This information is intended for use by health professionals

 This medicinal product is subject to additional monitoring. This will allow quick identification of new safety information. Healthcare professionals are asked to report any suspected adverse reactions. See section 4.8 for how to report adverse reactions.

1. Name of the medicinal product

Hemlibra 30 mg/mL solution for injection

Hemlibra 150 mg/mL solution for injection

2. Qualitative and quantitative composition

Hemlibra 30 mg/mL solution for injection

Each mL of solution contains 30 mg of emicizumab*

Each vial of 1 mL contains 30 mg of emicizumab at a concentration of 30 mg/mL.

Hemlibra 150 mg/mL solution for injection

Each mL of solution contains 150 mg of emicizumab*

Each vial of 0.4 mL contains 60 mg of emicizumab at a concentration of 150 mg/mL.

Each vial of 0.7 mL contains 105 mg of emicizumab at a concentration of 150 mg/mL.

Each vial of 1 mL contains 150 mg of emicizumab at a concentration of 150 mg/mL.

* produced by recombinant DNA technology using mammalian Chinese Hamster Ovary (CHO) cell culture

For the full list of excipients, see section 6.1

3. Pharmaceutical form

Solution for injection.

Colourless to slightly yellow solution.

4. Clinical particulars
4.1 Therapeutic indications

Hemlibra is indicated for routine prophylaxis of bleeding episodes in patients with haemophilia A with factor VIII inhibitors.

Hemlibra can be used in all age groups.

4.2 Posology and method of administration

Treatment should be initiated under the supervision of a physician experienced in the treatment of haemophilia and/or bleeding disorders.

Posology

Treatment (including routine prophylaxis) with bypassing agents (e.g. aPCC and rFVIIa) should be discontinued the day before starting Hemlibra therapy (see section 4.4).

The recommended dose is 3 mg/kg once weekly for the first 4 weeks (loading dose), followed by 1.5 mg/kg once weekly (maintenance dose), administered as a subcutaneous injection.

The patient dose (in mg) and volume (in mL) should be calculated as follows:

● Loading dose (3 mg/kg) once weekly for the first 4 weeks:

Patient bodyweight (kg) x dose (3 mg/kg) = total amount (mg) of emicizumab to be administered

● Followed by a maintenance dose (1.5 mg/kg) once weekly from week 5 on:

Patient bodyweight (kg) x dose (1.5 mg/kg) = total amount (mg) of emicizumab to be administered

The total volume of Hemlibra to be injected subcutaneously is calculated as follows:

Total amount (mg) of emicizumab to be administered ÷ vial concentration (mg/mL) = total volume of Hemlibra (mL) to be injected.

Different Hemlibra concentrations (30 mg/mL and 150 mg/mL) should not be combined when making up the total volume to be administered.

A volume greater than 2 mL per injection should not be administered.

Examples:

Patient's bodyweight of 60 kg:

● Loading dose (first 4 weeks) example: 60 kg x 3 mg/kg = 180 mg of emicizumab needed for the loading dose.

● To calculate the volume to be administered divide calculated dose 180 mg by 150 mg/mL: 180 mg of emicizumab ÷ 150 mg/mL = 1.20 mL of 150 mg/mL Hemlibra concentration to be injected.

● Choose appropriate dosage and volume from vial strengths available.

● Maintenance dose (from week 5 on) example: 60 kg x 1.5 mg/kg = 90 mg of emicizumab needed for the maintenance dose.

● To calculate the volume to be administered divide calculated dose 90 mg by 150 mg/mL: 90 mg of emicizumab ÷ 150 mg/mL = 0.6 mL of 150 mg/mL Hemlibra concentration to be injected.

● Choose appropriate dosage and volume from vial strengths available.

Patient's bodyweight of 16 kg:

● Loading dose (first 4 weeks) example: 16 kg x 3 mg/kg = 48 mg of emicizumab needed for the loading dose.

● To calculate the volume to be administered divide calculated dose 48 mg by 150 mg/mL: 48 mg of emicizumab ÷ 150 mg/mL = 0.32 mL of 150 mg/mL Hemlibra concentration to be injected.

● Choose appropriate dosage and volume from vial strengths available.

● Maintenance dose (from week 5 on) example: 16 kg x 1.5 mg/kg = 24 mg of emicizumab needed for the maintenance dose.

● To calculate the volume to be administered divide calculated dose 24 mg by 30 mg/mL: 24 mg of emicizumab ÷ 30 mg/mL = 0.8 mL of 30 mg/mL Hemlibra concentration to be injected.

● Choose appropriate dosage and volume from vial strength available.

Duration of treatment

Hemlibra is intended for long-term prophylactic treatment.

Dosage adjustments during treatment

No dosage adjustments of Hemlibra are recommended.

Delayed or missed doses

If a patient misses a scheduled weekly subcutaneous injection of Hemlibra, the patient should be instructed to take the missed dose as soon as possible, up to a day before the day of the next scheduled dose. The patient should then administer the next dose on the usual scheduled dosing day. The patient should not take a double dose to make up for a missed dose.

Special populations

Paediatric

No dose adjustments are recommended in paediatric patients (see section 5.2). There are no data in patients less than 1 year of age.

Elderly

No dose adjustments are recommended in patients ≥ 65 years of age (see section 5.2). There are no data in patients over 75 years old.

Renal and hepatic impairment

No dose adjustments are recommended in patients with mild renal or mild and moderate hepatic impairment (see section 5.2). Emicizumab has not been studied in patients with moderate or severe renal impairment or severe hepatic impairment

Management in the perioperative setting

The safety and efficacy of emicizumab have not been formally evaluated in the surgical setting. If bypassing agents (e.g. aPCC and rFVIIa) are required in the perioperative period, please refer to the dosing guidance on the use of bypassing agents in section 4.4.

Immune tolerance induction (ITI)

The safety and efficacy of emicizumab in patients receiving ongoing immune tolerance induction have not yet been established. No data are available.

Method of administration

Hemlibra is for subcutaneous use only, and it should be administered using appropriate aseptic technique (see section 6.6).

The injection should be restricted to the recommended injection sites: the abdomen, the upper outer arms and the thighs (see section 5.2).

Administration of Hemlibra subcutaneous injection in the upper outer arm should be performed by a caregiver or healthcare professional.

Alternating the site of injection may help prevent or reduce injection site reactions (see section 4.8). Hemlibra subcutaneous injection should not be administered into areas where the skin is red, bruised, tender or hard, or areas where there are moles or scars.

