Department of Health & Human Services Public Health Service

Food and Drug Administration

Center for Biologics Evaluation and Research

1401 Rockville Pike

Rockville, MD 20852

 

 

Division of Clinical Trial Design and Analysis

 

 

Date: April 18, 2003

 

 

 

BRIEFING DOCUMENT ON SAFETY

BLA STN 103976/0

 

Genentech, Inc.

 

 

Omalizumab (Xolair)

 

(recombinant humanized monoclonal antibody to IgE)

 

for treatment of allergic asthma

 

 

 

 

 




Table of Contents

Heading

 

Page

1. Introduction

 

4

A. Regulatory history

 

4

B. Materials reviewed

 

4

2. Proposed Indication, dose and dose regimen

 

4

3. Product background (chemistry, manufacturing, controls)

 

6

4. Clinical background

 

6

5. Summary of major safety observations

 

8

A. Overview

 

8

B. Serious adverse events

 

10

1. Malignancy

 

10

2. Anaphylaxis

 

11

C. Adverse events

 

12

1. Adverse events among all subjects in controlled studies

 

12

2. Adverse events among the geriatric population in controlled studies

 

14

D. Other observations

 

14

1. Clinical laboratory

 

14

2. Antibody formation

 

15

6. Preclinical studies

 

15

7. Clinical studies overview

 

16

A. Exploratory Studies

 

16

B. Major Studies

 

17

8. Safety database composition

 

19

A. Sample size

 

19

B. Baseline characteristics and drug exposure

 

20

C. In-study characteristics

 

22

D. Subject disposition

 

23

1. Controlled studies

 

23

2. Uncontrolled studies

 

25

9. Serious adverse events

 

26

A. Deaths

 

26

B. Nonfatal serious adverse events

 

26

C. Neoplasia (malignant and benign)

 

29

1. Malignancies

 

29

2. Benign neoplasms

 

33

10. Adverse events

 

33

A. Adverse events by body system and preferred term

 

33

B. Adverse events by severity

 

35

C. Digestive, respiratory and female genitourinary AE

 

37

1. Digestive system adverse events, including appendicitis

 

37

2. Respiratory system adverse events

 

39

3. Female genitourinary adverse events

 

39

D. Hypersensitivity adverse events

 

40

1. Anaphylaxis

 

40

2. Skin rash, including urticaria

 

41

3. Urticaris with bronchospasm

 

42

4. Serum sickness

 

42

E. Other immune-type adverse events

 

43

F. Bleeding-related AE and hemoglobin data

 

43

G. Injection site reactions/adverse events

 

46

11. Laboratory results

 

48

A. Hematology

 

48

1. Hemoglobin and leukocytes

 

48

2. Platelets

 

48

B. Blood chemistry

 

53

1. Liver tests

 

53

2. Renal tests

 

53

12. Antibody formation

 

54

13. Retreatment data

 

54

14. Pregnancy

 

55

15. Drug-level effects and adverse events

 

55

16. Adverse events by drug exposure duration

 

56

17. Adverse events by demographic subsets / in-study characteristics

 

57

18. Summary

 

58

19. Appendix A: AE tables cited in text

 

62

20. Appendix B: Narratives for subjects with anaphylaxis/anaphylactoid AE

 

69

21. Appendix C: Summaries of clinical study designs

 

70

22. Appendix D: Narratives for subjects with appendicitis

 

83

23. Appendix E: Narratives for subjects with malignancy

 

85

24. Appendix F: Laboratory data tables cited in text

 

93

25. Appendix G: Narratives for subjects with other AE

 

96

 

Abbreviations

Acronym

Definition

AA

Allergic asthma

AD

Allergic dermatitis

AE

Adverse event

BLA

Biological license application

CV

Cardiovascular

DB

Double blind

ER

Emergency room

GU

Genito-urinary

ICS

Inhaled corticosteroids

IV

Intravenous

LLN

Lower limit of normal

LTR

Leukotriene

MDI

Metered dose inhaler

MV

Mechanical ventilation

OCS

Oral corticosteroids

PAR

Perennial allergic rhinitis

PC

Placebo controlled

PI

Package insert

PK

Pharmacokinetic

SAR

Seasonal allergic rhinitis

SAE

Serious adverse experience

SC

Subcutaneous

SEER

Surveillance, Epidemiology and End Results database

SIR

Standardized incidence ratio

STC

Standard therapy controlled

ULN

Upper limit of normal

 


1. Introduction:

 

This document is a summary of the safety information contained within the Biological License Application (BLA) submitted to FDA by Genentech, Inc for Omalizumab, a recombinant humanized monoclonal antibody proposed for treatment of allergic asthma.

 

Omalizumab binds to human IgE at the same epitope as that of the high affinity IgE receptor (FceRI) on mast cells and basophils. This IgE binding blocks IgE from binding to mast cells and basophils. Once bound to Omalizumab, IgE is proposed to be cleared from the body by macrophage endocytotic clearance. Although the proposed indication is for the treatment of allergic asthma, the sponsor has performed a series of clinical studies that examine the effect of Omalizumab upon patients with allergic asthma (AA), seasonal allergic rhinitis (SAR), perennial allergic rhinitis (PAR) and allergic dermatitis (AD). This review encompasses the entire safety database, however the AA safety findings form a special focus of this review.

 

A. Regulatory history:

 

The major milestones in the regulatory history of this BLA are summarized in Table 1.

 

Table 1. Regulatory history

Date

Action

June 2, 2000

Original BLA submission

July 5, 2001

FDA issued Complete Review Letter

December 18, 2002

Sponsor submitted response to Complete Review Letter

 

Within the original BLA submission, the sponsor had sought licensure for use of Omalizumab in the prophylaxis and treatment of asthma and SAR. The July 5, 2001 Complete Review Letter from the FDA highlighted a number of limitations within the original submission including the limited size of the clinical safety database and the inability to meaningfully assess certain safety signals. The letter noted that substantially greater safety information was necessary in order to form a judgement of the risks and benefits related to the proposed asthma indication and that even greater amounts of clinical safety information were necessary for the proposed SAR indication. In response to this letter, the sponsor filed a December 18, 2002 BLA amendment (Complete Response to Complete Review Letter) that included clinical data from approximately three-fold more subjects exposed to Omalizumab than were originally submitted in June, 2000. The December 18, 2002 Complete Response also limited the proposed indication to AA.

 

B. Materials reviewed:

 

This review is focused upon the safety data contained in the June 2, 2002 and December 18, 2002 license application submissions as well as the safety data submitted to the pertinent investigational new drug applications.

 

2. Proposed indication, dose and dose regimen:

 

The information listed below are quotes from the proposed package insert (PI).

 

Proposed Indication:

 

"XOLAIR is indicated as maintenance therapy for the prophylaxis of asthma exacerbations and control of symptoms in adults and adolescents (12 years and above) with moderate to severe allergic asthma that is inadequately controlled despite the use of inhaled corticosteroids."

 

Proposed regimen:

 

"Xolair 150 to 375 mg is administered SC every 2 or 4 weeks. Doses (mg) and dosing frequency are determined by baseline serum total IgE level (IU/mL), measured before the start of treatment, and body weight (kg)." Table 2 duplicates the package insert's dose determination charts.

 

Table 2. Dose determination charts from package insert

 

 

ADMINISTRATION EVERY 4 WEEKS

 

XOLAIR Doses (milligrams) Administered by Subcutaneous Injection

Every 4 Weeks for Adults and Adolescents (12 Years of Age and Older)

 

 

with Allergic Asthma

 

 

 

 

 

 

 

 

Baseline IgE

(IU/mL)

Body Weight (kg)

30 - 60

> 60 - 70

> 70 - 80

> 80 - 90

> 90 - 150

30- 100

150

150

150

150

300

> 100- 200

300

300

300

300

 

> 200- 300

300

 

 

 

 

> 300- 400

 

ADMINISTRATION EVERY 2 WEEKS

> 400- 500

 

 

(SEE TABLE BELOW)

 

> 500- 600

 

 

 

 

 

 

 

ADMINISTRATION EVERY 2 WEEKS

 

XOLAIR Doses (milligrams) Administered by Subcutaneous Injection

Every 2 Weeks for Adults and Adolescents (12 Years of Age and Older)

 

 

with Allergic Asthma

 

 

 

 

 

 

 

Baseline IgE

(IU/mL)

Body Weight (kg)

30 - 60

> 60 - 70

> 70 - 80

> 80 - 90

> 90 - 150

30- 100

ADMINISTRATION EVERY 4 WEEKS

 

> 100- 200

 

(SEE TABLE ABOVE)

 

225

> 200- 300

 

225

225

225

300

> 300- 400

225

225

300

300

 

> 400- 500

300

300

375

375

 

> 500- 600

300

375

 

DO NOT DOSE

 

> 600- 700

375

 

 

 

 

Once reconstituted, a vial of Omalizumab contains a 150 mg dose within 1.2 mL (125 mg/mL). For the maximum monthly dose of 750 mg, three injections (a total of 3.0 mL) must be given every two weeks.

 

Comment: Serum IgE normally accounts for a very small proportion of total serum immunoglobulin concentration. Within one large population[1], the average serum IgE concentration was estimated to be 32 IU/mL. The upper limit of a "normal" serum IgE concentration may reach 90 IU/mL[2]. Most of the sponsor's multi-dose studies enrolled only subjects with baseline serum IgE concentrations 30 IU/mL but 700 IU/mL (Studies 006, 007, 008, 009, 010, IA04, 012, 014 and 011). However, some studies allowed enrolled subjects with baseline IgE 30 IU/mL but 1300 IU/mL (Studies Q2143g, D01 and 013). In all the major multi-dose studies, the Omalizumab doses did not exceed 750 mg/month. As will be noted, one of the sponsor's exploratory multi-dose studies (Study Q0694g), examined a higher Omalizumab dose than that proposed for marketing.

