1a. Medicinal product: The medicinal product is Raptiva (Efalizumab). Marketed biological entity. Is a recombinant, humanized IgG1 monoclonal antibody that targets CD11a, the Efalizumab-PrOACT-URL subunit of leucocyte function associated antigen 1 (LFA-1). Mechanism of action may lead to inhibition of leucocyte migration, similarly to natalizumab.

1b. Indication(s) for use: efalizumab is indicated in the treatment of "high need" adult patients with moderate to severe chronic plaque psoriasis who have failed to respond to, or who have a contraindication to, or are intolerant to other systemic therapies including ciclosporine, methotrexate and PUVA. The duration of initial therapy is 12 weeks. Therapy may be continued only in patients who responded to treatment (PGA good or better).

Together, the clinical pharmacology of efalizumab and the safety and efficacy data (including 2 phase 3 studies with 1.0 mg/kg/week and 2.0 mg/kg/week) support the selection of 1.0 mg/kg/week SC as the optimal dose for efalizumab. (EPAR scientific discussion)

1c. The therapeutic area and disease epidemiology: Moderate to severe chronic plaque discoid psoriasis. Psoriasis is a common chronic, squamous dermatosis with polygenic inheritance and a fluctuating course. Principal histological findings are Munro microabscesses and spongiform pustules; also seen are rounded, circumscribed, erythematous, dry, scaling patches of various sizes, covered by greyish white or silvery white, umbilicated and lamellar scales, usually on extensor surfaces, nails, scalp, genitalia and the lumbosacral region.

1d. The unmet medical need: At the time of initial Market Authorisation, there are well established systemic treatments (cyclosporine, methothrexate, PUVA) all of which with serious Adverse Effects (but B-R of the drugs is well established for a long time). At the time of the reevaluation of the B-R of efalizumab (Jan 2009) there are more recent alternative therapies(biologic treatments for moderate to severe psoriasis in "high need" adult patients e.gadalimubab, etanercept, infliximab, ustekinumab. with established efficacy but long term safety still uncertain in the psoriasis indication (although with longer experience in other indications such as RA)

Severity of condition: Psoriasis is a chronic disease, leading in its severe forms to a significant social disability impacting both professional and social life. Although psoriasis is a serious disease, with potential severe negative impact on the patient's social life, it is not a life-threatening disease apart from rare erythrodermic forms which were excluded from the clinical trials population and was not part of the approved indication (nor were pustular forms of the disease and psoriatic arthritis)

Affected population: "high need" adult patients with moderate to severe chronic plaque psoriasis who have failed to respond to, or who have a contraindication to, or are intolerant to other systemic therapies including cyclosporine, methotrexate and PUVA

Patient concerns: impact on quality of life, physical appearance and social functioning

Physician concerns: chronic and incurable with unpredictable flare ups, interested in long term efficacy

Time frame for health outcomes: 12 weeks for PASI 75 (efficacy/favourable effects)), 3 years for PML (safety/unfavourable effects). PASI 75 (primary endpoint) is a 75% reduction of the PASI score at week 12.

1e. What is to be decided: Re-evaluation of benefit-risk of efalizumab was prompted by incidence of emerging adverse events in the post-marketing period, i.e. presentation of PML (Progressive Multifocal Leucoencephalopathy) in addition to other serious risks (cardiotoxicity, neurotoxicity, serious infections including tuberculosis). The question to be addressed is: are there in January 2009 any risk minimisation measures which could be rapidly implemented, thus maintaining the B-R balance of the drug as positive? If not, should the Market Authorisation be suspended/revoked?

By whom: the Case study takes the regulator's perspective (1ST step of the efalizumab Task Force); next perspective to be addressed is the psoriasis patient's perspective, given the significant social impact of the severe forms of the disease.

