Gordon KB, Papp KA, Hamilton TK, Walicke PA, Dummer W, Li N, Bresnahan BW, Menter A, for the Efalizumab Study Group . Efalizumab for Patients With Moderate to Severe Plaque PsoriasisA Randomized Controlled Trial. JAMA. 2003;290(23):3073-3080. doi:10.1001/jama.290.23.3073
Author Affiliations: Loyola University Medical Center, Maywood, Ill (Dr Gordon); Probity Medical Research, Waterloo, Ontario (Dr Papp); Atlanta Dermatology, Vein and Research Center LLC, Atlanta, Ga (Dr Hamilton); Genentech Inc, South San Francisco, Calif (Drs Walicke, Dummer, Li, and Bresnahan); and Baylor University Medical Center, Dallas, Tex (Dr Menter).
Context Because T-cell interactions are involved in the pathophysiology of psoriasis,
therapy with a T-cell modulator may have beneficial effects on psoriasis severity
and health-related quality of life (HRQL).
Objective To assess the efficacy and safety of efalizumab, a T-cell modulator,
in patients with plaque psoriasis.
Design, Setting, and Patients Phase 3 randomized, double-blind, parallel-group, placebo-controlled
trial involving 556 adult patients with stable, moderate to severe plaque
psoriasis and conducted at 30 study centers in the United States and Canada
between January and July 2002.
Interventions Patients were randomly assigned in a 2:1 ratio to receive 12 weekly
doses of subcutaneous efalizumab, 1 mg/kg (n = 369), or placebo equivalent
(n = 187).
Main Outcome Measures At least 75% improvement on the Psoriasis Area and Severity Index (PASI-75);
improvement on the overall Dermatology Life Quality Index (DLQI), Itching
Visual Analog Scale (VAS), and Psoriasis Symptom Assessment (PSA) at week
12 vs baseline.
Results Efalizumab-treated patients experienced significantly greater improvement
on all end points than placebo-treated patients. Twenty-seven percent of efalizumab-treated
patients achieved PASI-75 vs 4% of the placebo group ( P<.001). Efalizumab-treated patients exhibited significantly greater
mean percentage improvement than placebo-treated patients on the overall DLQI
(47% vs 14%; P<.001), Itching VAS (38% vs −0.2%; P<.001), and PSA frequency and severity subscales (48%
vs 18% and 47% vs 17%, respectively; P<.001 for
both) at the first assessment point. Efalizumab was safe and well tolerated,
with primarily mild to moderate adverse events.
Conclusion In this 12-week study, efalizumab resulted in significant improvements
in clinical end points, including physician-assessed and dermatology-specific
patient-reported HRQL measures, in patients with moderate to severe plaque
Psoriasis, a disease affecting millions of persons worldwide, is a chronic
inflammatory disease that has a profound adverse effect on patients' physical,
social, and mental well-being.1- 6 In
the National Psoriasis Foundation survey of more than 17 000 respondents,
79% of patients with severe disease reported that psoriasis adversely affected
their lives and limited activities of daily living (eg, work and school participation).3
The physical symptoms of psoriasis adversely affect daily functioning,
with the most frequently reported symptoms including scaling, itching, and
inadequate control of these symptoms, physical, social, and mental functioning
and overall health-related quality of life (HRQL) are compromised.8,9 For example, the physical appearance
of lesions can cause patients to experience stress and embarrassment6,10 and have adverse effects on emotional
aspects and normal functioning.11 Up to 10%
of psoriasis patients, especially younger patients, harbor suicidal thoughts
compared with approximately 3% of general medical patients, indicating a significant
unmet medical need.3,12
Efalizumab is a humanized monoclonal IgG1 antibody that targets T-cell
interactions central to the pathophysiology of psoriasis.13- 23 In
this study, the effect of efalizumab on dermatology-related HRQL in patients
with moderate to severe plaque psoriasis was examined using a broad set of
outcome measures including physicians' assessments and patients' perceptions.