During treatment with Hemlibra, other medicinal products for subcutaneous administration should, preferably, be injected at different anatomical sites.

Administration by the patient and/or caregiver

Hemlibra is intended for use under the guidance of a healthcare professional. After proper training in subcutaneous injection technique, a patient may self-inject Hemlibra, or the patient's caregiver may administer it, if their physician determines that it is appropriate.

The physician and the caregiver should determine the appropriateness of the child self-injecting Hemlibra. However, self-administration is not recommended for children below 7 years of age.

For comprehensive instructions on the administration of Hemlibra, see section 6.6 and package leaflet.

4.3 Contraindications

Hypersensitivity to the active substance or to any of the excipients listed in section 6.1.

4.4 Special warnings and precautions for use

Traceability

In order to improve traceability of biological medicinal products, the name and the batch number of the administered product should be clearly recorded.

Thrombotic microangiopathy associated with Hemlibra and activated prothrombin complex concentrate

Cases of thrombotic microangiopathy (TMA) were reported from a clinical trial in patients receiving Hemlibra prophylaxis when on average a cumulative amount of >100U/kg/24 hours of activated prothrombin complex concentrate (aPCC) for 24 hours or more was administered (see section 4.8). Treatment for the TMA events included supportive care with or without plasmapheresis and haemodialysis. Evidence of improvement was seen within one week following discontinuation of aPCC and interruption of Hemlibra. This rapid improvement is distinct from the usual clinical course observed in atypical hemolytic uremic syndrome and classic TMAs, such as thrombotic thrombocytopenic purpura (see section 4.8). One patient resumed Hemlibra following resolution of TMA and continued to be treated safely.

Patients receiving Hemlibra prophylaxis should be monitored for the development of TMA when administering aPCC. The physician should immediately discontinue aPCC and interrupt Hemlibra therapy if clinical symptoms and/or laboratory findings consistent with TMA occur, and manage as clinically indicated. Physicians and patients/caregivers should weigh the benefits and risks of resuming Hemlibra prophylaxis following complete resolution of TMA on a case-by-case basis. In case a bypassing agent is indicated in a patient receiving Hemlibra prophylaxis, see below for dosing guidance on the use of bypassing agents.

Caution should be used when treating patients who are at high risk for TMA (e.g. have a previous medical history or family history of TMA), or those who are receiving concomitant medications known to be a risk factor for the development of TMA (e.g. ciclosporin, quinine, tacrolimus).

Thromboembolism associated with Hemlibra and activated prothrombin complex concentrate

Serious thrombotic events were reported from a clinical trial in patients receiving Hemlibra prophylaxis when on average a cumulative amount of >100U/kg/24 hours of aPCC for 24 hours or more was administered (see section 4.8). No cases required anticoagulation therapy. Following discontinuation of aPCC and interruption of Hemlibra, evidence of improvement or resolution was seen within one month (see section 4.8). One patient resumed Hemlibra following resolution of thrombotic event and continued to be treated safely.

Patients receiving Hemlibra prophylaxis should be monitored for the development of thromboembolism when administering aPCC. The physician should immediately discontinue aPCC and interrupt Hemlibra therapy if clinical symptoms, imaging, and/or laboratory findings consistent with thrombotic events occur, and manage as clinically indicated. Physicians and patients/caregivers should weigh the benefits and risks of resuming Hemlibra prophylaxis following complete resolution of thrombotic events on a case-by-case basis. In case a bypassing agent is indicated in a patient receiving Hemlibra prophylaxis, see below for dosing guidance on the use of bypassing agents.

Guidance on the use of bypassing agents in patients receiving Hemlibra prophylaxis

Treatment with bypassing agents should be discontinued the day before starting Hemlibra therapy.

Physicians should discuss with all patients and/or caregivers the exact dose and schedule of bypassing agents to use, if required while receiving Hemlibra prophylaxis.

Hemlibra increases the patient's coagulation potential. The bypassing agent dose required may therefore be lower than that used without Hemlibra prophylaxis. The dose and duration of treatment with bypassing agents will depend on the location and extent of bleeding, and the patient's clinical condition. Use of aPCC should be avoided unless no other treatment options/alternatives are available. If aPCC is indicated in a patient receiving Hemlibra prophylaxis, the initial dose should not exceed 50 U/kg and laboratory monitoring is recommended (including but not restricted to renal monitoring, platelet testing, and evaluation of thrombosis). If bleeding is not controlled with the initial dose of aPCC up to 50 U/kg, additional aPCC doses should be administered under medical guidance or supervision with consideration made to laboratory monitoring for the diagnosis of TMA or thromboembolism and verification of bleeds prior to repeated dosing. The total aPCC dose should not exceed 100 U/kg in the first 24-hours of treatment. Treating physicians must carefully weigh the risk of TMA and thromboembolism against the risk of bleeding when considering aPCC treatment beyond a maximum of 100 U/kg in the first 24-hours.

In clinical trials, no cases of TMA or thrombotic events were observed with use of activated recombinant human FVII (rFVIIa) alone in patients receiving Hemlibra prophylaxis.

Bypassing agent dosing guidance should be followed for at least 6 months following discontinuation of Hemlibra prophylaxis (see section 5.2).

Effects of emicizumab on coagulation tests

Emicizumab replaces the tenase cofactor activity of activated factor VIII (FVIIIa). Coagulation laboratory tests based on intrinsic clotting, including the activated clotting time (ACT), activated partial thromboplastin time (e.g. aPTT), measure the total clotting time including time needed for activation of FVIII to FVIIIa by thrombin. Such intrinsic pathway based tests will yield overly shortened clotting times with emicizumab, which does not require activation by thrombin. The overly shortened intrinsic clotting time will then disturb all single factor assays based on aPTT, such as the one stage FVIII activity assay (see section 4.4, Table 1). However, single factor assays utilising chromogenic or immuno-based methods are not affected by emicizumab and may be used to monitor coagulation parameters during treatment, with specific considerations for FVIII chromogenic activity assays as described below.

Chromogenic factor VIII activity tests may be manufactured with either human or bovine coagulation proteins. Assays containing human coagulation factors are responsive to emicizumab but may overestimate the clinical haemostatic potential of emicizumab. In contrast, assays containing bovine coagulation factors are insensitive to emicizumab (no activity measured) and can be used to monitor endogenous or infused factor VIII activity, or to measure anti FVIII inhibitors.

