CLINICAL PHARMACOLOGY REVIEW

BLA

125276, supplement 22

Submission Date:

10/15/2010

Brand Name

ACTEMRA

Submission Type

Pediatric Efficacy Supplement

Generic Name

Tocilizumab (RO4877533 or myeloma receptor antibody),

recombinant humanized anti-human monoclonal antibody

OCP Reviewer

Partha Roy, Ph.D.

Team Leader (Acting)

Yun Xu, Ph.D.

Pharmacometrics Reviewer

Atul Bhattaram, Ph.D.

Pharmacometrics Team Leader

Yaning Wang, Ph.D.

OCP Division

Clinical Pharmacology 2 (DCP2)

OND Division

Pulmonary, Allergy and Rheumatology Products (DPARP)

Sponsor

Genentech
Single-use vial of concentrated solution of 20 mg/mL;
Intravenous infusion

Formulation; Strength(s);
Administration Route
Approved Indication

Indicated for Adult patients with moderately to severely

active rheumatoid arthritis who have had an inadequate
response to one or more TNF antagonist therapies.

Purpose of this Efficacy

Supplement

To provide data to support an indication for the treatment of

Systemic Juvenile Idiopathic Arthritis (sJIA) in patients 2
years and older
The recommended dose is 12 mg/kg for patients <30 kg and
8 mg/kg for patients 30 kg once every 2 weeks

Proposed Dosage Regimen for

the sJIA indication:

Table of Contents
1

EXECUTIVE SUMMARY

1.1

Recommendation

1.2

Phase IV Commitments

1.3

Summary of Clinical Pharmacology Findings

QUESTION-BASED REVIEW (QBR)

2.1
General Attributes
7
2.1.1
What are the highlights of the chemistry and physico-chemical properties of the drug substance, and the formulation of the
drug product?
7
2.1.2
What are the approved therapeutic indication, dosage and route of administration?
7
2.2
General Clinical Pharmacology
2.2.1
What are the clinical pharmacology and clinical trials used to support the proposed claims?
2.2.2
Were the active moieties in the plasma appropriately identified and measured to assess pharmacokinetic parameters?
2.2.3
What was the rationale for the dose and dosing regimen of TCZ in sJIA patients?
2.2.4
What are the PK characteristics of TCZ in sJIA pediatric patients, particularly after long-term treatment with Actemra
administered once every 2 weeks?
2.3

Intrinsic Factors

PRELIMINARY LABELING RECOMMENDATION

4.1

17
17
17

APPENDIX

18

Pharmacometric Review

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

9
14

Systemic Juvenile Idiopathic Arthritis

Immunogenicity

7
7
8
9

18

Executive Summary

1.1 Recommendation
From a Clinical Pharmacology perspective, the application is acceptable provided that the Sponsor and the
Agency come to a mutually satisfactory agreement regarding the language in the package insert.
1.2
None.

Phase IV Commitments

1.3 Summary of Clinical Pharmacology Findings

Tocilizumab (RO4877533, TCZ), also referred to as myeloma receptor antibody (MRA), is a recombinant
humanized anti-human monoclonal antibody of the immunoglobulin G1 (IgG1) sub-class directed against the
soluble and membrane-bound interleukin 6 receptor (IL-6R). Currently, TCZ (Actemra, 4 mg/kg with an
increase to 8 mg/kg based upon clinical response) is approved in the US for RA patients who have had an
inadequate response to TNF antagonist therapies. This is a pediatric supplement to add the indication of
systemic onset juvenile idiopathic arthritis (sJIA) to the current tocilizumab label for the treatment of active
sJIA in patients aged 2 years of age and older. TCZ is presently approved in Japan and India for sJIA and there
are no other approved medicinal products globally to treat both the systemic features and the arthritis associated
with sJIA.
The clinical data pivotal to the current application is derived from a three-part 5-year Phase III study
(WA18221). Specifically, the application includes: 1) WA18221-Part 1: 12-week efficacy and safety study of
TCZ (12 mg/kg dose for patients with body weight <30 kg and 8 mg/kg dose for patients with body weight 30
kg) in sJIA patients (n=112); 2) Cut of WA18221 data (cut-off date: 10th May 2010) when 50 patients reach one
year in Part II of WA18221; 3) additional supportive safety and efficacy data from Japanese sJIA trials
containing data for at least 56 patients through to one year.
The clinical pharmacology program included PK and PD data and PK-PD relationships from the following
studies: Roches pivotal study WA18221 and five supportive Chugai studies (LRO320, MRA011JP,
MRA316JP, MRA317JP, and MRA324JP). In addition, two population PK analyses were also submitted: 1)
development of alternative dosing regimen recommendations for pediatric patients with sJIA and low body
weight and 2) Verification of the alternative dosing regimen recommendation using data from study WA18221.
Dose Selection
The dosing regimen is primarily selected based on pharmacokinetic analysis of data from sJIA patients
(MRA316JP- Japanese Pediatric). It was observed that after 6 weeks of treatment with TCZ 8 mg/kg every 2
weeks, the proportion of patients who reached the ACR50 response was lower in patients < 30 kg (83%) than in
patients weighing 30 kg (100%). Similarly, the proportion of patients who reached ACR70 was also lower in
patients < 30 kg (63%) than in patients 30 kg (85%). This difference was explained by the visible trend
toward lower systemic exposure to TCZ in patients with lower BW. To compensate for that, the sponsor
contemplated a higher dose for the patients with lower BW (<30 kg). Therefore, a population PK modeling and
simulation was performed using pooled PK data from Chugai studies MRA316JP and LRO320, with the goal to
explore an alternative dosing regimen to achieve uniform exposure across the entire BW range. The post-hoc
estimates of systemic exposures of TCZ using final population PK model predicted that the 12 mg/kg dose in
BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