This phase 3 randomized, double-blind, parallel-group, placebo-controlled
multicenter trial evaluated the efficacy and safety of efalizumab (anti-CD11a;
Raptiva, Genentech Inc) in adults with moderate to severe psoriasis. Eligible
patients were 18 to 75 years of age, diagnosed as having plaque psoriasis
for at least 6 months with at least 10% of total body surface area (BSA) affected,
had a minimum Psoriasis Area and Severity Index (PASI) score of 12.0 at screening,
and were candidates for systemic therapy. Patients could withdraw or be withdrawn
from the study at any time. All sites received institutional review board
approval prior to initiating the study, and all patients provided written
Patients were randomly assigned to receive 12 weekly doses of either
subcutaneous efalizumab, 1 mg/kg, or placebo equivalent, as per the original
study protocol (Figure 1). Allocation
sequences were generated by an unblinded statistician at Genentech Inc, the
study sponsor, who then passed them to the interactive voice response system
vendor. The sites enrolled and randomized patients by calling the interactive
voice response system, which assigned patients to either efalizumab or placebo
using a random permuted-block design to obtain approximately a 2:1 ratio within
categories defined by the stratification variables: baseline PASI score (≤16.0
vs ≥16.1), prior treatment for psoriasis (yes vs no), and study center.
Each patient received an initial conditioning dose of 0.7 mg/kg followed
by 11 weekly doses of 1 mg/kg of study drug (efalizumab or placebo equivalent).
After 12 weeks of placebo-controlled treatment, all patients were enrolled
in a separate long-term open-label extension study.
Patients received efalizumab monotherapy, with all phototherapy and
systemic therapy excluded. Patients were permitted to use emollients. Tar
and salicylic acid preparations were permitted for the scalp, and low-potency
topical corticosteroids were allowed for the face, hands, feet, groin, and
axillae. Patients, investigators, sponsor, and the contract research organization
were blinded regarding treatment assignment or active study drug until the
analyses of all data were complete.
The primary efficacy outcome measure was the proportion of patients
with at least 75% improvement in PASI (PASI-75) at week 12 relative to baseline
(Box). The PASI is a physician-assessed
score, recognized by the US Food and Drug Administration24 to
assess efficacy of psoriasis therapies in clinical trials, that takes into
account the extent of involved skin surface area and severity of erythema,
desquamation, and plaque induration.25 The
composite score ranges from 0 to 72, with higher numbers indicating more severe
disease and a reduction in score representing improvement. The PASI components
of lesion thickness, erythema, and scaling were examined separately. The PASI-75
is the currently recognized benchmark of end points used in psoriasis clinical
trials.26 For each study site, PASI assessments
were performed by a single investigator who was blinded to randomization.
Psoriasis Area and Severity Index (PASI): Assesses
the extent of psoriasis on 4 body surface areas (head, trunk, and upper and
lower limbs) and the degree of plaque erythema, scaling, and thickness. The
PASI scores account for both the extent of body surface area affected by the
erythema, scaling, and thickness and the severity of these measures. The score
ranges from 0 (no disease) to 72 (maximal disease).Overall Lesion Severity Scale (OLS): Static global assessment with
6 categories (clear, minimal, mild, moderate, severe, and very severe) based
on the characteristics of plaque elevation, scaling, and erythema.Physician's Global Assessment (PGA): Captures and categorizes
the global response to therapy of all clinical signs and symptoms of disease
relative to baseline. Physicians use all available information for the assessment,
including subjective information gathered from the patient and photographs
taken at baseline. The categories are worse, unchanged, slight, fair, good,
excellent, and cleared.