 

3. Product background (chemistry, manufacturing, controls):

 

Omalizumab is produced by a stably transfected Chinese hamster ovary (CHO) cell line. The final drug product is a lyophilized formulation that, upon reconstitution with Sterile Water for Injection (SWFI), delivers a 150 mg dose in 1.2 mL for subcutaneous administration.

 

Formulation changes during the clinical development program included a shift from a liquid solution to a lyophilized formulation, changes in the reconstitution directions for the lyophilized formulation and changes in the manufacturing following the completion of certain major clinical studies. Early versions of Omalizumab were examined in pilot studies that examined intravenous (IV), subcutaneous (SC) and aerosolization administrations and culminated in the development of a lyophilized product for SC administration. Certain changes in manufacturing occurred during terminal product development. Biochemical comparability data and clinical pharmcokinetic/pharmacodynamic were used to support the findings that the definitive premarketing clinical studies used an Omalizumab product comparable to that proposed for marketing.

 

4. Clinical background:

The sponsor has performed clinical studies that primarily examine the use of Omalizumab in two major clinical settings: AA and SAR. More limited clinical studies examined the use of Omalizumab in PAR and AD. These various manifestations of allergic disease are generally thought to share certain common pathophysiological correlates. The most notable of these correlates is the role of IgE. Indeed, the adjective, "allergic," is a qualifier that has been used synonymously with "IgE-mediated" disease by many clinicians.[3] The term "atopy" is also commonly used to describe IgE-mediated diseases, diseases such as AA, SAR, and AD. The presence of "atopy" is (generally) clinically characterized by the presence of an immediate skin reaction response, an IgE-mediated response, to a SC injection of a suspected allergen. Consequently, certain "allergic" diseases may be loosely designated as "IgE-mediated" diseases.

 

IgE-mediated diseases are thought to have a strong hereditary component and studies have shown that certain genetic mutations are detected more commonly among patients with IgE-mediated diseases than among the general population. However, efforts to identify the genes responsible for abnormal or excessive IgE production is an intense area of clinical investigation.

 

Allergic reactions result, in part, from the release of preformed granule-associated mediators from mast cells or basophils. This release is prompted by the binding of an allergen to the IgE already coating these cells. Omalizumab is thought to act by binding to certain epitopes on circulating IgE and preventing this IgE from binding to the mast cell and basophil. In effect, Omalizumab is proposed to act as a "sump" for the systemic burden of circulating IgE.

 

The symptoms and manifestations of IgE-mediated diseases overlap with many non-IgE-mediated disease. Consequently, it is difficult or impossible to distinguish some "allergic" diseases from "non-allergic" diseases based on clinical examination findings alone. This difficulty is especially notable for rhinitis and asthma--very common diseases/syndromes. In addition, the difficulty of distinguishing "allergic" from "non-allergic" disease has resulted in imprecise estimates of the prevalence of these two categories of diseases. The complexity of making "allergic" diagnoses is compounded by the finding that up to 22% of "normal" humans may have positive skin test responses to allergens--skin test reactions that suggest the individuals have preformed (and conceivably, pathologic) IgE formation.

 

Comment: Given the imprecision of diagnosis of "allergic" diseases, identification of the pertinent patient population for use of Omalizumab may hinge, in part, upon the criteria and definitions used for "allergic" asthma. Notably, all the sponsor's major safety and efficacy studies examined subjects with skin test reactivity to certain aeroallergens. The proposed PI does not describe the criteria for diagnosis of "allergic" asthma and does not directly refer to skin testing for hypersensitivity to aeroallergens.

 

The National Heart, Lung, and Blood Institute (NHLBI) defines asthma as "a chronic inflammatory disorder of the airways in which many cells and cellular elements play a role, in particular, mast cells, eosinophils, T lymphocytes, macrophages, neutrophils and epithelial cells. In susceptible individuals, this inflammation causes recurrent episodes of wheezing, breathlessness, chest tightness and coughing, particularly at night or in the early morning. These episodes are usually associated with widespread but variable airflow obstruction that is often reversible either spontaneously or with treatment. The inflammation also causes an associated increase in the existing bronchial hyperresponsiveness to a variety of stimuli."[4]

 

The NHLBI definition notes that there are multiple potential cellular and soluble mediator components to the inflammatory process associated with asthma. This report also notes that the specific role and impact of each of these various components upon the clinical manifestations of asthma remain hypothetical. Within clinical publications and textbooks, asthma has been subdivided into several loosely defined categories. These various categories or "types" of asthma include the following: allergic asthma, non-allergic asthma, exercise-induced asthma, status asthmaticus, refractory asthma, glucocorticoid-dependent asthma, nocturnal asthma and a host of other less common forms of asthma that are identified by various adjectives and eponyms. There are no well accepted epidemiologic data assessing the prevalence of each of the various "types" of asthma.

 

No precise definition of AA appears to exist even though the general clinical field has long recognized that asthma patients may be broadly grouped into categories of "allergic-type" versus "non-allergic-type." For practical and investigational purposes, one of the most common AA definitions employs the requirement of skin test positivity to a suspected aeroallergen along with a diagnosis of asthma. The subjectivity involved in making the diagnosis of AA is illustrated by the finding that skin test positivity to a suspected allergen is common (approximately 20% of the general population) while the incidence of asthma is considerably smaller (approximately 5% of the general population)--an observation that suggests factors other than atopy are involved in causing asthma.[5] Estimates for the prevalence of "asthma" (without qualifying adjectives) note that approximately 14 to 18 million Americans have the condition.

 

Asthma generally has two peaks of onset during human life, in childhood and after age 40. It is generally recognized that asthma in children commonly resolves with time and that asthma developing in older adults is frequently associated with complications.

 

Overall, death from asthma is very uncommon (estimated at less than 1% of all asthmatic patients dying annually). Most deaths from asthma occur among adults even though deaths among children with asthma may receive a larger amount of public attention. Epidemiological studies performed among residents of Rochester, Minnesota and Philadelphia, Pennsylvania have shown that the overwhelming majority of deaths due to asthma occur over the age of 50 years.[6] This observation is especially pertinent because studies have suggested that the patient population most vulnerable to death from asthma (patients with "refractory" asthma) may have airway inflammatory processes that are uniquely different from those inflammatory processes active in milder forms of asthma (i.e., the role of IgE-mediated processes may be different between patients with milder forms of asthma and those with refractory asthma).[7]

 

Important aspects of Omalizumab action are the potential consequences of inducing a form of IgE-deficiency. Following the administration of Omalizumab, the blood content of free IgE is markedly reduced. Chronic dosing with Omalizumab has the potential for maintaining, on a long term basis, an almost complete lack of any free IgE within the blood. While IgE has generally been implicated as a pro-inflammatory mediator with little, if any positive impact upon health, certain clinical observations have raised questions about a protective role of the immunoglobulin in mucosal defenses and in the defense against neoplasia.

 

One especially notable clinical report is that of a family with an inherited (isolated) deficiency of IgE. This deficiency was associated with severe sinopulmonary disease (focal emphysema and bronchitis).[8] The report's authors hypothesized that IgE may function in a protective role by inducing a mucosal "anaphylaxis" response to invading microorganisms and other antigens. This mucosal anaphylaxis might produce increased vascular permeability and result in the localized release of protective cell scavengers and soluble mediators.

 

Another concern with IgE physiology is its role in neoplasia survey. A number of epidemiological studies have described inverse correlations between the incidence of atopy and the development of neoplasms.[9],[10],[11] At least one study has suggested that patients with atopy may have more favorable prognoses in association with certain neoplasms.[12] Overall, these publications are far from definitive in establishing a protective role for IgE in the defense against neoplasia. However, the publications suggest that modification of the body's content of IgE may, in some manner, impact the immunity to cancer.

 

Comment: The clinical literature associated with the protective role of IgE is extremely limited. To a certain extent this may be because deficiencies of IgE are generally found in association with other immunoglobulin deficiencies. Nevertheless, the isolated reports raise questions about the potential for long term administration of Omalizumab to alter the susceptibility to diseases related to altered mucosal immunity and neoplasia.

 

5. Summary of major safety issues:

 

A. Overview:

 

The safety database consists of detailed information from 3,507 subjects exposed to Omalizumab within the major multi-dose AA, SAR, PAR and AD studies (see Appendix C for a summary of the design of these studies). Approximately 70% of the subjects (2,359) were enrolled in AA studies and approximately 60% (2,076) were enrolled in controlled AA studies. The AA studies used the Omalizumab dosages proposed for marketing and were, with the exception of one small study, of at least six months duration. SAR and PAR studies contribute safety data for 1,132 subjects and the one AD study provides safety data for 16 subjects. In general, the SAR, PAR and AD studies were controlled studies of short durations that, in some cases, examined Omalizumab dosages lower than those proposed for use in AA.

 

Consequently, the group of "all controlled clinical studies" is a combination of the AA data with the SAR, PAR and AD data, a combination that may not be fully informative with respect to the proposed AA indication because of the SAR, PAR and AD study design limitations. The group of AA studies, especially the controlled AA studies, provide the most directly applicable safety data for the AA indication. Throughout this review, the safety data from controlled studies are divided into two major groups: "AA controlled studies" and "all controlled studies."

 

All uncontrolled studies examined Omalizumab use in AA and these studies contribute 283 subjects to the safety database.

 

Overall, safety information from controlled and uncontrolled studies is generally limited to no more than one year of total Omalizumab exposure for any single subject, although some subjects in on-going studies supply preliminary data following three years of Omalizumab exposure.