patients < 30 kg would yield similar systemic TCZ exposure as the patients 30 kg. With this knowledge
gained from the modeling and simulation exercise, the sponsor evaluated TCZ 8 mg/kg (patients 30 kg) and
12 mg/kg (patients < 30 kg) every 2 weeks in the pivotal study WA18221. Indeed, the dosing regimen, 8 mg/kg
for patients with BW 30 kg and 12 mg/kg for patients weighing < 30 kg, in study WA18221, resulted in
similar serum concentrations at sampling time points over time (Figure 1) and similar systemic exposure (Table
1) between the BW categories (< 30 kg and 30 kg). Additionally, 94.7% and 75.5% of patients achieved the
primary efficacy endpoint of JIA ACR30 response and absence of fever at Week 12 (ITT population) for 12
mg/kg (<30 kg bodyweight) and 8 mg/kg ( 30 kg bodyweight) dosing, respectively compared to only 24.3% in
the placebo group. (refer to Pharmacometric Review in Appendix).

Tociluzimab Concentrations (ug/mL)

Figure 1. Tocilizumab serum concentrations (Mean SD) in sJIA patients with BW <30 kg (12 mg/kg) and
30 kg (8 mg/kg) in study WA18221.
400
350
300
250
8mg/kg

200

12mg/kg

150
100
50
0
0

500

1000

1500

2000

2500

Time(hr)

Table 1. Summary of model-predicted TCZ PK measures at week 12 by dosing groups

Pharmacodynamic Findings: Inflammatory pharmacodynamic markers such as C-reactive protein (CRP),

Eryththrocyte sedimentation rate (ESR), Acute phase serum amyloid A (SAA) were monitored in study
BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

WA18221 while IL-6 and sIL-6R, directly linked to the mechanism of action of tocilizumab, were also
measured. Following administration of TCZ, a rapid decline in mean CRP, ESR, and SAA was observed in the
two dose groups (8 mg/kg and 12 mg/kg) with significant decreases seen by Week 2 and remaining suppressed
through 12 weeks of treatment with tocilizumab.
Following administration of tocilizumab, IL-6 levels initially increased rapidly and then generally decreased
with time, although mean concentrations did not reach baseline levels by Week 12 in either dose group. Mean
sIL-6R increased rapidly by Week 2 and continued to increase toward a plateau. The observed changes in IL-6
and sIL-6R were similar between dose groups. These changes were not observed in patients receiving placebo.
The changes of inflammatory markers (CRP, ESR, and SAA) and markers of the TCZ mechanism of action (IL6 and sIL-6R) were similar between two dose groups providing support for the proposed BW based dosing
regimen in this pediatric patient population. In addition, there was no appreciable relationship between TCZ
trough concentrations and PD markers tested as PD responses are generally complete at the range of steadystate clinical concentrations of TCZ in both dose groups.
Exposure-Response Evaluation
There was no clear trend towards higher systemic exposures (AUC2weeks, Cmin, and Cmax) in efficacy
responders compared to non-responders. There was also no clear difference in mean PK exposures across
responders for ACR30, 50, 70, and 90 responses. There were a similar proportion of patients (83-89%) within
each exposure quartile who achieved the primary endpoint, ACR30 response and absence of fever at week 12.
This indicates a lack of correlation between systemic exposure and efficacy response within the range of
exposure achieved in the study. However, this data should be interpreted with caution for couple of reasons: 1)
only a narrow exposure range was evaluated, and 2) there were a limited number of subjects in each systemic
exposure quartile. However, it can be reasonably concluded that the range of systemic exposure estimated in the
patient population following TCZ administration was sufficient to achieve the desired efficacy for the
indication.
Similar to efficacy evaluation above, the sponsor conducted exposure-response analysis with safety by
evaluating adverse events across systemic exposure (AUC2weeks) quartiles. There was no trend towards increased
incidence in Adverse Events (AEs) or Serious Adverse Events (SAEs) with increasing TCZ exposure, however,
it should be noted that there are only small number of subjects in each exposure quartile to draw definitive
conclusions.

