 (2 patients in cohort 1 and 5 patients in cohort 2). One placebo-treated patient in cohort 3 (0.5%) had a CD4+ T-cell count below 300 cells/µL.

Course 2

We found no evidence of a cumulative effect of alefacept on total lymphocyte count or CD4+, CD8+, and CD45RO+ lymphocyte subset counts. Similar maximum reductions of blood lymphocyte counts were reached during courses 1 and 2 (Table 2 and Figure 2). The nadirs (mean count at the time of the most marked reduction during the dosing period) for cohort 1 were not significantly lower for course 2 vs course 1 with respect to total lymphocyte (1303 vs 1284 cells/µL) or CD4+ (467 vs 479 cells/µL), CD8+ (234 vs 250 cells/µL), CD4+CD45RO+ (212 vs 232 cells/µL), or CD8+CD45RO+ cell counts (52 vs 63 cells/µL). Thus, a nadir was reached during course 1 and did not change during course 2. As observed during the first alefacept course, changes in CD19+ B and CD16+/CD56+ natural killer cell counts were negligible during the second alefacept course.

Rates of recovery in course 2 were similar to those in course 1. In cohort 1, 97% of patients had total lymphocyte counts, 83% had CD4+ T-cell counts, and 65% had CD8+ T-cell counts above the lower limit of normal at 12 weeks after the last dose in the second alefacept course. In the same cohort, the mean total lymphocyte count relative to the course 1 baseline was −13% at the end of the 12-week follow-up interval of course 2. Corresponding percentages were −23% for CD4+ and −31% for CD8+ T-cell counts.

In course 2, placebo was substituted for alefacept treatment at least once in 9% of patients in cohort 1 and 5% of patients in cohort 3. There was 1 permanent placebo substitution (0.3%) in course 2 (cohort 3). No patient who received 2 courses of alefacept required permanent placebo substitution during the second course.

RELATIONSHIP BETWEEN PHARMACODYNAMICS AND EFFICACY

In courses 1 and 2, the relationship between the magnitude of change in CD4+CD45RO+ T-cell counts and probability of response was consistent, regardless of the definition of response (Figure 3). Patients with the most profound and sustained reductions in this T-cell subset showed the greatest reduction in disease activity. For example, in course 1, the percentages of alefacept-treated patients in the lowest quartile (Q1, 0-2209, cumulative reduction in T cells [CRT]) who achieved at least a 75% reduction from baseline PASI at any time during treatment or follow-up were 13% (6/46) (cohort 1) and 20% (9/46) (cohort 2). These percentages increased to 41% (18/44) and 47% (22/47), respectively, in the highest quartile (Q4, 4723-7660 CRT). In course 1, among alefacept-treated patients (cohorts 1 and 2 combined), there was a significant trend for increasing response rates (≥75% reduction in PASI) with increasing CD4+CD45RO+ EAUC quartile (P<.001). Analysis of changes in CD8+CD45RO+ T-cell counts and response rates showed a similar trend (P<.001). A similar association was observed between quartiles of memory T-cell effects and the percentages of alefacept-treated patients (cohorts 1 and 2 combined) who achieved at least a 50% reduction from baseline PASI at any time during treatment or follow-up: 33% in Q1 (30/92), 51% in Q2 (47/92), 62% in Q3 (57/92), and 75% in Q4 (68/91) for CD4+CD45RO+ T cells and 36% in Q1 (33/92), 51% in Q2 (47/92), 59% in Q3 (55/93), and 77% in Q4 (69/90) for CD8+CD45RO+ T cells.

In course 2, despite receiving placebo, cohort 2 patients continued to benefit from the first course of alefacept, which is indicative of the remittive effects of the study drug (Figure 3). For example, in the highest CD4+CD45RO+ EAUC quartile, 47% (22/47) of cohort 2 patients achieved at least a 75% reduction from baseline PASI at any time after the first alefacept dose in course 1, whereas 36% (10/28) of cohort 2 patients met this criterion at any time after the initiation of placebo treatment in course 2. Cohort 1 patients had additional clinical improvement during the second alefacept course relative to the first alefacept course, particularly among patients in the highest CD4+CD45RO+ EAUC quartile (Figure 3). In this latter quartile, more cohort 1 patients achieved at least a 75% reduction from baseline PASI in course 2 (62%; 37/60) than in course 1 (41%; 18/44). For PGA, an assessment of clear or almost clear was also achieved in 21% more cohort 1 patients in the second (53%; 32/60) vs the first (32%; 14/44) alefacept courses. The results for CD8+CD45RO+ T cells paralleled those of CD4+CD45RO+ T cells.