The clinical development program for Omalizumab use in AA collected substantial safety data from unblinded studies that used control subjects receiving standard therapy. Only 20% of the Omalizumab-exposed adult/adolescent AA subjects had their safety data collected within one of the three placebo controlled studies designed to assess safety and efficacy (Studies 008, 009 and 011). The two unblinded, standard therapy controlled studies (Studies Q2143g and IA04) enrolled a total of 2,211 subjects. Hence, a large portion of the safety data were obtained from studies in which the investigators were aware of subjects' treatment assignment, a design feature that might have influenced certain aspects of safety reporting, especially investigators' assessments of AE causal associations with Omalizumab.

 

The safety database consists of a predominance of Caucasian subjects (85%) and subjects aged between 18 and 64 years (75%). Adolescents and geriatric subjects account for only nine and four percent of the safety database, respectively.

 

The most noteworthy safety findings are presented in summary immediately below and a comprehensive safety review follows. The noteworthy findings are:

 

 

The noteworthy findings are derived from the comparisons of adverse event rates between the study groups in the controlled clinical studies. Due to the potentially large numbers of patients exposed to a product marketed for AA, even small differences in adverse event rates within the controlled studies suggest that many patients may ultimately experience drug-related adverse events.

 

B. Serious adverse events:

 

During the completed clinical studies, two Omalizumab-exposed subjects died, both of events that appeared unrelated to Omalizumab exposure (a motor vehicle accident in one case and ischemic heart disease in another). One additional death has been reported for an Omalizumab-exposed subject in an on-going clinical study. This subject died of rapidly progressive meningococcal sepsis following the onset of fever and chills. The subject had received approximately one year of Omalizumab administration within Study 011Ext. The impact of Omalizumab exposure upon this death event is unclear.

 

Nonfatal SAE occurred at a slightly higher rate among Omalizumab-exposed subjects in both the group of all controlled studies (4.2% versus 3.8%) and the group of AA adolescent/adult controlled studies (5.6% versus 4.6%). No single type of event or cluster of events appeared to account for the small excess of SAE among Omalizumab-exposed subjects. However, two types of SAE are of special concern: malignancy and anaphylaxis.

 

1. Malignancy:

 

Among all completed studies, malignant neoplasms occurred in 20/4127 (0.5%) Omalizumab-exposed subjects compared to 5/2236 (0.2%) control subjects. Excluding non-melanoma skin cancer, malignancies were detected among 16 (0.4%) Omalizumab-exposed subjects and two (0.1%) control subjects, a difference in rate of 0.3%. The duration of Omalizumab exposure varied among the clinical studies and was generally less than one year for most subjects. Consequently, the rates are better expressed in terms related to exposure times. Table 3 summarizes the malignancy rates in terms of events per 1000 patient years of exposure.

 

Table 3. Malignancy rates (events/1000 patient years) in all complete studies

Malignancy type

Omalizumab

(n = 4127)

Control

(n = 2236)

Rate difference

(95% CI)

Rate ratio

(95% CI)

First malignancy

(events/patient-years)

6.3

20/3160

3.3

(5/1513)

3.0

(-1.0, 7.0)

1.9

(0.7, 6.5)

First malignancy

excluding non-melanoma skin cancer

(events/patient-years)

5.1

(16/3160)

1.3

(2/1513)

3.7

(0.7, 6.8)

3.8

(0.9, 34.3)

all rates and their differences are calculated as per 1000 patient years

 

In addition to these malignancies, two Omalizumab-exposed subjects (2/1420 patient years of exposure) in on-going clinical studies have been diagnosed with malignancies (colon cancer, prostate cancer). Because most of the on-going studies are uncontrolled, the corresponding exposure time for controls was only 374 patient years.

 

The overall pattern of malignancies within the Omalizumab group is remarkable for a predominance of solid organ/epithelial cancers and only one case of a hematological/lymphatic cancer.

 

Comparisons of malignancy rates suggest (but do not definitively establish) an increased rate for the Omalizumab-exposed subjects. Table 3 suggests that, on average, the cancer rate might double (see rate ratio) with Omalizumab exposure--the confidence interval suggesting that the potential change might result in a rate ranging from one-third lower than baseline to one that is six fold higher than baseline. Of more concern is the comparison of rates for malignancies, exclusive of non-melanoma skin cancer. Table 3 suggests that Omalizumab administration might be associated with a four fold increase in the rate of these malignancies--the confidence interval suggesting that rate might be considerably higher. Comparisons of the numbers of subjects diagnosed with cancer within the sponsor's clinical studies to the numbers expected based upon a large epidemiological database (SEER) suggested that, on average, the Omalizumab group had a higher number of malignancies than might be expected while the control group had a lower number than might be expected.

 

No submitted or published data establish a direct pathophysiological link between anti-IgE therapy and cancer development or progression. However, the potential for alterations in IgE-mediated effector cell function must be considered. For example, innate immunity to neoplasia may be altered by effects on eosinophil or mast cell function. Eosinophils are known to bear IgE receptors. Increased malignant tissue eosinophil infiltration has been correlated with increased survival, an effect that may be related to eosinophil function in tumor cell lysis.[13],[14] Potentially, anti-IgE therapy may alter IgE-mediated eosinophil priming and effector cell function. Similarly, alteration of mast cell IgE-mediated signaling may impact the immunity to cancer. Mast cells and their mediators are known to alter tissue structure, blood flow, and immunologic milieu.[15] Alteration of these functions by anti-IgE therapy may, conceptually, impact the resistance to cancer. Overall, no direct biological evidence links IgE antagonism and oncogenesis, although several pathophysiological pathways may be cited in order to provide a biological plausibility for an association of anti-IgE therapy and cancer risks.

 

Overall, there was a numeric excess of malignancies among Omalizumab-exposed subjects--an observation that raises the concern the excess may be study drug-related.

 

2. Anaphylaxis:

 

Within all the sponsor's clinical studies, anaphylaxis or anaphylactoid reactions were experienced by four subjects exposed to Omalizumab and one subject in a control group. Table 4 briefly summarizes the Omalizumab-exposed subjects with anaphylaxis/anaphylactoid reactions. One Omalizumab-exposed subject experienced an anaphylaxis/anaphylactoid reaction following receipt of IV Omalizumab in an exploratory study, while the three other Omalizumab subjects had reactions following SC dosing within a major clinical study. One (peanut-allergic) control subject experienced an anaphylactoid reaction/anaphylaxis after the accidental ingestion of peanuts.

 

All subjects experiencing anaphylaxis/anaphylactoid reactions survived. The anaphylaxis reactions were managed with various combinations of epinephrine, steroids and antihistamine therapies (see Appendix B). No subject developed respiratory failure. Notably, one Omalizumab-exposed subject experienced anaphylaxis following exposure to Levaquin. This subject continued in the clinical study and received subsequent Omalizumab dosages with no recurrence of anaphylaxis. Overall, the data suggest that Omalizumab may, on a rare basis, be associated with life-threatening anaphylactic reactions.

 

 

 

 

 

 

Table 4. Anaphylaxis or anaphylactoid reactions following Omalizumab exposure

Subject

Study

Features

Outcome

2712

Q0694

30 y o male experienced anaphylactoid reactions 1.5 hours after first Omalizumab dose (IV)

discontinued Omalizumab

4621

008

39 y o female with a history of penicillin and trimethoprim-sulfa allergy developed facial edema, hives, dyspnea 30 minutes after a Levaquin tablet ingestion and 21 days after last Omalizumab dose

continued Omalizumab

12411

Q2143g

19 y o female developed hives, itching and dyspnea 90 minutes after her first Omalizumab dose

discontinued Omalizumab

11756

Q2143g

28 y o female developed injection site edema, throat and tongue edema 2 hours after her fourth Omalizumab dose

discontinued Omalizumab

y o = year old

 

C. Adverse events:

 

1. AE among all subjects in controlled studies:

 

The tables within Appendix A summarize the common AE that occurred in the controlled studies. Most of the common AE rates were similar between the study groups, although the Omalizumab group had a higher rate for the following events: rash, bleeding-related AE, various digestive system AE, and certain female genitourinary AE. Underscoring the disproportion in the rates of certain AE is the observation that the proportions of subjects discontinuing a study due to AE was higher among the Omalizumab groups than control groups (3% versus 1%).

 

a. Rash:

 

All grades of rash AE occurred in excess among Omalizumab-exposed subjects in the group of all controlled studies (Table 5). The excess appeared related to the occurrence of a broad range of non-urticarial reactions among subjects in the open label studies. The extent to which knowledge of treatment assignment may have impacted the reporting of rash AE is unclear. Nevertheless, the incidence of rash AE paralleled increases in blood Omalizumab concentrations, a pattern suggestive of correlation with Omalizumab exposure. Table 6 summarizes the rates of rash according to quartiles of end-of-study, trough Omalizumab blood concentrations.

 

Table 5. Rash in controlled studies, by severity

Severity

Any rash, n (%)

Omalizumab,

n = 3224

Control,

n = 2019

Any event

211 (6.5)

98 (4.9)

Mild

127 (3.9)

62 (3.1)

Moderate

74 (2.3)

34 (1.7)

Severe

10 (0.3)

2 (0.1)

 

Table 6. Rash AE in controlled studies, by quartile of terminal serum

Omalizumab trough concentration

AE term

Omalizumab Quartile

Control

n = 1216

1 (low)

n = 426

2

n = 419

3

n = 423

4 (high)

n = 422

Rash

10 (2.3)

13 (3.1)

13 (3.1)

22 (5.2)

25 (2.1)

 

Overall, the data suggest that, compared to controls, Omalizumab administration may be associated with a higher incidence of non-urticarial rashes, some of which are severe.