Immunogenicity:
In study WA18221, patients were tested for anti-TCZ antibodies according to the standard testing paradigm that
started with an initial screening assay and if positive followed by the confirmatory assay. If patient samples
were positive for the confirmatory assay, the samples were further tested for neutralizing antibodies. All
patients (N = 112) were tested at baseline and Week 12 for anti-TCZ antibodies.
All patients with assay results were negative at baseline for both confirmative and neutralizing assays. Although
from a small database of TCZ treated patients (n=72), two patients (#1664 and #1005) with assay results were
positive for both confirmation assay as well as neutralizing assay at week 12. Both patients discontinued from
study treatment at or immediately after the week 8 infusion due to SAEs.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

In Part I of study WA18221, patient 1664 had positive anti-TCZ antibody for the confirmation and the
neutralizing assay. The model-predicted systemic exposure of this patient appeared to be lower than the mean
values for all patients. In Part II, patient 1005 at Weeks 12 and 20 had positive anti-TCZ antibody for
neutralizing assay. Since patient 1005 did not receive a full dose of 12 mg/kg TCZ at Weeks 4 and 6 due to
infusion like reaction, it is not clear whether the pre-dose TCZ concentrations measured as BLQ are due to the
formation anti-TCZ neutralizing antibody and/or insufficient doses received.
None of patients who missed consecutive infusions were positive for anti-TCZ antibodies after restarting
dosing. None of patients with JIA ACR50 response withdrew due to loss of efficacy. Four patients, 2 from the
TCZ 8 mg/kg group and 2 from the TCZ 12 mg/kg group, prematurely discontinued study treatment for lack of
efficacy. None of these patients had a positive anti-TCZ neutralizing assay.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

Question-Based Review (QBR)

2.1

2.1.1

General Attributes

What are the highlights of the chemistry and physico-chemical properties of the drug substance, and
the formulation of the drug product?

Chemistry and Physico-Chemical Properties: Tocilizumab (RO4877533) is a humanized anti-human IL-6

receptor (IL-6R) monoclonal antibody of the immunoglobulin (Ig) IgG1 subclass produced using recombinant
DNA technology.
The tocilizumab molecule is composed of two heterodimers. Each of the heterodimers is composed of a heavy
(H) and a light (L) polypeptide chain. The four polypeptide chains are linked intra- and inter-molecularly by
disulfide linkages. The Molecular formula for TCZ is C6428H9976N1720O2018S42 (polypeptide moiety only).
TCZ has a molecular weight of approximately 149 kDa

(b) (4)

Formulation: Tocilizumab is supplied as a sterile liquid concentrate for solution for intravenous (iv) infusion
available at a concentration of 20 mg/mL.
2.1.2

What are the approved therapeutic indication, dosage and route of administration?

Indication: ACTEMRA (tocilizumab) is indicated for the treatment of adult patients with moderately to
severely active rheumatoid arthritis who have had an inadequate response to one or more TNF antagonist
therapies.
ACTEMRA can be used as monotherapy or concomitantly with methotrexate or other DMARDs.
Dosage and Route of Administration: When used in combination with DMARDs or as monotherapy the
recommended starting dose is 4 mg/kg followed by an increase to 8 mg/kg based on clinical response.

2.2
2.2.1

General Clinical Pharmacology

What are the clinical pharmacology and clinical trials used to support the proposed claims?

This pediatric development program consists of one 3-part 5-year pivotal Phase 3 trial (WA18221) and
additional supportive Phase 2 and Phase 3 studies from Japanese development for the same indication, as
described below.
1. Clinical Study Report for WA18221-Part I consisting of a 12-week randomized, double blind,
placebo-controlled, parallel group, 2-arm study to evaluate the efficacy and safety of tocilizumab (12
mg/kg < 30 kg and 8 mg/kg 30 kg) in patients (n=112) with active sJIA;
2. Cut of WA18221 data (cut-off point: May 10, 2010) when 50 patients reach one year in Part II;

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

3. Additional supportive data (safety and dose-finding information) from Chugais (Japanese) long-term
sJIA studies (Table 2)
Table 2. Summary of supportive clinical studies for TCZ in sJIA

In addition, data from two population PK analyses are provided:

1. Development of alternative dosing regimen recommendations for pediatric patients with sJIA and low
body weight.
2. Population PK analysis of study WA18221: Verification of an alternative dosing regimen
recommendation for pediatric patients with sJIA and low body weight.
2.2.2

Were the active moieties in the plasma appropriately identified and measured to assess
pharmacokinetic parameters?

Yes. Concentrations of TCZ were determined in human serum samples with validated sandwich enzyme
immunoassay (EIA), which was the same assay used for determining plasma levels in the original submission.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

For complete review of the assay validation, refer to the Clinical Pharmacology Review by Dr. Lei Zhang dated
08/21/2008 for the original BLA submission (BLA 125276).

2.2.3

What was the rationale for the dose and dosing regimen of TCZ in sJIA patients?