Dermatology Life Quality Index (DLQI): The
DLQI is a 10-item questionnaire that incorporates patients' assessments of
itch, pain, feelings of embarrassment and self-consciousness, problems with
their psoriasis treatment, and interference of their psoriasis with daily
activities, relationships, and sexual activity. The DLQI scores range from
0 (no impairment) to 30 (maximal impairment).Itching
Visual Analog Scale (VAS): The 10-point Itching VAS is a modified VAS
with scores ranging from 0 (no itching) to 10 (severe itching).Psoriasis Symptom Assessment (PSA): The PSA is a 16-item
measure of 8 psoriasis-related cutaneous symptoms (hurt, burning or stinging,
itched, bothered by water, irritated, sensitive, skin condition bled, scaling).
The PSA contains 2 subscales, one measuring the frequency of the 8 symptoms
and the other assessing how troublesome or bothersome psoriasis symptoms are.
Secondary efficacy measures included 2 global physician assessments:
a static measure designated the Overall Lesion Severity Scale (OLS) and a
dynamic measure designated the Physician's Global Assessment (PGA). Additional
secondary outcome measures were the proportion of patients with at least 50%
improvement on the PASI (PASI-50) relative to baseline, the percentage of
PASI improvement over time, PASI thickness component, and the percentage of
BSA affected by psoriasis.
Patient-reported outcomes were evaluated using prospectively defined
mean improvement in the overall Dermatology Life Quality Index (DLQI), Itching
Visual Analog Scale (VAS), and Psoriasis Symptom Assessment (PSA) frequency
and severity subscales. The overall improvement for all patient-reported outcomes
was assessed at week 12 relative to baseline.
The DLQI measures the impact of skin disorders and treatment on patient
functioning and well-being using a 7-day recall period.27 The
psychometric characteristics relating to the validity and reliability of the
DLQI are well established.27,28 The
Itching VAS used in previous trials of efalizumab has been shown to be a valid
and reliable measure.28 The PSA is an adaptation
of a validated skin disorder instrument, the Skindex,29,30 with
2 additional questions related to the frequency and severity of skin scaling
using a 2-week recall period. The PSA, also used in prior efalizumab studies,
has been established to be a valid, reliable, and responsive measure.28
Baseline measurements were those made closest but prior to administration
of the first dose of study drug. The PASI was assessed at baseline and every
2 weeks thereafter. The patient-reported outcomes (DLQI, Itching VAS, and
PSA) were performed prior to obtaining physician assessments for measures
(eg, PASI) at baseline and at weeks 4, 8, and 12.
The safety and tolerability of efalizumab were monitored by adverse
events, vital signs, physical examination, clinical laboratory evaluation,
and serum human antihuman antibody measurement. Adverse events were assessed
weekly. Standard methods to assess safety were used.
The sample size was based primarily on safety considerations. With a
planned sample size of 333 patients in the efalizumab group and 167 patients
in the placebo group, using a 2-sided Fisher exact test, the study had 99%
power to detect a difference between the assumed placebo response rate of
5% and the assumed efalizumab response rate of 25% at an α = .05 significance
level. Analyses were conducted with nQuery Advisor software, version 4.0 (Statistical
Solutions, Saugus, Mass).
The main analysis population for primary and secondary end points was
the intention-to-treat population, consisting of all patients who were randomized
whether or not they received any study drug or completed the full course of
treatment. Patients were analyzed according to their randomized treatment
group. Safety analyses were performed on an as-treated population.
The primary end point was evaluated by comparing the proportion of those
in the efalizumab group and those in the placebo group who achieved PASI-75
at week 12 relative to baseline using a 2-sided Fisher exact test for binomial
outcome at the .05 level of significance; the exact 95% confidence interval
for response rates within each treatment group and the difference in response
rates between the efalizumab and placebo group were calculated. Patients who
discontinued early from the study or whose week 12 PASI score was missing
were classified as nonresponders.