 

b. Digestive system events:

 

A small excess of digestive system AE was noted among Omalizumab-exposed subjects in the controlled studies. As summarized in Table 7 and shown in Appendix A, Omalizumab-exposed subjects experienced a modestly higher rate for a broad range of specific digestive system AE (nausea, vomiting, diarrhea, abdominal pain). The excess appeared largely related to AE of mild to moderate severity. However, digestive system SAE also occurred at a slightly higher rate among Omalizumab-exposed subjects. Although uncommon, appendicitis occurred among more Omalizumab-exposed subjects than control subjects (six versus three subjects). Increases in end-of-study, trough blood Omalizumab concentrations generally paralleled the increased incidence of digestive system AE.

 

Table 7. Digestive system AE by severity grade

Severity grade

All controlled studies, n (%)

AA controlled studies*, n (%)

Omalizumab

n = 3224

Control

n = 2019

Omalizumab

n = 2076

Control

n = 1383

Any event

612 (19.0)

360 (17.8)

444 (21.4)

260 (18.8)

Mild

273 (8.5)

163 (8.1)

201 (9.7)

117 (8.5)

Moderate

271 (8.4)

153 (7.6)

190 (9.2)

109 (7.9)

Severe

68 (2.1)

44 (2.2)

53 (2.6)

34 (2.5)

*adolescent/adult subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

 

c. Bleeding-related AE:

 

A small excess of bleeding-related AE among Omalizumab-exposed subjects was observed in the controlled studies (Table 8). The excess was largely attributable to mild to moderate severity grades of the following AE: epistaxis, menorrhagia and hematoma. Analyses of platelet changes showed that, compared to controls, a higher rate of Omalizumab-exposed subjects had mild decreases in platelet counts. The magnitude of the platelet count decreases appeared clinically unremarkable, both for the Omalizumab-exposed subjects with and without bleeding-related AE. These observations suggest that the Omalizumab group's platelet count changes, alone, did not account for the group's excess in bleeding related AE.

 

Table 8. Bleeding-related AE, by severity

Outcome/grade

All controlled studies

AA controlled studies*

Omalizumab

n = 3224

Control

n = 2019

Omalizumab

n = 2076

Control

n = 1383

Any event

81 (2.5)

33 (1.6)

60 (2.8)

24 (1.7)

Mild

55 (1.7)

20 (1.0)

38 (1.8)

14 (1.0)

Moderate

23 (0.7)

9 (0.4)

19 (0.9)

7 (0.5)

Severe

3 (0.1)

4 (0.2)

3 (0.1)

3 (0.2)

*adolescent/adult subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

 

d. Female genitourinary (GU) AE:

 

Although uncommon, female GU AE appeared at a higher rate among Omalizumab-exposed subjects than control subjects (Table 9). This excess appeared related, in part, to more Omalizumab-exposed subjects experiencing severe dysmenorrhea AE (0.2% versus 0) and severe grade urinary tract infection AE (0.2% versus 0), as well as a broad variety of milder GU AE. A correlate of these comparisons is the observation that menorrhagia bleeding AE were more common among Omalizumab-exposed subjects than controls.

 

 

Table 9. Female GU and reproductive system AE (females 12 years)

by severity grade

Severity grade

All controlled studies, n (%)

AA controlled studies*, n (%)

Omalizumab

n = 1662

Control

n = 1042

Omalizumab

n = 1239

Control

n = 794

Any event

187 (11.3)

108 (10.4)

162 (13.1)

98 (12.3)

Mild

78 (4.7)

33 (3.2)

66 (5.3)

27 (3.4)

Moderate

93 (5.6)

68 (6.5)

80 (6.5)

64 (8.1)

Severe

16 (1.0)

7 (0.7)

16 (1.3)

7 (0.9)

*adolescent/adult female subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

 

2. AE among the geriatric population in controlled studies:

 

The safety database consists of information from Omalizumab exposure among 151 geriatric subjects (subjects 65 years of age), 142 of whom received Omalizumab in a controlled study. The controlled studies show an excess of multiple types of AE among these subjects. Table 10 summarizes the AE rates by clusters according to body system. The bolded rows denote body systems where the rates for Omalizumab-exposed subjects exceed those of control subjects.

 

Table 10. AE by body system, rates for ages > 65 years in controlled studies

System

All controlled studies, n (%)

AA controlled studies*, n (%)

Omalizumab,

n = 142

Control, n = 71

Omalizumab,

n = 134

Control, n = 65

Any event

102 (71.8)

54 (76.1)

99 (73.9)

51 (78.5)

Body as whole

28 (19.7)

6 (8.5)

28 (20.9)

5 (7.7)

Cardiovascular

14 (9.9)

3 (4.2)

14 (10.4)

3 (4.6)

Digestive

20 (14.1)

7 (9.9)

20 (14.9)

7 (10.9)

Hemic/lymphatic

2 (1.4)

1 (1.4)

2 (1.5)

1 (1.5)

Infections/infestations

26 (18.3)

16 (22.5)

26 (19.4)

15 (23.1)

Lab abnormality

4 (2.8)

2 (2.8)

4 (3.0)

2 (3.1)

Musculoskeletal

11 (7.7)

3 (4.2)

29 (21.6)

13 (20.0)

Nervous

23 (16.2)

6 (8.5)

23 (17.2)

6 (9.2)

Respiratory

67 (47.2)

35 (49.3)

66 (49.3)

34 (52.3)

Skin/appendages

17 (12.0)

10 (14.1)

17 (12.7)

9 (13.8)

Special senses

10 (7.0)

6 (8.5)

9 (6.7)

6 (9.2)

GU/reproductive

9 (6.3)

2 (2.8)

9 (6.7)

2 (3.1)

*Studies 008, 009, 011, 012, IA04 and Q2143g

 

When the body system AE clusters were further analyzed according to preferred terms (ie. the specific types of AE), the numbers of subjects experiencing each specific event was very small such that it is not feasible to attribute the excesses cited in Table 10 to a disproportionate occurrence of specific events. Overall, the available data do not rule out the possibility of an excess of some types of AE among geriatric subjects exposed to Omalizumab.

 

D. Other observations:

 

1. Clinical laboratory:

 

The most remarkable clinical laboratory finding was the observation of a disproportionate number of Omalizumab-exposed subjects with decreases in hemoglobin. Overall, approximately 14% Omalizumab-exposed subjects and 10% control subjects had a hemoglobin value lower than baseline detected at some point during follow-up. Shift analyses of hemoglobin changes are summarized in Table 11. Within this table, "notably low" refers to a hemoglobin value of < 0.8 X LLN. The decreases in hemoglobin were generally mild for both study groups and could not be related to alterations in blood platelet counts or to bleeding-related AE. The basis for the modest excess in Omalizumab-exposed subjects with decreases in hemoglobin is unclear.

 

Table 11. Shift analyses of hemoglobin

Category

All controlled studies, n (%)

AA controlled studies,* n (%)

Omalizumab,

n = 3125

Control,

n = 1946

Omalizumab,

n = 2004

Control,

n = 1328

Normal or high at baseline

2883

1820

1857

1236

Shift to low

290 (10.1)

151 (8.3)

221 (11.9)

109 (8.8)

Shift to notably low

2 (0.1)

1 (0.1)

1 (0.1)

1 (0.1)

Low at baseline

242

126

147

92

Shift to even lower

131 (54.1)

51 (40.5)

95 (64.6)

42 (45.7)

Low at baseline, but not notably low

227

123

135

89

Shift to notably low

7 (3.1)

4 (3.3)

5 (3.7)

4 (4.5)

*adolescent/adult female subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g

 

The clinical laboratory data review examined in detail platelet count changes because of a preclinical finding of thrombocytopenia in monkeys receiving high doses of Omalizumab. The Omalizumab doses associated with thrombocytopenia in monkeys were considerably in excess of those proposed for use in humans. All clinical subjects exposed to the Omalizumab dose proposed for marketing had blood Omalizumab concentrations well below those concentrations associated with thrombocytopenia in monkeys. No Omalizumab-exposed subject in the clinical studies with normal or high baseline platelet counts developed thrombocytopenia, although a mild decrease in blood platelet count was observed in more Omalizumab-exposed subjects than control subjects. Overall, the clinical studies suggest Omalizumab administration may have been associated with mild decreases in platelet counts in approximately 7% of exposed subjects. The decrease in platelet counts was largely by a magnitude of < 100 X 109/L and resulted in counts that were still within normal limits.

 

2. Antibody formation:

 

No antibody formation to Omalizumab was reported in any clinical study. However, FDA review of antibody formation data is pending the submission of additional information regarding the antibody assay performance characteristics. No conclusions can be formed regarding the antibody formation data at this time.


 

6. Preclinical studies:

 

The sponsor's preclinical studies are especially notable for the demonstration of dose/serum concentration-related thrombocytopenia in monkeys. This finding was documented most extensively within cynomologus monkeys but was also observed in rhesus, African green monkeys and chimpanzees. The notable findings include the following:

 

 

 

 

 

 

 

7. Clinical studies overview:

 

Overall, the sponsor submits clinical data from 26 completed clinical studies and preliminary clinical data from nine on-going clinical studies. These 35 studies may be broadly grouped into exploratory studies (generally phase 1 or 2), major studies assessing safety and/or efficacy or on-going studies. The completed studies consist of 11 exploratory studies and 15 major studies, as described below. The major studies supply the most meaningful clinical safety data and form the safety database. Notable findings from the exploratory studies and on-going studies are also summarized in this review.

 

A. Exploratory Studies:

 

Most of the Exploratory Studies used early manufacturing iterations of the product and examined either single dose or a small number of repeat doses (Table 12). Many of these studies examined IV administration. For simplicity, Study 2203 (a PK study assessing the comparability of two late-development product iterations) is listed among the Exploratory Studies in Table 12.