The sponsor provided rationale for dosing based on efficacy response and systemic exposure data they collected
in the early studies. Initially, the dose was selected based on biomarker response (C-reactive protein) as well as
achieving and maintaining a minimum serum TCZ concentration. The rationale for dosing every 2 weeks came
from the observation that clearance of TCZ appeared higher in sJIA patients compared to adults where the
dosing is once every 4 weeks. The rationale for dosing 12 mg/kg in patients with body weight <30 kg and 8
mg/kg in patients with bodyweight 30 kg was based on achieving uniform exposure between the two
bodyweight groups. Following 8 mg/kg dose of TCZ, the lower bodyweight children reported lower clinical
response compared to children with bodyweights 30 kg. Detailed analysis and justification of dose and dosing
regimen selection can be found in the Pharmacometric Review by Dr. Atul Bhattaram attached with this review
(Appendix).
2.2.4

What are the PK characteristics of TCZ in sJIA pediatric patients, particularly after long-term
treatment with Actemra administered once every 2 weeks?

The mean ( SD) pre-dose concentration-time profile of TCZ by treatment group up to Week 52 of treatment in
study WA18221 is illustrated in Figure 2.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

Figure 2. Observed Mean (SD) pre-dose serum TCZ concentration time profile by treatment group (8 and 12
mg/kg) from baseline to week 52 (study WA18221)

Following TCZ administration, mean TCZ pre-dose concentrations trended upwards over time until Week 10
and stabilized after Week 12 in both treatments (Figure 2), indicating that steady-state has been reached
consistent with the half-life of 18-23 days. Pre-dose concentrations remained stable between Weeks 12 to 52
with observed serum TCZ concentrations of 69.527.7 and 73.826.9 g/mL, respectively. Comparing TCZ
serum concentrations between the two treatment groups, 8 mg/kg and 12 mg/kg, revealed similar serum
concentrations at all sampling points over time through Week 52 (Figure 2).
Besides collecting sparse PK samples from study WA18221, the sponsor submitted two population PK analyses
to support dose and dosing regimen and also to evaluate the influence of various covariates on the PK
parameters. The detailed review of the population PK modeling and analyses is captured in the Pharmacometric
Review by Dr. Atul Bhattaram (Appendix).
The posthoc estimated PK exposures by treatment group are summarized in Table 3. Mean computed PK
exposures (AUC2weeks, Cmax and Cmin) were similar between the two treatment groups (8 mg/kg for body weight
30 kg and 12 mg/kg for bodyweight <30 kg).

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

10

Table 3. Summary of predicted TCZ PK exposures at week 12 by treatment group (Study WA18221)

Using population PK analyses, the total clearance of TCZ was concentration-dependent and is the sum of the
linear clearance and the nonlinear clearance. The linear clearance estimated in population PK analysis was 0.17
L/day (or 7.1 mL/hour) in sJIA pediatric patients. In sJIA patients, the volume of distribution at steady state is
estimated to be 2.54 L. The half-life of TCZ is concentration dependent and is estimated to range from 18.4 to
22.7 days for 8 mg/kg (body weight 30 kg) and 19.2 to 23 days for 12 mg/kg (body weight < 30 kg) treatment
groups at week 12.
2.2.5. What pharmacodynamic markers were evaluated?
Following administration of TCZ, a rapid decline in mean markers of inflammation (CRP, ESR and SAA) was
observed across all studies. CRP concentration normalized or decreased dramatically after TCZ dosing and
remained within reference ranges through 12 weeks of treatment with TCZ (Figure 3 below). Similarly with
ESR and SAA, rapid decreases in mean levels were observed following administration of TCZ with declines
observed by 2 weeks and remaining suppressed through 12 weeks of treatment with TCZ. The changes of CRP,
ESR and SAA are comparable between the 8 mg/kg and 12 mg/kg treatment groups.
As shown in Figure 3, after TCZ administration, mechanistic markers such as IL-6 and sIL-6R increased rapidly
upon administration of TCZ with IL-6 subsequently declining gradually trending towards baseline while sIL-6R
continued to increase, approaching a plateau through week 12 when steady-state systemic exposure of TCZ is
attained.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

11

Figure 3. Relationship of TCZ PK and mean PD markers in all patients who received TCZ treatment in study
WA18221.

