On December 27, 2002 Genentech submitted to the Center for Biologics Evaluation and Research a Biologics License Application (STN BL 125075/0) for efalizumab for the treatment of psoriasis.

1.2 Drug Product

Efalizumab is a recombinant humanized IgG1 kappa isotype monoclonal antibody that selectively binds to human CD11a and has an approximate molecular weight of 150 kD.

The protein is produced by Chinese hamster ovary cells.

The drug product is provided as a sterile lyophilized powder to deliver 125 mg of efalizumab. Reconstitution with 1.3 mL of supplied Sterile Water for Injection yields a clear to slightly opalescent solution containing 100 mg/mL efalizumab, 0.2% polysorbate 20, 40 mM histidine, 240 mM sucrose, and SWFI at a pH of 6.2. Although the drug is produced in a suspension culture containing gentamicin, gentamicin is not detectable in the final product.

1.3 Rationale and Hypothesis

CD11a is the a subunit of lymphocyte function-associated antigen (LFA-1), a b2 integrin, and is expressed on all leukocytes. Efalizumab binds specifically to the CD11a alpha chain of LFA-1 and blocks the binding of LFA-1 to its ligand intercellular adhesion molecule 1 (ICAM-1). Binding of CD11a by efalizumab results in saturation of available CD11a binding sites and down-modulation of cell surface CD11a expression. This event is believed to decrease the activation of lymphocytes and reduce their translocation to peripheral tissues (such as in psoriatic plaques).

Activated T lymphocytes may play a role in autoimmune diseases including plaque‑type psoriasis. Blocking or reducing T lymphocyte activation and migration may improve the clinical manifestations of psoriasis.

1.4 Proposed Indication: Plaque Psoriasis

Psoriasis is a chronic skin disorder characterized by erythematous, scaly papules and plaques with a predisposition for the scalp, extensors of the limbs, lumbosacral area and genitalia. The condition affects between 1 and 3% of the general population. However, it is relatively infrequent among African-Americans, in Japanese populations and in the Native American population. Men and women are equally affected.

Psoriasis has a bimodal peak of onset, one in adolescents and young adults (at 16 to 22 years of age) and the second in older persons (at age 57-60). Onset is before the age of 15 in 27% of cases. Early onset disease is strongly linked to HLA -Cw6 and DR7, while late onset disease is linked to HLA-Cw2. The predisposition to psoriasis is thought to be polygenic with expression triggered by environmental factors such as streptococcal infection, stress, certain drugs, and HIV. The cause of psoriasis is not fully known.

Psoriasis is characterized by excessive proliferation of keratinocytes and inflammation. There is evidence that activated T cells are involved in the pathogenesis of psoriasis. In addition, abnormalities in cytokine expression, intracellular signaling, and polyamine metabolism may mediate psoriasis.

Plaque psoriasis is the most common form. The lesions are indurated/raised, erythematous and scaly. Approximately 1/3 of patients have moderate to severe disease. The disease waxes and wanes. Spontaneous remissions and relapses are the rule. Spontaneous durable remissions may occur.

Guttate (drop-like) psoriasis is sometimes triggered by streptococcal infection and is associated with development of chronic psoriasis. Pustular psoriasis varies in severity from localized to generalized forms with fever, malaise, and a relatively high mortality after prolonged courses. Erythroderma can be complicated by sepsis, temperature instability and high output cardiac failure. Psoriatic arthritis is a complication in approximately 10% of all psoriasis patients.

Patients with psoriasis report reduction in mental and physical functioning comparable to that seen in patients with cancer or arthritis. The chief complaints of patients with psoriasis are scaling, itching, redness and tightness of the skin, bleeding and burning sensations. In a 1998 National Psoriasis Foundation Patient-Membership survey, patients reported depression, difficulties in the workplace and socialization caused by psoriasis.

The goal of treatment of psoriasis is to decrease the severity and extent of psoriasis to the point that it no longer interferes with the patient’s occupation, personal or social life, or well-being.

1.5 Licensed Therapies for Psoriasis

Topical Therapy

The initial treatment of stable plaque psoriasis affecting less than 10-20% of body surface area is topical. Topical therapies include emollients, corticosteroids, anthralin, tar, retinoids, calcipotriene, and salicylic acid. The mainstay of treatment is topical corticosteroids. Topical corticosteroids induce skin atrophy, striae, purpura and may be absorbed systemically leading to suppression of the hypothalamic-pituitary-adrenal axis. Another possible limiting factor to their use is tachyphylaxis. Other commonly used topical agents include calcipotriene (a vitamin D analogue), tazarotine (a retinoid prodrug) and anthralin. Salicylic acid is used as a keratolytic agent. Skin irritation is the most common adverse effect of these topical agents.

Phototherapy

Phototherapy for psoriasis includes UVB, narrow band UVB, and psoralen, a photosensitizer, plus UVA (PUVA). PUVA induces responses in a high proportion of patients and can induce long-term remissions. PUVA causes premature aging of skin and increases the risk of cutaneous malignancy in a dose-related fashion. The relative increase in risk of a person with sun-sensitive skin (e.g. Fitzpatrick Type I or II skin; always burn; tan never/sometime) developing squamous cell carcinoma is at least 5 times greater than that of control.

Systemic Therapy

Methotrexate, cyclosporin, and retinoids, in general, induce moderate improvement in the majority of treated patients. These products are recommended for severe and/or recalcitrant psoriasis because they induce serious toxicities. Methotrexate, an antimetabolite folate analogue, may cause bone marrow toxicity with leukopenia, dose-dependent development of cirrhosis of the liver, severe pneumonitis and lymphomas. Methotrexate is also fetotoxic and an abortifacient. Cyclosporine, an immunosuppressant calcineurin inhibitor, induces hypertension, nephrotoxicity, increased risk of malignancy (especially B cell lymphoma) and infection. Retinoids are the treatment of choice for pustular psoriasis and have also been used in the treatment of erythrodermic psoriasis. Of major consideration in women of childbearing potential is teratogenicity of retinoids. Other serious adverse events are hepatotoxicity, pancreatitis, depression, visual impairment, and hyper-triglyceridemia.

Alefacept, an immunosuppressive and the first biologic agent to receive FDA approval for the treatment of moderate-to-severe psoriasis, results in 75% clearing in 10% (by IV route) 16% (by IM route) of patients. Remissions may last for months. The drug induces lymphopenia and requires monitoring of CD4+ T lymphocyte counts on a regular basis.

Immunosuppressive Agents and Anti-metabolites: Risk/benefit in Psoriasis

Psoriasis is a serious chronic disease associated with significant morbidity and impairment. The disease is usually not life threatening and does not induce irreversible injury to skin or other organs, with the exception of psoriatic arthritis. A number of serious toxicities are associated with the use of immunosuppressants and antimetabolites. These include serious infections, and neoplasms. In the case of neoplasms there may be a lag in the time to clinical detection and long-term follow-up of treated patients may be required to assess the excess risk. Therapies associated with significant risk of serious irreversible toxicity or mortality should be reserved for patients with severe, recalcitrant psoriasis. The goal of therapy is to bring disease under control and change to the least toxic therapy.

1.6 Licensing Status of Drug Product

At the time this application was submitted, efalizumab was not licensed in any country, nor had it been withdrawn from the market in any country.

1.7 Disclosure of Financial Interests and Arrangements of Clinical Investigators

At the time this application was submitted, none of the clinical investigators (from whom a response was received) disclosed financial interest in either Genentech or Genentech’s partner, XOMA, Ltd.

1.8 Debarment Certification

Genentech has provided certification that it did not and will not use the services of anyone debarred under Subsections A or B of Section 306 of the Food, Drug and Cosmetics Act in connection with this application.

2 CLINICAL STUDIES OF EFALIZUMAB AND REGULATORY HISTORY

Two sponsors, XOMA, Ltd. and Genentech, Inc., participated in the development of efalizumab for moderate to severe plaque psoriasis.

XOMA sponsored the phase 1 and 2 clinical studies and manufactured efalizumab used in those studies.
Genentech sponsored the phase 3 studies and is the current manufacturer of the to-be-marketed efalizumab product.
Study ACD2058g, the first phase 3 study, studied exclusively efalizumab manufactured by XOMA.
Most of the patients in study ACD2059g, received XOMA-manufactured efalizumab. In the latter part of the study Genentech-manufactured product was introduced.