Emicizumab remains active in the presence of inhibitors against factor VIII and so will produce a false negative result in clotting based Bethesda assays for functional inhibition of factor VIII. Instead, a chromogenic Bethesda assay utilising a bovine based factor VIII chromogenic test that is insensitive to emicizumab may be used.

These two pharmacodynamic markers do not reflect the true haemostatic effect of emicizumab in vivo (aPTT is overly shortened and reported factor VIII activity may be overestimated) but provide a relative indication of the pro-coagulant effect of emicizumab.

In summary, intrinsic pathway clotting-based laboratory test results in patients treated with Hemlibra should not be used to monitor its activity, determine dosing for factor replacement or anti-coagulation, or measure factor VIII inhibitors titers. Caution should be taken if intrinsic pathway clotting based laboratory tests are used, as misinterpretation of their results may lead to under-treatment of patients experiencing bleeding episodes, which can potentially result in severe or life-threatening bleeds.

Laboratory tests unaffected by emicizumab are also shown in Table 1 below. Due to its long half-life, these effects on coagulation assays may persist for up to 6 months after the last dose (see section 5.2).

Table 1 Coagulation test results affected and unaffected by emicizumab

Results Affected by emicizumab

Results Unaffected by emicizumab

- Activated partial thromboplastin time (aPTT)

- Bethesda assays (clotting-based) for FVIII inhibitor titers

- One-stage, aPTT-based, single-factor assays

- aPTT-based activated protein C resistance (APC-R)

- Activated clotting time (ACT)

- Bethesda assays (bovine chromogenic) for FVIII inhibitor titers

- Thrombin time (TT)

- One-stage, prothrombin time (PT)-based, single-factor assays

- Chromogenic-based single-factor assays other than FVIII1

- Immuno-based assays (e.g. ELISA, turbidimetric methods)

- Genetic tests of coagulation factors (e.g. Factor V Leiden, Prothrombin 20210)

1For important considerations regarding FVIII chromogenic activity assays, see section 4.4.

Paediatric population

There are no data in children <1 year of age. The developing hemostatic system in neonates and infants is dynamic and evolving, and the relative concentrations of pro- and anticoagulant proteins in these patients should be taken into consideration when making a benefit-risk assessment, including potential risk of thrombosis (e.g. central venous catheter-related thrombosis).

4.5 Interaction with other medicinal products and other forms of interaction

No adequate or well-controlled drug-drug interaction studies have been conducted with emicizumab.

Clinical experience indicates a drug interaction exists with emicizumab and aPCC (see sections 4.4 and 4.8).

There is a possibility for hypercoagulability with rFVIIa or FVIII with emicizumab based on preclinical experiments. Emicizumab increases coagulation potential, therefore the coagulation factor dose required to achieve hemostasis may be lower than when used without Hemlibra prophylaxis.

Experience with concomitant administration of anti-fibrinolytics with aPCC or rFVIIa in patients receiving emicizumab prophylaxis is limited. However, the possibility of thrombotic events should be considered when systemic anti-fibrinolytics are used in combination with aPCC or rFVIIa in patients receiving emicizumab.

4.6 Fertility, pregnancy and lactation

Women of childbearing potential/Contraception

Women of childbearing potential receiving Hemlibra should use effective contraception during, and for at least 6 months after cessation of Hemlibra treatment (see section 5.2).

Pregnancy

There are no clinical studies of emicizumab use in pregnant women. Animal reproduction studies have not been conducted with Hemlibra. It is not known whether emicizumab can cause fetal harm when administered to a pregnant woman or can affect reproductive capacity. Hemlibra should be used during pregnancy only if the potential benefit for the mother outweighs the potential risk to the fetus taking into account that, during pregnancy and after parturition, the risk for thrombosis is increased and that several pregnancy complications are linked to an increased risk for disseminated intravascular coagulation (DIC).

Breast-feeding

It is not known whether emicizumab is excreted in human milk. No studies have been conducted to assess the impact of emicizumab on milk production or its presence in breast milk. Human IgG is known to be present in human milk. A decision must be made whether to discontinue breast-feeding or to discontinue/abstain from Hemlibra therapy taking into account the benefit of breast-feeding for the child and the benefit of therapy for the woman.

Fertility

Animal studies do not indicate direct or indirect harmful effects with respect to reproductive toxicity (see section 5.3). No fertility data are available in humans. Thus, the effect of emicizumab on male and female fertility is unknown.

4.7 Effects on ability to drive and use machines

Hemlibra has no influence on the ability to drive and use machines.

4.8 Undesirable effects

Summary of the safety profile

The most serious adverse drug reactions (ADRs) reported from the clinical trials with Hemlibra were thrombotic microangiopathy (TMA) and thrombotic events, including cavernous sinus thrombosis (CST) and superficial vein thrombosis contemporaneous with skin necrosis (see below and section 4.4).

The most common ADRs reported in ≥ 10% of patients treated with at least one dose of Hemlibra were: injection site reactions (19%), headache (15%) and arthralgia (10%).

In total four patients (2.1%) in the clinical trials receiving Hemlibra prophylaxis withdrew from treatment due to ADRs, which were TMA, skin necrosis and superficial thrombophlebitis, and injection site reaction.

Tabulated list of adverse drug reactions

The following adverse drug reactions (ADRs) are based on pooled data from two phase III clinical trials (Study BH29884 and Study BH29992) and one phase I/II clinical trial (Study ACE002JP), in which a total of 189 male patients with haemophilia A received at least one dose of Hemlibra as routine prophylaxis. Ninety-four patients (50%) were adults. Seven out of the 189 patients (4%) included in the safety population were patients without FVIII inhibitors from the phase I/II clinical trial. The median duration of exposure across the studies was 38 weeks (range: 0.8 to 177.2 weeks).

ADRs from clinical trials in patients who received Hemlibra are listed by MedDRA system organ class (Table 2). The corresponding frequency categories for each ADR are based on the following convention: very common (≥ 1/10), common (≥ 1/100 to < 1/10), uncommon (≥ 1/1,000 to < 1/100), rare (≥ 1/10,000 to < 1/1,000), very rare (< 1/10,000) and not known (cannot be estimated from the available data).