2.2.6. What was exposure-response relationship of tocilizumab in terms of efficacy and safety?
There was no clear trend toward higher PK exposures (AUC2weeks, Cmin, and Cmax) in responders compared to
non-responders. There was also no clear difference in mean PK exposures across responders for ACR30, 50, 70,
and 90 responses (Table 4). One of the difficulties in making a definite conclusion about this is the fact that
there were limited number of subjects in the non-responder group although one can argue that consistently,
exposure in responders, on average, are numerically greater (5-20%) than non-responders. Also important to
note is that there is a similar proportion of patients within each exposure quartile who achieved the primary
endpoint, ACR30 response and absence of fever at week 12 (Table 5). Again, there are only limited number of
patients in each quartile to draw a definitive conclusion. However, these data collectively indicate lack of
correlation between TCZ systemic exposure and efficacy endpoints, suggesting that even the lowest systemic
exposure within the range of exposures (Table 3 above) in the sJIA patient population was sufficient to achieve
the desired efficacy. Refer to the Pharmacometric Review for further analyses of the exposure-response
relationship for efficacy.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

12

Table 4. Summary of TCZ PK exposures at week 12 by ACR response status for all patients

Table 5. Summary of Percentage of Patients Achieving JIA ACR30 and Absence of Fever Response Status at Week 12 by
Exposure Quartiles

Note: Mean AUC2weeks (g.day/mL) for Q1: 849, Q2: 1178, Q3: 1445, Q4: 1925

Similar to efficacy evaluation above, the sponsor conducted exposure-response analysis with safety by
evaluating adverse events across systemic exposure (AUC2weeks) quartiles. There was no trend towards increased
incidence in Adverse Events (AEs) with increasing TCZ exposure (Table 6). Similar conclusion can be drawn
for serious adverse events (SAEs) where most SAEs occurred in patients in the first exposure quartile (Q1). It
should be noted that there are only small number of subjects in each exposure quartile to draw definitive
conclusions about exposure-safety relationship.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

13

Table 6. Summary of Percentage of patients reporting AEs by body system and preferred term to week 12 by
systemic exposure quartiles

2.3

Intrinsic Factors

2.3.1. Is there any effect of 1) age, 2) gender and 3) race on the pharmacokinetics of TCZ in the sJIA patient
population?
Nonlinear mixed effects modeling was used to analyze the serum TCZ concentration-time data collected over
12 weeks of treatment (see Appendix for the Pharmacometric Report). Systemic exposure (AUC2weeks, Cmax and
Cmin at Week 12), was estimated for all patients who had provided samples.
A summary of estimated systemic exposures (AUC2weeks, Cmin and Cmax) of TCZ at Week 12 by age
category is provided in Table 7. Mean PK exposures were similar among age categories (2-5 yrs, 6-12 yrs, and
13-18 yrs), indicating lack of age effect on the pharmacokinetics of TCZ across the entire pediatric age range 2
years and above.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

14

Table 7. Mean PK parameters at week 12 by age categories across the pediatric age range

Model-independent PK parameters in Japanese pediatric sJIA patients and Japanese adult RA patients are
compared in Table 8. BW-normalized clearance of TCZ was higher in sJIA patients (0.78 0.34 mL/kg in
Group 1 and 0.89 0.18 mL/kg in Group 2) than in adult RA patients (0.16 0.12 mL/kg) at 8 mg/kg.
Table 8. Mean PK Parameters of TCZ after Single Infusion of 8 mg/kg to sJIA pediatric vs. adult RA
patients

Population PK analyses confirmed the findings of greater clearance in pediatric sJIA patients. Following 8
mg/kg IV infusion (once every 2 weeks in children compared to once every 4 weeks in adults), the Cmin is 6-fold
higher in sJIA patients than in adult RA patients (54.5 g/mL vs 8.6 g/mL), and AUC2weeks in sJIA patients is
similar to AUC4weesk in adult RA patients (1337 g/mL vs 1417 gday/mL).
Estimated TCZ systemic exposures (AUC2weeks, Cmin and Cmax) at week 12 by gender is provided in Table 9.
Mean PK exposures were similar between males and females, indicating no effect of gender on
pharmacokinetics of TCZ.
Table 9. Mean PK parameters at week 12 by gender in the sJIA patient population

Apart from the PK data from study WA18221, early trials (LRO320 and MRA316JP) also reported PK
parameters based on non-compartmental analysis of rich PK data. PK parameters from WA18221 are not
compared directly to PK parameters from the earlier studies due to different treatment durations. However,
BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

15

Cmin at Week 18 in study MRA316JP (56.7 18.8 g/mL) is comparable to the Cmin at Week 12 in study
WA18221 (54.5 20.7 g/mL) following 8 mg/kg every 2 weeks dosing, indicating that PK is comparable
between Japanese and Caucasian populations. The population PK parameters from WA18221 (predominately
Caucasian) were also found to be similar to the PK parameters from MRA316JP (Japanese sJIA patients). These
observations indicate that PK is similar between Caucasian and Japanese populations, therefore no apparent
effect of race on TCZ PK in these populations of sJIA patients.