The following secondary end points were compared between treatment groups
using a 2-sided Fisher exact test for binomial outcome performed at the .05
level of significance: the proportion of patients who achieved an OLS rating
of minimal or clear at week 12, the proportion of patients with PASI-50 at
week 12 relative to day 0, and the proportion of patients who achieved a PGA
rating of excellent or cleared at week 12. Using the 2-sided t test at the .05 significance level, mean improvement from baseline
in the PASI thickness score at week 12 and mean improvement from baseline
in the percentage of BSA affected by psoriasis at week 12 were compared. The
PASI score and percentage of PASI improvement from baseline were summarized
by time point. At each point, the mean percentage improvement from baseline
was compared between the treatment groups using the t test.
A hierarchical testing procedure determined the earliest point at which a
statistically significant difference (at the .05 level) between the treatment
groups was observed. No multiple-comparison adjustment for the type I error
was made in the hierarchical testing procedure, since there was no inflation
of the type I error. To ensure an overall type I error rate of α = .05
(2-sided) for all secondary efficacy analyses, the Hochberg-Bonferroni31 multiple-comparisons procedure was used to adjust
for multiple comparisons of secondary end points; all secondary end points
retained statistical significance after multiplicity adjustments.
Mean improvements from baseline to week 12 on the overall DLQI and each
of the 2 PSA scales were compared between treatment groups using the Wilcoxon
rank sum test. The mean improvement from baseline in the Itching VAS at week
12 was compared between treatment groups using the t test.
The overall scores and improvements from baseline for all patient-reported
outcome scales were summarized by time point. At each point, the mean percentage
improvement from baseline was compared between the treatment groups using
the Wilcoxon rank sum test (DLQI and PSA) or t test
(Itching VAS). A hierarchical testing procedure was used, as for efficacy.
Between January and July 2002, 556 patients were enrolled at 30 study
centers and randomized into the study, 369 patients into the 1-mg/kg-per-week
efalizumab group and 187 patients into the placebo group (Figure 1). There were no statistically significant differences between
the 2 treatment groups with respect to demographics, baseline characteristics,
and disease severity (Table 1).
The mean PASI score at baseline for the entire study cohort was 19, and mean
BSA with psoriasis was 28% for the efalizumab-treated group and 27% for the
placebo-treated group. The proportion of patients who did not complete treatment
was similar in both groups (6.5% vs 6.4%); the most common reasons were patient
decision (n = 10), loss to follow-up (n = 9), adverse event (n = 9), use of
an excluded medication (n = 5), and investigator decision (n = 3) (Figure 1). Four hundred seventeen patients
(75%) received all 12 doses of assigned study drug.
Physician-Assessed Outcomes. At the end of
the 12-week treatment course, 27% of patients treated with efalizumab (98/369)
achieved PASI-75 compared with 4% of patients who received placebo (8/187; P<.001). The treatment effect, defined as the difference
in the proportion of patients who achieved PASI-75 between the efalizumab
and placebo groups, was 22.3% (95% confidence interval, 15.8%-29.5%). Fifty-nine
percent of efalizumab-treated patients (216/369) achieved PASI-50 compared
with 14% of patients receiving placebo (26/187; P<.001).
The mean PASI improvement at week 12 in the efalizumab-treated patients relative
to baseline was about 52% compared with 19% in the patients who received placebo
The proportion of patients with an OLS rating of minimal or clear at
week 12 in the efalizumab group was significantly higher than that in the
placebo group (26% vs 3%; P<.001). Similarly,
the proportion of patients with a PGA rating of excellent or cleared at week
12 was significantly greater in the efalizumab group compared with the placebo
group (33% vs 5%; P<.001).
Patient-Reported Outcomes. At week 12, the
mean percentage improvement from baseline in DLQI overall score was greater
in the efalizumab-treated group than in the placebo group (47% vs 14%; P<.001). Patients in the efalizumab-treated group showed
improvement in mean DLQI scores compared with the placebo-treated group at
28 days (P<.001). Given the 1-week recall period
associated with the DLQI, this difference in score between the 2 groups indicates
that reduced dermatology-related limitations and improved functioning occurred
fairly soon after initiating efalizumab therapy. There was a statistically
significant trend toward a positive efalizumab treatment effect across all
DLQI components (Table 2), with
the greatest difference between the treatment groups at week 12 observed in
the "symptoms and feelings" domain (48% vs 18%). Additional analyses showed
the summary statistics on DLQI improvement stratified by PASI response at
week 12 in the 369 patients who were randomized to receive efalizumab, with
the greatest improvement observed in the patients who achieved PASI-50 (Table 3).