Table 2 Summary of adverse drug reactions from pooled clinical trials with Hemlibra

System Organ Class (SOC)

Adverse reactions

(preferred term, MedDRA)

Frequency

Infections and infestations

Cavernous sinus thrombosis

Uncommon

Blood and lymphatic system disorders

Thrombotic microangiopathy

Common

Nervous system disorders

Headache

Very common

Vascular disorders

Thrombophlebitis superficial

Uncommon

Gastrointestinal disorders

Diarrhoea

Common

Skin and subcutaneous tissue disorders

Skin necrosis

Uncommon

Musculoskeletal and connective tissue disorders

Arthralgia

Common

Myalgia

Common

General disorders and administration site conditions

Injection site reaction

Very common

Pyrexia

Common

Description of selected adverse drug reactions

Thrombotic microangiopathy

Thrombotic microangiopathy (TMA) events were reported in 1.6 % of patients (3/189) from clinical trials and in 8.3% of patients (3/36) who received at least one dose of aPCC while being treated with emicizumab. All 3 TMAs occurred when on average a cumulative amount of > 100 U/Kg/24 hours of aPCC for 24 hours or more was administered during a treatment event (see section 4.4). Patients presented with thrombocytopenia, microangiopathic hemolytic anemia, and acute kidney injury, without severe deficiencies in ADAMTS13 activity.

Thrombotic events

Serious thrombotic events were reported in 1.1% of patients (2/189) from clinical trials and in 5.6% of patients (2/36) who received at least one dose of aPCC while being treated with emicizumab. Both serious thrombotic events occurred when on average a cumulative amount of > 100 U/Kg/24 hours of aPCC for 24 hours or more was administered during a treatment event (see section 4.4).

Characterization of the interaction between emicizumab and aPCC treatment in pivotal clinical trials

There were 79 instances of aPCC treatment in patients receiving emicizumab prophylaxis, of which eight instances (10.1%) consisted of on average a cumulative amount of >100 U/kg/24 hours of aPCC for 24 hours or more; two of the eight were associated with thrombotic events and three of the eight were associated with TMA (Table 3). No TMA or thrombotic events were associated with the remaining instances of aPCC treatment. Of all instances of aPCC treatment, 67.1% consisted of only one infusion < 100 U/kg.

Table 3 Characterisation of aPCC treatment* in studies BH29884 and BH29992

Duration of aPCC treatment

Average cumulative amount of aPCC over 24 hours (U/kg/24 hours)

<50

50–100

>100

<24 hours

6

47

13

24-48 hours

0

3

1b

>48 hours

1

1

7a,a,a,b

* An instance of aPCC treatment is defined as all doses of aPCC received by a patient, for any reason, until there was a 36-hour treatment-free break.

a Thrombotic microangiopathy

b Thrombotic event

Injection site reactions

Injection site reactions (ISRs) were reported very commonly from clinical trials. All ISRs observed in the Hemlibra clinical trials were reported as being non-serious and generally mild to moderate in intensity. Most ISRs resolved without treatment. The most commonly reported ISR symptoms were injection site erythema (7.4 %), injection site pruritus (5.3%) and injection site pain (5.3%).

Paediatric population

The paediatric population studied comprises a total of 95 patients, of which 2 (2%) were infants (1 months to less than 2 years of age), 55 (58%) were children (from 2 to less than 12 years of age) and 38 (40%) were adolescents (from 12 to less than 18 years old).

The safety profile of Hemlibra was overall consistent between infants, children, adolescents, and adults.

Reporting of suspected adverse reactions

Reporting suspected adverse reactions after authorisation of the medicinal product is important. It allows continued monitoring of the benefit/risk balance of the medicinal product. Healthcare professionals are asked to report any suspected adverse reactions (see details below).

Ireland

HPRA Pharmacovigilance

Earlsfort Terrace

IRL - Dublin 2

Tel: +353 1 6764971

Fax: +353 1 6762517

Website: www.hpra.ie

e-mail: medsafety@hpra.ie

Malta

ADR Reporting

Website: www.medicinesauthority.gov.mt/adrportal

United Kingdom

Yellow Card Scheme

Website: www.mhra.gov.uk/yellowcard or search for MHRA Yellow Card in the Google Play or Apple App Store

4.9 Overdose

There is limited experience with overdose of Hemlibra.

Symptoms

Accidental overdose may result in hypercoagulability.

Management

Patients who receive an accidental overdose should immediately contact their physician and be monitored closely.

5. Pharmacological properties
5.1 Pharmacodynamic properties

Pharmacotherapeutic group: antihemorrhagics, other systemic hemostatics; ATC code: B02BX06

Mechanism of action

Emicizumab bridges activated factor IX and factor X to restore the function of missing activated factor VIII that is needed for effective haemostasis.

Emicizumab has no structural relationship or sequence homology to factor VIII and, as such, does not induce or enhance the development of direct inhibitors to factor VIII.

Pharmacodynamics

Prophylactic therapy with Hemlibra shortens the aPTT and increases the reported factor VIII activity (using a chromogenic assay with human coagulation factors). These two pharmacodynamic markers do not reflect the true haemostatic effect of emicizumab in vivo (aPTT is overly shortened and reported factor VIII activity may be overestimated) but provide a relative indication of the pro-coagulant effect of emicizumab.

Clinical efficacy and safety

Patients (aged 12 to 75 years old) with haemophilia A with factor VIII inhibitors (Study BH29884)

Hemlibra prophylaxis was evaluated in a randomised, multicentre, open-label clinical study in 109 adolescent and adult males (aged 12 to 75 years old) with haemophilia A with factor VIII inhibitors who had previously received either episodic or prophylactic treatment with bypassing agents (aPCC and rFVIIa). In the study, patients received weekly Hemlibra prophylaxis (Arms A, C, and D) — 3 mg/kg once weekly for 4 weeks followed by 1.5 mg/kg once weekly thereafter — or no prophylaxis (Arm B). Dose up-titration to 3 mg/kg once weekly was allowed after 24 weeks on Hemlibra prophylaxis in case of suboptimal efficacy (i.e. ≥ 2 spontaneous and clinically significant bleeds). During the study, two patients underwent up-titration of their maintenance dose to 3 mg/kg once weekly.

Fifty-three patients previously treated with episodic (on-demand) bypassing agents were randomised in a 2:1 ratio to receive Hemlibra prophylaxis (Arm A) or no prophylaxis (Arm B), with stratification by prior 24-week bleed rate (< 9 or ≥ 9). Patients randomised to Arm B could switch to Hemlibra after completing at least 24 weeks without prophylaxis.