2.3.2. What were the immunogenicity findings for tocilizumab? What was the impact of immunogenicity on
exposure and/or safety and efficacy?
In study WA18221, patients were tested for anti-TCZ antibodies according to the standard testing paradigm that
started with an initial screening assay and if positive followed by the confirmatory assay. If patient samples
were positive for the confirmatory assay, the samples were further tested for neutralizing antibodies. All
patients (N = 112) were tested at baseline and Week 12 for anti-TCZ antibodies.
All patients with assay results were negative at baseline for both confirmative and neutralizing assays. Although
from a small database of TCZ treated patients (n=72), only two patients (#1664 and #1005) with assay results
were positive for both confirmation assay as well as neutralizing assay at week 12. Both patients discontinued
from study treatment at or immediately after the week 8 infusion due to SAEs. The last treatment of TCZ 12
mg/kg for patient# 1664 was 60% of the dose at Week 8 and then dropped out of the study due to the
occurrence of SAE of angioedema. The model predicted AUC2weeks and Cmin exposure of patient# 1664 at Week
12 (844 gday/mL and 31.6 g/mL, respectively) appeared to be lower than the mean values for all patients
(1341 gday/mL and 57.5 g/mL, respectively). The model predicted Cmax at Week 12 was 265 g/mL which
was comparable to the mean Cmax for all patients (245 g/mL).
Patient 1005 had reported positive results for confirmatory as well as neutralizing assay at Week 12 and Week
20. This patient was randomized at baseline to placebo treatment and then escaped to open-label TCZ 12 mg/kg
at Week 2. The patient received a complete dose at Week 2 without issue, but only received 12% of the dose at
Week 4 and 78% at Week 6 due to infusion-like reactions. After pre-medication and a slowed infusion rate, the
patient received a complete dose at Week 8. The patient was discontinued from treatment following Week 8 due
to macrophage activating syndrome. As patient #1005 was originally randomized to placebo, model-predicted
PK parameters were not estimated. This patients observed serum TCZ levels were BLQ at all sampling points
taken (excluding one post dose sample at Week 2 immediately after completion of the infusion). Since patient
1005 did not receive a full dose of 12 mg/kg at Weeks 4 and 6, it is not clear whether the pre-dose
TCZ concentration of below the limit of detection is due to the formation of anti-TCZ antibody and/or
insufficient doses received.
None of patients who missed consecutive infusions were positive for anti-TCZ antibodies after restarting
dosing. None of patients with JIA ACR50 response withdrew due to loss of efficacy. Four patients, 2 from the
TCZ 8 mg/kg group and 2 from the TCZ 12 mg/kg group, prematurely discontinued study treatment for lack of
efficacy. None of these patients had a positive anti-TCZ neutralizing assay.

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

16

Preliminary Labeling Recommendation

1. Under section 6.1 Clinical Trials Experience
(b) (4)

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

17

Appendix
4.1

Pharmacometric Review

APPEARS THIS WAY ON

ORIGINAL

BLA 125276/22
ACTEMRA (Tocilizumab)
Clin Pharm Review

18

Office of Clinical Pharmacology:

Pharmacometric review
1

Summary of Findings

1.1 Key Review Questions

The purpose of this review is to address the following key questions:
1.1.1

Did the sponsor provide rationale for the dose and dosing regimen of tocilizumab (TCZ) in patients with
sJIA?

Yes, the sponsor conducted studies that provided rationale for choice of dose and dosing regimen
of TCZ in patients with sJIA.
Table 1 lists the summary of early clinical studies for TCZ in sJIA. These studies provided the
basis for proposed dose and dosing regimen of TCZ in the registration trial (WA18221).
Table 1. Summary of supportive clinical studies for TCZ in sJIA

Sponsors Rationale for 8 mg/kg dose every 2 weeks

Patients in MRA011JP received three doses of 2 mg/kg TCZ and the dose was escalated
to 4 mg/kg and 8 mg/kg as required to normalize CRP. When the serum TCZ
concentration was at or above the limit of quantification (1 g/mL) in this study, CRP

was virtually normal. Based on results from dose escalation during the course of the
study to maintain TCZ at measurable levels, the TCZ dose needed to be increased to 8
mg/kg in the patients in whom an adequate serum TCZ concentration could not be
maintained at 4 mg/kg. The 8 mg/kg dose was therefore proposed for the Phase III study
MRA316JP in Japan.
Sponsors Rationale for every 2 weeks dosing interval
Following single dose infusions of TCZ to sJIA patients in Study LR0320, PK and PD results
showed that TCZ clearance appeared higher in sJIA patients than in adult RA patients, and that
serum IL-6 concentrations were also higher before and after dosing in these children than in
adult RA patients. The PD response (IL-6) appeared to decline over a 4-week period following
the infusion of TCZ and TCZ was undetectable in plasma by 2 weeks after dosing in the majority
of patients. Thus, the TCZ dose frequency of every 2 weeks was appropriate for future studies.
Sponsors Rationale for 8 mg/kg dose every 2 weeks in patients with body weight 30 kg and 12
mg/kg every 2 weeks in patients with body weight <30 kg
A Phase III study (MRA316JP) assessing the PK, efficacy, and safety of TCZ
with 8 mg/kg every 2 weeks was conducted in Japanese sJIA patients. It was observed
that the clinical response was lower in children with a low body weight (BW), compared to
patients with a higher BW. After 6 weeks of treatment with TCZ 8 mg/kg every 2 weeks, the
proportion of patients who reached the ACR50 response was lower in patients < 30 kg (83%)
than in patients weighing 30 kg (100%). Similarly, the proportion of patients who reached
ACR70 was also lower in patients < 30 kg (63%) than in patients 30 kg (85%). This difference
was explained by the visible trend toward lower systemic
exposure to TCZ in patients with lower BW (Figure 1).