The finding that the pharmacodynamic effects of alefacept drive the clinical response is further supported by the results of the time-course analysis (Figure 4). The decline in CD4+ T-cell counts in alefacept-treated patients preceded the improvement in PASI. The maximal mean percentage change from baseline in CD4+ T-cell count occurred during the treatment period, whereas the maximal mean percentage change from baseline PASI occurred during the follow-up period after treatment had been completed. The time-course analysis of CD8+ T-cell count reduction and PASI improvement yielded similar results (data not shown).

The mechanism of action of alefacept would be expected to result in prolonged remissions of psoriasis symptoms after treatment cessation.9 Kaplan-Meier analysis demonstrated that the duration of response to alefacept was indicative of a remittive therapy and suggested a relationship between the duration of therapeutic effect and the magnitude of the reduction in CD4+ and CD8+ memory T-cell counts. Figure 5 shows the median duration of at least a 50% reduction in PASI by CD4+CD45RO+ (Figure 5A) and CD8+CD45RO+ (Figure 5B) EAUC quartiles in cohort 2 patients who achieved at least a 75% reduction from baseline PASI at any time during treatment or follow-up in course 1. In this subset of 52 patients, those who experienced the greatest reductions in memory T-cell count (ie, Q4) had a longer median duration of response to alefacept (271.1 days for CD4+ memory T cells; 240.6 days for CD8+ memory T cells) relative to patients who experienced a smaller reduction in memory T-cell count (173.2, 194.5, and 193.9 days for Q1, Q2, and Q3, respectively, for CD4+ memory T cells; 173.2, 228.0, and 221.4 days, for Q1, Q2, and Q3, respectively, for CD8+ memory T cells). Due to the limited sample size in this analysis, a statistically significant result was not achieved.

COMMENT

One or 2 courses of alefacept induced a decrease in circulating absolute counts of the CD4+ and CD8+ memory T-cell subsets from baseline while sparing the naïve population. These results are consistent with the preferential binding of alefacept to memory T cells, which have a higher level of expression of CD2 compared with that observed in the naïve compartment.15,16 A cumulative effect on lymphocyte subsets was not observed in this trial or in an earlier multiple-course study in patients with psoriasis.19 In this study, similar maximum reductions in memory T-cell counts and comparable rates of recovery were observed during both alefacept courses for cohort 1 patients. There was also a consistent lack of effect on naïve T cells during 2 courses of alefacept therapy.

Previous studies have demonstrated that most lymphocytes in psoriatic lesions are memory T cells3,20 and that these T cells are distinct from naïve T cells in that they have been previously activated by an antigen or primed.3 If memory T cells are the effector cells that drive psoriatic activity,3,20 then a relationship would be expected between the extent of reduction in this population and clinical outcome of alefacept therapy. The results of this trial strongly support a critical role for memory T cells in the pathogenesis of psoriasis. First, reductions in the memory T-cell subset counts were related to improvements in clinical symptoms, as measured by the PASI. Most patients who experienced the greatest reductions in memory T-cell counts (≥75%) achieved a clinically meaningful response (≥50% PASI reduction) to therapy. These results suggest that alefacept-induced T-cell changes in the blood are a surrogate for changes in the skin. Second, most of the alefacept-induced reductions in T-cell count preceded the peak improvement in PASI, which was not unexpected considering that alefacept interferes with memory T-cell levels and activity, but not further downstream in the T-cell–mediated inflammatory pathway (eg, inhibition of cytokines).