 

 


Table 12. Exploratory studies

Study

Design

Subjects, n

Omalizumab Dose

Subjects

Total

Omalizumab

Q0572g

OL, SD with a non-dosed control group

77

59

0.005 mg/kg to 1.0 mg/kg, IV or SC

atopic and non-atopic adults

Q0619g

OL, MD, uncontrolled

25

25

0.015 mg/kg to 0.5 mg/kg, IV or SC, X 5 to 6 doses

AA or SAR adults

Q0626g

Single blind, SD or MD, placebo-controlled

34

21

0.15 to 0.5 mg/kg, IV or SC, X 1 - 3 doses

pediatric AA

Q0637g

Single blind, MD, placebo-controlled

12

8

0.15 to 0.5 mg/kg

IV or SC, X 3 doses

AA adults with elevated IgE

Q0673g

OL, MD, uncontrolled

47

47

2.4 to 10.0 mg/kg IV q 2 weeks for ~ 1 year

adult SAR patients w or w/o AA

Q0723g

OL, MD, uncontrolled

46

46

0.014 mg/kg/IU/mL weekly, IV or SC, X 4 doses

adult and pediatric AA patients

Q0624g

Randomized, DB, placebo-controlled

240

181

0.15 or 0.5 mg/kg Q 2 weeks X 12, IV or SQ

adult SAR

Q0630g

Randomized, DB, placebo-controlled

20

11

2 mg/kg IV on day 0, then 1 mg/kg IV on day 7, 14 and Q 2 weeks X total of 10 doses

adult AA

Q0634g

Randomized, DB, placebo-controlled

19

10

0.5 mg/kg IV weekly X 9 doses

adult AA

Q0694g

Randomized, DB, placebo-controlled

317

212

0.006 or 0.014 mg/kg/IU/mL IV, q 2 weeks for 20 weeks

adult and pediatric (12 yrs) AA

2203

Randomized, OL, PK

87

87

150 or 300 mg, SC

Adult healthy volunteers

Total assigned to Omalizumab

707

-

Design information: OL = open label, SD = single dose, MD = multiple dose, DB = double blind; PK = pharmacokinetic

Subject information: AA = allergic asthma, SAR = seasonal allergic rhinitis;

Dose information: Q = every, yrs = years

 

B. Major Studies:

 

Table 13 summarizes the 15 major studies composing the safety database. Brief summaries of each study design are provided in Appendix C.

 


Table 13. Major studies

Study

Design

Subjects

Features

Oma-lizumab

Cont

008C/E

R, DB, PC

268

257

AA definitive study, ages 12 - 74, one year

009C/E

R, DB, PC

274

272

AA, definitive study, ages 12 - 76, one year

010C

R, DB, PC

225

109

AA, ages 5 - 12, 7 month controlled study

010E

OL, UC

309

(99)*

-

AA, ages 5 - 12, 4 month extension

011C

R, DB, PC

176

165

AA, ages 12 - 75, 8 month study focused on ability to decrease high doses of ICS

012

R, DB, PC

22

23

AA, ages 18 - 50, 4 month bronchoscopic study focused on sputum & biopsy findings

IA04

OL, STC

206

106

AA, ages 12 - 75, 1 year European study focused on comparison of asthma events among subjects who had to have an ER visit or hospitalization + oral steroids in past year

Q2143g

(ALTO)

OL, STC

1261

638

"asthma"--did not require skin test +, ages 6 - 75, 6 month study focused on comparison of SAE among subjects who had to be receiving ICS or OCS + another med

Q 2195g

(ALTO E)

OL, UC

613

(188)*

-

"asthma", ages 6 - 75, 6 month extension study

006

R, DB, PC

400

136

SAR, ages 12 - 75, 3 month study

007

R, DB, PC

165

86

SAR, ages 17 - 66, 3 month study

D01

R, DB, PC

114

111

SAR, ages 6 - 17, 6 month study

006 E

OL, UC

287

(0)*

-

SAR, ages 12 - 75, 3 month retreatment

014

R, DB, PC

144

145

PAR, ages 12 - 75, 4 month study

013

R, DB, PC

16

9

AD, ages 6 - 16, 6 month study

Total subjects

3558

2057

-

Cont = control; R = randomized; DB = double-blind; PC = placebo controlled; STC = standard therapy controlled; OL = open label; UC = uncontrolled; ICS = inhaled corticosteroids; OCS = oral corticosteroids

*newly exposed

 

The subject numbers cited within Table 13 refer to the numbers of subjects assigned to Omalizumab or control within each specific study, not the numbers of subjects who received the study drug or provided safety data. All subjects within the safety database (the "safety analyzable" population) either received some study drug or (for a standard therapy controlled, STC subject) had been randomized and provided some post-randomization data. Consequently, the sample sizes within the safety database are slightly smaller than the total subject numbers cited within Table 13. Overall, the major studies consist of 12 controlled studies and 3 uncontrolled studies.

 


8. Safety database composition:

 

A. Sample size:

 

The safety database consists of 3507 "safety analyzable" subjects from the major studies, as shown in Table 14.

 

Table 14. Safety database composition

# subjects from Major Studies

Omalizumab

Control

Randomized within a controlled study

(Studies 008C/E, 009C/E, 010C, 011C, 012, 006, 007, D01, 014, 013, IA04 and Q2143g)

3558

2057

Randomized within a controlled study but received no study drug or (if in STC control group) had no f/u data

334

38

"Safety analyzable" subjects from controlled studies

3224

2019

Enrolled in uncontrolled study having received no Omalizumab in past (Studies Q2195g and 010E)

287

not applicable

Enrolled in uncontrolled study but received no Omalizumab or had no f/u data

4

not applicable

"Safety analyzable" subjects from uncontrolled studies

283

not applicable

Total Safety Database

3507

2019

f/u = follow-up

 

For summary purposes, this review will focus upon the safety database's subjects from the controlled studies. However, important findings from the safety database's uncontrolled studies will be noted. The controlled study findings will be summarized according to those from:

 

-all controlled studies (all 12 major controlled studies, all indications)

 

-all adolescent/adult controlled AA studies (Subjects 12 yrs of age in

studies 008C/E, 009C/E, 011C, 012, IA04 and Q2143g)

Additionally, at times, the placebo controlled studies will specifically be cited, as follows:

-all placebo controlled studies (all 10 placebo controlled studies, all indications)

-all placebo controlled adolescent/adult controlled AA studies (Subjects 12 yrs

of age in studies 008C/E, 009C/E, 011C and 012)

 

Comment: Among the 12 major controlled studies, three were designed specifically for the assessment of safety and efficacy in adult/adolescent AA--studies 008, 009 and 011. These three studies used extensive exclusion criteria (especially studies 008 and 009) to limit enrollment to a fairly select subset from moderate-to-severe persistent AA patients. Indeed, approximately 50% of all screened subjects were excluded from enrollment in these three studies. These exclusions were based upon many criteria including:

-use of certain common maintenance AA medications

-screening blood IgE concentrations outside the applicable dosage limits

-presence of certain common co-morbid conditions.

The subject selectivity involved in these three studies may limit the ability to generalize their findings to a broad range of patients as regards certain issues. To address this problem, the sponsor submits clinical data from several other controlled clinical studies ("safety studies")--especially from Study Q2143g, an open label study that used relatively broad asthma eligibility criteria and had no requirement for documentation of skin test reactivity to aeroallergens. Notably, approximately 40% of all controlled safety data comes from the Study Q2143g, a study with two major limitations with respect to the ability to directly compare findings from the active treatment group to the standard therapy control group:

-the study design entailed a difference in the process for collection of adverse event (AE)

data (subjects in the active treatment group had more AE ascertainment visits than subjects in the control group). Hence, the finding of more AE in the active treatment group may relate to greater efforts at ascertainment.

-the open label nature of the study. Investigators may have been more or less likely to

record AE given knowledge of a subject's treatment assignment.

 

The most readily interpretable, comparative AA safety data are derived from the placebo controlled studies, Studies 008, 009, 011C, and, to a lesser extent, Study 012 (a small bronchoscopic study). This Omalizumab-exposure safety sample accounts for approximately 22% (738/3224) of all the controlled clinical safety data. These placebo controlled, AA studies are important because of the rigor of their methodology, the use of market-applicable clinical Omalizumab dosages and the administration of Omalizumab over relatively prolonged periods of time. These are also important considerations because inclusion of clinical safety data from several of the supportive, controlled clinical studies examining Omalizumab in other indications (e.g., SAR, PAR and AD) may not provide a clinically meaningful representation of comparative Omalizumab safety. Some of these non-AA controlled studies involved short Omalizumab exposure times (3 months or less), relatively low Omalizumab dosages and uneven randomization ratios.

 

B. Baseline characteristics and drug exposure:

 

Baseline characteristics and drug exposure data for the safety database of all subjects who received Omalizumab are summarized in Tables 15 and 16, respectively.

 

Table 15. Baseline characteristics in safety database, n (%)

Parameter

Safety Database,

n = 3507

Sex

Female

1930 (55)

Male

1577 (45)

Race

Caucasian

2988 (85)

Black

288 (8)

Oriental

48 (1)

Other

183 (5)

Age

6 - 11 yrs

443 (13)

12 - 17 yrs

318 (9)

18 - 64 yrs

2595 (74)

65 yrs

151 (4)

Total IgE, IU/mL

mean

212.2

range

20 - 1612

 

Comment: Notably, approximately 75% of the safety data comes from subjects between 18 and 64 years of age. Of concern is the limited extent of exposure among younger subjects (approximately 300 adolescent subjects) and geriatric subjects (approximately 150 subjects). These considerations are important because younger subjects may ultimately have the greatest life-long exposure to Omalizumab while the older subjects may be at higher risk of AE because of comorbid illnesses and other considerations unique to the elderly.