Figure 1. Dosing with 8mg/kg results in low exposure in pediatric patients with low
bodyweight

Therefore, for the registration trial (WA18221), population PK modeling and simulation was
performed using pooled PK data from Chugai studies MRA316JP and LRO320, with the goal to
explore an alternative dosing regimen to achieve uniform exposure across a wide BW range. The
post-hoc estimates of systemic exposures of TCZ using final population PK model, clearly
showed that the 12 mg/kg dose in patients < 30 kg would gain similar exposures as the patients
30 kg.

Table 2 shows the observed TCZ exposure with studied dose of 8 mg/kg and predicted TCZ
exposure with the proposed dose of 12 mg/kg for study MRA316JP.
Table 2. TCZ observed exposure with studied dose of 8 mg/kg and predicted
exposure with the proposed dose of 12 mg/kg for study MRA316JP

Source: Table 39 from page 93 in summary-clin-pharm.pdf

Is the proposed dosing regimen for tocilizumab (TCZ) based on body weight cut off of 30 kg acceptable?

Yes, the proposed dosing regimen for TCZ as shown in Table 3, based on body weight cut off of
30 kg, is acceptable if there are no major safety issues.
Table 3. Proposed dosing regimen for TCZ in sJIA
Recommended sJIA Dosage Every 2 Weeks
Patients < 30 kg
12 mg/kg
Patients 30 kg
8 mg/kg
Source: Proposed drug label

Tociluzimab Concentrations (ug/mL)

Figure 2 shows that the TCZ concentrations are similar in patients with body weight <30 kg
(treated with 12 mg/kg every 2 weeks) and 30 kg (treated with 8 mg/kg every 2 weeks) in
WA18221.
Figure 2. TCZ concentrations (meanstandard deviation) in patients with body
weight <30 kg (treated with 12 mg/kg every 2 weeks) and 30 kg (treated with 8
mg/kg every 2 weeks) in WA18221
400
350
300
250
8mg/kg

200

12mg/kg

150
100
50
0
0

500

1000

1500

2000

2500

Time(hr)

Source: Data from Page 697-704 from wa18221.pdf report.

The predicted Cmax, Cmin, AUC2weeks for 8 mg/kg and 12 mg/kg dose groups are shown in Table
4.
Table 4. Summary of predicted TCZ PK exposures at week 12 by treatment group

Source: Table 4 on Page 27 of summary-clin-pharm.pdf

Table 5 shows the percentage of patients with a JIA ACR30 response and absence of fever at
Week 12 (ITT Population) in Study WA18221.

Table 5. Percentage of Patients with a JIA ACR30 Response and Absence of Fever
(Primary endpoint) at Week 12 (ITT Population)
Placebo
Body weight <30 kg
Body Weight 30 kg
(treated with 12 mg/kg) (treated with 8
(N=37)
(N=38)
mg/kg)
(N=37)
Percentage of
24.3
94.7
75.5
Patients with a JIA
ACR30 response and
absence of fever at
Week 12 (ITT
population)
Source: Table 12 on Page 129 of WA18221.pdf
Figure 3 shows the relationship between TCZ AUC2weeks and percentage of responders, nonresponders based on JIA ACR30 Response,LOCF at week 12 in study WA18221.
Figure 3. Relationship between TCZ (tocilizumab) AUC2weeks and percentage of
responders (+), non-responders () based on week 12 JIA ACR30 Response, LOCF.
Shown also are the number of responders, non-responders in each AUC quartile

..\ER Analyses\ERAnalysis.sas

Table shows that Cmax, Cmin, AUC2weeks of TCZ are not different in responders and nonresponders based on ACR response status.
Table 6. Summary of TCZ PK exposures at week 12 by ACR response status for all
patients

Source: Table 13 on Page 45 of summary-clin-pharm.pdf

Table 7 shows that the the percentage of patients achieving JIA ACR30 and absence of fever
response status at Week 12 are similar across exposure quartiles.
Table 7. Summary of percentage of patients achieving JIA ACR30 and absence of
fever response status at Week 12 by exposure quartiles

Quartiles are defined as those patients falling within 0 - 25%, >25 - 50%, >50 - 75%,
>75-100% of exposures.
Source: Table 5 on Page 27 of summary-clin-pharm.pdf
Table 8 shows the summary of adverse events (AEs) with an incidence rate of 5% to Week 12
by Preferred Term and Trial Treatment (Safety Population).