The remittive effects of alefacept also are consistent with a key role of memory T cells in psoriasis pathogenesis. Once memory T-cell levels are reduced by alefacept, it likely takes a prolonged period of time to build them up again to provoke symptom reappearance. Although it was previously speculated that the mechanism of action of alefacept would result in lasting remissions,9 no definitive data have been available until now. However, several observations of the present study results confirm the prolonged benefit of alefacept on the symptoms of psoriasis. For example, cohort 2 patients who were followed up for approximately 36 months after a single 12-week dosing period continued to benefit from the initial alefacept course during the second placebo course. Moreover, the clinical response to alefacept peaked and was sustained during the observation period well after the last dose of study drug. In cohort 2, the median duration during both courses of at least a 50% reduction from baseline PASI was 7 to 8 months in responders, defined as patients who achieved at least a 75% reduction from baseline PASI or a PGA of clear or almost clear at any time during the study.21 The present analysis suggested that the duration of clinical benefit was longer among patients who had the greatest reduction in memory T-cell levels. Although this will need further investigation, it is an expected result if memory T cells drive the psoriatic process.

The results of the present trial parallel those of previous studies investigating the effect of alefacept on T-cell subsets in the skin of patients with psoriasis.22,23 Administration of intravenous alefacept, 7.5 mg once weekly for 12 weeks, reduced activated T-cell subset counts (CD3+CD69+ and CD3+CD25+) in lesional skin, and these changes correlated significantly with PASI improvement (r = 0.95 and r = 0.79, respectively). Moreover, a decrease in the number of interferon-γ–producing T cells in lesional skin correlated with PASI improvement (r = 0.80). Interferon-γ T-cell production was similarly inhibited in the blood (serum) of patients treated with alefacept.24 Further study of the relationship between T-cell changes in blood and skin and clinical outcome is under investigation.

In the present study, T-cells began to return toward baseline levels during the treatment-free follow period after alefacept administration, although recovery was not complete for all patients. Despite the alefacept-induced reductions in T-cell counts, no related safety concerns were identified during this 2-course study.21 As described elsewhere,21 no opportunistic infections were reported, and there was no association between infections and CD4+ T-cell counts. In a separate study, a single course of alefacept therapy did not produce detectable alterations in antibody responses to an experimental neoantigen or recall antigen challenge.25 Further longer-term, multiple-course studies are needed to provide a more complete assessment of the safety of alefacept as it relates to reduction of T-cell counts.

CONCLUSIONS

This phase 3 trial indicates that the reduction in levels of circulating memory T-cell subsets with alefacept treatment correlates to improvement in psoriasis and suggests a relationship between the length of response to alefacept and alterations in the memory T-cell population. These data strongly support the central role of memory T cells in psoriasis and the potential efficacy advantages of targeting the effector cells in this disease. Although this study does not evaluate individual clinical responses in relation to the pharmacodynamic effects of alefacept, it clearly suggests that further study of the T-cell effects of alefacept with particular emphasis on treatment protocols is warranted.

Corresponding author and reprints: Kenneth B. Gordon, MD, Division of Dermatology, Loyola University Medical Center, 2160 S First Ave, Maywood, IL 60153 (e-mail: kgordon@lumc.edu).

Accepted for publication April 30, 2003.

This study was funded by Biogen, Inc, Cambridge, Mass.

Preliminary results of this study were presented at the 44th Annual Meeting of the American Society of Hematology; December 2002; Philadelphia, Pa; and the Hawaii Dermatology Seminar; February 2003; Maui.

We thank Dan Freedman, Kursten Haggett, Jeff Haney, Gina Leone, Frances Lynn, Arthur McAllister, John O'Gorman, Jimmy Scaramucci, and Gloria Vigliani of Biogen, Inc.