When: 16th January 2009. Experts believed the margin of benefits over risks had narrowed since approval, i.e. modest efficacy and increased risks. The European Commission initiated a procedure under Article 20 of Regulation (EC) No 726/2004 and requested the Committee to assess the above concerns and its impact on the benefit/risk balance for efalizumab, and to give its opinion on measures necessary to ensure the safe and effective use of efalizumab and on whether the marketing authorisation for this product should be maintained, varied, suspended or revoked. The CHMP also took advice from the Scientific Advisory Group before making a decision.

The efalizumab case study intends to replicate the decision made by the CHMP in February 2009, but using various frameworks and quantitative models.

EPAR: EU authorisation on 20th September 2004. Suspended Feb 2009, withdrawn June 2009;

Standard Text Books

EfalizumabRMP update Nov 2008 pages 30-40

CHMP Opinion EMEA/CHMP/3552/2009;

Serono internal data: Serono analysis of patients treated with efalizumab after previous treatment with anti-TNF (26 Jan 2009)

Rapporteurs' Final Assessement Report EMEA/H/C/00542

Marketing authorisation, pivotal studies

efalizumab RMP update Nov 2008

Responses of the Scientific Advisory Group CNS to the CHMP list of questions on efalizumab 7 Jan 2009.

EMA/24463/2009

Scientific Conclusions

EMEA/H/C/000542/A20/0028

EMA/CHMP/3552/2009

2a. Problem of uncertainty, multiple conflicting objectives, combination of the two, or something else?

The 4 PML cases are not strongly confounded. The positive diagnosis is serologically confirmed in 3/4. There are no alternative diagnosis.

The uncertainty relates mainly on the relationship between duration of treatment (time on exposure) and the occurrence of PML. The impact is on the possible risk minimisation measure if this had been confirmed.

In addition to the PML risk (potentially fatal Adverse Effect), some other risks emerged during post-marketing period.

Risk has increased, documented with several SPC amendments over the 4 years marketing.

Long term treatment: some studies (ACD2058g)included a retreatment period (RT) or extended treatment (ET); there were 2 observation periods without treatment: Observation period (OB) and Follow-up (FU); ACD2059g included only 3 periods (FT, ET and FU); the results suggest that patients not responding within 3 months will be less likely to respond to prolonged treatment for another 3 months.

In total, data from extended treatment (more than 12 weeks) have been obtained from 4,311 patients in open label uncontrolled studies. Over 600 patients have been treated for more than 1 year including 166 patients treated for more than 2 years and up to 3 years.

2b. The factors to be considered in solving the problem:

Study design: no direct comparison with any systemic treatment (standard treatments or new biological). Topical symptomatic treatment was allowed as per investigator in all RCTs.

Adequacy of data sources: Efficacy data was obtained from 5 double blind, placebo controlled Phase III clinical trials designed to evaluate efficacy of efalizumab as a systemic monotherapy. Safety data is obtained via the number of adverse event (AE) reports received in post-marketing setting from spontaneous sources (health care professionals, literature, regulatory authorities, etc.) Safety data is based on reported events and so can potentially under represent the number of events. This may be due to poor reporting and sensitivity, and there may be an insufficient timeframe to allow for development of adverse events post long term exposure to efalizumab. However, underreporting of PML is likely to be minimal due to widely circulated documentation to physicians warning the risk of PML.

Disease epidemiology: efalizumab is indicated in the treatment of "high need" adult patients with moderate to severe chronic plaque psoriasis who have failed to respond to, or who have a contraindication to, or are intolerant to other systemic therapies including cyclosporine, methotrexate and PUVA. It needs to be considered how important/essential it is that efalizumab is available to "high need" patients where other medications may not have worked.

Presence of alternative treatments:In September 2004, 2 biologic medicines (i.e., etanercept and efalizumab) were approved in the EU for the treatment of plaque psoriasis. Subsequently, infliximab was approved for this indication in September 2005, followed by adalimumab in December 2007. All 4 biologic therapies licensed in the EU are indicated for adult psoriasis. PML cases have been reported with some of these biologicals, but not in their psoriasis indication.