 

Table 16 summarizes the Omalizumab safety database for two groups of subjects: 1) all subjects in the safety database and 2) all subjects in the safety database plus all subjects with some follow-up data from on-going studies (not thoroughly verified clinical data) up to the safety date cut-off of January 17, 2003. In general, clinical data after 52 weeks of exposure is limited to the collection of SAE reports. This preliminary data from the on-going, open label, uncontrolled clinical studies consist of information from 294 subjects newly exposed to Omalizumab. However, the safety database of 3507 subjects from the completed studies provides the basis for overall safety assessment. The duration of Omalizumab exposure within this safety database is limited to approximately one year.

 

Table 16. Drug exposure in safety database and on-going studies*, n (%)

Exposure

Safety Database,

n = 3507

Safety Database +

Preliminary Data from On-going Studies

n = 3790

8 weeks

3338 (94)

3619 (96)

12 weeks

3122 (89)

3401 (90)

24 weeks

2301 (66)

2590 (68)

36 weeks

1305 (37)

1822 (48)

52 weeks

715 (20)

1150 (30)

76 weeks

0

418 (11)

100 weeks

0

326 (9)

124 weeks

0

262 (7)

148 weeks

0

171 (5)

156 weeks

0

158 (4)

*includes data from all major studies and follow-up data from the extension studies through the safety cut-off date of January 17, 2003

 

Comment: The clinical data within the safety database have been audited by the sponsor for accuracy. The clinical data from the on-going studies are preliminary data. Nevertheless, it is notable that the sponsor has some experience with subjects receiving Omalizumab for over three years.

 

To facilitate comparative interpretations of safety findings in all controlled studies and the AA controlled studies, Table 17 shows the major baseline characteristics for subjects within these two groups of studies.

 

Table 17. Baseline characteristics in controlled studies

Parameter

All controlled studies

AA controlled studies*

Omalizumab,

n = 3224

Cont, n = 2019

Omalizumab,

n = 2076

Cont, n = 1383

Sex

Male

1440 (45%)

905 (45%)

837 (40%)

589 (43%)

Female

1784 (55%)

1114 (55%)

1239 (60%)

794 (57%)

Race

Caucasian

2764 (86%)

1741 (86%)

1758 (85%)

1171 (85%)

Black

253 (8%)

143 (7%)

158 (8%)

99 (7%)

Oriental

44 (1%)

25 (1%)

35 (2%)

24 (2%)

Other

165 (5%)

110 (5%)

125 (6%)

89 (6%)

Age

6 - 11 yrs

345 (11%)

197 (10%)

-

-

12 - 17 yrs

296 (9%)

190 (9%)

151 (7%)

97 (7%)

18 - 64 yrs

2441 (76%)

1561 (77%)

1791 (86%)

1221 (88%)

65 yrs

142 (4%)

71 (4%)

134 (7%)

65 (5%)

*adolescent/adult subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

Cont = control

 

Comment: Table 17 suggests that subjects were generally well balanced between the active treatment arm and the control arm in the controlled studies--for certain major baseline characteristics. In order to interpret the AA safety findings more clearly, Table 18 shows certain asthma-related baseline characteristics for subjects in all controlled AA studies and all placebo controlled AA studies (adult/adolescent subjects only).

 

Table 18. Asthma-related baseline characteristics in adult/adolescent

AA controlled studies

Parameter

AA controlled studies*

AA placebo controlled studies**

Omalizumab,

n = 2076

Cont, n = 1383

Omalizumab,

n = 738

Placebo, n = 717

Any prior year overnight asthma hospitalization

246 (12%)

147 (11%)

44 (6%)

49 (7%)

Any prior year asthma ICU adm

96 (5%)

61 (4%)

8 (1%)

11 (2%)

No. of prior year ER visits for asthma

Only 1

260 (13%)

142 (10%)

68 (9%)

58 (8%)

More than 1

258 (12%)

144 (10%)

48 (7%)

45 (6%)

Any prior intubation or MV for asthma

100 (5%)

57 (4%)

11 (2%)

11 (2%)

FEV1 % predicted

n

2076

1383

738

717

Mean

71

70

70

71

Range

(12 - 139)

(14 - 130)

(12 - 126)

(22 - 127)

*adolescent/adult subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

**adolescent/adult subjects (ages 12 yrs in Studies 008, 009, 011C and 012)

ICU = intensive care unit; adm = admission; MV = mechanical ventilation

 

Comment: Table 18 shows that "AA controlled studies" provide substantially more safety information than "AA placebo controlled studies" for the subset of subjects with the most refractory forms of AA: those with a history of overnight hospitalization or ICU admission or >1 ER visit in the past year for an asthma exacerbation. This is a special concern in interpreting the AA placebo controlled studies (mainly Studies 008 and 009) because these studies may not comprehensively characterize the risks associated with Omalizumab among refractory patients.

 

The safety findings from "AA controlled studies" (especially Studies IA04 and Q213g) are especially important for providing information from subjects at potentially higher risk for misadventures and these studies may provide a more general population-applicable description of the risks of Omalizumab than "AA placebo controlled studies."

 

C. In-study characteristics:

 

Comment: One of the concerns conveyed to the sponsor in FDA's July, 2001 Complete Review letter related to the limited amount of clinical data from Omalizumab-exposed subjects who received a variety of concomitant medications. As noted above, the sponsor's BLA resubmission attempts to resolve this concern by supplying additional clinical data from studies with relatively broad eligibility criteria--observations evident in comparisons of concomitant medication use between "All AA controlled studies" and "AA placebo controlled studies," as shown in Table 19. For example, the total number of Omalizumab-exposed subjects receiving concomitant oral steroids is over three fold larger in the group of "AA controlled studies" than "AA placebo controlled studies." The difference is also vivid for comparisons of the number of subjects receiving leukotriene modifying agents and/or xanthines--the number receiving these concomitant medications is almost negligible in "AA placebo controlled studies."

 

Table 19 summarizes the extent of concomitant medication exposure in the AA controlled clinical studies.

 

Table 19. Concomitant medication use in AA controlled studies

Medication

AA controlled studies*

AA placebo controlled studies

Omalizumab,

n = 2076

Control, n = 1383

Omalizumab,

n = 738

Placebo, n = 717

Oral steroids

654 (32%)

514 (37%)

194 (36%)

259 (36%)

Xanthines

226 (11%)

130 (9%)

6 (<1%)

29 (4%)

LTR modifiers

675 (33%)

372 (27%)

22 (3%)

34 (5%)

Long-acting beta agonists

1339 (65%)

785 (57%)

163 (22%)

191 (27%)

Cromolyns

77 (4%)

43 (3%)

7 (<1%)

8 (1%)

Penicillins

338 (16%)

221 (16%)

147 (20%)

131 (18%)

Fluoroquinolones

312 (15%)

166 (12%)

85 (12%)

75 (11%)

Sulfa Antibiotics

42 (2%)

17 (1%)

9 (1%)

4 (<1%)

ACE inhibitors

129 (6%)

72 (5%)

28 (4%)

26 (4%)

H2 antagonists

129 (6%)

90 (7%)

42 (6%)

41 (6%)

Calcium-channel antagonists

134 (7%)

79 (6%)

30 (4%)

29 (4%)

Iodinated contrast

6 (< 1%)

1 (< 1%)

1 (< 1%)

0

*adolescent/adult subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

LTR = leukotriene

 

Comment: Table 19 is also notable for showing minimal use of certain non-steroid controller medications among subjects in the placebo controlled studies. This is important because these studies are described as enrolling subjects with inadequately controlled asthma symptoms despite high doses of inhaled corticosteroids. Conceivably, some of these subjects could have had their symptoms controlled through concomitant use of non-steroid controller medications, such as LTR modifiers, cromolyns and/or long acting beta blockers.

 

Another basis for emphasizing the importance of "All AA controlled studies" relates to the construct of the safety database. When making comparisons to control groups for ascertaining effects in AA, it is important to note that the group of studies composing "All AA controlled studies," probably provides the most pertinent clinical data, as compared to the group of "All controlled studies" because "All controlled studies" includes several studies in which Omalizumab exposure times are brief and uneven randomization ratios were used to enrich the Omalizumab groups--designs which, in the composite of "All controlled studies," may dilute safety signals when only numbers of subjects are compared.

 

D. Disposition of subjects in Safety Database:

 

1. Controlled studies:

 

Table 20 summarizes the disposition of subjects in controlled clinical studies (shown are the numbers for all randomized subjects). Similar data for the placebo controlled studies are shown in Table 21. Discontinuations for AE are cited in bold text.

 

 

 

 

 

 

 

 

Table 20. Subject disposition

Disposition/Reason

All controlled studies

AA controlled studies*

Omalizumab,

n = 3274

Control,

n = 2054

Omalizumab,

n = 2115

Control,

n = 1414

Completed

2912 (88.9%)

1781 (86.7%)

1835 (86.8%)

1194 (84.4%)

Discontinued:

362 (11.1%)

273(13.3%)

280 (13.2%)

220 (15.6%)

AE

62 (1.9%)

19 (0.9%)

55 (2.6%)

16 (1.1%)

Abnormal lab value

3 (0.1%)

3 (0.1%)

3 (0.1%)

2 (0.1%)

Other reasons

297 (9.1%)

251 (12.2%)

222 (10.5%)

202 (14.3%)

*adolescent/adult subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

Cont = control; f/u = follow-up; Other reasons include: unsatisfactory therapy, protocol violation, consent withdrawal, lost to follow-up, administrative problems, death, physician's or sponsor's decision to withdraw subject or reason not stated

 

Comment: Table 20 shows more discontinuations for AE among the Omalizumab group than the control group (especially notable within the group of AA controlled studies). No specific AE or group of clinically similar AE accounts for the excess among the Omalizumab group. The larger number of Omalizumab discontinuations for AE is derived from the open label, standard care AA controlled studies, as shown by comparisons of Table 20 to Table 21. Conceivably, knowledge of the treatment assignment in the open label studies may have impacted the decision to discontinue a subject because of AE.