When patients with a baseline Itching VAS score of 0 (9 patients in
the efalizumab group and 3 in the placebo group) were excluded from the analysis,
efalizumab treatment was shown to produce a 38% improvement in Itching VAS
score, compared with a slight worsening in score (−0.2%) in the placebo
group at 12 weeks (P<.001). A significant improvement
in the mean Itching VAS in the efalizumab group relative to the placebo group
was observed at the first time point measured (P<.001).
Comparisons of the mean percentage improvement from baseline in the
efalizumab group vs the placebo group were statistically significant for both
frequency of symptoms (48% vs 18%; P<.001) and
severity of symptoms (eg, troubling or bothersome) (47% vs 17%; P<.001). Improvement in both mean PSA frequency and mean PSA severity
subscale scores was evident in the efalizumab group vs the placebo group at
the first time point at which PSA was measured (P<.001)
(Figure 3). There was a trend toward
improvement across all symptoms in both frequency and severity in the efalizumab
group, as evidenced by the mean improvement in PSA scores (Table 2). With respect to the frequency of symptoms, differences
between absolute improvements were similar for most symptoms (itching, irritation,
sensitivity, bleeding, and scaling). With respect to severity, or how bothersome
the symptoms were, the largest absolute improvements occurred in both the
itching (mean improvement of 1.1 for efalizumab vs 0.3 for placebo; P<.001) and scaling (1.2 for efalizumab vs 0.4 for placebo; P<.001) components.
Efalizumab therapy was generally well tolerated. All adverse events
that occurred in at least 5% of all patients are shown in Table 4. Five types of adverse events (headache, chills, fever,
myalgia, and pain) occurred at least 5% more frequently in the efalizumab
group than in the placebo group. These events tended to be a portion of the
complex of mild to moderate flulike symptoms that occur following the first
1 to 2 injections of efalizumab.
Serious adverse events were infrequent, occurring in 2% of efalizumab-treated
patients (9/368) and 1% of placebo-treated patients (1/187). No deaths were
reported for patients during the study. Fourteen patients (3%) (12 [3%] in
the efalizumab group and 2 [1%] in the placebo group) experienced adverse
events that resulted in withdrawal of study drug.
No clinically significant laboratory abnormalities or pattern of changes
in vital signs were observed during efalizumab treatment. Eight patients (2%)
developed positive antibodies to efalizumab after exposure to the drug. None
of these patients experienced serious adverse events or discontinued treatment.
No patients in the placebo group developed anti-efalizumab antibodies.
The rate of diagnosed infections was 27% for the efalizumab group and
23% for the placebo group. Infections that occurred at least 1% more frequently
in the efalizumab-treated patients than in the placebo-treated patients included
mild to moderate viral infection (primarily mild to moderate viral upper respiratory
tract infections), bacterial infection (eg, impetigo, streptococcal pharyngitis),
cellulitis, and fungal (yeast) infection. Infections graded as severe occurred
in 0.5% of patients in both the efalizumab and placebo groups. Efalizumab-treated
patients did not exhibit increased susceptibility to any one type of pathogen,
nor was there evidence of infection characteristic of opportunistic infections
observed in immunocompromised hosts.
No cases of anaphylaxis were observed during efalizumab therapy. During
the study, 2 cases of malignancy were diagnosed (1 case each of squamous cell
cancer at day 2 and basal cell cancer at day 77) in efalizumab-treated patients,
although neither was judged as being related to efalizumab by investigators.