In this study, the PK, PD and clinical activity of the XOMA-manufactured and Genentech-manufactured efalizumab products were compared.

Subsequent phase 3 studies, ACD2390g and ACD2600g, studied exclusively the Genentech-manufactured product.

September 2001:

The agency expressed concerns about the comparability of the efalizumab manufactured by XOMA and Genentech and recommended that a PK comparability study (ACD2389) be performed in healthy volunteers.

June 2002:

Study ACD2389g showed that the XOMA- and Genentech- produced efalizumab were equivalent pharmacodynamically, but were not pharmacokinetically equivalent. The Genentech-manufactured efalizumab appeared to have higher bioavailability and/or slower clearance. The ratio of geometric means for AUCinf of Genentech and XOMA

efalizumab was 1.32, with a 90% confidence interval of 1.19–1.47, above the 0.80–1.25 range specified for comparability.

These results prompted the FDA to request additional phase 3 studies for safety and efficacy of the Genentech-manufactured product.

November 2002:

Study ACD2390g showed that 1 mg/kg/wk SC of the Genentech-manufactured efalizumab was superior to placebo. The Agency agreed that the data were adequate for filing a licensing application and recommended that the XOMA- and Genentech-manufactured efalizumab databases be analyzed separately and also be pooled for the BLA submission.

Table 1 provides a listing of the clinical studies of efalizumab in patients with psoriasis and summarizes the number of patients treated and the duration of treatment as of May 2003.

Table 1 Efalizumab Studies: Psoriasis Subjects Receiving at Least One Dose of Efalizumab

				No. of Subjects
		Dose	Treatment	Treated 1^st Time With
		(mg/kg) and	Duration	Treated 1^st Time With
Study	Phase and Design	Route	(wk)	XOMA	GNE
HU9602	1, open-label	0.03–10.0 IV	1	31	NA
HUPS249	1, open-label	0.1–1.0 IV	7	39	NA
HUPS252	2, placebo-controlled	0.1, 0.3 IV	8	97	NA
HUPS254	1, open-label	0.5–2.0 SC	1–8	52	NA
HUPS256	1, open-label	0.3–1.0 IV	12	11	NA
		1.0–4.0 SC	12	57	NA
ACD2058g	3, placebo-controlled	1.0–2.0 SC	12–24	462	NA
ACD2059g	3, placebo-controlled	1.0–4.0 SC	12–24	442	137
ACD2062g	3, open-label extension study to ACD2058g	1.0–2.0 SC	12	28	6
ACD2142g	1, open-label	1.0–2.0 SC	12	NA	70
ACD2243g	3, open-label	2.0 SC, then	12	NA	339
		1.0–2.0 SC	³ 48
ACD2390g	3, placebo-controlled	1.0 SC	12	NA	368
ACD2391g	3, open-label extension to ACD2390g	1.0 SC	24	NA	174
ACD2600g	3, placebo-controlled	1.0 SC	12	NA	449
Subjects with psoriasis receiving efalizumab by manufacturer				1219	1543
Subjects with psoriasis receiving efalizumab					2762

In addition to phase 1 and 2 trials, four phase 3 double blinded, randomized, placebo controlled trials were conducted. Long-term exposure data were provided by studies ACD2058g and ACD2059g (24 weeks of treatment), by study ACD2243g (48 weeks of treatment), and by the open-label extension studies. The total safety database consisted of over 2500 patients exposed to efalizumab.

3 SUMMARY OF THE PHASE 1 AND 2 CLINICAL EXPERIENCE

XOMA conducted the Phase 1 studies (trials HU9602, HUPS249, HUPS254, and HUPS256), which characterized efalizumab’s intravenous (IV) and subcutaneous (SC) pharmacokinetic and pharmacodynamic properties in patients with psoriasis and obtained preliminary evidence of activity in psoriasis. Xoma also conducted one Phase 2 clinical study (HUPS252). It was determined from single-dose studies that adverse events including fever, headache and nausea were seen shortly after the intravenous infusion of efalizumab. In multiple-dose studies, these adverse events were most common after the first dose, hence the phenomenon was called a “first-dose” effect. These adverse events were also dose-related. This led to the development of an initial low “tolerization dose” that decreased the incidence and severity of the adverse events associated with dosing.

An efalizumab-treated patient experienced acute unilateral hearing loss in the phase 2 study. This finding lead to the inclusion of audiologic testing during the first of the phase 3 trials, Study ACD2058g (See Appendix 1).

4 PHASE 3 CLINICAL TRIALS

The four Phase 3, placebo-controlled studies were as follows:

Study ACD2058g: a randomized, double-blind study evaluating 12 weeks of therapy with SC administered XOMA efalizumab, followed by a placebo-controlled period with either continued treatment for 12 additional weeks or retreatment for an additional 12 weeks after relapse

Study ACD2059g: a randomized, placebo-controlled, double-blind study evaluating 12 weeks of therapy with SC administered efalizumab (~75% XOMA, ~25% Genentech), followed by a second placebo-controlled period with either continued active treatment or placebo for 12 weeks

Study ACD2390g: a randomized, double-blind study evaluating 12 weeks of therapy with SC administered Genentech efalizumab

Study ACD2600g: a randomized, double-blind, placebo-controlled trial, was conducted to provide additional controlled safety data with the Genentech-manufactured efalizumab.

The sponsor has also conducted two open-label phase 3 clinical trials.

Study ACD2062 was an open-label trial to assess the safety and efficacy of retreatment with efalizumab.

Study ACD2243g is an ongoing trial to evaluate the safety and efficacy of long-term maintenance with efalizumab. The interim data from this study provide the 1-year safety data to support this BLA submission.

4.1 Issues Explored in the Efficacy Trials

Some of the issues explored in the efficacy trials were as follows:

The lack of comparability of the pharmacokinetics of the XOMA-manufactured and the Genentech-manufactured efalizumab made it necessary to further study the Genentech material for safety and efficacy.
The safety and efficacy of long-term continuous treatment and retreatment upon relapse were explored.
The safety of treatment discontinuation was evaluated.
Correlations of efficacy as measured by PASI and other measures such as static and dynamic physician’s global assessment were performed.

Studies ACD2058g and ACD2059g were designed to evaluate the safety and efficacy of continuous therapy for a total of 6 months. In addition, retreatment upon relapse among patients classified as responders after the first treatment course was studied in Study ACD2058g.

4.2 Psoriasis Outcomes used in the Clinical Efficacy Trials

4.2.1 Primary efficacy outcome

A 75% improvement from baseline in the PASI (Psoriasis Area and Severity Index) score (Fredriksson et al, 1978) was the primary efficacy outcome used in the clinical trials. PASI scoring is discussed below.

PASI Scoring

PASI can range from 0 to 72. Dermatologic disease severity is scored as follows:

Body Areas

Four main body areas are assessed, the head (h), the trunk (t), the upper

extremities (u), and the lower extremities (l) corresponding to 10%, 30%, 20%,

and 40% of the total body surface area, respectively.

The area of psoriatic involvement for each body area (Ah, At, Au, Al) is

assigned a numerical value according to degree of involvement as follows:

0 = no involvement

1 = <10% involvement

2 = 10% to <30% involvement

3 = 30% to <50% involvement

4 = 50% to <70% involvement

5 = 70% to <90% involvement

6 = 90% to 100% involvement

The severity of the psoriatic lesions in three main signs—erythema (E), thickness (T), and scaling (S)—are assessed for each body area according to a scale (0–4) in which 0 represents a complete lack of cutaneous involvement and 4 represents the most severe possible involvement.

Calculating PASI

To calculate the PASI, the sum of the severity rating for the three main signs are multiplied with the numerical value of the area affected and with the various

percentages of the four body areas. These values are then added to complete the formula as follows:

PASI = 0.1 (Eh + Th + Sh) Ah + 0.3 (Et + Tt + St) At +0.2 (Eu + Tu + Su) Au + 0.4 (El + Tl + Sl) Al

4.2.2 Secondary efficacy outcomes

The principal secondary efficacy outcome used in the clinical efficacy trials was a static physician’s global assessment, the Overall Lesion Severity (OLS) Scale. The scoring system is depicted in the following table (Table 2).