Table 8. Summary of AEs with an Incidence Rate of 5% to Week 12 by Preferred

Term and Trial Treatment (Safety Population)

Source: Table 46 on Page 194 of wa18221.pdf

1.2 Recommendations
None
1.3 Label Statements
Labeling statements to be removed are shown in red strikethrough font and suggested labeling to
be included is shown in underline blue font.
Sponsors Proposal
Section 12.3 Pharmacokinetics
(b) (4)

(b) (4)

2 Pertinent regulatory background

Tocilizumab (RO4877533, TCZ), also referred to as myeloma receptor antibody (MRA),
is a recombinant humanized anti-human monoclonal antibody of the immunoglobulin G1
(IgG1) sub-class directed against the soluble and membrane-bound interleukin 6 receptor
(sIL-6R and mIL-6R). TCZ (Actemra, 4 mg/kg with an increase to 8 mg/kg based upon
clinical response) is approved in the US for RA patients who have had an inadequate response to
anti-TNFs. sJIA is a subset of juvenile idiopathic arthritis (JIA) that is characterized by the
presence of arthritis, intermittent fever, and rash and comprises between 4% and 17% of all cases
of JIA. Currently, there are no products approved for treatment of sJIA.
In this application, sponsor is seeking approval of TCZ for sJIA based on data from Part-I of
study WA18221 which is a 12-week randomized, double blind, placebo-controlled, parallel
group, 2-arm study to evaluate the efficacy and safety of TCZ (12 mg/kg in patients with body
weight< 30 kg and 8 mg/kg in patients with body weight 30 kg) in patients (n=112) with active
sJIA. The dosing regimen is selected based on pharmacokinetic analysis of data from sJIA
patients (MRA316JP).
Sponsor collected samples for characterizing TCZ pharmacokinetics in Study WA18221 as
shown below:
one sample at randomization (baseline), before the intake of the first dose of the
study medication,
one pre-dose sample (trough drug levels) on study days 8, 15, 29, 43, 57, 71, 78,
and 85,
One post-dose sample (within 15 minutes of completion of the infusion) on study
days 1, 15, 29 and 71.
3 Results of Sponsors Analysis
Sponsor analyzed TCZ concentration data obtained from Study WA18221 using the
pharmacokinetic model as shown in Figure 4. The model was developed based on data from
LR0320 and MRA316JP studies.

Figure 4. Pediatric structural pharmacokinetic model for TCZ with combined 1storder and Michaelis-Menten elimination.

Source: Figure 43 on Page 100 of summary-clin-pharm.pdf

Table 9 shows the estimates of pharmacokinetic parameters derived based on the model as
shown in Figure 4.

Table 9. Summary of population PK parameters from MRA316 & LRO320.

Source: Table 41 on Page 101 of summary-clin-pharm.pdf

Using POSTHOC option in NONMEM, sponsor fitted the time course of TCZ concentrations
from study WA18221. Figure 5 shows the observed, population and individual predicted
tocilizumab concentration data in representative patients.

Figure 5. Superimposition of observed tocilizumab serum concentrations (open

circles), population prediction (dotted line), and individual predicted tocilizumab
serum concentrations (solid line) by subject (patient ID and the average dose are
indicated on each graph)

Source: Figure 12 on Page 161 of summary-clin-pharm.pdf

Reviewers Comments: The population pharmacokinetic analysis conducted by the sponsor is
acceptable. Figure 2 shows that the mean tocilizumab concentrations are similar between
patients (< 30kg) treated with 12 mg/kg and patients (30 kg) treated with 8 mg/kg. Similar
findings are reported based on the population pharmacokinetic analysis as shown in Table 4.
4

Reviewers Analysis

The reviewer was able to reproduce the sponsors results with NONMEM. No additional analysis
was conducted. The following datasets were used in making Figure 3.
Table 10. Analysis Data Sets
Study
Number
wa18221

Name

Link to EDR

DEMPK.XPT

wa18221

Nonmem.xpt

\\cberfs3\m\eCTD Submissions\STN125276\0025\m5\datasets\wa18221\analysis
\\cber-fs3\M\eCTD Submissions\STN125276\0025\m5\datasets\wa18221pk\analysis

Listing of Analyses Codes and Output Files

File Name
Run1 mod

Description
NONMEM control stream

Run1.out

NONMEM output file

Location in \\cdsnas\pharmacometrics\
Z:\Reviews\Ongoing PM
Reviews\Tocilizumab_BLA125276_VAB\PPK
Analyses\Model\Run1\
Z:\Reviews\Ongoing PM
Reviews\Tocilizumab_BLA125276_VAB\PPK
Analyses\Model\Run1

Signatures
Partha Roy
Clinical Pharmacology Reviewer________________________________________

Atul Bhattaram
Pharmacometrics Reviewer_____________________________________________

Yaning Wang
Pharmacometrics TL__________________________________________________

Yun Xu
Clinical Pharmacology Acting TL________________________________________