Forty-nine patients previously treated with prophylactic bypassing agents were enrolled in Arm C to receive Hemlibra prophylaxis. Patients previously treated with episodic (on-demand) bypassing agents who had participated in the non-interventional study (NIS) prior to enrolment but were unable to enroll in Study BH29884 prior to the closure of Arms A and B were enrolled in Arm D to receive Hemlibra prophylaxis. The NIS is an observational study with the main objective of capturing detailed clinical data on the bleeding episodes and haemophilia medication use of patients with haemophilia A outside of an interventional trial setting.

The primary objective of the study was to evaluate, among patients previously treated with episodic (on-demand) bypassing agents, the treatment effect of weekly Hemlibra prophylaxis compared with no prophylaxis (Arm A vs. Arm B) on the number of bleeds requiring treatment with coagulation factors over time (minimum of 24 weeks or date of discontinuation). Other secondary objectives of the randomised comparison of Arms A and B were the efficacy of weekly Hemlibra prophylaxis in reducing the number of all bleeds, spontaneous bleeds, joint bleeds and target joint bleeds, as well as assessing patients' health-related quality of life and health status. The mean exposure time (+SD) for all patients on study was 21.38 weeks (12.01). For each treatment arm, the mean exposure times (+SD) were 28.86 weeks (8.37) for Arm A, 8.79 (3.62) for Arm B, 21.56 (11.85) for Arm C and 7.08 (3.89) for Arm D. One patient in Arm A withdrew from study prior to initiation of Hemlibra.

The study also evaluated the efficacy of weekly Hemlibra prophylaxis compared with previous episodic (on-demand) and prophylactic bypassing agents (separate comparisons) in patients who had participated in the NIS prior to enrolment (Arms A and C, respectively). Only patients from the NIS were included in this comparison, because bleed and treatment data were collected with the same level of granularity in both periods.

In Study BH29884, all primary and secondary objectives were met (see below Tables 4 and 5).

Table 4 Study BH29884: Overview of efficacy (Intent-To-Treat population)

Endpoint

Arm B: no prophylaxis

Arm A: 1.5 mg/kg Hemlibra weekly

N=18

N=35

Treated bleeds

ABR (95% CI)

23.3 (12.33; 43.89)

2.9 (1.69; 5.02)

% reduction (RR), p-value

87% (0.13), < 0.0001

% patients with 0 bleeds (95% CI)

5.6 (0.1; 27.3)

62.9 (44.9; 78.5)

Median ABR (IQR)

18.8 (12.97;35.08)

0 (0; 3.73)

All bleeds

ABR (95% CI)

28.3 (16.79; 47.76)

5.5 (3.58; 8.60)

% reduction (RR), p-value

80% (0.20), < 0.0001

% patients with 0 bleeds (95% CI)

5.6 (0.1; 27.3)

37.1 (21.5; 55.1)

Treated spontaneous bleeds

ABR (95% CI)

16.8 (9.94; 28.30)

1.3 (0.73; 2.19)

% reduction (RR), p-value

92% (0.08), < 0.0001

% patients with 0 bleeds (95% CI)

11.1 (1.4; 34.7)

68.6 (50.7; 83.1)

Treated joint bleeds

ABR (95% CI)

6.7 (1.99; 22.42)

0.8 (0.26; 2.20)

% reduction (RR), p-value

89% (0.11), 0.0050

% patients with 0 bleeds (95% CI)

50.0 (26.0; 74.0)

85.7 (69.7; 95.2)

Treated target joint bleeds

ABR (95% CI)

3.0 (0.96; 9.13)

0.1 (0.03; 0.58)

% reduction (RR), p-value

95% (0.05), 0.0002

% patients with 0 bleeds (95% CI)

50.0 (26.0; 74.0)

94.3 (80.8; 99.3)

Rate ratio, and confidence interval (CI) come from negative binomial regression (NBR) model and p-value from Stratified Wald test, comparing bleed rate between specified arms.

Arm B: includes no prophylaxis period only.

Bleed definitions adapted based on ISTH criteria.

Treated bleeds = bleeds treated with bypassing agents.

All bleeds = bleeds treated and not treated with bypassing agents.

Includes data before up-titration only, for patients whose dose was up-titrated.

Patients exposed to emicizumab started with a loading dose of 3 mg/kg/week for 4 weeks.

ABR= Annualised Bleed Rate; CI= confidence interval; RR= rate ratio; IQR= interquartile range, 25th percentile to 75th percentile.

In the intra-patient analysis, Hemlibra prophylaxis resulted in statistically significant (p = 0.0003) and clinically meaningful reduction (79%) in bleed rate for treated bleeds compared with previous bypassing agent prophylaxis collected in the NIS prior to enrolment (see Table 5).

Table 5 Study BH29884: Annualised Bleed Rate for Hemlibra prophylaxis intra-patient comparison – treated bleeds (NIS patients)

Endpoint

Arm CNIS: previous treatment with prophylactic bypassing agent

Arm C: Hemlibra 1.5 mg/kg weekly

N=24

N=24

Treated bleeds

ABR (95% CI)

15.7 (11.08; 22.29)

3.3 (1.33; 8.08)

% patients with 0 bleeds (95% CI)

12.5 (2.7; 32.4)

70.8 (48.9; 87.4)

Median ABR (IQR)

12.0 (5.73; 24.22)

0.0 (0.00; 2.23)

% reduction

(RR), p-value

79%

(0.21), 0.0003

Rate ratio and confidence interval (CI) comes from negative binomial regression (NBR) model and p-value from Stratified Wald test, comparing ABR between specified arms.

Intra-patient comparator data from the NIS.

Only patients who participated in the NIS and in study BH29884 are included.

Includes data before up-titration only, for patients whose dose was up-titrated.

Treated bleeds = bleeds treated with bypassing agents.

Bleed definitions adapted based on ISTH criteria.

ABR= Annualised Bleed Rate; CI= confidence interval; RR= rate ratio; IQR=interquartile range, 25th percentile to 75th percentile

In Study BH29884, health-related quality of life for patients aged ≥ 18 years was evaluated at Week 25 based on the Haemophilia-specific Quality of Life (Haem-A-QoL) questionnaire for adults. Baseline Total Scores (mean = 41.14 and 44.58, respectively) and Physical Health scale scores (mean = 52.41 and 57.19, respectively) were similar for Hemlibra prophylaxis and no prophylaxis. Table 6 provides a summary of the comparison between the Hemlibra prophylaxis arm (Arm A) and the no prophylaxis arm (Arm B) on the Haem-A-QoL Total Score and Physical Health scale after 24 weeks of treatment adjusting for baseline. Weekly Hemlibra prophylaxis showed a statistically significant and clinically meaningful improvement compared with the no prophylaxis in the pre-specified endpoints of Haem-A-QoL Total Score and Physical Health Scale score at the Week 25 assessment.