 

Table 21. Disposition in placebo controlled studies

Disposition/Reason

All placebo controlled studies

AA placebo controlled studies*

Omalizumab,

n = 1806

Control,

n = 1311

Omalizumab,

n = 740

Control,

n = 717

Completed

1666 (92.2%)

1145 (87.3%)

663 (89.6%)

594 (82.8%)

Discontinued:

140 (7.8%)

166 (12.7%)

77 (10.4%)

123 (17.2%)

AE

13 (0.7%)

15 (1.1%)

6 (0.8%)

12 (1.7%)

Abn lab value

3 (0.2%)

3 (0.2%)

3 (0.4%)

2 (0.3%)

Other reasons

124 (6.9%)

148 (11.3%)

68 (9.2%)

109 (15.2%)

*adolescent/adult subjects (ages 12 yrs in Studies 008, 009, 011C and 012)

Abn = abnormal; Cont = control; f/u = follow-up

 

Within the group of all controlled studies, the specific AE prompting study discontinuation was known for 54 Omalizumab subjects and 16 control subjects. The AE that prompted study discontinuation in more than one of the Omalizumab group subjects within the group of all controlled studies are summarized in Table 22, along similar data for AA controlled studies.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 22. AE leading to study discontinuation in > 1 Omalizumab subjects

in controlled studies (number of subjects with AE)

Main clinical AE*

All controlled studies

AA controlled studies

Oma-lizumab

n

Control

n

Oma-lizumab

n

Control

n

Rash (including urticaria, dermatitis, facial rash, etc)

10

0

8

0

Pregnancy

6

1

5

1

Injection site reaction

4

0

3

0

Fatigue

3

1

3

0

Diarrhea, vomiting, gastroenteritis

3

0

-

-

Asthma exacerbation

2

0

-

-

Upper respiratory infection

2

0

2

0

Pneumonia

2

1

-

-

Anaphylaxis

2

0

2

0

Headache

2

2

-

-

Ischemic heart disease

2

2

-

-

Atrial fibrillation

2

0

2

0

Cancer

2

0

-

-

Dry mouth

2

0

2

0

Rhinitis

2

0

2

0

Eye edema

2

0

2

0

*several subjects had preferred term AE that represented multiple manifestations of what may be regarded as the main clinical AE; the table lists the main clinical AE that appears most clinically pertinent to this reviewer; the symbol "-" notes where no more than 1 OMALIZUMAB subject experienced the AE

 

Comment: Table 22 shows the AE of rash and injection site reaction appearing notably more commonly among discontinuing Omalizumab subjects than control subjects--a pattern not found when the analysis is limited to subjects within only the placebo controlled studies. However, this numeric excess (alone) does not account for the larger number of Omalizumab subjects discontinuing in non-placebo controlled studies than in the placebo controlled studies. Table 22 also illustrates the relatively broad spectrum of AE among discontinuing subjects in all controlled and AA controlled studies. No single AE other than rash and injection site reaction appears prominent in frequency. Consequently, it appears that Omalizumab group discontinuations due to AE within the non-placebo controlled studies were related to a very broad variety of AE, but especially prominent are rash and injection site reactions.

 

2. Uncontrolled studies:

 

Three major studies were uncontrolled: Studies 10E, ,Q2195g and 006E. Together, these three studies enrolled 1209 subjects and 1137 (94%) completed the studies. The reasons for study discontinuation included:

-AE in 17 subjects (5 from Study 006E, 12 from Q2195g)

-Other reasons in 55 subjects

 

Of the five AE prompting discontinuation from Study 006E, two are especially notable because of their temporal associated with Omalizumab administration and the investigators' assessment of suspected association with the Omalizumab injection. Both AE consisted of facial flushing and skin erythema shortly following the Omalizumab injection. Both AE resolved with antihistamine therapy.

 

Of the 12 AE prompting discontinuation from Study Q2195g, four are especially notable because of cancer (discussed in the malignancy section of this document). Most other discontinuations were related to asthma exacerbations.

 

Comment: The overall pattern of study discontinuations suggests that Omalizumab was associated with more AE prompting study discontinuation when compared to a control group. Chief among the specific types of AE prompting discontinuation were rash and injection site reaction.

 

9. Serious adverse events (SAE):

 

A. Deaths:

 

No deaths occurred during the exploratory studies. During the major studies, three subjects died, as summarized below:

 

-Subject 4145 (Omalizumab) in Study 014 died after a motor vehicle accident

-Subject 11 (Omalizumab) in Study IA04 died of ischemic heart disease

-Subject 2581 (Placebo) in Study 008 died of cardiac arrest.

 

One additional Omalizumab subject died during an on-going study. Subject 5613 died of meningococcal sepsis after receiving approximately one year of Omalizumab administrations within Study 011Ext.


 

One placebo subject also died during a post-study follow-up period. Subject 2322 (Placebo) died after a motor vehicle accident, an event occurring during the 12 weeks of observation after the completion of the study.

 

B. Nonfatal SAE:

 

1. Major studies:

 

Table 23 summarizes SAE in all controlled and all AA controlled studies. Hospitalizations for asthma-related events were efficacy endpoints in most placebo controlled studies and were not recorded as SAE.

Table 23. SAE in controlled studies

Study group

Placebo and standard therapy controlled studies

Placebo controlled studies

Omalizumab

Control

Omalizumab

Control

All controlled studies

Total n

3224

2019

1801

1310

Subjects with SAE, n (%)

135 (4.2)

76 (3.8)

44 (2.4)

35 (2.7)

AA adolescent/adult studies*

Total n

2076

1383

738

717

Subjects with SAE, n (%)

117 (5.6)

64 (4.6)

32 (4.3)

25 (3.5)

*adolescent/adult female subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

 

Comment: Table 23 shows a small excess of Omalizumab subjects with SAE when compared to all control subjects, an observation that, for the AA studies was present in both the placebo controlled and non-placebo controlled studies.

 

Table 24 summarizes the SAE by body system clusters.

 

 

 

 

 

Table 24. SAE by body system in controlled studies

System

All controlled studies, n (%)

AA controlled studies*, n (%)

Omalizumab

n = 3224

Control

n = 2019

Omalizumab

n = 2076

Control

n = 1383

Any event

135 (4.2)

76 (3.8)

117 (5.6)

64 (4.6)

Body as a whole

17 (0.5)

7 (0.3)

15 (0.7)

7 (0.5)

CV system

13 (0.4)

8 (0.4)

12 (0.6)

8 (0.6)

Digestive system

30 (0.9)

10 (0.5)

23 (1.1)

8 (0.6)

Endocrine

0

1

0

1 (0.1)

Fetal/neonatal

1 (0)

2 (0.1)

1 (0)

2 (0.1)

Hemic and Lymphatic

1 (0)

0

1 (0)

0

Infections/infestations

3 (0.1)

2 (0.1)

2 (0.1)

2 (0.1)

Lab abnormality

3 (0.1)

2 (0.1)

2 (0.1)

2 (0.1)

Metabolic/nutritional

1 (0)

0

1 (0)

0

Musculoskeletal

10 (0.3)

10 (0.5)

9 (0.4)

7 (0.5)

Nervous

7 (0.2)

4 (0.2)

7 (0.3)

4 (0.3)

Respiratory

24 (0.7)

18 (0.9)

23 (1.1)

13 (0.9)

Skin and appendages

1 (0)

2 (0.1)

1 (0)

2 (0.1)

Special senses

6 (0.2)

2 (0.1)

4 (0.2)

1 (0.1)

Med/surg procedures

15 (0.5)

5 (0.2)

14 (0.7)

5 (0.4)

GU and reproductive

15 (0.5)

9 (0.4)

14 (0.7)

8 (0.6)

*adolescent/adult female subjects (ages 12 yrs in Studies 008, 009, 011C, 012, IA04 and Q2143g)

CV = cardiovascular, Med/surg = medical or surgical procedures; GU = genitourinary

 

Comment: Table 24 suggests no single body system cluster of SAE accounts for the excess of Omalizumab subjects with SAE, although subjects with digestive system SAE appear to account for a large portion of the excess. The broad variety of SAE are illustrated by the preferred term mappings, as shown in Table 25 for subjects in all controlled studies.

 

Table 25 summarizes the most common SAE (by preferred term) for subjects in all controlled studies.

 

Table 25. SAE (other than asthma-related events) occurring in 3 subjects within any group, by preferred term, all controlled studies, n

Term

Omalizumab,

n = 135

Control, n = 76

Surgery

15

5

Pneumonia

8

2

Fracture

6

5

Pregnancy

6

1

Appendicitis

5

3

Cholelithiasis

4

0

Chest infection

4

1

Anaphylaxis

3

1

Abdominal pain

3

0

Chest pain

3

1

Supraventricular tachycardia

3

0

Bronchospasm

2

4

Cholecystitis

0

4

Bronchitis

0

4

 

Comment: Table 25 suggests that the specific types of SAE within the controlled studies covered a broad range of events, with the number of subjects experiencing any single type of SAE very small. No single type of SAE appears to account for the excess of Omalizumab subjects with SAE but surgeries, pneumonia, pregnancy and cholelithiasis were the most disparate.