CHMP opinion EMEA/CHMP/3552/2009;

Rapporteurs' Final Assessement Report EMEA/H/C/00542

PSURs and SPC Variations

efalizumab RMP update Nov 2008 pages 30-40

3.

4. Identify:

a) favourable effects

b) unfavourable effects

4a. Favourable effects (i.e. efficacy): The primary efficacy endpoint is the proportion of subjects with a 75% or more improvement from baseline in the PASI score (PASI75). This endpoint is strongly recommended in conjunction with a validated standardised global score (e.g. PGA) in the EMA GUIDELINE ON CLINICAL INVESTIGATION OF MEDICINAL PRODUCTS INDICATED FOR THE TREATMENT OF PSORIASIS. Five pivotal clinical studies evaluating efficacy of efalizumab in moderate to severe psoriasis primarily as systemic monotherapy were submitted (ACD2058g, ACD2059g, ACD2390g, ACD2600g and IMP24011). These studies were double blind, placebo-controlled Phase III trials. In total 2714 patients received efalizumab subcutaneously (SC). These trials with efalizumab all had similar study design. In addition study 24011 had a prospectively defined "high need" population (patients with moderate to severe chronic plaque psoriasis who have failed to respond to, or who have a contraindication to, or are intolerant to other systemic therapies including cyclosporine, methotrexate and PUVA.)

The inclusion and exclusion criteria were comparable. The main inclusion criteria were a minimum Psoriasis Area and Severity Index (PASI) score of 12.0 at screening, a plaque psoriasis covering >=10% of total body surface area (BSA) and a need for systemic treatment.

Other outcomes to be considered include PGA (percentage of patients achieving Physician's Global Assessment clear/almost clear at week 12), OLS (percentage of subjects with Overall Lesion Severity (OLS) rating of Minimal or Clear at week 12).

In some studies (ACD 2058g, ACD 2059g, and ACD 2390g) additional endpoints included mean improvement in DLQI (dermatology life quality index) and mean improvement in the frequency and severity subscales of Psoriasis Symptom Assessment (PSA).

In study 24011, an additional endpoint was PASI 50 (proportion of subjects with a 50% improvement from baseline in the PASI score (partial responders).

4b. Unfavourable effects (i.e. safety): adverse events reported to be associated or caused by efalizumab(spontaneously reported Adverse Effects are deemed to be causally related to the drug per reporter).

Safety issues added to the SPC or strengthened warnings since the initial MAA of efalizumab in the EU are as follows: aseptic meningitis, (opportunistic) infections including tuberculosis, immune mediated haemolytic anaemia, decreased antibody development with vaccinations, interstitial pneumonitis, arthritis, erythema multiforme, inflammatory polyradiculoneuropathy including GuillainBarre like syndrome and Miller Fisher syndrome, facial palsy and Bells palsy during long-term use, severe infections and malignancies, PML.

Other unfavourable effects may include overall incidence of AEs per SOC in Clinical Trials at week 12.

At the time of the CHMP assessment report, the efalizumab worldwide exposed population was estimated 47,000 patient-years. An evaluation of the exposed population per duration of exposure is available (e.g. estimated number of patients exposed to efalizumab for more than 3 years ranges from 2,236 to 3,832)

CHMP Assessment Report EMEA/H/542/A20/28 (See Table 1);

Market Authorisation/EPAR

CHMP Assessment Report EMEA/H/542/A20/28 (See Table 2 and additional notes for summary);

PSURs and SPC variations (See Table 3 and additional notes for summary)

5a. Pre-approval:N/A

• do nothing, if the B-R assessment is still positive

• limit duration, to 2 years (proposed by MAH based on the observed delay of onset of the 4 reported cases of PML)

• Limit duration AND restrict indication to a subset of patients where B-R would still be positive

• Suspend/revoke Market Authorisation.

N/A

CHMP Opinion EMA/CHMP/3552/2009

Alternative: Do nothing:implies that B-R balance still considered positive by Rapporteur and CHMP using MCDA quantitative model based on above data.