Table 2 Overall Lesion Severity Scale

Score	Category	Description
0	Clear	Plaque elevation = 0 (no elevation over normal skin) Scaling = 0 (no scale) Erythema = ± (hyperpigmentation, pigmented macules, diffuse faint pink or red coloration)
1	Minimal	Plaque elevation = ± (possible but difficult to ascertain whether there is a slight elevation above normal skin) Scaling = ± (surface dryness with some white coloration) Erythema = up to moderate (up to definite red coloration)
2	Mild	Plaque elevation = slight (slight but definite elevation, typically edges are indistinct or sloped) Scaling = fine (fine scale partially or mostly covering lesions) Erythema = up to moderate (up to definite red coloration)
3	Moderate	Plaque elevation = moderate (moderate elevation with rough or sloped edges) Scaling = coarser (coarse scale covering most of all of the lesions) Erythema = moderate (definite red coloration)
4	Severe	Plaque elevation = marked (marked elevation typically with hard or sharp edges) Scaling = coarse (coarse, non-tenacious scale predominates covering most or all of the lesions) Erythema = severe (very bright red coloration)
5	Very severe	Plaque elevation = very marked (very marked elevation typically with hard sharp edges) Scaling = very coarse (coarse, thick tenacious scale over most of all of the lesions; rough surface) Erythema = very severe (extreme red coloration; dusky to deep red coloration)

Clinical response was defined as “clear” or “minimal” at 12 weeks.

The Dermatology Life Quality Index (DLQI) (Finlay et al,1994) was one of the secondary efficacy outcomes. The DLQI is a 10-item questionnaire that attempts to assess how much a skin condition has affected the subject’s quality of life during the previous 7 days. Each question has four possible responses: “not at all,” “a little,” “a lot,” or “very much,” with scores of 0, 1, 2, and 3, respectively. The DLQI represents the sum of the scores, ranging from 0 to 30 points.

Dermatology Life Quality Index Questionnaire

1. Over the last week, how itchy, sore, painful Very much

or stinging has your skin been? A lot

A little

Not at all

2. Over the last week, how embarrassed or self Very much

conscious have you been because of your A lot

skin? A little

Not at all

3. Over the last week, how much has your skin Very much

interfered with you going shopping or looking A lot

after your home or garden? A little

Not at all Not relevant

4. Over the last week, how much has your skin Very much

influenced the clothes you wear? A lot

A little

Not at all Not relevant

5. Over the last week, how much has your skin Very much

affected any social or leisure activities? A lot

A little

Not at all Not relevant

6. Over the last week, how much has your skin Very much

made it difficult for you to do any sport? A lot

A little

Not at all Not relevant

7. Over the last week, has your skin prevented Yes

you from working or studying? No Not relevant

If “No,” over the last week how much has your A lot

skin been a problem at work or studying? A little

Not at all

8. Over the last week, how much has your skin Very much

created problems with your partner or any of A lot

your close friends or relatives? A little

Not at all Not relevant

9. Over the last week, how much has your skin Very much

caused any sexual difficulties? A lot

A little

Not at all Not relevant

10 Over the last week, how much of a problem Very much

. has the treatment for your skin been, for A lot

example by making your home messy, or by A little

taking up time? Not much at all Not relevant

4.3 Protocol ACD2058g

4.3.1 Study Title

“A Phase III, Randomized, Double-Blind, Parallel-Group, Placebo-Controlled, Multicenter, Multidose Study to Evaluate the Efficacy and Safety of Subcutaneously Administered Anti-CD11a in Adults with Moderate to Severe Plaque Psoriasis”

4.3.2 Study Objectives

To investigate the efficacy of weekly subcutaneous (SC) dosing with either 1.0 mg/kg or 2.0 mg/kg efalizumab relative to placebo as measured by the proportion of subjects achieving a ³75% decrease from baseline in PASI at the end of the initial 12-week treatment period (First Treatment, or FT Day 84)
To evaluate the safety and tolerability of 12 weekly SC doses of 1.0 mg/kg or 2.0 mg/kg efalizumab relative to placebo

The primary objective of the study was to assess the safety and efficacy of a 12-week treatment of efalizumab. The study was also designed to explore a number of secondary safety and efficacy questions with special emphasis on issues relevant to patients with psoriasis. These questions included:

optimization of dose
duration of treatment-free response
potential for treatment-withdrawal phenomena (e.g. flaring, psoriasis variants)
safety and efficacy of retreatment following relapse
safety and activity of extended treatment for partial responders and non-responders

4.3.3 Study Design

Study ACD2058g was a randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of efalizumab in subjects with moderate to severe psoriasis. Following the initial 84-day blinded placebo-controlled treatment period, responders were observed on no treatment until they relapsed or until 168 days (OB period). Upon relapse, patients who had initially received efalizumab were rerandomized to receive placebo or efalizumab at the same dose previously received. Responding patients in the placebo arm who relapsed during the OB period were treated with efalizumab during retreatment. Patients who were initially randomized to receive efalizumab who had no response or a partial response at Day 84 were rerandomized to continue efalizumab at the same dose they had received previously or to receive placebo in the extended treatment regimen (ET) protocol for an additional contiguous 84 days. Partial responders and non-responders who received placebo initially went on to receive efalizumab in the ET period. The study also included a follow-up period (FU) and an extended follow-up period (EFU), for people who had not relapsed by the FU Day 84, to assess duration of effect and safety. Figure 1 below shows the study schema.

Figure 1 Design of Study ACD2058g

4.3.3.1 Open Label Extension Study

Protocol ACD2062g served as the open-label extension for Study ACD2058g. Subjects who did not experience a ³50% improvement in PASI by ET day 56 (as compared to FT day 0) could transfer to open-label treatment.

4.3.3.2 Study Drug

Subjects in the active group received XOMA-manufactured efalizumab. Dosing volumes and schedules were identical during the FT, RT, and ET periods. Subjects received an initial conditioning dose of 0.7 mg/kg study drug followed by 11 weekly doses of 1.0 or 2.0 mg/kg SC study drug (efalizumab or placebo).

4.3.3.3 Randomization

During the FT period, subjects were randomized to low-dose (1.0 mg/kg) efalizumab, high dose (2.0 mg/kg) efalizumab or low dose or high dose placebo in a 2:2:1:1 ratio. Randomization was stratified by FT Day 0 PASI (£ 16, ³ 16.1), by prior treatment for psoriasis (naïve to systemic treatment vs. history of prior systemic therapy) and by study site through an IVRS. The random assignment to efalizumab or placebo was blinded to subjects, investigators and the sponsor.

At the start of the ET period, subjects who previously received efalizumab were rerandomized within the low- or high- dose group to active therapy or placebo in a 2:1 ratio. Randomization was balanced for subjects who were partial responders and non-responders.

All patients assigned to placebo in the FT period were reassigned to receive efalizumab, whether they participated in the RT or ET periods and regardless of response status during the first treatment period.

4.3.3.4 Withholding Treatment

Subjects were discontinued from efalizumab treatment if they met any of the following criteria: diagnosis of any cancer; anaphylaxis; opportunistic infection; or any medical condition that the investigator determined could jeopardize the subject’s safety if he or she continued in the study. Other reasons for discontinuation included pregnancy, administration of live virus or bacteria vaccine, or concurrent treatment with an excluded systemic or topical therapy.

If a subject had an atypical severe relapse or emergence of a new psoriatic morphology (e.g., pustular, rupioid, guttate), the investigator was to contact the Medical Monitor. If, in the judgment of the investigator, this flare required treatment, the subject had to discontinue from the study.

Treatment options for these subjects included the following:

· Immediate transfer to the open-label study, ACD2062g, for treatment with efalizumab upon approval from the Medical Monitor

· Early discontinuation from the study to begin excluded psoriasis medications. Subjects were to undergo end-of-treatment-period evaluations and immediately enter the follow-up period.

Concomitant Medications

The only concomitant psoriasis treatments that could be used until study Day 84 were Eucerin cream and tar or salicylic acid preparations (for scalp psoriasis only). Potency Group VII topical corticosteroids could be used in small amounts on psoriatic lesions on the face, hands, feet, groin, or axillae, if required (except for the day of the scheduled visit).

After FU Day 84, any topical or systemic psoriasis therapies could be used at the investigator’s discretion, even for subjects continuing in the extended follow-up EFU period through FU Day 252.