Article
The Alefacept Clinical Study Group

Jerold Powers (Scottsdale, Ariz), Marc Epstein (Tucson, Ariz), Dow Stough (Hot Springs, Ark), David Tashjian (Fresno, Calif), Nick Lowe (Santa Monica, Calif), Karl Beutner (Vallejo, Calif), J. Michael Maloney (Denver, Colo), Ronald Savin (New Haven, Conn), Armand Cognetta, Jr (Tallahassee, Fla), Donna Zeide (West Palm Beach, Fla), Tiffani Hamilton (Atlanta, Ga), James Aton (Martinez, Ga), Michael Tharp (Chicago, Ill), Holly Hahn (Indianapolis, Ind), Larry Millikan (New Orleans, La), Owen Reynolds (North Andover, Mass), Daniel Stewart (Clinton Township, Mich), Marie Hordinsky (Minneapolis, Minn), John Ervin (Kansas City, Mo), Craig Leonardi (St Louis, Mo), Linda Ford (Papillion, Neb), W. Thomas Garland (Lawrenceville, NJ), Alice Gottlieb (New Brunswick, NJ), Mark Lebwohl, Matthew Stiller, and Ken Washenik (New York, NY), Lauren Gandhi (Winston-Salem, NC), Neil Korman (Cleveland, Ohio), Michelle Chambers (Columbus, Ohio), Mark Naylor and Steven Sutter (Oklahoma City, Okla), Diane Baker (Lake Oswego, Ore), Harold Farber (Philadelphia, Pa), John DiGiovanna (Providence, RI), John Humeniuk (Greer, SC), Keith Loven (Goodlettsville, Tenn), Michael Gold (Nashville, Tenn), Julius van Bavel (Austin, Tex), Frank Hampel, Jr (New Braunfels, Tex), Gerald Krueger (Salt Lake City, Utah), and Bernard Goffe (Seattle, Wash) in the United States; and Jason Rivers (Vancouver, British Columbia), Marc Bourcier (Moncton, New Brunswick), Wayne Gulliver (St John's, Newfoundland), Richard Langley (Halifax, Nova Scotia), Lynn Guenther (London, Ontario), Charles Lynde (Markham, Ontario), Kim Papp (Waterloo, Ontario), Jerry Tan (Windsor, Ontario), Robert Bissonnette and David Gratton (Montreal, Quebec); Yves Poulin (Sainte-Foy, Quebec); and Peter Hull (Saskatoon, Saskatchewan) in Canada.