 

Appendix A summarizes the asthma-related hospitalizations within the AA controlled studies, including those studies where asthma-related hospitalizations were recorded as efficacy endpoints and not AE.

 

Comment: Few subjects (< 3%) were hospitalized for asthma in the controlled studies. However, a larger proportion of control subjects required hospitalization for asthma than Omalizumab subjects. This pattern was present both in the placebo controlled AA studies and the standard therapy AA controlled studies.

 

2. Exploratory studies:

 

In the phase 1/2 studies, 16 subjects experienced SAE, nine in the Omalizumab group and seven in the control groups. Most had SAE related to asthma exacerbations (7 Omalizumab subjects and 4 control subjects). Other SAE were isolated events consistent with the subjects' other underlying diseases.

 

3. Uncontrolled studies:

 

Three of the 15 major studies examined Omalizumab use in an uncontrolled study design (Studies Q2195g, 006E and 010E). No SAE were reported in Study 010E and the only SAE reported in Study 006E was appendicitis. Consequently, Study Q2195g provided the most notable SAE findings. SAE were reported for 39 of the 609 (6%) treated subjects within Study Q2195g (eight subjects in the new Omalizumab treatment group and 31 subjects in the continued Omalizumab treatment group). None of the SAE were assessed as related to Omalizumab by the site investigators. The most remarkable SAE related to the reporting of nine malignancies among seven subjects (as reviewed subsequently).

 

4. On-going studies:

 

The sponsor had eight on-going studies (and one recently discontinued study) at the time of the BLA CR amendment submission (December, 2002). Most of these studies are uncontrolled. A 120 day safety update was submitted to the BLA with a safety cut-off date of January 17, 2003. As of this cut-off date, all eight studies were still on-going. From these studies, the sponsor received reports of SAE among 154 Omalizumab-exposed subjects (out of a known, on-going Omalizumab exposure database of 1,184 subjects). Three of the on-going studies are double blinded studies and the treatment assignment of subjects with SAE in these studies has not been unblinded.

 

SAE related to the respiratory system accounted for almost half of the SAE. Most (nearly three-quarters) of the respiratory system SAE related to asthma exacerbations. GU system SAE were the second most frequently involved body system and most of these events related to pregnancies. Of the 15 GU SAE, 12 were pregnancies. Other GU events included isolated cases of kidney calculus, hysterectomy and prostate cancer. Digestive system SAE accounted for the third most frequently involved body system. Except for three cases of appendicitis, specific digestive system SAE covered a broad range with no more than two subjects experiencing any specific preferred term event (e.g., abdominal pain, vomiting, cholecystitis, nausea, etc).

 

SAE of special note from on-going studies include four cases of malignancy and one case of meningococcal sepsis. Malignancy was diagnosed within one subject each, as follows: prostate cancer, colon adenocarcinoma, basal cell skin cancer and squamous cell skin cancer. These events are summarized within the Neoplasia section of this review.

 

C. Neoplasia (malignant and benign):

 

This section will summarize all benign and malignant neoplasia events.

 

1. Malignancies:

 

Comment: In reviewing the malignancy data, it is important to note that subjects with a history of malignancy (more than 3 months prior to enrollment) were not excluded from the sponsor's clinical studies. Consequently, some of the cases of malignancy represent recurrent disease.

 

Among all completed studies, malignant neoplasms occurred in 20/4127 (0.5%) Omalizumab subjects compared to 5/2236 (0.2%) control subjects. Among the 20 Omalizumab subjects with malignancies, one subject (#21/4005) was determined, after discontinuation from the study, to have had a breast lump present prior to study entry (although the malignancy diagnosis was unknown) and another subject (#1230/10045) was suspected of having had a diagnosis of recurrent thyroid cancer made in error during the clinical study.

 

Comment: There were special circumstances cited for three Omalizumab malignancy subjects. One subject, an Omalizumab subject who was suspected of having a recurrent optic glioma, had the event clarified when recurrent follow-up imaging studies documented no recurrence. The two other special Omalizumab subjects included the woman who, following study enrollment, reported that she had noted a breast lump prior to enrollment and another subject who, according to a reviewing physician, may not have had a correct diagnosis of recurrent thyroid cancer (see Appendix E). These two subjects are counted as having malignancies because the cancer diagnosis was made in the breast cancer subject only after Omalizumab exposure and, for the thyroid cancer subject, Omalizumab administration was followed by the development of evidence (but not histological confirmation) of recurrent thyroid cancer, a finding that prompted radiation therapy.

 

Table 26 summarizes the subjects with a malignancy diagnosis within all completed clinical studies (the major studies plus exploratory studies).

 

Table 26. Malignant neoplasms in all completed studies, n (%)

Neoplasm

Omalizumab n = 4127

Control,

n = 2236

Any event

20 (0.5)*

5 (0.2)

Skin, non-melanoma

5

3

Breast

5

0

Prostate

2

0

Melanoma

2

0

Bladder

1

0

Glioma

0

1

Non-Hodgkin's lymphoma

1

0

Pancreas

1

0

Rectum

1

0

Parotid gland

2

0

Thyroid

1

0

Testis

0

1

*One subject with melanoma also had a basal cell skin cancer and appears in both the Skin, non-melanoma and Melanoma rows. This subject is counted only once in the total number of subjects with any event.

 

Excluding locally resected squamous or basal cell carcinomas, malignancies were detected in 16 (0.4%) Omalizumab group subjects and 2 (0.1%) control subjects. Due to variation in Omalizumab exposure times between the study groups, the occrrence data are best

expressed in terms of Omalizumab exposure duration (Table 27). The table refers to first malignancy because some subjects with non-melanoma skin cancer had multiple skin cancers.

 

Table 27. Malignancy rates (events/1000 patient year) in all complete studies

Malignancy type

Omalizumab

(n = 4127)

Control

(n = 2236)

Rate difference

(95% CI)

Rate ratio

(95% CI)

First malignancy

(event/patient years)

6.3

20/3160

3.3

(5/1513)

3.0

(-1.0, 7.0)

1.9

(0.7, 6.5)

First malignancy

excluding non-melanoma skin cancer

(event/patient years)

5.1

(16/3160)

1.3

(2/1513)

3.7

(0.7, 6.8)

3.8

(0.9, 34.3)

*all rates and their differences are calculated as per 1000 patient years

 

In addition to these malignancies, two Omalizumab-exposed subjects (2/1420 patient years of exposure) in on-going clinical studies have been diagnosed with malignancies (colon cancer, prostate cancer). Because most of the on-going studies are uncontrolled, the corresponding exposure time for controls was only 374 additional patient years.

 

Comment: The Omalizumab group experienced an absolute 0.3 percentage point increase in the rate of malignancy per year. Table 27 suggests that the cancer rate might double with Omalizumab exposure. However, the confidence intervals are broad and exclude neither no increase nor an increase of many fold. Most concerning is the comparison of rates for malignancies exclusive of non-melanoma skin cancer. Table 27 suggests that the cancer rate for these types of malignancies might increase considerably. The numbers of subjects with malignancy in the studies was very small such that it is very difficult to form conclusions.

 

The National Cancer Institute's (NCI) SEER (Surveillance, Epidemiology and End Results) database was used to compare the incidence of malignancy in the Omalizumab clinical studies to the expected incidence within the general USA population. SEER is a continuing project of NCI in which cancer statistics are collected from approximately 14% of the US population. These statistics are used to estimate overall cancer statistics in the USA. The SEER database provides cancer incidence rates adjusted for age, sex and race. Demographics of the SEER database are generally thought to parallel the demographics for the USA as a whole. The sponsor's comparisons to the SEER data were age and gender-adjusted and included calculation of the SIR (standardized incidence ratio), a ratio comparing the observed number of cases to the expected number. The sponsor's comparisons to SEER were not adjusted for race. Notably, non-melanoma skin cancers are excluded from these analyses because SEER doesn't contain comparison data. Table 28 summarizes the sponsor's SIR findings for the group of all completed controlled studies and all completed studies (some controlled, some uncontrolled).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 28. Observed and expected number of malignancies

(excluding non-melanoma skin cancer) in completed studies

Study group

Omalizumab

Control

No. of events

SIR

(95%CI)

No. of events

SIR

(95% CI)

observed

expected

observed

expected

All completed controlled studies

Female

4

4.1

1.0

(0.3 - 2.5)

1.0

2.8

0.4

(0.0 - 2.0)

Males

6

3.1

2.0

(0.7 - 4.3)

1.0

2.0

0.5

(0.0 - 2.9)

Total

10

7.1

1.4

(0.7 - 2.6)

2.0

4.7

0.4

(0.1 - 1.6)

All completed studies

Female

7

5.3

1.3

(0.5 - 2.8)

1.0

2.8

0.4

(0.0 - 2.0)

Male

9

3.9

2.3

(1.1 - 4.4)

1.0

2.0

0.5

(0.0 - 2.9)

Total

16

9

1.8

(1.0 - 2.9)

2.0

4.7

0.4

(0.1 - 1.6)

 

Comment: The SEER comparisons suggest, on average, the Omalizumab group had a higher number of malignancies than might be expected while the control group had a lower number than might be expected. These data support the concerns raised by direct comparisons of cancer rates between the study groups. However, several limitations are pertinent to interpreting the malignancy comparisons to the SEER database, as follows:

 

a. Demographics of subjects within the SEER database may be meaningfully different from demographics within the sponsor's safety and efficacy studies. The sponsor's two longest duration clinical studies assessing safety and efficacy (Studies 008 and 009) used eligibility criteria that were rigorous in eliminating subjects at substantial risk for co-morbidities, including malignancy. The subject demographics within these two studies also appear substantially different from those of the general USA population for certain notable characteristics. For example, the USA Census Bureau estimates that approximately 75% of the USA population is Caucasian an