4.3.3.5 Disallowed treatments

The following were not allowed: Systemic treatments for psoriasis (e.g., PUVA, cyclosporine, corticosteroids, methotrexate, oral retinoids, MMF, thioguanine, hydroxyurea, sirolimus, azathioprine, 6-MP, etanercept) and immunosuppressive medications for any indication other than psoriasis.

Treatment with UVB phototherapy and all other topical treatments for psoriasis (e.g., topical corticosteroids, calcipotriene, tazarotene, anthralin, tar) were excluded from Day –14 through Day 84, with the exceptions noted previously. Tanning booths or nonprescription UV light sources were not to be used.

Use of live virus or bacteria vaccines were prohibited.

4.3.3.6 Eligibility

Inclusion

· Plaque psoriasis covering ³ 10% of total BSA

· A minimum PASI score of 12.0 at screening

· Plaque psoriasis diagnosed for at least 6 months

· Plaque psoriasis clinically stable for at least 3 months prior to screening

· Candidate for systemic therapy for psoriasis who had not been previously treated or history of prior treatment with systemic therapy for psoriasis (e.g., PUVA, cyclosporine, corticosteroids, methotrexate, oral retinoids

· Body weight £ 140 kg

· 18 to 70 years old

· Women of childbearing potential had to use an acceptable method of contraception to prevent pregnancy and agree to continue to practice an acceptable method of contraception for the duration of their participation in the study.

· Willingness to hold sun exposure reasonably constant and to avoid use of tanning booths or other UV light sources for the duration of the trial

Exclusion

Subjects who met any of the following exclusion criteria were ineligible for

study entry:

· Guttate, erythrodermic, or pustular psoriasis as sole or predominant form of psoriasis

· History of severe allergic or anaphylactic reactions to humanized monoclonal antibodies

· Clinically significant psoriasis flare during screening or at the time of enrollment

· History of or ongoing uncontrolled bacterial, viral, fungal, or atypical mycobacterial infection

· History of opportunistic infections (e.g., systemic fungal infections, parasites)

· History of hepatitis B or C infection

· Hepatic enzymes 3× the upper limits of normal (ULN) Subjects with hepatic enzymes elevated above the ULN but <3× the ULN had to be tested for hepatitis B and C. Only subjects with negative viral tests could be enrolled.

· Active tuberculosis (TB) or currently undergoing treatment for TB. Purified protein derivative (PPD) testing and/or a chest X-ray were required for high-risk subjects.

Presence or history of malignancy within the past 5 years, including lymphoproliferative disorders. Subjects with a history of fully resolved basal cell or squamous cell skin cancer could be enrolled.

· Previous treatment with efalizumab

· Initiation or change in treatment regimen of b-blockers, angiotensin-converting enzyme (ACE) inhibitors, interferons, quinidine, antimalarial drugs, or lithium within the past month

· Seropositivity for human immunodeficiency virus (HIV). Subjects underwent mandatory testing at screening.

· Pregnancy or lactation

· White blood cell (WBC) count <4000/µL or >14,000/µL

· Serum creatinine ³ 2× the ULN

· Progressive hearing loss

· Any medical condition that, in the judgment of the investigator, could have jeopardized the subject’s safety following exposure to study drug

4.3.3.7 Efficacy Outcomes

The primary efficacy outcome measure was the proportion of subjects with a ³ 75% improvement in PASI score between FT Day 0 and FT Day 84 (the end of the FT period).

The principal secondary efficacy outcome measure was the proportion of subjects who achieved an OLS rating of Minimal or Clear at FT Day 84.

4.3.3.8 Assessment of Safety

The following were performed: physical examinations (including vital signs and body weight); monitoring for adverse events; blood chemistry, hematology (CBC, platelet, differential); urinalysis; antibodies to efalizumab (HAHA) and urine pregnancy test (females of childbearing potential). TB skin testing was done for high risk subjects only at day 84. Audiograms were also performed in this study prior to the first study drug administration and at the end of FT (FT Day 84) ET and RT. (See appendix 1)

Drug concentration measurements, steady-state trough concentration, were performed on samples collected on FT Day 84 for HAHA analysis. Patients who were positive for anti-efalizumab antibodies were excluded from the summary.

Adverse events and concomitant medications were recorded. Adverse events were defined as any sign, symptom, data or medical diagnosis, regardless of relationship to study drug, that began or worsened after the start of study drug treatment and were recorded in the subject’s adverse event case report form. Definitions of seriousness, severity and causality were included in the protocol. Provisions were made for reporting serious adverse events to sponsor, to IRB, and to FDA.

4.3.3.9 Statistical Considerations

Two treatment comparisons were of interest during the FT period of this study: 1.0 mg/kg efalizumab versus placebo and 2.0 mg/kg efalizumab versus placebo. The placebo groups for each of the two dose levels were combined for all statistical comparisons following investigation of baseline comparability of the two placebo groups.

Analysis of Treatment Group Comparability:

Treatment groups were assessed for comparability at the beginning of the FT, RT and ET periods with respect to demographic (i.e., age, sex, race/ethnicity) and baseline characteristics. The baseline value of any variable was defined as the last available value prior to the first administration of study drug. Continuous variables were analyzed using ANOVA, and categorical variables were assessed using appropriate contingency table methodology.

Efficacy Analyses: First Treatment Period

Analysis Population:

The intent-to-treat (ITT) population consisted of all subjects who were randomized, whether or not they received any study drug or completed the full course of treatment. The ITT population was the primary analysis population for the primary and secondary endpoints.

Primary Endpoint:

Response status at the end of the FT period was determined as follows:

· Responder: Any subject whose PASI score decreased ³ 75% from FT Day 0 to FT Day 84

· Partial responder: Any subject whose PASI score decreased ³ 50% but <75% from FT Day 0 to FT Day 84

· Non-responder: Any subject whose PASI score decreased <50% from FT Day 0 to FT Day 84

The evaluation of the primary endpoint consisted of the pairwise comparison of the proportion of responders in each efalizumab dose group (1.0 mg/kg efalizumab and 2.0 mg/kg efalizumab) versus placebo by Fisher’s exact test for the ITT population. Partial responder and non-responder categories were combined for the primary analysis. The placebo groups from the FT period for each of the two dose levels were also combined for all statistical comparisons.

Conventions for missing data imputation were as follows:

For all study endpoints, if a subject discontinued from the study prior to FT Day 84 but after receiving the final scheduled dose of study drug on FT Day 77, data from the early termination visit were to be used for analysis, i.e., the Day 84 data was not considered to be missing in this case.

A formal interim analysis of the primary efficacy endpoint was performed by an independent data monitoring committee (DMC) once approximately half (~225) of the projected 450 subjects completed the FT period. The stopping rules established prior to review of the results by the DMC allowed the trial to be stopped for futility only; stopping for efficacy was not allowed. Nonetheless, a penalty rule adjusting the critical value for the final analysis was established. Therefore, the analysis of the primary efficacy endpoint was carried out at the 0.049 level rather than at the 0.05 level. To maintain a type I error rate for the primary analysis of a=0.049 (two sided), the Hochberg-Bonferroni multiple comparisons procedure was used to adjust for the two comparisons. If both comparisons had p<0.049 in favor of efalizumab over placebo, both active treatment groups were considered significantly different from placebo. If one comparison had a

p>0.049, the other active treatment was considered statistically significantly different from placebo only if its associated p-value was <0.0245 in favor of efalizumab over placebo.

4.3.3.10 Protocol Amendments

The protocol was amended twice. The first amendment was to ensure that only patients who were clinically stable could enroll. The principal secondary objective was changed from the dynamic physician’s global assessment to the static physician’s global assessment scale based on discussions with the FDA. This change was also reflected in the secondary and other endpoints. In the second protocol amendment, the unblinding of treatment assignment was allowed to be performed separately for the FT period and the RT and ET periods because the availability of the primary efficacy results at the earliest possible time was necessary to justify the continued exposure to efalizumab in the ongoing open-label studies.

4.3.4 Study Results

4.3.4.1 Patient Disposition

The first subject was enrolled into the study on January 4, 2000, and the last subject completed the study on October 15, 2001. Twenty nine sites in the United States and Canada participated. A total of 498 patients were enrolled into this study (planned enrollment 450). The disposition of the patients who enrolled is shown in Table 3.