Table 6 Study BH29884: Haem-A-QoL Scores in patients ≥ 18 years after 24 weeks

Haem-A-QoL scores after 24 weeks

Arm B: no prophylaxis (n=14)

Arm A: 1.5 mg/kg Hemlibra weekly (n=25)

Total score

Adjusted mean

43.21

29.2

Difference in adjusted means (95% CI)

14.01 (5.56, 22.45)

p-value

0.0019

Physical health

Adjusted mean

54.17

32.61

Difference in adjusted means (95% CI)

21.55 (7.89, 35.22)

p-value

0.0029

Arm B: includes no prophylaxis period only.

Includes data before up-titration only, for patients whose dose was up-titrated.

Patients exposed to emicizumab started with a loading dose of 3 mg/kg/week for 4 weeks.

Haem-A_QoL scales range from 0 to 100; lower scores are reflective of better HRQoL.

Clinically meaningful difference: Total score: 7 points; Physical Health: 10 points.

In Study BH29884, patients' health status was assessed according to the EuroQoL Five-Dimension-Five Levels Questionnaire (EQ-5D-5L). Table 7 provides a summary of the comparison between the Hemlibra prophylaxis arm (Arm A) and the no prophylaxis arm (Arm B) on the EQ-5D-5L index utility scale and visual analog scale after 24 weeks of treatment adjusting for baseline. Weekly Hemlibra showed a statistically significant and clinically meaningful improvement compared with no prophylaxis in the pre specified endpoints of EQ-5D-5L index utility scale and visual analogue scale at the Week 25 assessment.

Table 7 Study BH29884: EQ-5D-5L scores in patients ≥ 12 years after 24 weeks

EQ-5D-5L scores after 24 weeks

Arm B: no prophylaxis (n=16)

Arm A: 1.5 mg/kg Hemlibra weekly (n=29)

Visual Analogue Scale

Adjusted mean

74.36

84.08

Difference in adjusted means (95% CI)

-9.72 (-17.62, -1.82)

p-value

0.0171

Index Utility Score

Adjusted mean

0.65

0.81

Difference in adjusted means (95% CI)

-0.16 (-0.25, -0.07)

p-value

0.0014

Arm B: includes no prophylaxis period only.

Includes data before up-titration only, for patients whose dose was up-titrated.

Patients exposed to emicizumab started with a loading dose of 3 mg/kg/week for 4 weeks.

Higher scores indicate better quality of life.

Clinically meaningful difference: VAS: 7 points, Index Utility Score: 0.07 points

Paediatric patients (age < 12 years old, or 12 to 17 years old weighing < 40 kg) with haemophilia A with factor VIII inhibitors (Study BH29992) (Interim Analysis)

Hemlibra weekly prophylaxis was evaluated in a single-arm, multicentre, open-label clinical study in paediatric patients (age < 12 years old, or 12 to 17 years old weighing < 40 kg) with haemophilia A with factor VIII inhibitors. Patients received Hemlibra prophylaxis at 3 mg/kg once weekly for the first 4 weeks followed by 1.5 mg/kg once weekly thereafter.

The study evaluated the pharmacokinetics, safety, and efficacy including the efficacy of weekly Hemlibra prophylaxis compared with previous episodic and prophylactic bypassing agent treatment in patients who had participated in the NIS prior to enrolment (intra-patient comparison).

At the time of the interim analysis, the clinical study had enrolled 60 male patients. Two patients aged < 2 years old, 17 patients aged 2 to < 6 years, 38 patients aged 6 to < 12 years and 3 patients aged ≥ 12 years, resulting in 57 patients that were < 12 years old and evaluable for efficacy. The annualised bleed rate and percent of patients with zero bleeds were calculated for 23 patients <12 years old who received weekly Hemlibra prophylaxis for at least 12 weeks (see Table 7). The median observation time for these patients was 38.1 weeks (range: 12.7 to 41.6 weeks).

The interim analysis efficacy results for Study BH29992 are summarised below (see Tables 8 and 9). In total 20 of 23 (87%) patients had zero treated bleeds and 8 of 23 (34.8%) did not have any bleeds while receiving Hemlibra prophylaxis (see Table 8).

Table 8 Study BH29992: Overview of efficacy (interim analysis)

Endpoint

ABR (95% CI)

N = 23

Median ABR (IQR)

N = 23

% Zero Bleeds

(95% CI)

N = 23

Treated bleeds

0.2 (0.06; 0.62)

0 (0; 0)

87 (66.4; 97.2)

All bleeds

2.9 (1.75; 4.94)

1.5 (0; 4.53)

34.8 (16.4; 57.3)

Treated spontaneous bleeds

0.1 (0.01; 0.47)

0 (0; 0)

95.7 (78.1; 99.9)

Treated joint bleeds

0.1 (0.01; 0.47)

0 (0; 0)

95.7 (78.1; 99.9)

Treated target joint bleeds

Not Estimable*

0 (0; 0)

100 (85.2; 100)

*No treated target joint bleeds reported

ABR = annualised bleed rate; CI = confidence interval; IQR = interquartile range, 25th percentile to 75th percentile

In the intra-patient analysis, Hemlibra weekly prophylaxis resulted in a clinical meaningful reduction (99%) in treated bleed rate in thirteen paediatric patients after at least 12 weeks of treatment compared to their bleed rate collected in the NIS prior to enrolment.

Table 9 Study BH29992: Annualised Bleed Rate for Hemlibra prophylaxis intra-patient comparison in paediatric patients < 12 years of age (interim analysis) – treated bleeds (NIS patients)

Endpoint

Previous bypassing agent treatment* (N = 13)

Hemlibra prophylaxis (N = 13)

Treated bleeds

ABR (95% CI)

17.2 (12.38; 23.76)

0.2 (0.06; 0.76)

% reduction

99%

RR (95% CI)

0.01 (0.004; 0.044)

% patients with zero bleeds (95% CI)

7.7 (0.2; 36)

84.6 (54.6; 98.1)

Median ABR (IQR)

14.3 (11.02; 24.35)

0 (0; 0)

ABR = annualised bleed rate; CI = confidence interval; RR = rate ratio

* Previous prophylactic treatment for 12 patients; previous episodic (on-demand) treatment for 1 subject

There is limited experience with bypassing agent use during surgeries and procedures. Bypassing agent use during surgeries and procedures was determined by the investigator.