References
1.
Wrone-Smith  TNickoloff  BJ Dermal injection of immunocytes induces psoriasis. J Clin Invest. 1996;981878- 1887
PubMedArticle
2.
Gottlieb  SLGilleaudeau  PJohnson  R  et al.  Response of psoriasis to a lymphocyte-selective toxin (DAB389IL-2) suggests a primary immune, but not keratinocyte, pathogenic basis. Nat Med. 1995;1442- 447
PubMedArticle
3.
Bos  JDHagenaars  CDas  PKKrieg  SRVoorn  WJKapsenberg  ML Predominance of "memory" T cells (CD4+, CDw29+) over "naïve" T cells (CD4+, CD45R+) in both normal and diseased human skin. Arch Dermatol Res. 1989;28124- 30
PubMedArticle
4.
Szabo  SKHammerberg  CYoshida  YBata-Csorgo  ZCooper  KD Identification and quantitation of interferon-γ producing T cells in psoriatic lesions: localization to both CD4+ and CD8+ subsets. J Invest Dermatol. 1998;1111072- 1078
PubMedArticle
5.
Austin  LMOzawa  MKikui  TWalters  IBKrueger  JG The majority of epidermal T cells in psoriasis vulgaris lesions can produce type I cytokines, interferon-γ, interleukin-2, and tumor necrosis factor-α, defining TC1 (cytotoxic T lymphocyte) and TH1 effector populations: a type 1 differentiation bias is also measured in circulating blood T cells in psoriatic patients. J Invest Dermatol. 1999;113752- 759
PubMedArticle
6.
Friedri  MKrammig  SHenze  MDöcke  WDSterry  WAsadullah  K Flow cytometric characterization of lesional T cells in psoriasis: intracellular cytokine and surface antigen expression indicates an activated, memory/effector type 1 immunophenotype. Arch Dermatol Res. 2000;292519- 521
PubMedArticle
7.
Jeffes 3rd  EWLee  GCSaid  S  et al.  Elevated numbers of proliferating mononuclear cells in the peripheral blood of psoriatic patients correlate with disease severity. J Invest Dermatol. 1995;105733- 738
PubMedArticle
8.
Sigmundsdóttir  HGudjónsson  JEJónsdóttir  ILúdvíksson  BRValdimarsson  H The frequency of CLA+ CD8+ T cells in the blood of psoriasis patients correlates closely with the severity of their disease. Clin Exp Immunol. 2001;126365- 369
PubMedArticle
9.
Singri  PWest  DPGordon  KB Biologic therapy for psoriasis: the new therapeutic frontier. Arch Dermatol. 2002;138657- 663
PubMedArticle
10.
Lebwohl  MAli  S Treatment of psoriasis, I: topical therapy and phototherapy. J Am Acad Dermatol. 2001;45487- 498
PubMedArticle
11.
Lebwohl  MAli  S Treatment of psoriasis, I: systemic therapies. J Am Acad Dermatol. 2001;45649- 661
PubMedArticle
12.
Wingren  AGParra  EVarga  M  et al.  T cell activation pathways: B7, LFA-3, and ICAM-1 shape unique T cell profiles. Crit Rev Immunol. 1995;15235- 253
PubMedArticle
13.
Miller  GTHochman  PSMeier  W  et al.  Specific interaction of lymphocyte function-associated antigen 3 with CD2 can inhibit T cell responses. J Exp Med. 1993;178211- 222
PubMedArticle
14.
Majeau  GRMeier  WJimmo  BKioussis  DHochman  PS Mechanism of lymphocyte function-associated molecule 3-Ig fusion proteins inhibition of T cell responses: structure/function analysis in vitro and in human CD2 transgenic mice. J Immunol. 1994;1522753- 2767
PubMed
15.
Sanders  MEMakgoba  MWSharrow  SO  et al.  Human memory T lymphocytes express increased levels of three cell adhesion molecules (LFA-3, CD2, and LFA-1) and three other molecules (UCHL1, CDw29, and Pgp-1) and have enhanced IFN-γ production. J Immunol. 1988;1401401- 1407
PubMed
16.
Majeau  GRWhitty  AYim  KMeier  WHochman  PS Low affinity binding of an LFA-3/IgG1 fusion protein to CD2+ T cells is independent of cell activation. Cell Adhes Commun. 1999;7267- 279
PubMedArticle
17.
Ellis  CNKrueger  GGfor the Alefacept Clinical Study Group, Treatment of chronic plaque psoriasis by selective targeting of memory effector T lymphocytes. N Engl J Med. 2001;345248- 255
PubMedArticle
18.
Fredriksson  TPettersson  U Severe psoriasis—oral therapy with a new retinoid.  Dermatologica. 1978;157238- 244
PubMedArticle
19.
Ellis  CKrueger  GShrager  Dfor the Alefacept Clinical Study Group, Repeated courses of alefacept therapy in chronic plaque psoriasis provide consistent efficacy and safety [abstract]. J Eur Acad Dermatol Venereol. 2001;15 (suppl 2) 246Abstract P24-38
20.
Morganroth  GSChan  LSWeinstein  GDVoorhees  JJCooper  KD Proliferating cells in psoriatic dermis are comprised primarily of T cells, endothelial cells, and factor XIIIa+ perivascular dendritic cells. J Invest Dermatol. 1991;96333- 340
PubMedArticle
21.
Krueger  GGPapp  KAStough  DB  et al.  A randomized, double-blind, placebo-controlled phase III study evaluating efficacy and tolerability of 2 courses of alefacept in patients with chronic plaque psoriasis. J Am Acad Dermatol. 2002;47821- 833
PubMedArticle
22.
Gyulai  RSugiyama  HKobayashi  S  et al.  Alefacept treatment for psoriasis reduces the number of activated T cells in lesional skin and correlates with reduced PASI scores.  Poster presented at: International Psoriasis Symposium, European Congress on Psoriasis June 22, 2001 San Francisco, Calif
23.
Kobayashi  SSugiyama  HGyulai  R  et al.  Alefacept treatment for psoriasis reduces the number of infiltrating IFNγ+-producing T cells in lesional skin [abstract]. J Invest Dermatol. 2001;117546Abstract 938
24.
Sugiyama  HKobayashi  SGyulai  R  et al.  Alefacept inhibits IFNγ production by normal and psoriatic peripheral blood T cells and enhances the action of UVB [abstract]. J Invest Dermatol. 2001;117465Abstract 455
25.
Gottlieb  ABCasale  TBFrankel  E  et al.  CD4+ T-cell–directed antibody responses are maintained in patients with psoriasis receiving alefacept: results of a randomized study.  J Am Acad Dermatol. 2003;49816- 825
PubMedArticle
×