Table 3 Disposition of Subjects and Reasons for Discontinuation during the FT Period

		Efalizumab
	Placebo	1.0 mg/kg	2.0 mg/kg
Subject Status	(n=170)	(n=162)	(n=166)
Completed FT, n	151 (89%)	149 (92%)	145 (87%)
Entered ET	144	83	99
Entered OB	4	63	44
Entered FU	1	2	2
Discontinued from study	2	1	0
Discontinued FT, n	19 (13%)	13 (8%)	21 (13%)
Entered FU	7	5	8
Discontinued from study	12	8	13
Reason for discontinuation from FT	(n=19)	(n=13)	(n=21)
Adverse event	5	5	8
Lost to follow-up	3	3	5
Subject’s decision	8	2	5
Investigator’s decision	2	3	2
Use of excluded medication	1	0	1

During the first treatment period, the most common reason for discontinuation in the efalizumab treatment arm was for adverse events. In the placebo treatment arm, the most common reason for discontinuation was “subject’s decision.” Several, but not all, of the patients who discontinued the study due to “subject’s decision” were noted to have worsening of both the PASI score and the dynamic physician’s global assessment including some of the patients who were randomized to study drug and others to placebo (data not shown).

4.3.4.2 Demographics

Table 4 below depicts the demographics of subjects during the first treatment period of the study.

Table 4 Demographic Characteristics of Subjects in the FT Period

				FT Efalizumab
			FT Placebo	1.0 mg/kg/wk		2.0 mg/kg/wk
Characteristic, n			(n=170)	(n=162)		(n=166)
Age (yr)
Mean			42	45		46
Median			43	45		44
Range			18–68	18–75		20–74
Age group (yr)
18–40			73 (42.9%)	53 (32.7%)		63 (38.0%)
41–64			94 (55.3%)	98 (60.5%)		87 (52.4%)
³ 65			3 (1.8%)	11 (6.8%)		16 (9.6%)
Sex
Male			124 (73%)	118 (73%)		118 (71%)
Female			46 (27%)	44 (27%)		48 (29%)
Race/ethnicity
White			157 (92.4%)	147 (90.7%)		152 (91.6%)
Black			3 (1.8%)	5 (3.1%)		1 (0.6%)
Asian/ Pacific Islander			6 (3.5%)	6 (3.7%)		3 (1.8%)
Hispanic			4 (2.4%)	2 (1.2%)		8 (4.8%)
Other			0	2 (1.2%)		2(1.2%)
Weight (kg)
Mean			93	92		94
Median			91	90		91
Range			45–144	50–138		53–144
	BMI (kg/m²)
	Mean	31		31	31
	Median	30		30	30
	Range	14.8–60.2		18.7–52.0	18.5–53.6

More male than female patients participated in the study, whereas, in the general psoriasis population, men and women are equally affected. The population tends to have higher than average median weight and body mass index probably reflecting the overall U.S. psoriatic population.

Other than the gender distribution, the characteristics are reflective of the general psoriasis population in the United States. A higher proportion of patients were age 65 or older in the active treatment arms than in placebo. The numbers of patients older than 70 years of age are limited because this age group was excluded per the eligibility criteria (data not shown). The other baseline characteristics were well-balanced among the treatment groups.

4.3.4.3 Disease Characteristics at Baseline

Table 5 below contains the baseline disease characteristics for the study population.

Table 5 Baseline Psoriasis Characteristics of Subjects in the FT Period

		FT Efalizumab
	FT Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
Characteristic	(n=170)	(n=162)	(n=166)

Duration of psoriasis (yr)
Mean	18.5	19.1	16.7
Median	17	19	15
Range	1–56	1–58	1–60
Prior systemic therapy
Yes	91 (53.5%)	89 (54.9%)	93 (56.0%)
No	79 (46.5%)	73 (45.1%)	73 (44.0%)
PASI
Mean	19.02	18.63	18.86
Median	16.5	16.9	16.8
Range	9.6–57.6	11.9–50.1	10.0–55.6
PASI category
£ 16.0	79 (46.5%)	74 (45.7%)	74 (44.6%)
16.1–30.0	78 (45.9%)	77 (47.5%)	79 (47.6%)
>30	13 (7.6%)	11 (6.8%)	13 (7.8%)
PASI thickness component
Mean	6.26	6.07	6.08
Median	5.6	5.5	5.5
Range	2.4–19.2	2.1–18.2	3.0–18.6
Physician’s Global Assessment
Mild	6 (3.5%)	4 (2.5%)	4 (2.4%)
Moderate	86 (50.6%)	90 (56.3%)	86 (51.8%)
Severe	67 (39.4%)	61 (38.1%)	70 (42.2%)
Very severe	11 (6.5%)	5 (3.1%)	6 (3.6%)
% BSA of psoriasis
Mean Median	29.4 25.6	29.6 24.3	29.9 24.2
Range	10-85	10-72	10-83
Itching Scale
Mean	5.9	5.8	6.1
Median	6	6	7
Range	0-10	0-10	0-10

The median duration of disease was 17 years (range 1-60 years). Approximately 55% of all subjects had a history of prior systemic therapy; however, for a substantial proportion of the study stubjects, efalizumab was the first systemic therapy received. The median baseline PASI score was approximately 16.7. The treatment groups were well-balanced in all the measures of baseline disease activity. Overall, the baseline psoriasis characteristics indicate the study population had moderate-to-severe psoriasis.

Table 6 below contains information with regard to treatment compliance with study drug by treatment group.

Table 6 Treatment Compliance for Subjects during the FT Period

		Efalizumab
No. of Doses	Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
Received	(n=170)	(n=162)	(n=166)
All 12	122 (72%)	129 (80%)	119 (72%)
10–11	27 (16%)	20 (12%)	29 (18%)
<10	21 (12%)	13 (8%)	18 (11%)

Twenty-one subjects (4.2%) received two or more conditioning doses: 5.9% of subjects in the placebo group, 3.1% in the 1.0 mg/kg/wk efalizumab group, and 3.6% in the 2.0 mg/kg/wk efalizumab group.

The proportion of patients who received £ 10 doses, in part, reflects the proportion of patients discontinuing the first treatment period.

4.3.4.4 Use of Concomitant Medications

A total of 30 subjects, 6.0% of all patients, received an excluded medication or phototherapy during the first treatment period. Eight of these patients were in the placebo group, 14 in the 1.0 mg/kg dose efalizumab group, and 8 were in the 2.0 mg/kg efalizumab group.

4.3.4.5 Primary Efficacy Outcomes

Table 7 PASI Response to Treatment during the FT Period: All Randomized Subjects

		Efalizumab
PASI Response at FT Day 84	Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
	(n=170)	(n=162)	(n=166)
Responders, n	4 (2%)	63 (39%)	44 (26%)

Partial and non-responders, n *	166 (98%)	99 (61%)	122 (74%)
Fisher’s exact p-value
efalizumab vs. placebo	—	<0.001	<0.001

* Included subjects who discontinued.

The proportion of responders was statistically higher in the treatment groups than in placebo. The absolute difference was 37% for the 1.0 mg/kg/wk group and 24% for the 2 mg/kg/wk group. The response rate was not higher with the 2.0 mg/kg/wk vs. the 1.0 mg/kg/wk dose of efalizumab.

Reviewer’s comment: CD11a receptors of circulating lymphocytes are saturated at the 1.0 mg/kg/wk dose and would probably explain the lack of dose response.

Last observation carried forward and other sensitivity analyses for missing data did not change the estimate of the treatment effect.

Table 8 Mean Percent Improvement in PASI Thickness, Erythema, and Scaling Components during the FT Period

		FT Efalizumab
	FT Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
PASI Component	(n=170)	(n=162)	(n=166)
Thickness ^a	17.4	55.9	45.4
Erythema ^a	16.4	50.9	43.0
Scaling ^a	17.4	58.6	51.2
PASI total ^b	19.8	60.1	50.5

Note: Improvement in each component was reflected by a decrease in score.

a The last observation carried was used to impute missing FT Day 84 PASI data.

b Values from the early termination visits were assigned to the next scheduled visit for PASI evaluation.

The components of the PASI - thickness, erythema, scaling- each show higher mean percentage improvement in the efalizumab-treated patients as compared to placebo-treated patients. Therefore, the all of the components appear to contribute similarly to improvement in the overall score.