Immunogenicity

As with all therapeutic proteins, there is the potential for an immune response in patients treated with emicizumab. A total of 189 patients were tested for anti-emicizumab antibodies in the clinical trials. Four patients (2.1%) tested positive for anti-emicizumab antibodies in the phase I/II trials, all of which were non-neutralising.

The data reflect the number of patients whose test results were considered positive for antibodies to emicizumab using an enzyme-linked immunosorbent assay (ELISA). Immunogenicity assay results may be influenced by several factors including assay sensitivity and specificity, sample handling, timing of sample collection, concomitant medicinal products and underlying disease. For these reasons, comparison of incidence of antibodies to emicizumab with the incidence of antibodies to other products may be misleading.

In case of clinical signs of loss of efficacy, a change of treatment should be considered.

Paediatric population

The European Medicines Agency has deferred the obligation to submit the results of studies with Hemlibra in one or more subsets of the paediatric population in the treatment of hereditary factor VIII deficiency (see section 4.2 for information on paediatric use).

5.2 Pharmacokinetic properties

The pharmacokinetics of emicizumab was determined via non-compartmental analysis in healthy subjects and using a population pharmacokinetic analysis on a database composed of 141 patients with haemophilia A.

Absorption

Following subcutaneous administration in haemophilia A patients, the absorption half-life was 1.7 days.

Following multiple subcutaneous administrations of 3 mg/kg once weekly for the first 4 weeks in haemophilia A patients, mean (±SD) trough plasma concentrations of emicizumab increased to achieve 54.6±14.3 µg/mL at Week 5. Trough plasma concentrations of approximately 50 µg/mL were sustained thereafter with weekly dosing of 1.5 mg/kg (Figure 1).

Figure 1 Studies BH29884 (adult and adolescent study) and BH29992 (paediatric study): mean emicizumab trough plasma concentrations (µg/mL)

The predicted mean (±SD) Ctrough and Cmax at steady state were 52.2±13.5 μg/mL and 56.5±13.5 μg/mL, respectively. The mean (±SD) ratio of Cmax/Ctrough at steady state was 1.07±0.03.

In healthy subjects, the absolute bioavailability following subcutaneous administration of 1 mg/kg was between 80.4% and 93.1% depending on the injection site. Similar pharmacokinetic profiles were observed following subcutaneous administration in the abdomen, upper arm, and thigh. Emicizumab can be administered interchangeably at these anatomical sites (see section 4.2).

Distribution

Following a single intravenous dose of 0.25 mg/kg emicizumab in healthy subjects, the volume of distribution at steady state was 106 mL/kg (i.e. 7.4 L for a 70-kg adult).

The apparent volume of distribution (V/F), estimated from the population PK analysis, in haemophilia A patients following multiple subcutaneous doses of emicizumab was 11.4 L.

Metabolism

The metabolism of emicizumab has not been studied. IgG antibodies are mainly catabolised by lysosomal proteolysis and then eliminated from or reused by the body.

Elimination

Following intravenous administration of 0.25 mg/kg in healthy subjects, the total clearance of emicizumab was 3.26 mL/kg/day (i.e. 0.228 L/d for a 70-kg adult) and the mean terminal half-life was 26.7 days.

Following single subcutaneous injection in healthy subjects, the elimination half-life was approximately 4 to 5 weeks.

Following multiple subcutaneous injections in haemophilia A patients, the apparent clearance was 0.244 L/day and the elimination apparent half-life was 27.8 days.

Dose linearity

Emicizumab exhibited dose-proportional pharmacokinetics in patients with haemophilia A over a dose range from 0.3 to 3 mg/kg once weekly following subcutaneous administration.

Special populations

Paediatric

The effect of age on the pharmacokinetics of emicizumab was assessed in a population pharmacokinetic analysis which included 59 children (less than 12 years) and 38 adolescents (12 to 17 years) with haemophilia A. An additional descriptive analysis of pharmacokinetic data collected from Study BH29992 was performed in two infants (1 month to < 2 years), 55 children (≥ 2 years to < 12 years) and 3 adolescents (12 to 17 years).

Age did not affect the pharmacokinetics of emicizumab in paediatric patients.

Elderly

The effect of age on the pharmacokinetics of emicizumab was assessed in a population pharmacokinetic analysis which included three subjects aged 65 years and older (no subjects were older than 75 years of age). Clearance increased with older age, but no clinically important differences were observed in the pharmacokinetics of emicizumab between subjects < 65 years and subjects ≥ 65 years.

Race

Population pharmacokinetics analyses in patients with haemophilia A showed that race did not affect the pharmacokinetics of emicizumab. No dose adjustment is required for this demographic factor.

Renal impairment

No dedicated studies of the effect of renal impairment on the pharmacokinetics of emicizumab have been conducted.

The safety and efficacy of emicizumab have not been specifically tested in patients with renal impairment. There are limited data available on the use of Hemlibra in patients with mild renal impairment. No data are available on the use of Hemlibra in patients with moderate to severe renal impairment. Mild renal impairment did not affect the pharmacokinetics of emicizumab.

Emicizumab is a monoclonal antibody and is cleared via catabolism rather than renal excretion and a change in dose is not expected to be required for patients with renal impairment.

Hepatic impairment

No dedicated studies on the effect of hepatic impairment on the pharmacokinetics of emicizumab have been conducted. Most of the patients with haemophilia A in the population pharmacokinetic analysis had normal hepatic function (bilirubin and AST ≤ ULN, n=113) or mild hepatic impairment (bilirubin ≤ ULN and AST > ULN or bilirubin < 1.0 to 1.5 × ULN and any AST, n=17). Mild hepatic impairment did not affect the pharmacokinetics of emicizumab (see section 4.2). The safety and efficacy of emicizumab have not been specifically tested in patients with hepatic impairment. Patients with mild and moderate hepatic impairment were included in clinical trials. No data are available on the use of Hemlibra in patients with severe hepatic impairment.