Alternative: Restrictions:

• (i) 2 year treatment duration limitation: no available data; would require a prospective study; no guidance for transition to alternative treatment

• (ii) Target population change; however the indication in EU is already restricted to the defined "high need" population.

• (iii) Suspension/revocation of MA: dose tapering consequences; risk of rebound effect (rare erythrodermic forms reported upon treatment withdrawal); transition to alternative treatment (not documented, no available data nor guidance). Drug Recall Worldwide in case of revocation of MA in EU and US.

(i) Serono internal document: Risk of PML: analysis of incidence and risk reduction;

(ii) no efficacy and safety data, no subgroup analysis.

Judgement that was made about the benefit-risk balance: Negative Benefit-Risk Balance, voted by CHMP (20 out of 31). B-R assessment to be reiterated using the same data but with a MCDA quantitative method.

CHMP Opinion EMA/CHMP/3552/2009

Efficacy:Uncertainty on the extent of off-label use in patients with less severe conditions, decreasing the benefit part of the balance.

No direct comparison with any other systemic treatment, neither standard (cyclosporine, methotrexate, PUVA) nor biologicals. Assessors of B-R in Jan 2009 had indirect comparison with results of RCT for new biologicals.

Safety: Uncertainty on the shape of the risk function of PML over time (probably not linear), based on only 4 cases. No true incidence but only reporting rate, although under-reporting is unlikely or very limited due to large communication of this risk to patients and prescribers An internal document provides the patient exposure per duration of treatment based on Sales data.

Efficacy: no source data on off label use. Limited post-marketing studies.

Safety: Serono internal document: Risk of PML: analysis of incidence and risk reduction

The extent to which the benefit-risk balance in step 7 is reduced by considering all sources of uncertainty, to provide a benefit-risk balance:

Whichever the uncertainty on efficacy and safety data, all scenarii would decrease the benefit risk balance (underestimated risk, overestimated benefit). If all deterministic measures (derived from measures of central tendency on all the criteria) were set to the favourable limits of their confidence intervals, then, clearly, the B-R ratio would improve. However, considering the full range of uncertainty usually leads to a less favourable B-R balance. Thresholds are not considered in multi-criteria decision analysis because these models just compare the benefit-risk balances of the alternatives. Decisions based on single criteria can only be justified if the entire weight of 100% is assigned to that one criterion.

• Initial MA in 2004 was already controversial (no consensus between Rapporteur and co-Rapporteur)

• In January 2009, medical need is covered by several other therapeutic options, and efalizumab has modest efficacy when compared to alternative treatments (indirect comparison with similar endpoints from RCT with new biological)

• Psoriasis is not a life-threatening disease though it may have a serious impact on social and professional life

• Risk Management Plan with no obvious risk minimisation measures which could be easily and quickly implemented (sub population ?, limitation of treatment to 2 years).

11. The basis for the decision maker's decision as to how tolerable the benefit-risk balance is judged to be (taking into account stakeholders' views of risk?):

Safety Advisory Group (SAG, consisting of dermatologists and neurologists) was consulted shortly prior to the final decision. Some have voiced the patient's perspective.

Responses of the Scientific Advisory Group CNS to the CHMP list of questions on efalizumab 7 Jan 2009.

EMA/24463/2009

How this decision might set a precedent or make similar decisions in the future easier or more difficult:

Efalizumab is the first monoclonal antibody ever to be definitively revoked from the market for safety reasons (natalizumab came back on the market with a RMP).

The FDA made in US a similar decision to EMA, leading to a US withdrawal of efalizumab from market approximately at the same time as EU and rest of the world.

Benefit-Risk balance of immunosuppressive monoclonal antibodies with unknown long term effects in non life-threatening diseases with existing alternative treatments may be questionable over time. Development programmes to be adapted to this situation (design, duration, sub population analysis, etc.)