The effect on affected body surface area is shown below.

Table 9 Mean Improvement in Percent BSA of Psoriasis during the FT Period

		FT Efalizumab
	FT Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
	(n=170)	(n=162)	(n=166)
Percent BSA affected at FT Day 0	29.4	29.6	29.9
Percent BSA affected FT Day 84	27.6	15.8	19.9
Improvement ^a from baseline	1.8	13.8	10.0

^a Improvement was reflected by a decrease in the percent BSA score.

The mean percentage body surface area affected at the end of the 84-day treatment period improved more in the 1.0 mg/kg/wk group and 2.0 mg/kg/wk efalizumab groups than in in the placebo-treated patients.

Reviewer’s comment: The mean percentage improvement in affected BSA is smaller than that of the other measures of disease severity- erythema, thickness and scale.

The response among various subsets of the studied population is show in Table 10 below.

Table 10 PASI Responders by Subsets of Randomized Subjects: FT Period

		Efalizumab
	Placebo	1.0 mg/kg	2.0 mg/kg
Subject Subset	n=170	n=162	n=166
Gender
Men	1/124 (0.8%)	43/118 (36%)	29/118 (25%)
Women	3/46 (6.5%)	20/44 (46%)	15/48 (31%)
Age group (yr)
18–40	3/73 (4.1%)	17/53 (32%)	17/63 (27%)
41–64	1/94 (1.1%)	40/98 (41%)	24/87 (27%)
³ 65, n	0/3 (0%)	6/11 (55%)	3/16 (19%)
Baseline PASI category
£ 16.0	1/79 (1.3%)	32/74 (43%)	20/74 (27%)
16.1–30.0, n	2/78 (2.6%)	25/77 (33%)	20/79 (25%)
>30.0, n	1/13 (7.7%)	6/11 (55%)	4/13 (31%)
Prior systemic therapy
Yes, n	1/91 (1.1%)	32/89 (36%)	27/93 (29%)
No, n	3/79 (3.8%)	31/73 (43%)	17/73 (23%)

The results of the primary efficacy analysis are generalizable across gender, age, baseline PASI and history of prior systemic therapy subsets. There was a trend towards higher response rates in patients in the low dose group than the high dose group of efalizumab.

The data below show the response rate by geographic latitude by treatment group

(see Table 11). The Northern U.S. sites included those in the Northeast, Northcentral, and Northwest regions (Washington, Oregon, Utah, Colorado, and northern California).

The Southern U.S. sites included those in the south and southwest regions (Arizona, New

Mexico, and southern California).

Table 11 PASI Response to Treatment by Latitude (FT Period)

			Efalizumab	Efalizumab
		Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
Latitude	FT Day 84 Response	(N=170)	(N=162)	(N=166)

Canada	N	47	45	46
	Responder	2 (4.3%)	14 (31.1%)	11 (23.9%)
	95% Confidence Interval	[0.005, 0.145]	[0.182, 0.466]	[0.126, 0.388]

Northern United States	N	76	77	74
	Responder	2 (2.6%)	34 (44.2%)	19 (25.7%)
	95% Confidence Interval	[0.003, 0.092]	[0.328, 0.559]	[0.162, 0.372]

Southern United States	N	47	40	46
	Responder	(0.0%)	15 (37.5%)	14 (30.4%)
	95% Confidence Interval	[0.000, 0.075]	[0.227, 0.542]	[0.177, 0.458]

Clinical responses did not differ by geographic latitude.

Table 12 Covariates Potentially Predictive of PASI Response: FT Period

Model Predictor	Odds Ratio	95% CI
Sex
Female vs. male	1.725	1.021, 2.912
Prior systemic therapy
No vs. yes	1.108	0.689, 1.779
Geographic region
Canada vs. western United States	0.612	0.314, 1.186
North central vs. western United States	1.186	0.620, 2.276
Northeastern vs. western United States	0.205	0.045, 0.676
Southern vs. western United States	0.622	0.287, 1.312

The following covariates were examined in the model and did not have any relationship to treatment response: baseline PASI score, age, history of prior systemic therapy and season (spring vs. summer). There was a suggestion of higher responses in women, but this was not supported in subsequent studies. Also, a comparison of response by geographic region suggested a higher response in the Western United States vs. that seen in the Northeastern United States.

4.3.4.6 Secondary Efficacy Outcomes

Table 13 Principal Secondary Efficacy Endpoint: FT Period

		FT Efalizumab
	FT Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
OLS Response at FT Day 84	(n=170)	(n=162)	(n=166)
Minimal or Clear	5 (2.9%)	52 (32.1%)	37 (22.3%)
Mild to Very Severe *	165 (97.1%)	110 (67.9%)	129 (77.7%)
Fisher’s exact p-value efalizumab vs. placebo	—	<0.001	<0.001

* Included subjects who were classified as Mild, Moderate, Severe, and Very

Severe and those who discontinued.

The secondary efficacy outcomes also showed that efalizumab was superior to placebo.

From the figure, there is separation of the efalizumab curves from that of placebo by day 14 days of treatment.

The PGA, a physician’s global assessment, was used to measure the dynamic response of patients as compared baseline (See Table 14 below).

Table 14 PGA Response for Subjects during the FT Period

		FT Efalizumab
	Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
PGA Response at FT Day 84	(n=170)	(n=162)	(n=166)
Excellent or Cleared	7 (4.1%)	63 (38.9%)	50 (30.1%)
Good to Worse ^a	163 (95.9%)	99 (61.1%)	116 (69.9%)
Fisher’s exact p-value
efalizumab vs. placebo	—	<0.001	<0.001
^a Included subjects who were classified as Good, Fair, Slight, Poor, Unchanged, or Worse and those who discontinued.

A greater proportion of patients achieved excellent or cleared in both efalizumab treatment groups compared to placebo using the dynamic physician’s global assessment. The differences reached statistical significance.

4.3.4.7 Time-to-response and duration of treatment response

Time-to-onset of PASI-75 response was analyzed and the results are shown in Table 15 below.

Table 15 Time (days) to PASI-75 Response, Using Kaplan Meier Estimates Study ACD2058g (FT Period): Subjects Who Achieved a PASI-75 Response at Any Time


		Efalizumab	Efalizumab
Characteristic	Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk


Subjects Who Achieved PASI-75 at Any Time	9	74	52
Median	43.0	57.0	57.0
95% C.I. for Median	(41.0, 71.0)	(56.0, 59.0)	(55.0, 71.0)
25-75 %ile	41.0 - 71.0	43.0 - 72.0	45.5 - 79.5
Minimum - Maximum	29.0 - 74.0	28.0 - 89.0	28.0 - 92.0

Median time to achieve PASI 75 in patients who achieved PASI 75 at any time was approximately 2 months.

4.3.4.8 Duration

Time-to-relapse defined as a loss of ³ 50% improvement in PASI score achieved between baseline and the end of the 84-day treatment period was summarized by treatment group for treatment responders (³ PASI 75 at day 84). The results are shown in Table 16 below.

Table 16 Time (days) to Relapse during the OB Period, Using Kaplan-Meier Estimates: Subjects Treated with Efalizumab during the FT Period

	Efalizumab
	1.0 mg/kg	2.0 mg/kg
Characteristic	n=63	n=44
Events	55	37
Censored observations ^a	8	7
Median (95% CI)	60.0 (57, 66)	59.0( 57, 82)
25th–75th Percentile	43.0–85.0	49.0–87.0

^aData from subjects who discontinued prior to relapse or who did not relapse

during the OB period were censored.

The median time to relapse during the observation period was 60 days (67 days after the last dose of efalizumab) for the 1.0- mg/kg/wk group and 59 days for the 2-mg/kg/wk group.

An analysis of the distribution of percent improvement in PASI achieved during the first 84 days of treatment is shown in Table 17 below.

Table 17 PASI Response by Percent Improvement from Baseline for Subjects during the FT Period (% of total)

		Efalizumab
Percent Improvement from Baseline	Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
	(n=170)	(n=162)	(n=166)
³ 90%	1.2	12.3	4.8
³75% to < 90%	1.2	26.5	21.7
³ 50% to < 75%	12.4	22.2	24.7
³ 25% to < 50%	20.0	16.7	21.1
< 25%	54.1	14.2	15.7
Missing ^a	11.2	8.0	12.0

^aSubjects who were missing the FT Day 84 score were classified as non-responders for the analysis of the primary efficacy endpoint.