Emicizumab is a monoclonal antibody and cleared via catabolism rather than hepatic metabolism and a change in dose is not expected to be required for patients with hepatic impairment.

5.3 Preclinical safety data

Preclinical data reveal no special hazards for humans based on studies of acute and repeated dose toxicity, including safety pharmacology endpoints and endpoints for reproductive toxicity.

Fertility

Emicizumab did not cause any changes in the reproductive organs of male or female cynomolgus monkeys up to the highest tested dose of 30 mg/kg/week (equivalent to 11 times the human exposure at the highest dose of 3 mg/kg/week, based on AUC).

Teratogenicity

No data are available with respect to potential side effects of emicizumab on embryo-foetal development.

Injection site reactions

Reversible hemorrhage, perivascular mononuclear cell infiltration, degeneration/necrosis of subcutis and swelling of endothelium in the subcutis was noted in animals after subcutaneous injection.

6. Pharmaceutical particulars
6.1 List of excipients

L-Arginine

L-Histidine

L-Aspartic acid

Poloxamer 188

Water for injections

6.2 Incompatibilities

No incompatibilities between Hemlibra and polypropylene or polycarbonate syringes, polycarbonate vial adapters and stainless steel needles have been observed.

In the absence of compatibility studies, this medicinal product must not be mixed with other medicinal products.

6.3 Shelf life

Unopened vial

30 months.

Once removed from the refrigerator, unopened vials can be kept at room temperature (below 30°C) for up to 7 days.

After storage at room temperature, unopened vials may be returned to the refrigerator. If stored out of and then returned to refrigeration, the total combined time out of refrigeration should not exceed 7 days. The vials should never be exposed to temperatures above 30 °C. Vials that have been kept at room temperature for more than 7 days or exposed to temperatures above 30 °C should be discarded.

Pierced vial and filled syringe

From a microbiological point of view, once transferred from the vial to the syringe, the medicinal product should be used immediately. If not used immediately, in-use storage times and conditions are the responsibility of the user.

6.4 Special precautions for storage

Store in a refrigerator (2°C to 8°C).

Do not freeze.

Keep the vial in the outer carton in order to protect from light.

For storage conditions after first opening of the medicinal product, see section 6.3.

6.5 Nature and contents of container

Hemlibra 30 mg/mL solution for injection

3 mL clear glass type I vial with butyl rubber stopper laminated with a fluoro-resin film and crimped with an aluminium cap fitted with a plastic flip-off disk. Each vial contains 30 mg emicizumab in 1 mL of solution for injection. Each carton contains one vial.

Hemlibra 150 mg/mL solution for injection

3 mL clear glass type I vial with butyl rubber stopper laminated with a fluoro-resin film and crimped with an aluminium cap fitted with a plastic flip-off disk. Each vial contains 60 mg emicizumab in 0.4 mL of solution for injection. Each carton contains one vial.

3 mL clear glass type I vial with butyl rubber stopper laminated with a fluoro-resin film and crimped with an aluminium cap fitted with a plastic flip-off disk. Each vial contains 105 mg emicizumab in 0.7 mL of solution for injection. Each carton contains one vial.

3 mL clear glass type I vial with butyl rubber stopper laminated with a fluoro-resin film and crimped with an aluminium cap fitted with a plastic flip-off disk. Each vial contains 150 mg emicizumab in 1 mL of solution for injection. Each carton contains one vial.

Not all pack sizes may be marketed.

6.6 Special precautions for disposal and other handling

Hemlibra solution is a sterile, preservative-free, and ready to use solution for subcutaneous injection that does not need to be diluted.

Hemlibra should be inspected visually to ensure there is no particulate matter or discolouration prior to administration. Hemlibra is a colourless to slightly yellow solution. The solution should be discarded if particulate matter is visible or product is discoloured.

Do not shake.

Hemlibra solution for injection vials are for single-use only.

A syringe, a transfer needle (or vial adapter) and an injection needle are needed to withdraw Hemlibra solution from the vial and inject it subcutaneously (please see below recommended features).

A 1 mL syringe should be used for an injection up to 1 mL of Hemlibra solution, whereas a 2 to 3 mL syringe should be used for an injection greater than 1 mL and up to 2 mL.

Refer to the Hemlibra “Instructions for Use” for handling instructions when combining vials in a syringe. Different Hemlibra vial concentrations (30 mg/mL and 150 mg/mL) should not be used when combining vials to administer the prescribed dose.

1 mL syringe

Criteria: Transparent polypropylene or polycarbonate syringe with Luer-lock tip, graduation 0.01 mL.

When used together with a vial adapter, a low dead space plunger 1 mL syringe (transparent polypropylene or polycarbonate syringe with Luer-lock tip, graduation 0.01 mL) must be used.

2 to 3 mL syringe

Criteria: Transparent polypropylene or polycarbonate syringe with Luer-lock tip, graduation 0.1 mL.

When used together with a vial adapter, a low dead space plunger 3 mL syringe (transparent polypropylene or polycarbonate syringe with Luer-lock tip, graduation 0.1 mL) must be used.

Transfer needle or vial adapter

Criteria for transfer needle: Stainless steel with Luer lock connection, gauge 18 G, length 35mm (1½″), preferably semi-blunted tip.

Criteria for vial adapter: Polycarbonate with Luer-lock connection, sterile, fitting 15 mm vial neck outer diameter, single-use, latex-free and non-pyrogenic.

Injection needle

Criteria: Stainless steel with Luer-lock connection, gauge 26 G, length preferably 9 mm (3/8″) or maximally 13mm (½″), preferably including needle safety feature.

Please see section 4.2 and package leaflet (section 7 Instructions for Use), for additional information on administration.

Any unused medicinal product or waste material should be disposed of in accordance with local requirements.

7. Marketing authorisation holder

Roche Registration GmbH

Emil-Barell-Strasse 1

79639 Grenzach-Wyhlen

Germany

8. Marketing authorisation number(s)

EU/1/18/1271/001 (30 mg/1 ml)

EU/1/18/1271/002 (60 mg/0.4 ml)

EU/1/18/1271/003 (105 mg/0.7 ml)

EU/1/18/1271/004 (150 mg/1 ml)

9. Date of first authorisation/renewal of the authorisation

Date of first authorisation: 23 February 2018

10. Date of revision of the text

22 October 2018

Detailed information on this medicinal product is available on the website of the European Medicines Agency http://www.ema.europa.eu.