Limitations of currently available psoriasis therapies highlight the
need for effective and safe treatment options. Traditional systemic therapies
are associated with cumulative toxic effects, potentially increasing the risk
of end-organ damage or malignancy.32,33 In
this phase 3 randomized, placebo-controlled trial, efalizumab therapy resulted
in significant improvement in the primary end point and other prespecified
efficacy end points. The efficacy, accompanied by the safety profile and the
HRQL outcomes, provides evidence of potential benefit for efalizumab in patients
with psoriasis. By week 12, 27% of efalizumab-treated patients achieved a
PASI-75 response compared with 4% of placebo-treated patients. These results
are consistent with the physician-reported outcomes from another recently
reported trial of efalizumab.34 Efalizumab-treated
patients demonstrated significant benefit on the additional physician-assessed
global psoriasis scales, OLS and PGA, compared with placebo-treated patients.
Consistent and significant improvements on multiple dermatology- and psoriasis-specific
patient-reported end points were observed throughout the study.
The magnitude of the percentage improvement in HRQL was greater than
that for PASI in the early stage of treatment. Improvement in the individual
PASI components occurred early; however, it is likely that the BSA component,
which improves less rapidly, underestimates the improvement occurring in the
lesions. This suggests that concentrating solely on PASI improvement might
underestimate the timing and overall response to efalizumab.
Efalizumab treatment reduced the frequency and severity of psoriasis
symptoms, particularly in the severity of itching and scaling, the 2 most
frequently reported subjective symptoms.3 The
severity and frequency of other symptoms, including bleeding and burning or
stinging, also were improved by efalizumab treatment. The significant improvement
on the DLQI indicated consistent improvements across measures of social and
mental assessment (Table 2), areas
known to be adversely affected in patients with psoriasis. Efalizumab-treated
patients reported greater improvement in their attitudes about their disease,
their ability to participate in daily activities (including leisure and work),
and their personal relationships, and they reported fewer treatment-related
problems compared with placebo.
Efalizumab exhibited a favorable safety profile and was generally well
tolerated. The most commonly occurring adverse events during the initiation
of efalizumab treatment were mild to moderate flulike symptoms following the
first 1 to 2 injections. Serious adverse events were infrequent and occurred
at only a slightly greater frequency in the efalizumab group (2%) than in
the placebo group (1%). There was no evidence to suggest clinically relevant
increases in either infection or malignancy, including lymphoma, among efalizumab-treated
patients; however, longer follow-up will be needed to accurately characterize
the risk of infection and malignancy.
There were no clinically relevant laboratory abnormalities. Previous
efalizumab studies in psoriasis patients have shown that the discrete elevations
in lymphocyte counts, white blood cell counts, and other hematologic parameters
are transient, usually within normal ranges, with values returning to baseline
following efalizumab discontinuation.35- 37 There
was no evidence of general systemic toxicity reflected by changes in hepatic
or renal function.
This investigation is limited by the relatively short duration and by
lack of an active comparator. Since psoriasis is a chronic disease, outcomes
with longer-term use of efalizumab are important. The efficacy and safety
of longer-term efalizumab treatment has been further evaluated in an extension
of this study that analyzes up to 24 weeks of continuous efalizumab therapy.
The preliminary analysis of data from this study suggests continued improvement
in disease severity and maintenance of safety. Additionally, an ongoing open-label
study evaluating the efficacy and safety of up to 3 years of continuous efalizumab
therapy is currently being conducted. Efficacy with continued use of efalizumab
has also been suggested in a recently reported clinical trial.34 The
current study did not compare efalizumab directly with any of the currently
used therapies for psoriasis. Additional randomized studies would be needed
to compare this profile with other currently available therapies.
Efalizumab was recently approved for treatment of patients with chronic
moderate to severe plaque psoriasis.38 The
benefit across physician-assessed end points and multiple patient-reported
measures of HRQL observed in this study along with the favorable safety profile
suggest that efalizumab could provide a viable treatment option for patients
with moderate to severe plaque psoriasis.