This analysis demonstrates a general shift toward improvement in the efalizumab groups compared with the placebo group. Additionally, a trend toward higher percentage improvements in PASI in the low dose group than in the high dose group exists.

4.3.4.9 The OB Period

Patients who achieved PASI 75 at the end of the first treatment period could enter the observation period. A total of 111 patients entered the OB Period. Of these, 4 received placebo in the first treatment period and the remainder received treatment with efalizumab. Overall, 83% of the patients who discontiued the observation period met the endpoint of relapse during the observation period and entered retreatment. Overall, 9.9% of patients did not experience relapse during the 84-day observation period.

Table 18 Disposition of Subjects and Reasons for Discontinuation during the OB Period

		FT Efalizumab

	FT Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk	All Subjects
Subject Status	(n=4)	(n=63)	(n=44)	(n=111)
Completed OB, n	1 (25.0%)	5 (7.9%)	5 (11.4%)	11 (9.9%)
Enrolled in Study ACD2062g	0	0	1	1
Entered RT	0	1	1	2
Entered FU	1	3	3	7
Discontinued from study	0	1	0	1
Discontinued OB, n	3 (75.0%)	58 (92.1%)	39 (88.6%)	100 (90.1%)
Enrolled in Study ACD2062g	0	1	2	3
Entered RT	3	49	31	83
Entered FU	0	2	2	4
Discontinued from study	0	6	4	10
Reason for discontinuation from OB
Adverse event	0	1	1	2
Lost to follow-up	0	1	1	2
Subject’s decision	0	2	1	3
Investigator’s decision	0	1	4	5
Pregnancy	0	1	1	2
Relapse of psoriasis	3	52	31	86

A protocol amendment made it possible for patients experiencing severe psoriasis upon relapse to enter Study ACD2062g. Of the 100 patients who discontinued the observation period, 86 patients experienced a relapse of psoriasis. Of these patients, 83 entered the retreatment period and 3 entered Study ACD2062g. Among the 14 patients listed as having discontinued for reasons other than relapse of psoriasis, several discontinued for psoriasis variants and worsening of psoriasis.

Table 19 Psoriasis-Related Concomitant Medications Initiated during the OB Period:

		FT Period Responders
		Placebo	Drug 1.0 mg/kg/wk		Drug 2.0 mg/kg/wk		All Subjects
USAN Class	Generic Name	(N=4)	(N=63)		(N=44)		(N=111)

Subjects with completed medication forms		4		63		44	111
Subjects initiated at least one psoriasis- related medication		2 (50.0%)	12 (19.0%)		13 (29.5%)		27(24.3%)

Dermatologic agents	- Total -	(0.0%)	2	(3.2%)	2	(4.5%)	4	(3.6%)
	Acitretin	(0.0%)		(0.0%)	1	(2.3%)	1	(0.9%)
	Calcipotriene	(0.0%)	1	(1.6%)		(0.0%)	1	(0.9%)
	Coal tar	(0.0%)	1	(1.6%)	1	(2.3%)	2	(1.8%)

Steroids	- Total -	2 (50.0%)	11 (17.5%)		12 (27.3%)		25 (22.5%)
	Betamethasone valerate	(0.0%)		(0.0%)	1	(2.3%)	1	(0.9%)
	Cortisone acetate	(0.0%)		(0.0%)	1	(2.3%)	1	(0.9%)
	Dexamethasone	(0.0%)		(0.0%)	1	(2.3%)	1	(0.9%)
	Fluocinolone acetonide	(0.0%)		(0.0%)	1	(2.3%)	1	(0.9%)
	Fluticasone propionate	(0.0%)	2	(3.2%)		(0.0%)	2	(1.8%)
	Halobetasol propionate	(0.0%)	1	(1.6%)		(0.0%)	1	(0.9%)
	Hydrocortisone	1 (25.0%)	5	(7.9%)	5 (11.4%)		11	(9.9%)
	Mometasone furoate	(0.0%)	4	(6.3%)	1	(2.3%)	5	(4.5%)
	Prednisolone acetate	(0.0%)		(0.0%)	1	(2.3%)	1	(0.9%)
	Prednisone	1 (25.0%)		(0.0%)	2	(4.5%)	3	(2.7%)
	Triamcinolone acetonide	(0.0%)	1	(1.6%)	3	(6.8%)	4	(3.6%)

Overall, 24% of patients entering the observation period initiated at least one psoriasis-related medication (See Table 19). The majority of these patients initiated treatment with topical steroids. Three patients initiated treatment with systemic steroids (prednisone).

Reviewer’s comment: It is possible that the use of the disallowed concomitant medications during the observation period may have affected the estimate of the duration of response. Analyses are ongoing to evaluate.

4.3.4.10 Response to Second Treatment Course in Patients who Responded to the First Treatment

The only study which examined the response to retreatment was Study ACD2058. In this study, patients who were responders (achieved ³ 75% improvement in PASI) at day 84 were eligible to enter an observation period, no treatment for up to 168 days during which they were followed until relapse, and then rerandomized to a second treatment course. Upon relapse, placebo patients received a second course of efalizumab, while efalizumab-treated patients were rerandomized centrally to receive either the same dosage of efalizumab or placebo in a 2:1 ratio.

Subjects not responding to the second course of treatment by RT Day 56 were eligible to transfer to the open label study, ACD2062g. As discussed on page 33, the majority of both placebo- and efalizumab- treated patients discontinued the observation period due to relapse of their psoriasis.

Table 20 below reflects the disposition of the subset of patients who responded to treatment with efalizumab (active drug) during the first 84 days of treatment and were rerandomized during the observation period to retreatment.

Table 20 Disposition of RT-A Subjects and Reasons for Discontinuation

		Efalizumab
	Placebo	1.0 mg/kg	2.0 mg/kg
Subject Status	(n=27)	(n=32)	(n=23)
Completed RT	8 (29.6%)	26 (81.2%)	16 (69.6%)
Entered FU	5	23	13
Entered Study ACD2062g	2	3	1
Discontinued from study	1	0	2

Discontinued RT	19 (70.4%)	6 (18.8%)	7 (30.4%)
Entered FU	1	0	2
Entered Study ACD2062g	18	6	5

Reason for RT discontinuation	(n=19)	(n=6)	(n=7)
Subject’s decision	1	0	1
Investigator’s decision	1	0	0
Non-response to RT	16	6	6
Non-response to ET ^a	1	0	0

^a One subject should have been classified as a non-responder to retreatment, making the total number of non-responders 29 (90.6%).

Eighty-six subjects were eligible to enter RT. Of these, 82 were rerandomized to RT.

Most of the patients who were rerandomized to receive efalizumab completed the course of retreatment, while fewer than one-third (29.6%) of the patients who were rerandomized to placebo completed the retreatment period. Most of the latter patients discontinued due to non-response to retreatment.

Table 21 below shows the psoriasis characteristics of the subset of responders who entered retreatment upon relapse.

Table 21 Psoriasis Characteristics of RT-A Subjects

		Efalizumab		All
				Efalizumab-
		1.0	2.0	Treated
	Placebo	mg/kg/wk	mg/kg/wk	Subjects
Characteristic, n	(n=27)	(n=32)	(n=23)	(n=55)

Duration of psoriasis (yr)
Mean	20.6	20.0	18.9	19.5
Median	21	20	18	20
Range	3–46	3–43	2–43	2–43
Prior systemic therapy
Yes	15 (55.6%)	19 (59.4%)	13 (56.5%)	32 (58.2%)
No	12 (44.4%)	13 (40.6%)	10 (43.5%)	23 (41.8%)
Baseline PASI (FT Day 0)
Mean	18.7	17.7	19.2	18.3
Median	17.5	15.4	17.1	15.6
Range	11.9–29.7	12.0–35.3	12.1–36.0	12.0–36.0
RT Day 0 PASI
Mean	14.8	13.2	13.7	13.4
Median	12.4	11.9	11.6	11.6
Range	7.5–39.0	7.4–29.2	8.5–28.7	7.4–29.2
Baseline (FTDay 0) PASI category
£16.0	12 (44.4%)	19 (59.4%)	11 (47.8%)	30 (54.5%)
16.1–30.0	15 (55.6%)	10 (31.3%)	8 (34.8%)	18 (32.7%)
>30	0	3 (9.4%)	4 (17.4%)	7 (12.7%)
Baseline (FT Day 0) OLS
Moderate	18 (66.7%)	21 (65.6%)	12 (52.2%)	33 (60.0%)
Severe	9 (33.3%)	10 (31.3%)	11 (47.8%)	21 (38.2%)
Very severe	0	1 (3.1%)	0	1 (1.8%)

The rerandomized patients were reasonably well balanced in terms of the baseline disease severity. This subset had moderate-to-severe psoriasis at Day 0 of the first treatment period according to the static physician’s global assessment and a median baseline PASI of 15.6. As would be expected, the overall median PASI score at the beginning of retreatment, 11.6, was lower than that at Day 0 of the first treatment period.

Psoriasis Characteristics of RT-A Subjects (cont)

		Efalizumab		All
				Efalizumab-
		1.0	2.0	Treated
	Placebo	mg/kg/wk	mg/kg/wk	Subjects
Characteristic, n	(n=27)	(n=32)	(n=23)	(n=55)
RT Day 0 OLS
Mild	0	1 (3.2%)	4 (17.4%)	5 (9.3%)
Moderate	23 (85.2%)	25 (80.6%)	15 (65.2%)	40 (74.1%)
Severe	4 (14.8%)	4 (12.9%)	4 (17.4%)	8 (14.8%)
Very severe	0	1 (3.2%)	0	1 (1.9%)
% BSA of psoriasis
Mean	31.7	29.1	31.5	30.1
Median	32.0	24.6	28.0	24.8
Range	13.0–63.0	11.0–59.0	11.0–81.0	11.0–81.0
RT Day 0 % BSA of psoriasis
Mean	24.0	17.1	21.8	19.1
Median	20.0	14.0	15.5	14.8
Range	5.0–74.0	6.2–43.0	7.0–54.0	6.2–54.0
FT Day 84 PGA
Cleared	1 (3.7%)	2 (6.3%)	0	2 (3.6%)
Excellent	23 (85.2%)	27 (84.4%)	20 (87.0%)	47 (85.5%)
Good	2 (7.4%)	2 (6.3%)	3 (13.0%)	5 (9.1%)
Fair	1 (3.7%)	0	0	0
Missing	0	1 (3.1%)	0	1 (1.8%)
RT Day 0 PGA
Good	1 (3.7%)	2 (6.3%)	0	2 (3.6%)
Fair	12 (44.4%)	17 (53.1%)	7 (30.4%)	24 (43.6%)
Slight	4 (14.8%)	5 (15.6%)	12 (52.2%)	17 (30.9%)
Unchanged	2 (7.4%)	2 (6.3%)	1 (4.3%)	3 (5.5%)
Worse	8 (29.6%)	6 (18.8%)	3 (13.0%)	9 (16.4%)
Unless otherwise stated baseline was FT Day 0.

The subset of patients who subsequently received retreatment had moderate to severe psoriasis at baseline (FT Day 0). The baseline characteristics were comparable to the ITT population as a whole. The patients’ retreatment baseline were, as would be expected, better than their original baseline in all measures of disease severity including PASI and baseline BSA affected by psoriasis.

Treatment compliance in the patients randomized to retreatment is shown in

Table 22 below.

Table 22 Treatment Compliance for RT-A Subjects

		Efalizumab
No. of Doses	Placebo	1.0 mg/kg/wk	2.0 mg/kg/wk
Received	(n=27)	(n=32)	(n=23)
All 12	8 (30%)	24 (75%)	16 (70%)
10–11	0	2 (6.3%)	1 (4.3%)
<10	19 (70%)	6 (19%)	6 (26%)

Whereas, the majority of patients in the efalizumab groups received all 12 doses in the RT period, less than one-third of the placebo patients received all 12 doses. Again, the reasons for missed doses are probably reflective of the reasons for discontinuing the retreatment period. In the case of the placebo patients the most common reason was non-response to treatment.

Response to retreatment is described below. These were patients who responded to the first treatment period and then were rerandomized to either efalizumab or placebo upon relapse (loss of 50% of improvement in PASI). In this analysis, responses were compared to the original baseline at FT Day 0.

Table 23 PASI Response to Retreatment (% Improvement

from FT Day 0)

	Placebo	Efalizumab (Combined)
Response Category	(N=27)	(N=55)
³ 75%	0	17 ( 31%)
³ 50% -< 75%	5 (19%)	20 ( 36%)
0-50%	2 (7%)	4 ( 7%)
<0%	1 (4%)	1 (2 %)
Missing	19 (70%)	13 ( 24%)

Among patients who received retreatment with efalizumab, 34% of the 1 mg/kg group and 25% of the 2 mg/kg group responded at the PASI 75 level at the end of the retreatment period. This was in contrast to patients rerandomized to placebo who had no responders to retreatment. The majority of the patients receiving efalizumab upon retreatment, 72% and 61% the 1 mg/kg group and 2 mg/kg group, respectively, responded at the PASI 50 level.

While all patients had achieved a 75% improvement in PASI at Day 84 of the first treatment period, less than one-third of the combined efalizumab-treated patients achieved this level of response at Day 84 of retreatment. The patients’ state of active relapse at the beginning of the retreatment period may have contributed to the lower response rate to retreatment.

Figure 2 below depicts the PASI score at retreatment in relation to the minimum relapse PASI defined as a loss of 50% of the improvement achieved during the first treatment period.

Figure 2

Since PASI assessments were scheduled for OB Days 14, 28, 56, 84, 112, 140 and 168,

in most cases the retreatment baseline PASI score exceeded the minimum retreatment score. The magnitude of the difference between retreatment PASI and the minimum requirement ranged from 0 to nearly 25 points. In most cases the difference did not exceed 10 PASI points. (Of note one patient, entered retreatment with less than the minimum required PASI score.)

Reviewer’s comment: It is not known whether the severity of relapse may have played a role in inhibiting the ability of the drug to recapture PASI responses comparable to those achieved during the first treatment period.

Of note, there was a considerable proportion of patients in each group for whom these data are missing. The majority of the patients in the placebo group had missing data. The high proportion of missing data reflects the number of patients who transferred to Study ACD2062g with our completing the retreatment period, due to non-response.

The responses to retreatment when compared to the new baseline are shown in Table 24 below.

Table 24 PASI Response to Retreatment (% Improvement

from RT Day 0)

	Placebo	Efalizumab (Combined)
Response Category	(N=27)	(N=55)
³ 75%	0	12 (22%)
³ 55%	3 (11%)	27 (49%)
0-50	3 (11%)	12 (22%)
< 0	2 (7.4%)	3 (5%)
Missing	19 (70%)	13 (24%)

By comparison to the most recent baseline, the proportions of patients achieving PASI 50 and PASI 75 are fewer than when the comparison is to the patient’s own original baseline as would be expected.

The time course of response to retreatment is shown in Figure 3 below. Separation of the efalizumab curves from that of placebo took place by 28 days of retreatment.

Figure 3

The proportion of patients responding at retreatment day 84 by OLS minimal to clear was approximately 23.6% higher in the combined efalizumab group as compared to the placebo group. See Table 25 below.

Table 25 OLS Response to Treatment for RT-A Subjects

		Efalizumab
		1.0	2.0	All Efalizumab-	All
RT Day 84	Placebo	mg/kg/wk	mg/kg/wk	Treated	Efalizumab
Characteristic	(n=27)	(n=32)	(n=23)	Subjects (n=55)	vs. Placebo

Minimal or Clear	1 (3.7%)	9 (28.1%)	6 (26.1%)	15 (27.3%)	0.016

4.3.4.11 Response to Extended Treatment in Patients who were Non-responders or Partial Responders to the First Treatment Period

The group of non-responders and partial responders from the first treatment period who were re-randomized to extended treatment or placebo for the extended treatment period was also analyzed.

The patient disposition of this group is shown below.

Table 26 Disposition of ET-A Subjects and Reasons for Discontinuation

		Efalizumab
	Withdrawal/	1.0	2.0
	Placebo	mg/kg/wk	mg/kg/wk
Subject Status	(n=60) ^a	(n=57)	(n=66)